From bugzilla-daemon at llvm.org  Thu Jul  1 00:11:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 00:11:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6972] virtual derived constructor with "throw()" is
 seen as too lax when templates are involved
In-Reply-To: <bug-6972-206@http.llvm.org/bugs/>
References: <bug-6972-206@http.llvm.org/bugs/>
Message-ID: <20100701051104.D79AF2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6972

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-01 00:11:04 CDT ---
Fixed in r107385.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 00:18:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 00:18:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7539] New: llc crashes on functions with union as
	an argument
Message-ID: <bug-7539-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7539

           Summary: llc crashes on functions with union as an argument
           Product: libraries
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: matt.giuca at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5151)
 --> (http://llvm.org/bugs/attachment.cgi?id=5151)
Simplified test case which crashes llc

Any function I try with a union as an argument seems to compile OK until it
gets to llc, and then I get an error "UNREACHABLE executed!" An example
(bitcode attached):

target triple = "i386-pc-linux-gnu"

define void @union_fail(union { i32 } %x) nounwind readnone {
entry:
  ret void
}

Note that it's fine if you pass a pointer to a union, just not a union object,
struct containing a union, etc. With --march=x86:

UNREACHABLE executed!
0  libLLVM-2.7.so.1 0x017e4698
Stack dump:
0.    Program arguments: llc bugpoint-reduced-simplified.ll 
1.    Running pass 'X86 DAG->DAG Instruction Selection' on function
'@union_fail'
Aborted

I tried a number of architectures, and they all crash on different passes, but
with the same error. For example, --march=arm:

UNREACHABLE executed!
0  libLLVM-2.7.so.1 0x018a7698
Stack dump:
0.    Program arguments: llc --march=arm bugpoint-reduced-simplified.ll 
1.    Running pass 'ARM Instruction Selection' on function '@union_fail'
Aborted

This affects all architectures, so I have filed it under Common Code Generator
Code.

Running Ubuntu 10.04, using the version of LLVM that came with Ubuntu 10.04
(llvm 2.7-0ubuntu1).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 05:34:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 05:34:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7540] New: Assertion failed: (Emitted && "Unable to
 issue a copy instruction!\n"), function EmitSpecialNode,
 file InstrEmitter.cpp, line 807.
Message-ID: <bug-7540-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7540

           Summary: Assertion failed: (Emitted && "Unable to issue a copy
                    instruction!\n"), function EmitSpecialNode, file
                    InstrEmitter.cpp, line 807.
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu


I am getting this when trying to:


clang -c -m32 -O2 comconsole.c

....warnings...


Assertion failed: (Emitted && "Unable to issue a copy instruction!\n"),
function EmitSpecialNode, file InstrEmitter.cpp, line 807.
Stack dump:
0.      Program arguments: clang -c -m32 -O2 comconsole.c 
1.      <eof> parser at end of file
2.      Code generation
3.      Running pass 'X86 DAG->DAG Instruction Selection' on function
'@comc_probe'
Abort (core dumped)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 05:53:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 05:53:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7495] LLVM/Clang does not allow ivar access for
 synthesised ivars without @synthesized
In-Reply-To: <bug-7495-206@http.llvm.org/bugs/>
References: <bug-7495-206@http.llvm.org/bugs/>
Message-ID: <20100701105338.148132A6C130@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7495

Dave Verwer <dave.verwer at shinydevelopment.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
                 CC|                            |dave.verwer at shinydevelopmen
                   |                            |t.com
         Resolution|WORKSFORME                  |

--- Comment #4 from Dave Verwer <dave.verwer at shinydevelopment.com> 2010-07-01 05:53:37 CDT ---
I just found out about the plans to change this functionality from Martin and I
would like to suggest that the proposed change as mentioned by Fariborz is not
made.

I don't like the idea of accessing completely undeclared ivars, it makes it too
easy to write misleading code mistakenly accessing an ivar that is not
explicitly defined rather than accessing it through the accessors.

I would like to propose it works like this:

* If an ivar is undeclared (with or without synthesize), the ivar is not
directly accessible but the property is created and is accessible through the
getters and setters/dot syntax.
* When you do need to access an ivar directly it must be explicitly declared
and synthesized.

This way there is no magic at all and in the majority of cases the ivar and
synthesize can still be omitted.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 09:36:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 09:36:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7541] New: Exception-specification diagnostics
 missing critical information
Message-ID: <bug-7541-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7541

           Summary: Exception-specification diagnostics missing critical
                    information
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Keywords: quality-of-implementation
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dgregor at apple.com
                CC: llvmbugs at cs.uiuc.edu, sebastian.redl at getdesigned.at,
                    dgregor at apple.com


Several of the exception-specification diagnostics provide too little
information to be useful, particularly when we're checking for compatibility
between exception-specifications. For example:


t.cpp:11:39: error: target exception specification is not superset of source
  Y &(Y::*yptr1)(const Y&) throw(A) = &Y::operator=;

and


test/CXX/except/except.spec/p14.cpp:12:8: error: exception specification of
      overriding function is more lax than base version
struct X2 : public X0, public X1 { }; // expected-error 2{{exception ...
       ^
test/CXX/except/except.spec/p14.cpp:7:11: note: overridden virtual function is
      here
  virtual ~X0() throw(A); // expected-note{{overridden virtual function ...
          ^

should say what the actual exception specifications are or, perhaps, which
types are missing (in the "not a superset" diagnostic) or have been added (in
the "more lax" diagnostic).

As a secondary nit, we probably shouldn't use the term "lax". "Permissive",
perhaps?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 10:20:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 10:20:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7542] New: clang signal 6 (sometimes with "Bogus
 start position hint!")
Message-ID: <bug-7542-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7542

           Summary: clang signal 6 (sometimes with "Bogus start position
                    hint!")
           Product: new-bugs
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu, chenyang at cs.utah.edu


Created an attachment (id=5153)
 --> (http://llvm.org/bugs/attachment.cgi?id=5153)
failure-inducing C

Argh, here's another program that makes my clang behave non-deterministically. 
clang reliably crashes but doesn't always give the assert failure notice. 
Apologies for non-reduced testcase.

regehr at john-home:~/volatile/bugs/tmp322$ clang -v

clang version 2.8 (trunk 107352)
Target: i386-pc-linux-gnu
Thread model: posix

regehr at john-home:~/volatile/bugs/tmp322$ clang -w small.c -Os

clang: LiveInterval.cpp:128: bool llvm::LiveInterval::overlapsFrom(const
llvm::LiveInterval&, const llvm::LiveRange*) const: Assertion `(StartPos->start
<= i->start || StartPos == other.begin()) && StartPos != other.end() && "Bogus
start position hint!"' failed.
0  clang 0x090a4d08
Stack dump:
0.      Program arguments:
/home/regehr/z/compiler-install/llvm-gcc-r107352-install/bin/clang -cc1 -triple
i386-pc-linux-gnu -S -disable-free -main-file-name small.c -mrelocation-model
static -mdisable-fp-elim -mconstructor-aliases -target-cpu pentium4
-resource-dir
/home/regehr/z/compiler-install/llvm-gcc-r107352-install/lib/clang/2.8 -Os -w
-ferror-limit 19 -fmessage-length 91 -fgnu-runtime -fdiagnostics-show-option
-fcolor-diagnostics -o /tmp/cc-ZMnM8A.s -x c small.c 
1.      <eof> parser at end of file
2.      Code generation
3.      Running pass 'Simple Register Coalescing' on function '@func_85'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:02:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:02:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7543] New: exception not caught after rethrow
Message-ID: <bug-7543-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7543

           Summary: exception not caught after rethrow
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: release blocker
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: thom.heller at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Exceptions will not get caught after rethrowing.
The attached code illustrates the problem.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:13:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:13:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7544] New: Wrong offset for struct member
Message-ID: <bug-7544-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7544

           Summary: Wrong offset for struct member
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ofv at wanadoo.es
                CC: llvmbugs at cs.uiuc.edu


Consider this llvm assembly code that assigns values to the data members of a
struct:

; ModuleID = 'lp0'
target datalayout =
"e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:32-f32:32:32-f64:32:32-v64:32:32-v128:128:128-a0:0:64-f80:32:32"
target triple = "i686-pc-win32"

%0 = type { i32, double }

declare void @abort()

define void @mainf() {
  %sdafo_ = alloca %0                             ; <%0*> [#uses=2]
  %pts1_4 = getelementptr %0* %sdafo_, i32 0, i32 0 ; <int*> [#uses=1]
  store i32 15, i32* %pts1_4
  %pts1_1 = getelementptr %0* %sdafo_, i32 0, i32 1 ; <double*> [#uses=1]
  store double 2.100000e+001, double* %pts1_1
  ret void
}


All data is aligned to 32 bits (with the platform's default settings for the
datalayout, the result is the same) but the generated assembly code is:

    .def     _mainf;
    .scl    2;
    .type    32;
    .endef
    .text
    .globl    _mainf
    .align    16, 0x90
_mainf:                                 # @mainf
# BB#0:
    subl    $20, %esp
Ltmp0:
    movl    $15, (%esp)
    movl    $1077215232, 12(%esp)   # imm = 0x40350000
    movl    $0, 8(%esp)
    addl    $20, %esp
    ret


The second data member (with type `double') is 12 bytes away from the first
data member of type `int', leaving 8 bytes of padding between them.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:19:16 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:19:16 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7545] New: X86MCInstLower DWARF exception in -O0 -g
	test
Message-ID: <bug-7545-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7545

           Summary: X86MCInstLower DWARF exception in -O0 -g test
           Product: libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: stoklund at 2pi.dk
                CC: llvmbugs at cs.uiuc.edu


The attached bitcode files are from the  Geryon-PIC-O0-g__llvm-gcc_PROD__x86_64
nightly tester running on trunk.

The crash llc with an Assertion failed: (MI->getOperand(0).isReg() &&
MI->getOperand(3).isImm()), function getDebugValueLocation, file
lib/Target/X86/AsmPrinter/X86MCInstLower.cpp, line 520

llc -asm-verbose=false -O0 -relocation-model pic -disable-fp-elim
Output/office-stringsearch.llvm.bc

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:33:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:33:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7546] New: x86-64 -O0 -g miscompiles 253.perlbmk
Message-ID: <bug-7546-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7546

           Summary: x86-64 -O0 -g miscompiles 253.perlbmk
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: stoklund at 2pi.dk
                CC: llvmbugs at cs.uiuc.edu


Some of the benchmarks are failing:


diff -u 253.perlbmk.out-{nat,llc}
--- 253.perlbmk.out-nat    2010-07-01 03:42:02.000000000 -0700
+++ 253.perlbmk.out-llc    2010-07-01 03:42:24.000000000 -0700
@@ -31,7 +31,7 @@
 ok 18
 ok 19
 ok 20
-ok 21
+not ok 21
 ok 22
 ok 23
 ok 24
@@ -579,10 +579,10 @@
 ok 11
 ok 12
 ok 13
-ok 14
+not ok 14
 ok 15
-ok 16
-ok 17
+not ok 16
+not ok 17
 ok 18
 ok 19
 ok 20
@@ -600,20 +600,20 @@
 ok 32
 ok 33
 ok 34
-ok 35
-ok 36
+not ok 35
+not ok 36
 ok 37
-ok 38
-ok 39
-ok 40
-ok 41
+not ok 38
+not ok 39
+not ok 40
+not ok 41
 ok 42
 ok 43
-ok 44
-ok 45
+not ok 44
+not ok 45
 ok 46
 ok 47
-ok 48
+not ok 48 505550505
 ok 49
 ok 50
 ok 51
@@ -734,23 +734,23 @@
 ok 23
 # takeuchi(18, 12, 6) = 7
 1..18
-ok 1
-ok 2
-ok 3
-ok 4
-ok 5
-ok 6
-ok 7
-ok 8
-ok 9
-ok 10
-ok 11
-ok 12
-ok 13
-ok 14
-ok 15
-ok 16
-ok 17
+not ok 1
+not ok 2
+not ok 3
+not ok 4
+not ok 5
+not ok 6
+not ok 7
+not ok 8
+not ok 9
+not ok 10
+not ok 11
+not ok 12
+not ok 13
+not ok 14
+not ok 15
+not ok 16
+not ok 17
 ok 18
 1..51
 ok 1
@@ -902,10 +902,10 @@
 ok 11
 ok 12
 ok 13
-ok 14
+not ok 14
 ok 15
-ok 16
-ok 17
+not ok 16
+not ok 17
 ok 18
 ok 19
 ok 20
@@ -943,7 +943,7 @@
 ok 22
 ok 23
 ok 24
-ok 25
+not ok 25
 ok 26
 ok 27
 ok 28
@@ -956,11 +956,11 @@
 ok 35
 ok 36
 ok 37
-ok 38
+not ok 38
 ok 39
-ok 40
-ok 41
-ok 42
+not ok 40
+not ok 41
+not ok 42
 ok 43
 ok 44
 ok 45
@@ -975,10 +975,10 @@
 ok 54
 ok 55
 ok 56
-ok 57
-ok 58
-ok 59
-ok 60
+not ok 57
+not ok 58
+not ok 59
+not ok 60
 ok 61
 ok 62
 1..97

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:37:10 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:37:10 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7547] New: x86-64 -O0 -g miscompiles
	2004-03-15-IndirectGoto
Message-ID: <bug-7547-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7547

           Summary: x86-64 -O0 -g miscompiles 2004-03-15-IndirectGoto
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: stoklund at 2pi.dk
                CC: llvmbugs at cs.uiuc.edu


The test SingleSource/Regression/C/2004-03-15-IndirectGoto is miscompiled on
x86-64 when building with -O0 -g:

diff -u 2004-03-15-IndirectGoto.out-{nat,llc}
--- 2004-03-15-IndirectGoto.out-nat    2010-07-01 03:12:13.000000000 -0700
+++ 2004-03-15-IndirectGoto.out-llc    2010-07-01 03:12:18.000000000 -0700
@@ -2,11 +2,5 @@
 B
 C
 D
-B
-C
-D
-C
-D
-D
 E
 exit 0

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:42:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:42:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7549] New: x86-64 -O0 -g tests crash
Message-ID: <bug-7549-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7549

           Summary: x86-64 -O0 -g tests crash
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: stoklund at 2pi.dk
                CC: llvmbugs at cs.uiuc.edu


These tests crash when built with -O0 -g on x86-64, Darwin PIC:

External/SPEC/CINT2000/176_gcc/176_gcc [LLC, LLC compile]
MultiSource/Applications/JM/lencod/lencod [LLC, LLC compile]
SingleSource/Benchmarks/Shootout-C++/except [LLC, LLC compile]
SingleSource/Regression/C++/EH/ctor_dtor_count [LLC, LLC compile]
SingleSource/UnitTests/ObjC/exceptions [LLC, LLC compile]

TEST Output/exceptions.llc FAILED: process terminated by signal (exit status
134)!
TEST Output/ctor_dtor_count.llc FAILED: process terminated by signal (exit
status 134)!
TEST Output/except.llc FAILED: process terminated by signal (exit status 134)!
TEST Output/lencod.llc FAILED: process terminated by signal (exit status 139)!
TEST ../176.gcc.llc FAILED: process terminated by signal (exit status 139)!

Exit status 134 = SIGABRT = abort(3) call
Exit status 139 = SIGSEGV = segmentation violation

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:48:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:48:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7550] New: x86-64 -O0 -g consumer-lame loops
Message-ID: <bug-7550-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7550

           Summary: x86-64 -O0 -g consumer-lame loops
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: stoklund at 2pi.dk
                CC: llvmbugs at cs.uiuc.edu


The MultiSource/Benchmarks/MiBench/consumer-lame/consumer-lame test times out
after 500s when built with -O0 -g on x86-64, but the corresponding gcc-built
binary runs in less than a second.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 12:51:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 12:51:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7551] New: x86-64 -O0 -g miscompiles ClamAV test
Message-ID: <bug-7551-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7551

           Summary: x86-64 -O0 -g miscompiles ClamAV test
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: stoklund at 2pi.dk
                CC: llvmbugs at cs.uiuc.edu


MultiSource/Applications/ClamAV/clamscan produces different output when built
with -O0 -g on x86-64.

There is extra output on stderr:


--- clamscan.out-nat    2010-07-01 03:14:07.000000000 -0700
+++ clamscan.out-llc    2010-07-01 03:14:25.000000000 -0700
@@ -2,6 +2,15 @@
 LibClamAV Warning: ***  The virus database is older than 7 days!  ***
 LibClamAV Warning: ***   Please update it as soon as possible.    ***
 LibClamAV Warning: **************************************************
+ERROR: Scanning README
+ERROR: EADME: OK
+ERROR: Scanning clam-v2.rar
+LibClamAV Warning: RAR code not compiled-in
+ERROR: lam-v2.rar: OK
+ERROR: Scanning clam-v3.rar
+LibClamAV Warning: RAR code not compiled-in
+ERROR: lam-v3.rar: OK
+ERROR: Scanning clam.cab
 Initializing the engine (devel-20071218)

 Initializing phishcheck module
@@ -246,19 +255,10 @@

    * Submodule %10s:    %s

-Scanning README
-README: OK
-Scanning clam-v2.rar
-LibClamAV Warning: RAR code not compiled-in
 Recognized %s file

-clam-v2.rar: OK
-Scanning clam-v3.rar
-LibClamAV Warning: RAR code not compiled-in
 Recognized %s file

-clam-v3.rar: OK
-Scanning clam.cab
 Recognized %s file

 in cli_scanmscab()
@@ -314,7 +314,15 @@

 SizeOfCode: 0x%x

-SizeOfInitializedData: 0x%x
+SizeOfERROR: lam.cab: OK
+ERROR: Scanning clam.exe
+ERROR: lam.exe: OK
+ERROR: Scanning clam.exe.bz2
+ERROR: lam.exe.bz2: OK
+ERROR: Scanning clam.zip
+ERROR: lam.zip: OK
+ERROR: Scanning clamdoc.tar.gz
+InitializedData: 0x%x

 SizeOfUninitializedData: 0x%x

@@ -360,8 +368,6 @@

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 15:24:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 15:24:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7475] checker-242 doesn't reset nullity assumptions
 after function call
In-Reply-To: <bug-7475-206@http.llvm.org/bugs/>
References: <bug-7475-206@http.llvm.org/bugs/>
Message-ID: <20100701202445.B43F22A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7475

Ted Kremenek <kremenek at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Ted Kremenek <kremenek at apple.com> 2010-07-01 15:24:45 CDT ---
Fixed:

http://llvm.org/viewvc/llvm-project?view=rev&revision=107423

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 15:51:29 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 15:51:29 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6895] Vector literal/casting confusion.
In-Reply-To: <bug-6895-206@http.llvm.org/bugs/>
References: <bug-6895-206@http.llvm.org/bugs/>
Message-ID: <20100701205129.1AA7F2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6895

John Thompson <John.Thompson.JTSoftware at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from John Thompson <John.Thompson.JTSoftware at gmail.com> 2010-07-01 15:51:28 CDT ---
Fixed in r107347.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 16:28:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 16:28:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7552] New: Alignment on loads with
	__attribute__((packed))
Message-ID: <bug-7552-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7552

           Summary: Alignment on loads with __attribute__((packed))
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: arplynn at gmail.com
                CC: llvmbugs at cs.uiuc.edu


When emitting loads from __attribute__((packed)) structs, clang will happily
tell LLVM to use the natural alignment of the type. This is wrong - it should
work out the alignment for the struct, or at least use align 1.


Test case:

struct foo {
char a;
int b;
} __attribute__((packed));

int bees(struct foo* bar) { return bar->b; }

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 16:56:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 16:56:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7552] Alignment on loads with
	__attribute__((packed))
In-Reply-To: <bug-7552-206@http.llvm.org/bugs/>
References: <bug-7552-206@http.llvm.org/bugs/>
Message-ID: <20100701215623.7633D2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7552

Anton Korobeynikov <asl at math.spbu.ru> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |asl at math.spbu.ru
         Resolution|                            |DUPLICATE

--- Comment #1 from Anton Korobeynikov <asl at math.spbu.ru> 2010-07-01 16:56:23 CDT ---


*** This bug has been marked as a duplicate of bug 5598 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 17:35:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 17:35:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7554] New: .o generated by clang is over 2m larger
 than the one produced by gcc
Message-ID: <bug-7554-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7554

           Summary: .o generated by clang is over 2m larger than the one
                    produced by gcc
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: evan at chromium.org
                CC: llvmbugs at cs.uiuc.edu


I am looking into why gcc Chrome is ~600mb while clang builds of Chrome are
1.7gb.  (Both with debugging symbols.)  I've attached the preprocessed version
of one of the files that differs the most between the two compilers.

With the same input file, I can generate two output files:
g++ -pthread -O0 -g -c -o profile.gcc profile.ii
../llvm/Release/bin/clang++ -pthread -O0 -g -c -o profile.gcc profile.ii

Their sizes differ significantly:
-rw-r--r-- 1 evanm evanm 4709032 2010-07-01 15:30 profile.clang
-rw-r--r-- 1 evanm evanm 2570352 2010-07-01 15:30 profile.gcc

gzipped file attached.



llvm$ svn info . tools/clang/ | grep Revision
Revision: 107405
Revision: 107405

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 19:57:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 19:57:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7547] x86-64 -O0 -g miscompiles
	2004-03-15-IndirectGoto
In-Reply-To: <bug-7547-206@http.llvm.org/bugs/>
References: <bug-7547-206@http.llvm.org/bugs/>
Message-ID: <20100702005742.492BC2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7547

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |gohman at apple.com
         Resolution|                            |FIXED

--- Comment #1 from Dan Gohman <gohman at apple.com> 2010-07-01 19:57:42 CDT ---
This is fixed temporarily by r107393, and properly by r107454.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  1 21:46:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  1 Jul 2010 21:46:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7555] New: Warn for missing braces around indented
	statements
Message-ID: <bug-7555-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7555

           Summary: Warn for missing braces around indented statements
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jediknil at belkadan.com
                CC: llvmbugs at cs.uiuc.edu


Might be a dream, but Clang should have a warning for this sort of code:

if (!valid(str))
  log("Invalid input: %s", str);
  return NULL;
return process(str);

The "return NULL" line is clearly supposed to be part of the consequent block.

A fixit would be nice but not always correct, since maybe a user just forgot to
dedent in some cases.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  2 10:59:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  2 Jul 2010 10:59:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7546] x86-64 -O0 -g miscompiles 253.perlbmk
In-Reply-To: <bug-7546-206@http.llvm.org/bugs/>
References: <bug-7546-206@http.llvm.org/bugs/>
Message-ID: <20100702155921.2A1AE2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7546

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-02 10:59:20 CDT ---
This failure is no longer present in r107460

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  2 11:02:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  2 Jul 2010 11:02:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7550] x86-64 -O0 -g consumer-lame loops
In-Reply-To: <bug-7550-206@http.llvm.org/bugs/>
References: <bug-7550-206@http.llvm.org/bugs/>
Message-ID: <20100702160217.905BC2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7550

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-02 11:02:17 CDT ---
This test no longer fails in r107460.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  2 12:20:36 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  2 Jul 2010 12:20:36 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7556] New: Temporary object with trivial destructor
 but non-trivial destructor isn't getting destroyed
Message-ID: <bug-7556-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7556

           Summary: Temporary object with trivial destructor but
                    non-trivial destructor isn't getting destroyed
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dgregor at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


terfin:clang dgregor$ cat t.cpp 
struct A { ~A(); }; void foo() { A(); }

terfin:clang dgregor$ clang++ -cc1 -ast-dump t.cpp 
<snip>

void foo() (CompoundStmt 0x104c0fa60 <t.cpp:1:32, col:39>
  (CXXZeroInitValueExpr 0x104c10c30 <col:34, col:36> 'struct A'))

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  2 13:01:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  2 Jul 2010 13:01:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7557] New: Integrated assembler bug while building
 mozilla/js/src/jsnum.cpp
Message-ID: <bug-7557-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7557

           Summary: Integrated assembler bug while building
                    mozilla/js/src/jsnum.cpp
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: Nomis101 at web.de
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com
            Blocks: 5511


If I build Mozilla Firefox with clang on Mac OS X, than it stops at:

/Volumes/Developer/temp/src/mozilla/js/src/jsobj.cpp:6207:1: warning: unused
      function 'ReservedSlotIndexOK' [-Wunused-function]
ReservedSlotIndexOK(JSContext *cx, JSObject *obj, JSClass *clasp,
^
/Volumes/Developer/temp/src/mozilla/js/src/jsnum.cpp:675:9: error: unrecognized
      instruction
    asm("fstcw %0" : "=m" (control) : );
        ^
<inline asm>:1:2: note: instantated into assembly here
        fstcw -14(%ebp)
        ^
/Volumes/Developer/temp/src/mozilla/js/src/jsnum.cpp:675:9: error: unrecognized
      instruction
    asm("fstcw %0" : "=m" (control) : );
        ^
<inline asm>:1:2: note: instantated into assembly here
        fstcw -14(%ebp)
        ^
5 warnings and 2 errors generated.
make[5]: *** [jsnum.o] Error 1
make[5]: *** Waiting for unfinished jobs....


Eli told me this is a integrated assembler bug, which doesn't show up on Linux.
Affected file:
http://hg.mozilla.org/mozilla-central/file/d261faaa483a/js/src/jsnum.cpp

clang version 2.8 (trunk 107428), Target: x86_64-apple-darwin10, Mac OS X
10.6.4

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  2 15:22:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  2 Jul 2010 15:22:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5125] clang always generates stackframe when inline
 asm is present
In-Reply-To: <bug-5125-206@http.llvm.org/bugs/>
References: <bug-5125-206@http.llvm.org/bugs/>
Message-ID: <20100702202232.95E8E2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5125

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #19 from Dale Johannesen <dalej at apple.com> 2010-07-02 15:22:31 CDT ---
Fixed in 107506; at least, GCC-style inline asm should now work the same as
gcc.  Playing more with Apple gcc's asm blocks, gcc-4.2 is quite inconsistent
about whether it does this; for example, this case

void foo() {
  asm {
    int 3
  }
}

gets the stack aligned with -m32 -fomit-stack-pointer, but not with most other
options.  Proper emulation will wait until we have some actual asm-block code
to deal with.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  2 15:44:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  2 Jul 2010 15:44:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7558] New: msp430: llc runs away on simple program
 with long (32-bit) arithmetic
Message-ID: <bug-7558-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7558

           Summary: msp430: llc runs away on simple program with long
                    (32-bit) arithmetic
           Product: new-bugs
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ransford at cs.umass.edu
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5158)
 --> (http://llvm.org/bugs/attachment.cgi?id=5158)
bitcode program that makes llc run away

This program makes llc (tested against ToT) eat memory like mad.  My system has
limited resources, so I have no choice but to kill llc after its rsize tops
4GB.  I haven't seen it finish.  I've run llc under gdb and it seems to be
getting stuck in llvm::SUnit::ComputeHeight.

int main (void) {
    unsigned long S[2] = {1, 2};
    unsigned long i = 0;
    unsigned long j = 0;
    unsigned int ofs = 0;

    i += S[ofs++];
    j += S[ofs];
    return 0;
}

I've compiled this bitcode file with clang from the llvm-msp430 project
(https://www.fooe.net/trac/llvm-msp430), but since this seems to be a codegen
bug, I'm submitting only the bitcode.  I'd be happy to provide more on demand.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul  3 02:26:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat,  3 Jul 2010 02:26:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7559] New: __attribute((flatten)) not supported
Message-ID: <bug-7559-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7559

           Summary: __attribute((flatten)) not supported
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu


Per summary, __attribute((flatten)) isn't supported at the moment.  From gcc
docs, "For a function marked with this attribute, every call inside this
function will be inlined, if possible."  Implemented in gcc 4.1 (so anything
that uses it will assume clang supports it).

This attribute is used quite a bit in ffmpeg.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul  3 03:05:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat,  3 Jul 2010 03:05:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7560] New: Type merging not performed on local
 extern declarations in C++
Message-ID: <bug-7560-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7560

           Summary: Type merging not performed on local extern
                    declarations in C++
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rideau3 at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The following invalid program is accepted in C++ mode (though not in C mode):

extern int i[];
void foo () { extern int i [3]; }
void bar () { extern int i [4]; }

In C mode, we correctly diagnose that this involves declaring i with two
different types.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul  3 04:01:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat,  3 Jul 2010 04:01:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7561] New: wiki.llvm.org - Database down
Message-ID: <bug-7561-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7561

           Summary: wiki.llvm.org - Database down
           Product: Website
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: General Website
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: grosser at fim.uni-passau.de
                CC: llvmbugs at cs.uiuc.edu


The wiki does not work since two days.

I always get:

(Can't contact the database server: Can't connect to local MySQL server through
socket '/var/run/mysqld/mysqld.sock' (2) (localhost))

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul  3 10:15:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat,  3 Jul 2010 10:15:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7562] New: -mem2reg -indvars segfaults
Message-ID: <bug-7562-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7562

           Summary: -mem2reg -indvars segfaults
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Loop Optimizer
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: grosser at fim.uni-passau.de
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5160)
 --> (http://llvm.org/bugs/attachment.cgi?id=5160)
Segfault with opt -mem2reg -indvars

Hi,

running "opt -mem2reg -indvars" on the attached test case leads to a SIGSEGV
because of stack space exhaustion.

Neither opt -mem2reg, opt -indvars nor opt -mem2reg -loopsimplify -indvars
crashes.

During the crash Indvars is introducing more and more canonical induction
variables, but does not recognize that it has already inserted one. I believe
this is because the
LI->getCanonicalInductionVariable() function expects a loop header with just
two predecessors. However in this case there are three.

Two possible solutions I thought about are:

* Change LI->getCanonicalInductionVariable() to not depend on loop simplified
form
* Require all users of SCEVExpander to ask for LoopSimplified loops.

Part of the backtrace until it repeats:
--------------------------------------------------------------
#21891 0x00007ffff5f2c12f in llvm::SCEVExpander::expand (this=0x7fffffffe070,
S=0x672140) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1297
#21892 0x00007ffff5f2c1ee in llvm::SCEVExpander::expandCodeFor
(this=0x7fffffffe070, SH=0x672140, Ty=0x0) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1255
#21893 0x00007ffff5f2d787 in llvm::SCEVExpander::expandCodeFor
(this=0x7fffffffe070, SH=0x672140, Ty=0x0, I=0x672c00) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1250
#21894 0x00007ffff5f2d86b in
llvm::SCEVExpander::getOrInsertCanonicalInductionVariable (this=0x7fffffffe070,
L=0x66f760, Ty=0x65e9c8) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1344
#21895 0x00007ffff5f2f87f in llvm::SCEVExpander::visitAddRecExpr
(this=0x7fffffffe070, S=0x672140) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1138
#21896 0x00007ffff5f34423 in llvm::SCEVVisitor<llvm::SCEVExpander,
llvm::Value*>::visit (this=0x7fffffffe070, S=0x672140) at
/home/grosser/Projekte/llvm/git/include/llvm/Analysis/ScalarEvolutionExpressions.h:588
#21897 0x00007ffff5f2c12f in llvm::SCEVExpander::expand (this=0x7fffffffe070,
S=0x672140) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1297
#21898 0x00007ffff5f2f547 in llvm::SCEVExpander::visitAddRecExpr
(this=0x7fffffffe070, S=0x671a50) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1100
#21899 0x00007ffff5f34423 in llvm::SCEVVisitor<llvm::SCEVExpander,
llvm::Value*>::visit (this=0x7fffffffe070, S=0x671a50) at
/home/grosser/Projekte/llvm/git/include/llvm/Analysis/ScalarEvolutionExpressions.h:588
#21900 0x00007ffff5f2c12f in llvm::SCEVExpander::expand (this=0x7fffffffe070,
S=0x671a50) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1297
#21901 0x00007ffff5f2c1ee in llvm::SCEVExpander::expandCodeFor
(this=0x7fffffffe070, SH=0x671a50, Ty=0x65e9c8) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1255
#21902 0x00007ffff5f2edc4 in llvm::SCEVExpander::visitSMaxExpr
(this=0x7fffffffe070, S=0x671c80) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1206
#21903 0x00007ffff5f34436 in llvm::SCEVVisitor<llvm::SCEVExpander,
llvm::Value*>::visit (this=0x7fffffffe070, S=0x671c80) at
/home/grosser/Projekte/llvm/git/include/llvm/Analysis/ScalarEvolutionExpressions.h:590
#21904 0x00007ffff5f2c12f in llvm::SCEVExpander::expand (this=0x7fffffffe070,
S=0x671c80) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1297
#21905 0x00007ffff5f2c1ee in llvm::SCEVExpander::expandCodeFor
(this=0x7fffffffe070, SH=0x671c80, Ty=0x65e9c8) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1255
#21906 0x00007ffff5f2d604 in llvm::SCEVExpander::visitAddExpr
(this=0x7fffffffe070, S=0x6720d0) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:721
#21907 0x00007ffff5f343e4 in llvm::SCEVVisitor<llvm::SCEVExpander,
llvm::Value*>::visit (this=0x7fffffffe070, S=0x6720d0) at
/home/grosser/Projekte/llvm/git/include/llvm/Analysis/ScalarEvolutionExpressions.h:582
#21908 0x00007ffff5f2c12f in llvm::SCEVExpander::expand (this=0x7fffffffe070,
S=0x6720d0) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1297
#21909 0x00007ffff5f2c1ee in llvm::SCEVExpander::expandCodeFor
(this=0x7fffffffe070, SH=0x6720d0, Ty=0x65e9c8) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1255
#21910 0x00007ffff5f2d787 in llvm::SCEVExpander::expandCodeFor
(this=0x7fffffffe070, SH=0x6720d0, Ty=0x65e9c8, I=0x666560) at
/home/grosser/Projekte/llvm/git/lib/Analysis/ScalarEvolutionExpander.cpp:1250
#21911 0x00007ffff708b8e9 in RewriteLoopExitValues (this=0x665530, L=0x66f3a0,
Rewriter=...) at
/home/grosser/Projekte/llvm/git/lib/Transforms/Scalar/IndVarSimplify.cpp:301
#21912 0x00007ffff708c234 in runOnLoop (this=0x665530, L=0x66f3a0, LPM=...) at
/home/grosser/Projekte/llvm/git/lib/Transforms/Scalar/IndVarSimplify.cpp:504
#21913 0x00007ffff5eb06dd in llvm::LPPassManager::runOnFunction (this=0x66c670,
F=...) at /home/grosser/Projekte/llvm/git/lib/Analysis/LoopPass.cpp:269
#21914 0x00007ffff572f109 in llvm::FPPassManager::runOnFunction (this=0x6649b0,
F=...) at /home/grosser/Projekte/llvm/git/lib/VMCore/PassManager.cpp:1433
#21915 0x00007ffff572f361 in llvm::FPPassManager::runOnModule (this=0x6649b0,
M=...) at /home/grosser/Projekte/llvm/git/lib/VMCore/PassManager.cpp:1453
#21916 0x00007ffff572ed41 in llvm::MPPassManager::runOnModule (this=0x663bf0,
M=...) at /home/grosser/Projekte/llvm/git/lib/VMCore/PassManager.cpp:1507
#21917 0x00007ffff5731c67 in llvm::PassManagerImpl::run (this=0x6636a0, M=...)
at /home/grosser/Projekte/llvm/git/lib/VMCore/PassManager.cpp:1588
#21918 0x00007ffff5731cc9 in llvm::PassManager::run (this=0x7fffffffe850,
M=...) at /home/grosser/Projekte/llvm/git/lib/VMCore/PassManager.cpp:1631
#21919 0x0000000000437e9d in main (argc=4, argv=0x7fffffffea38) at
/home/grosser/Projekte/llvm/git/tools/opt/opt.cpp:550
----------------------------------------------------------------------------

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 14:19:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 14:19:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7561] wiki.llvm.org - Database down
In-Reply-To: <bug-7561-206@http.llvm.org/bugs/>
References: <bug-7561-206@http.llvm.org/bugs/>
Message-ID: <20100704191901.AC2A72A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7561

Daniel Berlin <dberlin at dberlin.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Daniel Berlin <dberlin at dberlin.org> 2010-07-04 14:19:01 CDT ---
Apparently i run it ;)
It had been running so long i didn't realize it was on google2.
I just restarted mysql and it's happy again.

google2 recently went through a distro upgrade for security reasons (from
ubuntu 8.04 LTS to 10.04 LTS), and somewhere along the way mysql got installed
(though all the data was still there).

Fixed now

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 17:25:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 17:25:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7544] Wrong offset for struct member
In-Reply-To: <bug-7544-206@http.llvm.org/bugs/>
References: <bug-7544-206@http.llvm.org/bugs/>
Message-ID: <20100704222509.90D652A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7544

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-04 17:25:09 CDT ---
This is behaving correctly.  The target settings override what is in the .ll
file.  We see this in X86Subtarget.h:

  std::string getDataLayout() const {
    const char *p;
    if (is64Bit())
      p = "e-p:64:64-s:64-f64:64:64-i64:64:64-f80:128:128-n8:16:32:64";
    else if (isTargetDarwin())
      p = "e-p:32:32-f64:32:64-i64:32:64-f80:128:128-n8:16:32";
    else if (isTargetMingw() || isTargetWindows())
      p = "e-p:32:32-f64:64:64-i64:64:64-f80:32:32-n8:16:32";
    else
      p = "e-p:32:32-f64:32:64-i64:32:64-f80:32:32-n8:16:32";

Which causes windows to get f64:64:64.

-Chris

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 17:41:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 17:41:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7537] Unable to issue a copy instruction!
In-Reply-To: <bug-7537-206@http.llvm.org/bugs/>
References: <bug-7537-206@http.llvm.org/bugs/>
Message-ID: <20100704224101.7AA4E2A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7537

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-04 17:41:00 CDT ---


Reduced testcase:


target datalayout =
"e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:32:32-n8:16:32"
target triple = "i386-pc-linux-gnu"

define void @int327(i32* nocapture %p_38) nounwind {
entry:
  %tmp1 = load i32* %p_38                         ; <i32> [#uses=1]
  %conv = trunc i32 %tmp1 to i16                  ; <i16> [#uses=1]
  %sext = mul i16 %conv, -14592                   ; <i16> [#uses=1]
  %conv315 = lshr i16 %sext, 8                    ; <i16> [#uses=1]
  %conv5 = trunc i16 %conv315 to i8               ; <i8> [#uses=1]
  %tmp7 = load i8* undef                          ; <i8> [#uses=1]
  %mul.i13 = mul i8 %conv5, %tmp7                 ; <i8> [#uses=1]
  %tobool = icmp eq i8 %mul.i13, 0                ; <i1> [#uses=1]
  br i1 %tobool, label %lor.rhs, label %lor.end

lor.rhs:                                          ; preds = %entry
  unreachable

lor.end:                                          ; preds = %entry
  ret void
}

Fails very very similarly to: PR7540, duping it.

*** This bug has been marked as a duplicate of bug 7540 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 17:47:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 17:47:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7514] clang signal 11
In-Reply-To: <bug-7514-206@http.llvm.org/bugs/>
References: <bug-7514-206@http.llvm.org/bugs/>
Message-ID: <20100704224702.E501C2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7514

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Chris Lattner <clattner at apple.com> 2010-07-04 17:47:02 CDT ---
Dan fixed this in r107071 but didn't close the bug.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 17:57:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 17:57:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7518] [x86, mmx] terrible codegen of <2 x float>
In-Reply-To: <bug-7518-206@http.llvm.org/bugs/>
References: <bug-7518-206@http.llvm.org/bugs/>
Message-ID: <20100704225724.3216C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7518

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #15 from Chris Lattner <clattner at apple.com> 2010-07-04 17:57:23 CDT ---
Fixed in r107600.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 18:20:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 18:20:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7486] clang: hard to understand error message for
 missing macro declaration
In-Reply-To: <bug-7486-206@http.llvm.org/bugs/>
References: <bug-7486-206@http.llvm.org/bugs/>
Message-ID: <20100704232059.AF16C2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7486

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-04 18:20:59 CDT ---
Thanks for filing this, but I don't think this is common enough to be worth
cluttering up the parser to support.  Stop using macros ;-)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul  4 18:21:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun,  4 Jul 2010 18:21:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7512] LLVM ERROR: Cannot yet select: 0x28ecb778:
 f80 = ConstantFP<APFloat()> [ID=8]
In-Reply-To: <bug-7512-206@http.llvm.org/bugs/>
References: <bug-7512-206@http.llvm.org/bugs/>
Message-ID: <20100704232145.036D42A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7512

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #9 from Chris Lattner <clattner at apple.com> 2010-07-04 18:21:44 CDT ---
Looks fixed, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 03:38:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 03:38:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7564] New: Assertion `... && "Size mismatch!"'
	failed.
Message-ID: <bug-7564-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7564

           Summary: Assertion `... && "Size mismatch!"' failed.
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jay.foad at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5163)
 --> (http://llvm.org/bugs/attachment.cgi?id=5163)
obfuscated test case

On the attached source file I get:

$ /home/jay/clang/objdir/Debug/bin/clang -cc1 -w -S obf.c
clang: /home/jay/svn/llvm-project/cfe/trunk/lib/CodeGen/CGExprConstant.cpp:401:
static
llvm::Constant*<unnamed>::ConstStructBuilder::BuildStruct(clang::CodeGen::CodeGenModule&,
clang::CodeGen::CodeGenFunction*, clang::InitListExpr*): Assertion
`llvm::RoundUpToAlignment(Builder.NextFieldOffsetInBytes,
Builder.getAlignment(Result)) == Builder.getSizeInBytes(Result) && "Size
mismatch!"' failed.
...
0.      Program arguments: /home/jay/clang/objdir/Debug/bin/clang -cc1 -w -S
obf.c 
1.      <eof> parser at end of file
2.      Per-file LLVM IR generation
3.      obf.c:795:13: Generating code for declaration 'ABHFBF'
4.      obf.c:796:2: LLVM IR generation of compound statement ('{}')

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 04:41:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 04:41:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7565] New: linker command failed due to signal 11
Message-ID: <bug-7565-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7565

           Summary: linker command failed due to signal 11
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: niebuhr at niebuhrt.de
                CC: llvmbugs at cs.uiuc.edu


I compiled my project with

clang++ -O4 -I ~/odemx/odemx/odemx/install/include/ -L
~/odemx/odemx/odemx/install/lib/ -lOdemX -lPocoXML -lPocoSQLite -lPocoData
-lPocoFoundation -lpthread -ldl -o main *.cpp

I attached the error log (generated with clang -v ... ). What do I have to do,
to give you the information that you need?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 04:48:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 04:48:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7566] New: clang++ crashes when includes are not
	found
Message-ID: <bug-7566-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7566

           Summary: clang++ crashes when includes are not found
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: niebuhr at niebuhrt.de
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5164)
 --> (http://llvm.org/bugs/attachment.cgi?id=5164)
error log generated with the -v flag

I just misspelled one of my include paths while compiling. Usually the compiler
just complains that the file was not found, but this time it crashes because an
assertion failed.

compiling with "clang++ -v -o main *.cpp" gives me the attached error log.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 10:18:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 10:18:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7567] New: clang generates movaps instruction for
 insufficiently aligned local on stack
Message-ID: <bug-7567-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7567

           Summary: clang generates movaps instruction for insufficiently
                    aligned local on stack
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jay.foad at gmail.com
                CC: llvmbugs at cs.uiuc.edu


If I compile this source:

typedef __attribute__ ((aligned(16))) struct { unsigned long long x[2]; } L;
int one(L x);
int two(L x, L y) { return one(x) || one(y); }

with "clang -cc1 -S", I get:

two:
        subq    $88, %rsp
        movq    %rdi, 72(%rsp)
        movq    %rsi, 80(%rsp)
        movq    %rdx, 56(%rsp)
        movq    %rcx, 64(%rsp)
        movaps  72(%rsp), %xmm0
        movaps  %xmm0, 32(%rsp)
...

Memory arguments to movaps are required to be 16-byte aligned, but the argument
"72(%rsp)" is only 8-byte aligned. This causes a segmentation fault at run
time.

I'm using LLVM and Clang built from trunk this morning, on Ubuntu 10.4 64-bit.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 10:20:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 10:20:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7568] New: linker errors compiling perl from
	spec2006
Message-ID: <bug-7568-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7568

           Summary: linker errors compiling perl from spec2006
           Product: new-bugs
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


This is odd... when I try to build perl from SPEC2006 I get very many linker
errors of this form:

Opcode.c:(.text+0x8a0): multiple definition of `__signbitf'

The preprocessed SPEC code gets this from math.h:

extern int __signbitf (float __value);

The function is never referenced by the C file.

The .o file in question does in fact improperly define the symbol, so the
linker is right to croak:

$ nm Opcode.o |grep signbitf
000008a0 T __signbitf

The behavior disappears when I manually preprocess the file and then compile
it.

Perhaps it's relevant that the SPEC header files destroy the attributes that
math.h was using?  For example here's what we get by including math.h from a
bare C program without all the SPEC crap:

extern int __signbitf (float __value) __attribute__ ((__nothrow__))
     __attribute__ ((__const__));

Is clang somehow getting confused about library functions here?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 10:45:00 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 10:45:00 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7569] New: clang optimizer produces bad instruction
 -O1 but not at -O0
Message-ID: <bug-7569-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7569

           Summary: clang optimizer produces bad instruction -O1 but not
                    at -O0
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sean at rogue-research.com
                CC: llvmbugs at cs.uiuc.edu


$ sw_vers
ProductName:    Mac OS X
ProductVersion: 10.6.2
BuildVersion:   10C2234

$ clang --version
clang version 2.8 (trunk 107463)
Target: x86_64-apple-darwin10
Thread model: posix

$ cat crash.c
int main() {
    *(int*)0 = 0;
    return 0;
}
-------------------------------------------------
$ gcc crash.c -O0 -o crash-gcc-O0
$ gcc crash.c -O0 -o crash-gcc-O0.s -S
$ gcc crash.c -O1 -o crash-gcc-O1
$ gcc crash.c -O1 -o crash-gcc-O1.s -S
$ ./crash-gcc-O0
[1]    23639 segmentation fault  ./crash-gcc-O0
$ ./crash-gcc-O1
[1]    23641 segmentation fault  ./crash-gcc-O1
$ diff crash-gcc-O0.s crash-gcc-O1.s
9,10c9
<       movl    $0, %eax
<       movl    $0, (%rax)
---
>       movl    $0, 0
-------------------------------------------------
$ clang crash.c -O0 -o crash-clang-O0
$ clang crash.c -O0 -o crash-clang-O0.s -S
$ clang crash.c -O1 -o crash-clang-O1
$ clang crash.c -O1 -o crash-clang-O1.s -S
$ ./crash-clang-O0
[1]    23642 segmentation fault  ./crash-clang-O0
$ ./crash-clang-O1
[1]    23644 illegal hardware instruction  ./crash-clang-O1
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ???

The optimizer seems to get confused and produces
a bad instruction:

$ diff crash-clang-O0.s crash-clang-O1.s
11,16c11
<       movl    $0, -4(%rbp)
<       movabsq $0, %rax
<       movl    $0, (%rax)
<       xorl    %eax, %eax
<       popq    %rbp
<       ret
---
>       ud2
        ^^^ Undefined Instruction!

Interestingly, changing the bad "pointer" value to '1'
fixes it:

$ cat crash.c
int main() {
    *(int*)1 = 0;
    return 0;
}

$ clang crash.c -O0 -o crash-clang-O0
$ clang crash.c -O0 -o crash-clang-O0.s -S
$ clang crash.c -O1 -o crash-clang-O1
$ clang crash.c -O1 -o crash-clang-O1.s -S
$ ./crash-clang-O0
[1]    23726 segmentation fault  ./crash-clang-O0
$ ./crash-clang-O1
[1]    23728 segmentation fault  ./crash-clang-O1
$ diff crash-clang-O0.s crash-clang-O1.s
11,13c11
<       movl    $0, -4(%rbp)
<       movabsq $1, %rax
<       movl    $0, (%rax)
---
>       movl    $0, 1

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 13:03:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 13:03:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7564] Assertion `... && "Size mismatch!"' failed.
In-Reply-To: <bug-7564-206@http.llvm.org/bugs/>
References: <bug-7564-206@http.llvm.org/bugs/>
Message-ID: <20100705180340.E682F2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7564

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-05 13:03:40 CDT ---
Fixed in r107621

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 15:21:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 15:21:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7567] clang generates movaps instruction for
 insufficiently aligned local on stack
In-Reply-To: <bug-7567-206@http.llvm.org/bugs/>
References: <bug-7567-206@http.llvm.org/bugs/>
Message-ID: <20100705202123.D595B2A6C133@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7567

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-05 15:21:23 CDT ---
Fixed in r107627, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 17:02:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 17:02:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7570] New: williams.c (from 2001 IOCCC) crashes
	analyzer
Message-ID: <bug-7570-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7570

           Summary: williams.c (from 2001 IOCCC) crashes analyzer
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: bruce.r.stephens at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5166)
 --> (http://llvm.org/bugs/attachment.cgi?id=5166)
failures from 2001 IOCCC

The 2004 entries all processed fine.  One file from 2001 caused a crash, and
(for what it's worth) I attach the failed directory.

This is using llvm 107625, clang 107628.

I just edited Makefile, changing the compiler lines and ran "scan-build make
all".

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 22:31:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 22:31:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7529] Missing memoperands prevents ARM load/store
 folding in a few cases
In-Reply-To: <bug-7529-206@http.llvm.org/bugs/>
References: <bug-7529-206@http.llvm.org/bugs/>
Message-ID: <20100706033123.DAE132A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7529

Evan Cheng <evan.cheng at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Evan Cheng <evan.cheng at apple.com> 2010-07-05 22:31:23 CDT ---
Fixed: 107591. Jakob, please verify.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 22:39:57 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 22:39:57 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7529] Missing memoperands prevents ARM load/store
 folding in a few cases
In-Reply-To: <bug-7529-206@http.llvm.org/bugs/>
References: <bug-7529-206@http.llvm.org/bugs/>
Message-ID: <20100706033957.0E33B2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7529

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #3 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-05 22:39:56 CDT ---
I think your patch in 107591 is definitely correct, and it should provide more
ldm/stm folding, but the original 107114 only changed the behavior for loads
and stores without any memoperands.


107591 doesn't add memoperands to to loads and stores that don't have any,
right?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul  5 22:52:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon,  5 Jul 2010 22:52:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7543] exception not caught after rethrow
In-Reply-To: <bug-7543-206@http.llvm.org/bugs/>
References: <bug-7543-206@http.llvm.org/bugs/>
Message-ID: <20100706035256.25D152A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7543

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from John McCall <rjmccall at apple.com> 2010-07-05 22:52:55 CDT ---
I've substantially revised IR generation in r107631.  That said, the given test
case should have passed before; the outer caught type is the same as the inner
caught type.

If the problem isn't fixed, please give us some more information to help us
reproduce the problem ? OS distribution and version, command line you're using,
etc.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 03:38:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 03:38:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7571] New: wctype.h: unknown type name 'wint_t'
Message-ID: <bug-7571-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7571

           Summary: wctype.h: unknown type name 'wint_t'
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Headers
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jay.foad at gmail.com
                CC: llvmbugs at cs.uiuc.edu


I get:

$ cat c.c
#include <stddef.h>
#include <wctype.h>
$ clang -c c.c
In file included from c.c:2:
/usr/include/wctype.h:112:22: error: unknown type name 'wint_t'
extern int iswalnum (wint_t __wc) __THROW;
                     ^
/usr/include/wctype.h:118:22: error: unknown type name 'wint_t'
extern int iswalpha (wint_t __wc) __THROW;
                     ^
...

Under bug 6691, Clang's stddef.h was fixed to define wint_t if __need_wint_t is
#defined. I think the problem is that that is done inside the "#ifndef
__STDDEF_H ... #endif" that wraps the whole file, so it only works if stddef.h
was never previously included.

I'm using Ubuntu 10.4 x86_64. The system headers come from package libc6-dev
version 2.11.1-0ubuntu7.1.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 06:23:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 06:23:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7572] New: Crash analyzing coreutils-6.9
Message-ID: <bug-7572-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7572

           Summary: Crash analyzing coreutils-6.9
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: bruce.r.stephens at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5167)
 --> (http://llvm.org/bugs/attachment.cgi?id=5167)
contents of failures from coreutils-6.9

This is with llvm @107641, clang @107646

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 08:50:16 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 08:50:16 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7573] New: Assertion `Emitted && "Unable to issue a
 copy instruction!\n"' failed
Message-ID: <bug-7573-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7573

           Summary: Assertion `Emitted && "Unable to issue a copy
                    instruction!\n"' failed
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jay.foad at gmail.com
                CC: llvmbugs at cs.uiuc.edu


I get:

$ cat jay.c
double f(void) {}
$ clang -O -mno-sse jay.c
...
clang:
/home/jay/svn/llvm-project/llvm/trunk/lib/CodeGen/SelectionDAG/InstrEmitter.cpp:807:
void llvm::InstrEmitter::EmitSpecialNode(llvm::SDNode*, bool, bool,
llvm::DenseMap<llvm::SDValue, unsigned int, llvm::DenseMapInfo<llvm::SDValue>,
llvm::DenseMapInfo<unsigned int> >&): Assertion `Emitted && "Unable to issue a
copy instruction!\n"' failed.


I'm running on Ubuntu 10.4 x86_64. I built clang and llvm from the trunk this
morning.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 09:26:06 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 09:26:06 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7574] New: Assertion `isStackEmpty() && "Stack not
 empty at end of basic block?"' failed
Message-ID: <bug-7574-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7574

           Summary: Assertion `isStackEmpty() && "Stack not empty at end
                    of basic block?"' failed
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jay.foad at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5169)
 --> (http://llvm.org/bugs/attachment.cgi?id=5169)
test case

With the attached source file I get:

$ clang -m32 -mno-sse -O -c jay.c
clang:
/home/jay/svn/llvm-project/llvm/trunk/lib/Target/X86/X86FloatingPoint.cpp:300:
bool<unnamed>::FPS::processBasicBlock(llvm::MachineFunction&,
llvm::MachineBasicBlock&): Assertion `isStackEmpty() && "Stack not empty at end
of basic block?"' failed.
...

I'm running on Ubuntu 10.4 x86_64. I built llvm and clang from trunk this
morning.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 09:26:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 09:26:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5094] clang -fstack-protector fails to link
 (undefined reference to `__stack_chk_guard')
In-Reply-To: <bug-5094-206@http.llvm.org/bugs/>
References: <bug-5094-206@http.llvm.org/bugs/>
Message-ID: <20100706142635.672E92A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5094

Nelson Elhage <nelhage at mit.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |nelhage at mit.edu
         Resolution|                            |FIXED

--- Comment #5 from Nelson Elhage <nelhage at mit.edu> 2010-07-06 09:26:34 CDT ---
This should be fixed by llvm r107640.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 10:40:37 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 10:40:37 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7575] New: Assertion " " from boost::array
Message-ID: <bug-7575-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7575

           Summary: Assertion " " from boost::array
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5170)
 --> (http://llvm.org/bugs/attachment.cgi?id=5170)
asserting array example

The attached code causes the following assertion:

While deleting: { { i8 } }* %
Use still stuck around after Def is destroyed:  %arrayidx6 = getelementptr
inbounds %"struct.std::basic_string"* <badref>, i64 %8 ;
<%"struct.std::basic_string"*> [#uses=1]
clang: Value.cpp:75: virtual llvm::Value::~Value(): Assertion `use_empty() &&
"Uses remain when a value is destroyed!"' failed.

The code is reduced from a testcase of boost::array. The test has been broken
(in clang) in the last few days, it works with a 2-week old version of the
compiler. I have not attempted to further reduce when it was broken.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 10:52:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 10:52:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7576] New: clang takes a long time to build a c
	file with -O2
Message-ID: <bug-7576-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7576

           Summary: clang takes a long time to build a c file with -O2
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Driver
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: vince at kyllikki.org
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5171)
 --> (http://llvm.org/bugs/attachment.cgi?id=5171)
Intermediate source from clang compile

The NetSurf project (http://netsurf-browser.org/) has a single file which takes
an unusually long time to compile with -O2 optimisation turned on.

The smallest test case I can currently reproduce the issue with is building the
file (/content/urldb.c within our codebase) as part of the test suite.

Here are the tests showing the compile time for clang and gcc with and without
-O2

$ clang --version
clang version 2.0 (trunk 106560)
Target: x86_64-unknown-linux-gnu
Thread model: posix

$ gcc --version
gcc (Ubuntu 4.3.3-5ubuntu4) 4.3.3
Copyright (C) 2008 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ time clang -std=c99 -g -O2 -D_BSD_SOURCE -D_POSIX_C_SOURCE -I.. `pkg-config
--cflags libxml-2.0 libcurl` ../content/urldb.c ../utils/url.c ../utils/utils.c
../utils/messages.c ../utils/hashtable.c ../utils/filename.c
../test/urldbtest.c -o urldbtest `pkg-config --libs libxml-2.0 libcurl`

real    1m55.421s
user    1m49.927s
sys    0m1.228s
$ time clang -std=c99 -g -O0 -D_BSD_SOURCE -D_POSIX_C_SOURCE -I.. `pkg-config
--cflags libxml-2.0 libcurl` ../content/urldb.c ../utils/url.c ../utils/utils.c
../utils/messages.c ../utils/hashtable.c ../utils/filename.c
../test/urldbtest.c -o urldbtest `pkg-config --libs libxml-2.0 libcurl`

real    0m1.577s
user    0m1.200s
sys    0m0.192s
$ time gcc -std=c99 -g -O2 -D_BSD_SOURCE -D_POSIX_C_SOURCE -I.. `pkg-config
--cflags libxml-2.0 libcurl` ../content/urldb.c ../utils/url.c ../utils/utils.c
../utils/messages.c ../utils/hashtable.c ../utils/filename.c
../test/urldbtest.c -o urldbtest `pkg-config --libs libxml-2.0 libcurl`

real    0m3.229s
user    0m2.696s
sys    0m0.308s
$ time gcc -std=c99 -g -O0 -D_BSD_SOURCE -D_POSIX_C_SOURCE -I.. `pkg-config
--cflags libxml-2.0 libcurl` ../content/urldb.c ../utils/url.c ../utils/utils.c
../utils/messages.c ../utils/hashtable.c ../utils/filename.c
../test/urldbtest.c -o urldbtest `pkg-config --libs libxml-2.0 libcurl`

real    0m1.242s
user    0m0.920s
sys    0m0.196s


The specific file that is taking the time is content/urldb.c

$ time clang -std=c99 -c -g -O2 -D_BSD_SOURCE -D_POSIX_C_SOURCE -I..
`pkg-config --cflags libxml-2.0 libcurl` ../content/urldb.c 

real    1m50.883s
user    1m47.823s
sys    0m0.764s
$ time clang -std=c99 -c -g -O0 -D_BSD_SOURCE -D_POSIX_C_SOURCE -I..
`pkg-config --cflags libxml-2.0 libcurl` ../content/urldb.c 

real    0m0.649s
user    0m0.520s
sys    0m0.060s

I saved the intermediate files (-save-temps) from the previous compile
(attached) and re-tested

time clang -c -O0 urldb.i 

real    0m0.645s
user    0m0.512s
sys    0m0.028s
[16:49 vince at derik ~/netsurf/netsurf-trunk/test]$ time clang -c -O2 urldb.i 

real    1m52.981s
user    1m48.683s
sys    0m0.892s

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 12:35:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 12:35:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7575] ICE in EH logic compiling boost::array
In-Reply-To: <bug-7575-206@http.llvm.org/bugs/>
References: <bug-7575-206@http.llvm.org/bugs/>
Message-ID: <20100706173559.403A72A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7575

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from John McCall <rjmccall at apple.com> 2010-07-06 12:35:58 CDT ---
Indeed it was.  Fixed in r107676.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 13:17:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 13:17:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7577] New: no debug info is generated for
	"find_strings"
Message-ID: <bug-7577-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7577

           Summary: no debug info is generated for "find_strings"
           Product: Runtime Libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: other
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: dpatel at apple.com
                CC: llvmbugs at cs.uiuc.edu


In this example, llvm is not generating debug info for "find_strings" variable.

@.str = private constant [4 x i8] c"one\00", align 1 ; <[4 x i8]*> [#uses=1]
@.str1 = private constant [4 x i8] c"two\00", align 1 ; <[5 x i8]*> [#uses=1]
@C.9.2167 = internal constant [2 x i8*] [i8* getelementptr inbounds ([4 x i8]*
@.str, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8]* @.str1, i64 0, i64
0)]
!38 = metadata !{i32 524329, metadata !"pbmsrch.c", metadata
!"/Users/grawp/LLVM/test-suite/MultiSource/Benchmarks/MiBench/office-stringsearch",
metadata !39} ; [ DW_TAG_file_type ]
!39 = metadata !{i32 524305, i32 0, i32 1, metadata !"pbmsrch.c", metadata
!"/Users/grawp/LLVM/test-suite/MultiSource/Benchmarks/MiBench/office-stringsearch",
metadata !"4.2.1 (Based on Apple Inc. build 5658) (LLVM build 9999)", i1 true,
i1 false, metadata !"", i32 0} ; [ DW_TAG_compile_unit ]
!46 = metadata !{i32 524303, metadata !38, metadata !"", metadata !38, i32 0,
i64 64, i64 64, i64 0, i32 0, metadata !47} ; [ DW_TAG_pointer_type ]!97 =
metadata !{i32 524334, i32 0, metadata !38, metadata !"main", metadata !"main",
metadata !"main", metadata !38, i32 73, metadata !98, i1 false, i1 true, i32 0,
i32 0, null, i1 false, i1 false, null} ; [ DW_TAG_subprogram ]!101 = metadata
!{[2 x i8*]* @C.9.2167}
!47 = metadata !{i32 524324, metadata !38, metadata !"char", metadata !38, i32
0, i64 8, i64 8, i64 0, i32 0, i32 6} ; [ DW_TAG_base_type ]
!97 = metadata !{i32 524334, i32 0, metadata !38, metadata !"main", metadata
!"main", metadata !"main", metadata !38, i32 73, metadata !98, i1 false, i1
true, i32 0, i32 0, null, i1 false, i1 false, null} ; [ DW_TAG_subprogram ]
!98 = metadata !{i32 524309, metadata !38, metadata !"", metadata !38, i32 0,
i64 0, i64 0, i64 0, i32 0, null, metadata !99, i32 0, null} ; [
DW_TAG_subroutine_type ]
!99 = metadata !{metadata !100}
!100 = metadata !{i32 524324, metadata !38, metadata !"int", metadata !38, i32
0, i64 32, i64 32, i64 0, i32 0, i32 5} ; [ DW_TAG_base_type ]
!101 = metadata !{[2 x i8*]* @C.9.2167}
!102 = metadata !{i32 524544, metadata !103, metadata !"find_strings", metadata
!38, i32 75, metadata !104} ; [ DW_TAG_auto_variable ]
!103 = metadata !{i32 524299, metadata !97, i32 73, i32 0} ; [
DW_TAG_lexical_block ]
!104 = metadata !{i32 524289, metadata !38, metadata !"", metadata !38, i32 0,
i64 85312, i64 64, i64 0, i32 0, metadata !46, metadata !105, i32 0, null} ; [
DW_TAG_array_type ]
!105 = metadata !{metadata !106}
!106 = metadata !{i32 524321, i64 0, i64 1332}    ; [ DW_TAG_subrange_type ]
!107 = metadata !{i32 73, i32 0, metadata !103, null}

define i32 @main() nounwind ssp {
bb.nph:
  tail call void @llvm.dbg.declare(metadata !101, metadata !102), !dbg !107
  ret i32 0, !dbg !107
}

declare void @llvm.dbg.declare(metadata, metadata) nounwind readnone

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 13:19:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 13:19:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7545] X86MCInstLower DWARF exception in -O0 -g test
In-Reply-To: <bug-7545-206@http.llvm.org/bugs/>
References: <bug-7545-206@http.llvm.org/bugs/>
Message-ID: <20100706181915.09DBE2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7545

Devang Patel <dpatel at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Devang Patel <dpatel at apple.com> 2010-07-06 13:19:14 CDT ---
Fixed the crash. r107678.
File PR 7577 to track missing debug info.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 13:38:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 13:38:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7578] New: undefined reference to
	`_Unwind_Resume_or_Rethrow'
Message-ID: <bug-7578-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7578

           Summary: undefined reference to `_Unwind_Resume_or_Rethrow'
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: physhh at googlemail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


I tested this on mingw32 4.5 and 3.4.5.

How to reproduce:
#include <set>
int main()
{
    std::set<int> foo;
    foo.insert(123);
    return 0;
}

Error:
undefined reference to `_Unwind_Resume_or_Rethrow'

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 14:52:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 14:52:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7417] fixits need to be de-duped somehow
In-Reply-To: <bug-7417-206@http.llvm.org/bugs/>
References: <bug-7417-206@http.llvm.org/bugs/>
Message-ID: <20100706195201.3E5852A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7417

Nick Lewycky <nicholas at mxc.ca> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #12 from Nick Lewycky <nicholas at mxc.ca> 2010-07-06 14:52:00 CDT ---
Fixed in r107690.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 18:27:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 18:27:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7528] [inline asm] spandsp compilation failure
In-Reply-To: <bug-7528-206@http.llvm.org/bugs/>
References: <bug-7528-206@http.llvm.org/bugs/>
Message-ID: <20100706232726.9DCD82A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7528

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Dale Johannesen <dalej at apple.com> 2010-07-06 18:27:26 CDT ---
fixed in 107727.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 18:47:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 18:47:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7454] Should not custom lower [EH_SJLJ_LONGJMP]!;
 UNREACHABLE executed at X86ISelLowering.cpp:7576!
In-Reply-To: <bug-7454-206@http.llvm.org/bugs/>
References: <bug-7454-206@http.llvm.org/bugs/>
Message-ID: <20100706234733.B1A382A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7454

Jim Grosbach <grosbach at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Jim Grosbach <grosbach at apple.com> 2010-07-06 18:47:33 CDT ---
Fixed in r107734. __builtin_longjmp() is effectively a nop for x86 now, just
like it is from llvm-gcc. __builtin_setjmp() is similarly just a constant '0'.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 18:53:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 18:53:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7572] Crash analyzing coreutils-6.9
In-Reply-To: <bug-7572-206@http.llvm.org/bugs/>
References: <bug-7572-206@http.llvm.org/bugs/>
Message-ID: <20100706235358.2E0282A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7572

Ted Kremenek <kremenek at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul  6 19:02:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue,  6 Jul 2010 19:02:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7570] williams.c (from 2001 IOCCC) crashes analyzer
In-Reply-To: <bug-7570-206@http.llvm.org/bugs/>
References: <bug-7570-206@http.llvm.org/bugs/>
Message-ID: <20100707000221.830322A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7570

Ted Kremenek <kremenek at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |kremenek at apple.com
         Resolution|                            |DUPLICATE

--- Comment #1 from Ted Kremenek <kremenek at apple.com> 2010-07-06 19:02:21 CDT ---
This no longer appears in TOT, and appears to have been fixed in r107738.  This
is the same issue as PR 7572.

*** This bug has been marked as a duplicate of bug 7572 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 00:10:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 00:10:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7528] [inline asm] spandsp compilation failure
In-Reply-To: <bug-7528-206@http.llvm.org/bugs/>
References: <bug-7528-206@http.llvm.org/bugs/>
Message-ID: <20100707051009.DFA7F2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7528

ojab <ojab at ojab.ru> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #9 from ojab <ojab at ojab.ru> 2010-07-07 00:10:09 CDT ---
Still the same crash, preprocessed source is the same.

clang version 2.8 (trunk 107753)
Target: x86_64-unknown-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 01:14:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 01:14:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7569] warn about store to null
In-Reply-To: <bug-7569-206@http.llvm.org/bugs/>
References: <bug-7569-206@http.llvm.org/bugs/>
Message-ID: <20100707061444.645DD2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7569

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Chris Lattner <clattner at apple.com> 2010-07-07 01:14:43 CDT ---
Warning implemented in r107756

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 04:42:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 04:42:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7581] New: SimpleRegisterCoalescing crash: Interval
 does not exist for register
Message-ID: <bug-7581-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7581

           Summary: SimpleRegisterCoalescing crash: Interval does not
                    exist for register
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu


Reduced by bugpoint from the x86-64 dragonegg buildbot failure.

$ llc coalesce.bc 
llc: llvm/include/llvm/CodeGen/LiveIntervalAnalysis.h:94: llvm::LiveInterval&
llvm::LiveIntervals::getInterval(unsigned int): Assertion `I != r2iMap_.end()
&& "Interval does not exist for register"' failed.
...
5  libc.so.6 0x00007fb6d1efba61 __assert_fail + 241
6  llc       0x0000000000a88208 llvm::LiveIntervals::getInterval(unsigned int)
+ 200
7  llc       0x0000000000aba593
llvm::SimpleRegisterCoalescing::RemoveCopyByCommutingDef(llvm::CoalescerPair
const&, llvm::MachineInstr*) + 1411
8  llc       0x0000000000abd3df
llvm::SimpleRegisterCoalescing::JoinCopy(llvm::CopyRec&, bool&) + 3807
9  llc       0x0000000000abd91f
llvm::SimpleRegisterCoalescing::CopyCoalesceInMBB(llvm::MachineBasicBlock*,
std::vector<llvm::CopyRec, std::allocator<llvm::CopyRec> >&) + 1151
10 llc       0x0000000000abdfce llvm::SimpleRegisterCoalescing::joinIntervals()
+ 750
11 llc       0x0000000000abf2f6
llvm::SimpleRegisterCoalescing::runOnMachineFunction(llvm::MachineFunction&) +
4566

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 04:42:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 04:42:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7582] New: Add triple support for Minix to llvm
Message-ID: <bug-7582-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7582

           Summary: Add triple support for Minix to llvm
           Product: new-bugs
           Version: trunk
          Platform: Other
        OS/Version: other
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: reeuwijk at few.vu.nl
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5174)
 --> (http://llvm.org/bugs/attachment.cgi?id=5174)
Patch for Minix triples in llvm

The attached patch adds support for Minix triples to llvm. This is needed for
clang support of Minix, which is submitted as a separate patch.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 04:43:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 04:43:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7583] New: Clang driver support for Minix
Message-ID: <bug-7583-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7583

           Summary: Clang driver support for Minix
           Product: clang
           Version: trunk
          Platform: Other
        OS/Version: other
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: reeuwijk at few.vu.nl
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5176)
 --> (http://llvm.org/bugs/attachment.cgi?id=5176)
A patch to add Minix driver support

The attached patch adds drivers for the Minix assembler and loader. This patch
needs a patch to llvm to support Minix triples. This patch has been submitted
separately.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 08:34:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 08:34:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7584] New: Bad code generated for Remainder
	operation
Message-ID: <bug-7584-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7584

           Summary: Bad code generated for Remainder operation
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: esteaqui at hotmail.com
                CC: llvmbugs at cs.uiuc.edu


When clang is fed an unsigned int as the divisor on the Remainder operand (%)
the generated code is faulty.

According to C99 specc, the sign of the remainder shall be the same as the sign
of the dividend.
Also, assuming n: dividend and d: divisor, "n / d x d - n == n % d"
(assuming a truncating division)

This bug is further explained by the attached source code.

As an example, following C99:
2 % 5 = 2
-2 % 5 = -2
But clang generates
2 % 5 = 2
-2 % 5 = 4

I got my C99 specification from 
http://www.open-std.org/jtc1/sc22/WG14/www/docs/n1256.pdf (relevant page is 82,
mark 6.5.5.6)
and also from 
http://en.wikipedia.org/wiki/Modulo_operation ("Integer modulo operators in
various programming languages")

If my interpretation of the standard is wrong, please let me know and mark the
bug as invalid.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 09:49:29 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 09:49:29 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7581] SimpleRegisterCoalescing crash: Interval does
 not exist for register
In-Reply-To: <bug-7581-206@http.llvm.org/bugs/>
References: <bug-7581-206@http.llvm.org/bugs/>
Message-ID: <20100707144929.EF9562A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7581

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |stoklund at 2pi.dk
         Resolution|                            |FIXED

--- Comment #2 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-07 09:49:29 CDT ---
Fixed in r107780
Thanks, Duncan!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 10:53:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 10:53:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7582] Add triple support for Minix to llvm
In-Reply-To: <bug-7582-206@http.llvm.org/bugs/>
References: <bug-7582-206@http.llvm.org/bugs/>
Message-ID: <20100707155314.BA6712A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7582

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-07 10:53:14 CDT ---
Thanks, applied in r107785.  Please send patches to the llvm-commits list
instead of filing bugs, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 11:02:16 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 11:02:16 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7583] Clang driver support for Minix
In-Reply-To: <bug-7583-206@http.llvm.org/bugs/>
References: <bug-7583-206@http.llvm.org/bugs/>
Message-ID: <20100707160216.86E1C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7583

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-07 11:02:16 CDT ---
Applied in r107788, thanks.  In the future, please stay within 80 columns (I
took care of it this time) and send patches to cfe-commits.  Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 11:07:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 11:07:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7584] Bad code generated for Remainder operation
In-Reply-To: <bug-7584-206@http.llvm.org/bugs/>
References: <bug-7584-206@http.llvm.org/bugs/>
Message-ID: <20100707160704.5D4422A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7584

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-07 11:07:04 CDT ---
Clang agrees with GCC here.  I don't have time to go language lawyer this, but
I'm pretty sure we're right. :)

If the quotient a/b is representable, the expression (a/b)*b + a%b shall equal
a.  Note that -2/5 is 0, not -1.

-Chris

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 12:03:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 12:03:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7585] New: Assertion failed: (Emitted && "Unable to
 issue a copy instruction!\n"), function EmitSpecialNode,
 file InstrEmitter.cpp, line 807.
Message-ID: <bug-7585-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7585

           Summary: Assertion failed: (Emitted && "Unable to issue a copy
                    instruction!\n"), function EmitSpecialNode, file
                    InstrEmitter.cpp, line 807.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu


on i386 with TOT:

witten delta# clang -O2 vga.c
vga.c:77:1: warning: declaration does not declare anything
typedef struct video_driver {
^~~~~~~
vga.c:108:250: warning: excess elements in struct initializer [-pedantic]
  ...vga_load_state, vga_set_origin, vga_read_hw_cursor, vga_set_hw_cursor,
vga_set_hw_cursor_shape, vga_blank_display, vga_mmap_buf, vga_dev_ioctl, ...
                                                                           
^~~~~~~~~~~~~~~~~~~~~~~
vga.c:110:19: warning: excess elements in struct initializer [-pedantic]
"vga", &vgavidsw, vga_configure};
                  ^~~~~~~~~~~~~
vga.c:121:1: warning: control reaches end of non-void function [-Wreturn-type]
}
^
Assertion failed: (Emitted && "Unable to issue a copy instruction!\n"),
function EmitSpecialNode, file InstrEmitter.cpp, line 807.
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
i386-unknown-freebsd9.0 -S -disable-free -main-file-name vga.c
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases -target-cpu
i486 -resource-dir /usr/local/lib/clang/2.8 -O2 -ferror-limit 19
-fmessage-length 152 -fgnu-runtime -fdiagnostics-show-option
-fcolor-diagnostics -o /tmp/cc-LPdrmk.s -x c vga.c 
1.    <eof> parser at end of file
2.    Code generation
3.    Running pass 'X86 DAG->DAG Instruction Selection' on function
'@vga_set_hw_cursor'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 12:55:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 12:55:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7586] New: possible integer wrong code bug on x64
Message-ID: <bug-7586-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7586

           Summary: possible integer wrong code bug on x64
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu, chenyang at cs.utah.edu


[regehr at gamow ~]$ clang -O1 foo.c -o foo
[regehr at gamow ~]$ ./foo
checksum g_166 = -10
[regehr at gamow ~]$ clang -O2 foo.c -o foo
[regehr at gamow ~]$ ./foo
checksum g_166 = -2
[regehr at gamow ~]$ clang -v
clang version 2.8 (trunk 107688)
Target: x86_64-unknown-linux-gnu
Thread model: posix
[regehr at gamow ~]$ cat small.c
extern int printf (__const char *__restrict __format, ...);

int g_14 = 0xE8B0D193L;
int g_66 = -10L;

short foo(short s1, short s2)
{
    return s1 * s2;
}

int  main(void)
{
    int *l_320 = &g_14;
    char l_441 = -9L;
    if (((foo(l_441, (*l_320))) <= -1L))
       g_66 = 1;
    else
       g_66 = 0;
    printf("checksum g_66 = %d\n", g_66);
    return g_66;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 13:09:29 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 13:09:29 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7585] Assertion failed: (Emitted && "Unable to
 issue a copy instruction!\n"), function EmitSpecialNode,
 file InstrEmitter.cpp, line 807.
In-Reply-To: <bug-7585-206@http.llvm.org/bugs/>
References: <bug-7585-206@http.llvm.org/bugs/>
Message-ID: <20100707180929.A215B2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7585

Roman Divacky <rdivacky at freebsd.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #2 from Roman Divacky <rdivacky at freebsd.org> 2010-07-07 13:09:29 CDT ---
nevermind. it's a dupe of my own bug report :(

*** This bug has been marked as a duplicate of bug 7540 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 13:32:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 13:32:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7587] New: segfault on parsing,
 related to a complex default value for a template parameter
Message-ID: <bug-7587-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7587

           Summary: segfault on parsing, related to a complex default
                    value for a template parameter
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: orzel at freehackers.org
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5180)
 --> (http://llvm.org/bugs/attachment.cgi?id=5180)
file to reproduce the problem

The problem appears with recent trunk (updated today, svn rev 107795, but has
been here for several weeks).

The original code is from eigen (homepage: http://eigen.tuxfamily.org/, code:
http://bitbucket.org/eigen/eigen/), but i've narrowed it to this file.

g++ and lot of other compilers have no problem with this. For this narrowed
problem, g++ will fail because of m_matrix being undeclared, but this is
something else.

clang segfault this way:

orzel at berlioz /home/orzel% /home/orzel/svn/llvm/Release/bin/clang++
~/clang.problem.cpp
clang: SemaTemplateInstantiate.cpp:881:
clang::QualType<unnamed>::TemplateInstantiator::TransformTemplateTypeParmType(clang::TypeLocBuilder&,
clang::TemplateTypeParmTypeLoc, clang::QualType): Assertion
`TemplateArgs(T->getDepth(), T->getIndex()).getKind() == TemplateArgument::Type
&& "Template argument kind mismatch"' failed.
0  clang           0x00000000015629cf
1  clang           0x000000000156351a
2  libpthread.so.0 0x00007f54da6461b0
3  libc.so.6       0x00007f54d995e555 gsignal + 53
4  libc.so.6       0x00007f54d995f9d6 abort + 390
5  libc.so.6       0x00007f54d99570c5 __assert_fail + 245
6  clang           0x0000000000818f98
7  clang           0x0000000000819407
8  clang           0x0000000000825ee2
9  clang           0x0000000000827cdc
10 clang           0x0000000000817892
11 clang           0x0000000000819407
12 clang           0x000000000081961d
13 clang           0x0000000000819e89
14 clang           0x000000000081a30e
15 clang           0x0000000000817ad7
16 clang           0x0000000000819407
17 clang           0x000000000081dda8
18 clang           0x00000000007e20f9
19 clang           0x00000000007f5ece
20 clang           0x00000000007f8e34
21 clang           0x00000000007f9440
22 clang           0x0000000000b0e1c8
23 clang           0x0000000000b034b6
24 clang           0x0000000000acf548
25 clang           0x0000000000addd6a
26 clang           0x0000000000aea810
27 clang           0x0000000000aed13d
28 clang           0x0000000000aee10d
29 clang           0x0000000000addf86
30 clang           0x0000000000b0c44b
31 clang           0x0000000000b0f74b
32 clang           0x0000000000b1115b
33 clang           0x0000000000ae15a7
34 clang           0x0000000000ad3850
35 clang           0x0000000000ad3fba
36 clang           0x0000000000678cbb
37 clang           0x00000000005275c4
38 clang           0x00000000004273ad
39 clang           0x00000000004096b5
40 clang           0x0000000000410195 main + 2629
41 libc.so.6       0x00007f54d994ab6d __libc_start_main + 253
42 clang           0x00000000004075c9
Stack dump:
0.      Program arguments: /home/orzel/svn/llvm/Release/bin/clang -cc1 -triple
x86_64-unknown-linux-gnu -S -disable-free -main-file-name clang.problem.cpp
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases
-munwind-tables -target-cpu x86-64 -resource-dir
/home/orzel/svn/llvm/Release/lib/clang/2.8 -ferror-limit 19 -fmessage-length
172 -fexceptions -fgnu-runtime -fdiagnostics-show-option -fcolor-diagnostics -o
/tmp/cc-UeX0i4.s -x c++ /home/orzel/clang.problem.cpp 
1.      /home/orzel/clang.problem.cpp:27:20: at annotation token 
2.      /home/orzel/clang.problem.cpp:10:50: parsing struct/union/class body
'VectorwiseOp'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 15:33:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 15:33:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7540] subreg related scheduler crash: "Unable to
 issue a copy instruction!"
In-Reply-To: <bug-7540-206@http.llvm.org/bugs/>
References: <bug-7540-206@http.llvm.org/bugs/>
Message-ID: <20100707203346.6DA3F2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7540

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-07 15:33:46 CDT ---
Fixed in r107809

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 17:35:31 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 17:35:31 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7195] [MC] unrecognized instruction "callw FOO"
In-Reply-To: <bug-7195-206@http.llvm.org/bugs/>
References: <bug-7195-206@http.llvm.org/bugs/>
Message-ID: <20100707223532.03BDB2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7195

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Chris Lattner <clattner at apple.com> 2010-07-07 17:35:31 CDT ---
Implemented in r107825/107826.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 17:35:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 17:35:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7556] Temporary object with trivial default
 constructor but non-trivial destructor isn't getting destroyed
In-Reply-To: <bug-7556-206@http.llvm.org/bugs/>
References: <bug-7556-206@http.llvm.org/bugs/>
Message-ID: <20100707223532.8974A2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7556

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-07 17:35:32 CDT ---
Fixed in r107827.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 18:17:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 18:17:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7399] LLVM doesn't properly parse section names (or
 emit them into .ll files)
In-Reply-To: <bug-7399-206@http.llvm.org/bugs/>
References: <bug-7399-206@http.llvm.org/bugs/>
Message-ID: <20100707231722.03A852A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7399

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Chris Lattner <clattner at apple.com> 2010-07-07 18:17:21 CDT ---
The asmprinter is now fixed in r107834 and we can now round trip through .ll
files successfully.  Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 18:18:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 18:18:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7467] Code ok in C++ mode does not parse in C++0x
	mode
In-Reply-To: <bug-7467-206@http.llvm.org/bugs/>
References: <bug-7467-206@http.llvm.org/bugs/>
Message-ID: <20100707231856.1E29E2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7467

Sebastian Redl <sebastian.redl at getdesigned.at> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sebastian.redl at getdesigned.
                   |                            |at
         Resolution|                            |FIXED

--- Comment #1 from Sebastian Redl <sebastian.redl at getdesigned.at> 2010-07-07 18:18:55 CDT ---
"Fixed" in r107853. Our C++0x handling of destructors was broken because the
specification is still in flux.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 18:24:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 18:24:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4499] clang rejects valid character constants
 beginning with three backslashes
In-Reply-To: <bug-4499-206@http.llvm.org/bugs/>
References: <bug-4499-206@http.llvm.org/bugs/>
Message-ID: <20100707232439.77F852A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4499

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-07 18:24:38 CDT ---
Applied in r107836, thanks Kyle!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 18:37:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 18:37:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7556] Temporary object with trivial default
 constructor but non-trivial destructor isn't getting destroyed
In-Reply-To: <bug-7556-206@http.llvm.org/bugs/>
References: <bug-7556-206@http.llvm.org/bugs/>
Message-ID: <20100707233748.62A602A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7556

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-07 18:37:48 CDT ---
Fix reverted in r107837 due to bootstrap breakage.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 19:28:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 19:28:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7437] Address space not being correctly preserved
In-Reply-To: <bug-7437-206@http.llvm.org/bugs/>
References: <bug-7437-206@http.llvm.org/bugs/>
Message-ID: <20100708002818.938392A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7437

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-07 19:28:18 CDT ---
Second half was a scalarrepl bug, fixed in r107846.  We now compile the example
to this:


define void @Test(%struct.anon addrspace(2)* nocapture %pPtr) nounwind ssp {
entry:
  %pPtr15.0 = getelementptr inbounds %struct.anon addrspace(2)* %pPtr, i64 0,
i32 0, i64 0 ; <float addrspace(2)*> [#uses=2]
  %tmp = load float addrspace(2)* %pPtr15.0, align 4 ; <float> [#uses=1]
  %sub = fadd float %tmp, -5.000000e+00           ; <float> [#uses=1]
  store float %sub, float addrspace(2)* %pPtr15.0, align 4
  ret void
}

Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul  7 23:15:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed,  7 Jul 2010 23:15:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7588] New: Wrong code for _mm_shufflehi_epi16
Message-ID: <bug-7588-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7588

           Summary: Wrong code for _mm_shufflehi_epi16
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Headers
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: astrange at ithinksw.com
                CC: llvmbugs at cs.uiuc.edu


Using checker-243 -O3 -m32 -march=pentium-m.

Source:
#include <xmmintrin.h>

__m128i a(__m128i b)
{
    return _mm_shufflehi_epi16(b, 1 | (0 << 2) | (3 << 4) | (2 << 6));
}

gcc:
_a:
    pushl    %ebp
    movl    %esp, %ebp
    subl    $8, %esp
    pshufhw    $177, %xmm0, %xmm0
    leave
    ret

clang:
_a:                                     ## @a
## BB#0:                                ## %entry
    pushl    %ebp
    movl    %esp, %ebp
    pextrw    $5, %xmm0, %eax
    movlhps    %xmm0, %xmm0
    pshufhw    $-76, %xmm0, %xmm0
    pinsrw    $4, %eax, %xmm0
    popl    %ebp
    ret

It looks like the macro definition in xmmintrin is wrong.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 01:14:25 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 01:14:25 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7556] Temporary object with trivial default
 constructor but non-trivial destructor isn't getting destroyed
In-Reply-To: <bug-7556-206@http.llvm.org/bugs/>
References: <bug-7556-206@http.llvm.org/bugs/>
Message-ID: <20100708061425.43F0F2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7556

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Douglas Gregor <dgregor at apple.com> 2010-07-08 01:14:25 CDT ---
Re-fixed in r107857.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 03:12:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 03:12:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7589] New: GVN incorrectly kills instructions
Message-ID: <bug-7589-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7589

           Summary: GVN incorrectly kills instructions
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: zhousheng00 at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Here is input ll code:

; ModuleID = 'error.bc'
target datalayout =
"e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
target triple = "x86_64-unknown-linux-gnu"

define i16 @BlockCol([65 x i16]* %row2col, [64 x i16]* %colout, i4 %i) nounwind
{
entry:
  %row2col.addr1 = getelementptr [65 x i16]* %row2col, i64 0, i64 64 ; <i16*>
[#uses=1]
  %row2col.load = load i16* %row2col.addr1, align 1 ; <i16> [#uses=1]
  %dc_scaler = trunc i16 %row2col.load to i8      ; <i8> [#uses=0]
  br label %bb

bb:                                               ; preds = %entry
  %0 = zext i4 %i to i64                          ; <i64> [#uses=0]
  %.cast28 = zext i4 %i to i7                     ; <i7> [#uses=0]
  %.cast27 = zext i4 %i to i6                     ; <i6> [#uses=0]
  %.cast27.cast = zext i4 %i to i5                ; <i5> [#uses=1]
  %.cast26 = zext i4 %i to i5                     ; <i5> [#uses=1]
  %sum532 = or i5 %.cast26, -16                   ; <i5> [#uses=1]
  %sum5.cast = zext i5 %sum532 to i64             ; <i64> [#uses=1]
  %row2col.addr.1.26 = getelementptr [65 x i16]* %row2col, i64 0, i64
%sum5.cast ; <i16*> [#uses=1]
  %row2col.load.1.2 = load i16* %row2col.addr.1.26, align 1 ; <i16> [#uses=1]
  %sum1334 = or i5 %.cast27.cast, -16             ; <i5> [#uses=1]
  %sum13.cast31 = sext i5 %sum1334 to i6          ; <i6> [#uses=1]
  %sum13.cast = zext i6 %sum13.cast31 to i64      ; <i64> [#uses=1]
  %row2col.addr.1.614 = getelementptr [65 x i16]* %row2col, i64 0, i64
%sum13.cast ; <i16*> [#uses=1]
  %row2col.load.1.6 = load i16* %row2col.addr.1.614, align 1 ; <i16> [#uses=1]
  %.ret = add i16 %row2col.load.1.6, %row2col.load.1.2 ; <i16> [#uses=1]
  ret i16 %.ret
}


After GVN, the LoadInst "%row2col.load.1.6" is replaced with
"%row2col.load.1.2", however the address %sum5.cast and %sum13.cast are
obviously different.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 03:17:08 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 03:17:08 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7590] New: C++ mangler output not correct in some
	cases
Message-ID: <bug-7590-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7590

           Summary: C++ mangler output not correct in some cases
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: thom.heller at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The C++ mangler seems to have problems with template-template parameters.

Here is the output of the attached test-case after piping through c++filt:

$ clang++ mangle_test.cpp && ./a.out | c++filt -t
A<B, C, void, void, void>
A<B, C, D, void, void>
A<B, C, D, C, void>
Z<B, void, void, void>
Z<B, D, void, void>
Z<B, D, D, void>

The correct output should be:
A<B, C, void, void, void>
A<B, C, D, void, void>
A<B, C, D, D, void>
Z<B, void, void, void>
Z<B, D, void, void>
Z<B, D, D, void>


Note, that a call to abi::__cxa_demangle gives the same results

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 04:10:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 04:10:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7591] New: exception Failure("There are two
 interfaces of module Llvm_executionengine.")
Message-ID: <bug-7591-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7591

           Summary: exception Failure("There are two interfaces of module
                    Llvm_executionengine.")
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ojab at ojab.ru
                CC: llvmbugs at cs.uiuc.edu


During `make install`

make[2]: Leaving directory `/sources/llvm/bindings/ocaml'
Fatal error: exception Failure("There are two interfaces of module
Llvm_executionengine.")
make[1]: *** [regen-ocamldoc] Error 2
make[1]: Target `install' not remade because of errors.

llvm svn r107860, ocaml-3.11.2, linux x86_64.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 05:25:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 05:25:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7592] New: (a == 0 || a < 0 || a > 0)
Message-ID: <bug-7592-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7592

           Summary: (a == 0 || a < 0 || a > 0)
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Scalar Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: Edmund.Grimley-Evans at arm.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5187)
 --> (http://llvm.org/bugs/attachment.cgi?id=5187)
test cases

Perhaps someone is interested in this. In the two functions defined below (the
code is also in the attachment) the call to unused() is unreachable. GCC 4.4.1
seems to get this right, but Clang/LLVM rev 107760 didn't when invoked like
this:

clang -cc1 -triple thumbv7-eabi -O3 -target-cpu cortex-a8 -S tri.c -o tri.s

int unused();

int f1(int a)
{
  return a == 0 ? 11 : a < 0 ? 22 : a > 0 ? 33 : unused();
}

int f2(int a)
{
  if (a == 0)
    return 11;
  if (a < 0)
    return 22;
  if (a > 0)
    return 33;
  return unused();
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 06:06:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 06:06:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7593] New: Assertion failed:
 (castIsValid(getOpcode(), S, Ty) && "Illegal BitCast")
Message-ID: <bug-7593-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7593

           Summary: Assertion failed: (castIsValid(getOpcode(), S, Ty) &&
                    "Illegal BitCast")
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: themiwi at users.sourceforge.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


While compiling a file on Mac OS X 10.6.4 with clang++ 2.8 (trunk 107757),
clang++ aborted with the following error message:

Assertion failed: (castIsValid(getOpcode(), S, Ty) && "Illegal BitCast"),
function BitCastInst, file
/Users/mwild/Downloads/llvm/lib/VMCore/Instructions.cpp, line 2647.
Stack dump:
0.    Program arguments: /Users/mwild/root/software/clang-svn-dbg/bin/clang
-cc1 -triple x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free
-main-file-name IPstream.C.i -pic-level 1 -mdisable-fp-elim -masm-verbose
-munwind-tables -target-cpu core2 -resource-dir
/Users/mwild/root/software/clang-svn-dbg/lib/clang/2.8 -ferror-limit 19
-fmessage-length 190 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o IPstream.C.o -x c++
CMakeFiles/OpenFOAM.dir/db/IOstreams/Pstreams/IPstream.C.i 
1.    <eof> parser at end of file
2.    Per-file LLVM IR generation
3.   
/Users/mwild/Projects/FreeFOAM/src/OpenFOAM/db/IOstreams/Pstreams/IPstream.C:71:29:
Generating code for declaration 'Foam::IPstream::readFromBuffer'
4.   
/Users/mwild/Projects/FreeFOAM/src/OpenFOAM/db/IOstreams/Pstreams/IPstream.C:72:1:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)


The preprocessed source is attached. Compile with

$ clang++ -c -x c++ IPstream.C.i

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 06:55:36 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 06:55:36 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7367] Opt crash on attached .bc file when built by
	msvc 2010
In-Reply-To: <bug-7367-206@http.llvm.org/bugs/>
References: <bug-7367-206@http.llvm.org/bugs/>
Message-ID: <20100708115536.D55AD2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7367

Duncan Sands <baldrick at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #19 from Duncan Sands <baldrick at free.fr> 2010-07-08 06:55:35 CDT ---
Fixed here:
 
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20100705/103839.html
Thanks for tracking it down!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 10:36:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 10:36:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7595] New: DWARF debug info does not handle forward
 declarations properly
Message-ID: <bug-7595-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7595

           Summary: DWARF debug info does not handle forward declarations
                    properly
           Product: clang
           Version: trunk
          Platform: Other
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: kan at freebsd.org
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5190)
 --> (http://llvm.org/bugs/attachment.cgi?id=5190)
Create DICompositeType node with FlagFwdDecl for forward declatations of
scructs and unions

Given the following trivial example:

struct a;
union  b;

void foo(struct a *ptr_a, union b *ptr_b)
{
}

clang 2.7 and trunk both end up generating the following DWARF info fragment do
describe pointers to struct a and union b:

<1><  229>      DW_TAG_structure_type
                DW_AT_name                  fwd.type.0
                DW_AT_byte_size             0
                DW_AT_decl_file             2 /home/kan/test/ctf/test.c
                DW_AT_decl_line             1
<1><  244>      DW_TAG_pointer_type
                DW_AT_type                  <229>
                DW_AT_byte_size             8
                DW_AT_decl_file             2 /home/kan/test/ctf/test.c
                DW_AT_decl_line             0
<1><  252>      DW_TAG_union_type
                DW_AT_name                  fwd.type.1
                DW_AT_byte_size             0
                DW_AT_decl_file             2 /home/kan/test/ctf/test.c
                DW_AT_decl_line             2
<1><  267>      DW_TAG_pointer_type
                DW_AT_type                  <252>
                DW_AT_byte_size             8
                DW_AT_decl_file             2 /home/kan/test/ctf/test.c
                DW_AT_decl_line             0

Note the use of fictitious zero-sized structures fwd.type.[12].

Attached patch attempts to fix this, by emitting proper struct and enum
definitions with DW_AT_declaration=1 for forward declarations. This results in
the following (correct) DWARF fragment:

<1><  229>      DW_TAG_structure_type
                DW_AT_name                  a
                DW_AT_declaration           yes(1)
<1><  233>      DW_TAG_pointer_type
                DW_AT_type                  <229>
                DW_AT_byte_size             8
                DW_AT_decl_file             2 /home/kan/test/ctf/test.c
                DW_AT_decl_line             0
<1><  241>      DW_TAG_union_type
                DW_AT_name                  b
                DW_AT_declaration           yes(1)
<1><  245>      DW_TAG_pointer_type
                DW_AT_type                  <241>
                DW_AT_byte_size             8
                DW_AT_decl_file             2 /home/kan/test/ctf/test.c
                DW_AT_decl_line             0

Note proper names and DW_TAG_declaration use and notice that respective
pointers now actually make sense.

As a side effect, this makes OpenSolaris' ctftools work properly on FreeBSD
kernel compiled with clang, instead of taking ages to compile, mostly sifting
though tons of fwd.type.<num> fictitious types.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 11:33:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 11:33:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7596] New: Assertion failed: (0 && "Unexpected
 Start and End labels for a inlined scope!")
Message-ID: <bug-7596-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7596

           Summary: Assertion failed: (0 && "Unexpected Start and End
                    labels for a inlined scope!")
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5192)
 --> (http://llvm.org/bugs/attachment.cgi?id=5192)
t.bc.gz

% llc t.bc
Assertion failed: (0 && "Unexpected Start and End  labels for a inlined
scope!"), function constructInlinedScopeDIE, file DwarfDebug.cpp, line 1506.
Stack dump:
0.    Program arguments:
/usr/home/buildbot/buildbot/clang-i686-freebsd/llvm/Debug+Asserts/bin/llc -o
bugpoint-test-program.bc-sJo88m.llc.s bugpoint-test-program.bc-sJo88m 
1.    Running pass 'Function Pass Manager' on module
'bugpoint-test-program.bc-sJo88m'.
2.    Running pass 'X86 AT&T-Style Assembly Printer' on function '@fpathconf'

When trying to reduce this with bugpoint it dies too:
bugpoint -run-llc t.bc
Read input file      : 't.bc'
*** All input ok
Initializing execution environment: Found gcc: /usr/bin/gcc
Running the code generator to test for a crash: <llc>
Error running tool:
  /usr/home/buildbot/buildbot/clang-i686-freebsd/llvm/Debug+Asserts/bin/llc -o
bugpoint-test-program.bc-sJo88m.llc.s bugpoint-test-program.bc-sJo88m
Assertion failed: (0 && "Unexpected Start and End  labels for a inlined
scope!"), function constructInlinedScopeDIE, file DwarfDebug.cpp, line 1506.
Stack dump:
0.    Program arguments:
/usr/home/buildbot/buildbot/clang-i686-freebsd/llvm/Debug+Asserts/bin/llc -o
bugpoint-test-program.bc-sJo88m.llc.s bugpoint-test-program.bc-sJo88m 
1.    Running pass 'Function Pass Manager' on module
'bugpoint-test-program.bc-sJo88m'.
2.    Running pass 'X86 AT&T-Style Assembly Printer' on function '@fpathconf'
*** Debugging code generator crash!

Checking to see if we can delete global inits: Unknown constant!
UNREACHABLE executed at BitcodeWriter.cpp:905!
Stack dump:
0.    Program arguments:
/home/buildbot/buildbot/clang-i686-freebsd/llvm/Debug+Asserts/bin/bugpoint
-run-llc t.bc 
Abort

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 11:35:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 11:35:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7594] Initializing static member variable with call
 to constructor causes assert to fail
In-Reply-To: <bug-7594-206@http.llvm.org/bugs/>
References: <bug-7594-206@http.llvm.org/bugs/>
Message-ID: <20100708163550.66D5D2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7594

Nico Weber <nicolasweber at gmx.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dgregor at apple.com,
                   |                            |llvmbugs at cs.uiuc.edu,
                   |                            |nicolasweber at gmx.de
          Component|Frontend                    |C++

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 13:19:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 13:19:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7597] New: Crash creating object without defined
	constructor.
Message-ID: <bug-7597-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7597

           Summary: Crash creating object without defined constructor.
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dlarimer at gmail.com
                CC: llvmbugs at cs.uiuc.edu


I was attempting to compile code with clang that compiled fine with gcc and
VC++.  I was unable to determine the exact "cause" of the clang crash because
it could be triggered by various arrangements of the code and showed no clear
"pattern".   

If I directly specify the type (instead of using the template parameter) it
works.  If I use a template parameter and then instantiate the class with the
template parameter set to the same type it crashes.

If i construct on the stack it works.  If I construct via new it crashes clang;


Given the following code
class SomeType{};

template<typename T, template <typename T> DelegateType = DefautlDelegateType >
struct stub
{
     // lots of nested struct declarations here.
}

namespace bb {
   class server
   {
       template<typename T>
       server( boost::shared_ptr<T>& ptr )
       {
            stub<SomeType>  x; // works
            new stub<SomeType>; // works
            stub<T> y; // works  
             new stub<T>; // with T == SomeType crashes clang, unless stub
defines a constructor
       }
   };
}

Assertion failed: ((Record->hasTrivialCopyConstructor() ||
Record->hasTrivialCopyAssignment()) && "Trying to aggregate-copy a type without
a trivial copy " "constructor or assignment operator"), function
EmitAggregateCopy, file
/Users/dlarimer/projects/llvm/tools/clang/lib/CodeGen/CGExprAgg.cpp, line 792.
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free -main-file-name
serv.cpp -pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables
-target-cpu core2 -resource-dir /usr/local/lib/clang/2.8 -D QT_NO_DEBUG_OUTPUT
-D __STDC_LIMIT_MACROS -I /Users/dlarimer/projects/fl/libraries -I
/Users/dlarimer/projects/fl/libraries/core -I
/Users/dlarimer/projects/fl/libraries/drivers -I /usr/local/include
-ferror-limit 19 -fmessage-length 270 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o CMakeFiles/bbs.dir/serv.cpp.o
-x c++ /Users/dlarimer/projects/fl/libraries/fl/backbone/serv.cpp -v 
1.    <eof> parser at end of file
2.    Per-file LLVM IR generation
3.    /Users/dlarimer/projects/fl/libraries/fl/backbone/server.hpp:20:5:
Generating code for declaration 'bb::server::server'
4.    /Users/dlarimer/projects/fl/libraries/fl/backbone/server.hpp:22:5: LLVM
IR generation of compound statement ('{}')

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 13:38:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 13:38:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7587] segfault on parsing,
 related to a complex default value for a template parameter
In-Reply-To: <bug-7587-206@http.llvm.org/bugs/>
References: <bug-7587-206@http.llvm.org/bugs/>
Message-ID: <20100708183822.241B52A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7587

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Douglas Gregor <dgregor at apple.com> 2010-07-08 13:38:21 CDT ---
Fixed in r107895.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 14:57:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 14:57:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7595] DWARF debug info does not handle forward
 declarations properly
In-Reply-To: <bug-7595-206@http.llvm.org/bugs/>
References: <bug-7595-206@http.llvm.org/bugs/>
Message-ID: <20100708195704.1C4AC2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7595

devang.patel <devang.patel at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from devang.patel <devang.patel at gmail.com> 2010-07-08 14:57:03 CDT ---
Thanks!  I applied your patch. r107900.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 15:10:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 15:10:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7588] Wrong code for _mm_shufflehi_epi16
In-Reply-To: <bug-7588-206@http.llvm.org/bugs/>
References: <bug-7588-206@http.llvm.org/bugs/>
Message-ID: <20100708201004.016DF2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7588

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #1 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-08 15:10:03 CDT ---
Fixed in r107902.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 16:07:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 16:07:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7598] New: Spurious error involving enums,
 conditional expressions, and const return values
Message-ID: <bug-7598-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7598

           Summary: Spurious error involving enums, conditional
                    expressions, and const return values
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: matthewbg at google.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5193)
 --> (http://llvm.org/bugs/attachment.cgi?id=5193)
test case

Below, v has type Enum; the return type of g() is 'const Enum', which is a
little suspicious, but not AFAICT illegal.

% clang -fsyntax-only conditional_enum.cc
conditional_enum.cc:10:8: error: cannot initialize a variable of type 'Enum'
     with an rvalue of type 'int'
 Enum e = false ? g() : v;
      ^   ~~~~~~~~~~~~~~~
1 error generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 16:37:30 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 16:37:30 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7566] clang++ crashes when includes are not found
In-Reply-To: <bug-7566-206@http.llvm.org/bugs/>
References: <bug-7566-206@http.llvm.org/bugs/>
Message-ID: <20100708213730.DEF202A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7566

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-08 16:37:30 CDT ---


*** This bug has been marked as a duplicate of bug 7481 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 17:39:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 17:39:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7596] Assertion failed: (0 && "Unexpected Start and
 End labels for a inlined scope!")
In-Reply-To: <bug-7596-206@http.llvm.org/bugs/>
References: <bug-7596-206@http.llvm.org/bugs/>
Message-ID: <20100708223934.1BEB02A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7596

devang.patel <devang.patel at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |devang.patel at gmail.com
         Resolution|                            |FIXED

--- Comment #1 from devang.patel <devang.patel at gmail.com> 2010-07-08 17:39:33 CDT ---
Fixed r107923.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 18:07:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 18:07:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7594] Initializing static member variable with call
 to constructor causes assert to fail
In-Reply-To: <bug-7594-206@http.llvm.org/bugs/>
References: <bug-7594-206@http.llvm.org/bugs/>
Message-ID: <20100708230755.7EA1F2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7594

Sebastian Redl <sebastian.redl at getdesigned.at> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Sebastian Redl <sebastian.redl at getdesigned.at> 2010-07-08 18:07:55 CDT ---
Fixed in r107927.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 20:44:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 20:44:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7600] New: Incorrect return type from
 __sync_val_compare_and_swap builtin
Message-ID: <bug-7600-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7600

           Summary: Incorrect return type from __sync_val_compare_and_swap
                    builtin
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chandlerc at gmail.com
                CC: llvmbugs at cs.uiuc.edu


% cat t.cc
void *foo(void **m) {
  void *newval = 0;
  void *oldval = 0;
  return __sync_val_compare_and_swap(m, oldval, newval);
}

% ./clang -fsyntax-only t.cc
t.cc:4:10: error: cannot initialize return object of type 'void *' with an
rvalue of type 'long long'
  return __sync_val_compare_and_swap(m, oldval, newval);
         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 22:05:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 22:05:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6889] memory unsafety crash bug
In-Reply-To: <bug-6889-206@http.llvm.org/bugs/>
References: <bug-6889-206@http.llvm.org/bugs/>
Message-ID: <20100709030522.C09702A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6889

John Regehr <regehr at cs.utah.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WORKSFORME

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 22:11:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 22:11:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7601] New: x64 clang segfault
Message-ID: <bug-7601-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7601

           Summary: x64 clang segfault
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


Valgrind says it's a null-ptr deref.

[regehr at bethe tmp602]$ clang -v

clang version 2.8 (trunk 107853)
Target: x86_64-unknown-linux-gnu
Thread model: posix

[regehr at bethe tmp602]$ clang -O2 -c small.c

small.c:36:1: warning: type specifier missing, defaults to 'int'
[-Wimplicit-int]
int326 (struct S0 p_7, struct S3 p_8, int16_t p_9, int16_t p_10, uint8_t p_11)
^
small.c:38:12: warning: incompatible pointer types initializing 'int32_t *'     
      (aka 'int *') with an expression of type 'int16_t *' (aka 'short *')
  int32_t *l_431 = &g_432;
           ^       ~~~~~~
0  clang           0x0000000001353a2f                                           
1  clang           0x0000000001355ac2
2  libpthread.so.0 0x00007fdd45ac6190
3  clang           0x0000000000fecdc9
4  clang           0x0000000000fee266
5  clang           0x0000000000fee64f
6  clang           0x0000000000feeefe
7  clang           0x0000000000ff0015
8  clang           0x00000000012d6e9d
9  clang           0x00000000012d707b
10 clang           0x00000000012d736c
11 clang           0x000000000050a67e
12 clang           0x000000000050678b
13 clang           0x00000000006159cf
14 clang           0x00000000005076f4
15 clang           0x000000000042711d
16 clang           0x000000000040a02b
17 clang           0x0000000000410281 main + 3185
18 libc.so.6       0x00007fdd44db7abd __libc_start_main + 253
19 clang           0x0000000000407779
Stack dump:
0.      Program arguments:
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r107853-install/bin/clang
-cc1 -triple x86_64-unknown-linux-gnu -S -disable-free -main-file-name small.c
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases
-munwind-tables -target-cpu x86-64 -resource-dir
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r107853-install/lib/clang/2.8
-O2 -ferror-limit 19 -fmessage-length 86 -fgnu-runtime
-fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-y4sLYQ.s -x c small.c 
1.      <eof> parser at end of file
2.      Code generation
3.      Running pass 'Simple Register Coalescing' on function '@int326'
clang: error: clang frontend command failed due to signal 11 (use -v to see
invocation)

[regehr at bethe tmp602]$ cat small.c

typedef short int int16_t;
typedef int int32_t;
typedef unsigned char uint8_t;
typedef unsigned int uint32_t;
typedef unsigned long int uint64_t;
static int16_t
safe_mod_func_int32_t_s_s (int32_t si1, int32_t si2)
{
  return si2 == 0 || si1 && si2 == 1 ? si1 : (si1 % si2);
}

static uint64_t
safe_div_func_uint64_t_u_u (uint64_t ui1, uint64_t ui2)
{
  return ui2 == 0 ? ui1 : (ui1 / ui2);
}

struct S0
{
  uint32_t f0;
};
struct S2
{
  const unsigned f1:1;
};
struct S3
{
};
int16_t g_51;
const int16_t *g_200 = &g_51;
const int16_t **volatile g_199 = &g_200;
struct S2 g_403[7] = {
};

int16_t g_432;
int326 (struct S0 p_7, struct S3 p_8, int16_t p_9, int16_t p_10, uint8_t p_11)
{
  int32_t *l_431 = &g_432;
  *l_431 =
    p_7.f0 <=
    safe_mod_func_int32_t_s_s (safe_div_func_uint64_t_u_u (**g_199, *l_431) |
                               +*l_431, *l_431) || g_403[2].f1;
  return 0;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul  8 23:36:25 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu,  8 Jul 2010 23:36:25 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7601] x64 clang segfault
In-Reply-To: <bug-7601-206@http.llvm.org/bugs/>
References: <bug-7601-206@http.llvm.org/bugs/>
Message-ID: <20100709043625.4B4452A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7601

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-08 23:36:24 CDT ---
Fixed in r107955

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 00:31:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 00:31:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7602] New: Cannot emit physreg copy instruction
Message-ID: <bug-7602-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7602

           Summary: Cannot emit physreg copy instruction
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5195)
 --> (http://llvm.org/bugs/attachment.cgi?id=5195)
bugpoint-reduced-simplified.bc

Fails when building freebsd/sbin/newfs/mkfs.c with clang version 2.8 (trunk
107952)

% llc bugpoint-reduced-simplified.bc 
Cannot emit physreg copy instruction
UNREACHABLE executed at
/Users/pwo/src/llvm/lib/CodeGen/TargetInstrInfoImpl.cpp:381!
0  llc               0x00000001005fecb2 std::_Rb_tree<llvm::sys::Path,
llvm::sys::Path, std::_Identity<llvm::sys::Path>, std::less<llvm::sys::Path>,
std::allocator<llvm::sys::Path>
>::_M_erase(std::_Rb_tree_node<llvm::sys::Path>*) + 8546
1  llc               0x00000001005ff169 std::_Rb_tree<llvm::sys::Path,
llvm::sys::Path, std::_Identity<llvm::sys::Path>, std::less<llvm::sys::Path>,
std::allocator<llvm::sys::Path>
>::_M_erase(std::_Rb_tree_node<llvm::sys::Path>*) + 9753
2  libSystem.B.dylib 0x00007fff85e6835a _sigtramp + 26
3  libSystem.B.dylib 0x0000000100900000 _sigtramp + 2057927872
4  llc               0x00000001005e3285 llvm::cl::opt<bool, true,
llvm::cl::parser<bool> >::~opt() + 8517
5  llc               0x00000001003c4b6e
std::vector<std::pair<llvm::MachineBasicBlock*, unsigned int>,
std::allocator<std::pair<llvm::MachineBasicBlock*, unsigned int> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::MachineBasicBlock*,
unsigned int>*, std::vector<std::pair<llvm::MachineBasicBlock*, unsigned int>,
std::allocator<std::pair<llvm::MachineBasicBlock*, unsigned int> > > >,
std::pair<llvm::MachineBasicBlock*, unsigned int> const&) + 5934
6  llc               0x00000001003194be llvm::Pass*
llvm::callDefaultCtor<llvm::LiveVariables>() + 3214
7  llc               0x000000010032b1d5 llvm::PassInfo::~PassInfo() + 117
8  llc               0x000000010058ff2b llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 21947
9  llc               0x00000001005901fb llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 22667
10 llc               0x0000000100590366 llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 23030
11 llc               0x0000000100590a8c llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 24860
12 llc               0x00000001005912fd llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 27021
13 llc               0x0000000100015a81 void
llvm::cl::ValuesClass<int>::apply<llvm::cl::opt<llvm::PassDebugLevel, false,
llvm::cl::parser<llvm::PassDebugLevel> > >(llvm::cl::opt<llvm::PassDebugLevel,
false, llvm::cl::parser<llvm::PassDebugLevel> >&) const + 12257
14 llc               0x0000000100014444 void
llvm::cl::ValuesClass<int>::apply<llvm::cl::opt<llvm::PassDebugLevel, false,
llvm::cl::parser<llvm::PassDebugLevel> > >(llvm::cl::opt<llvm::PassDebugLevel,
false, llvm::cl::parser<llvm::PassDebugLevel> >&) const + 6564
15 llc               0x0000000000000002 void
llvm::cl::ValuesClass<int>::apply<llvm::cl::opt<llvm::PassDebugLevel, false,
llvm::cl::parser<llvm::PassDebugLevel> > >(llvm::cl::opt<llvm::PassDebugLevel,
false, llvm::cl::parser<llvm::PassDebugLevel> >&) const + 4294890850
Stack dump:
0.    Program arguments: llc bugpoint-reduced-simplified.bc 
1.    Running pass 'Function Pass Manager' on module
'bugpoint-reduced-simplified.bc'.
2.    Running pass 'Subregister lowering instruction pass' on function '@mkfs'
Abort

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 00:55:51 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 00:55:51 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7603] New: Wiki - Cannot edit wiki pages
Message-ID: <bug-7603-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7603

           Summary: Wiki - Cannot edit wiki pages
           Product: Website
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: General Website
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: grosser at fim.uni-passau.de
                CC: llvmbugs at cs.uiuc.edu, dberlin at dberlin.org


After saving a changed page I always get this error message:

Set $wgShowExceptionDetails = true; at the bottom of LocalSettings.php to show
detailed debugging information.

Could you have a look?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 00:57:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 00:57:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7602] Cannot emit physreg copy instruction
In-Reply-To: <bug-7602-206@http.llvm.org/bugs/>
References: <bug-7602-206@http.llvm.org/bugs/>
Message-ID: <20100709055702.09E552A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7602

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-09 00:57:01 CDT ---
Fixed in r107957. Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 02:33:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 02:33:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7273] error in sse code with _mm_set1_ps
In-Reply-To: <bug-7273-206@http.llvm.org/bugs/>
References: <bug-7273-206@http.llvm.org/bugs/>
Message-ID: <20100709073340.C7BC02A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7273

Mikko Lehtonen <scoopr at iki.fi> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Mikko Lehtonen <scoopr at iki.fi> 2010-07-09 02:33:40 CDT ---
It would seem that this got fixed at some point (I'm testing with r107124).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 04:05:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 04:05:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7604] New: clang doesn't find gcc-4.4.4's headers
	on GenToo
Message-ID: <bug-7604-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7604

           Summary: clang doesn't find gcc-4.4.4's headers on GenToo
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jarausch at igpm.rwth-aachen.de
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Compiling any C++ program (e.g. hello world) fails because the C++ header files
are not found.

The -v command line option shows that it tries to locate 
/usr/lib/gcc/x86_64-pc-linux-gnu/V.V.V/include

with V.V.V upto 4.4.3

But I have gcc-4.4.4 (or 4.5.0  or 4.6-git) installed,
so clang fails with
fatal error: 'iostream' file not found


This is on a Gentoo system (bleeding edge)

Thanks for looking into it,
Helmut.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 05:04:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 05:04:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7605] New: CreateMalloc() and CreateFree() are not
	orthogonal
Message-ID: <bug-7605-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7605

           Summary: CreateMalloc() and CreateFree() are not orthogonal
           Product: libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: hvdieren at elis.ugent.be
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5196)
 --> (http://llvm.org/bugs/attachment.cgi?id=5196)
Proposed patch

CreateMalloc() takes an optional malloc()-function argument when inserting the
code at the end of a basic block, but not when inserting the code before an
instruction.

CreateFree() returns the code generated when inserting at the end of a basic
block but not when inserting before an instruction.

Better to make them orthogonal; works better for my project because I use a lot
of template function where the insertion point type (Instruction * or
BasicBlock *) is abstracted as a template argument.

Patch makes these changes. Impacts Instructions.h Instructions.cpp
GlobalOpt.cpp

Kind regards,
Hans Vandierendonck

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 06:00:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 06:00:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7606] New: Poor diagnostic for a typo of ','
 instead of '; ' at the end of an statement
Message-ID: <bug-7606-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7606

           Summary: Poor diagnostic for a typo of ',' instead of ';' at
                    the end of an statement
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Keywords: quality-of-implementation
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chandlerc at gmail.com
                CC: llvmbugs at cs.uiuc.edu


This diagnostic could be a *lot* better, at the least poniting at the ','
operator, and the two expression operands instead of the curly on the next
line. Still better would be to suggest a ';', at least if the next token is a
'}'.

% cat t3.cc 
int foo() {
  return 42,
}

% gcc -fsyntax-only t3.cc
t3.cc: In function 'int foo()':
t3.cc:3: error: expected primary-expression before '}' token
t3.cc:3: error: expected ';' before '}' token

% clang -fsyntax-only t3.cc
t3.cc:3:1: error: expected expression
}
^
t3.cc:3:2: error: expected '}'
}
 ^
2 errors generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 06:50:08 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 06:50:08 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7607] New: Assertion failure: Use still stuck
 around after Def is destroyed
Message-ID: <bug-7607-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7607

           Summary: Assertion failure: Use still stuck around after Def is
                    destroyed
           Product: new-bugs
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: wotte at dre.vanderbilt.edu
                CC: llvmbugs at cs.uiuc.edu


Reproduced with latest trunk.   I have been unable to narrow down a test case,
perhaps this will help someone with more expertise to track down the bug. 

I have attached the preprocessed source file.  




clang++ -ftemplate-depth-23  -Wall -Wpointer-arith  -g -pipe   -pipe  
-I/Users/Shared/workspace/DOC_ROOT_clang/ACE -DACE_HAS_EXCEPTIONS
-D__ACE_INLINE__ -I.. -DACE_BUILD_DLL  -c  -o .shobj/Filecache.o Filecache.cpp
While deleting: { [233 x i8], [7 x i8] }* %
Use still stuck around after Def is destroyed:  %arrayidx15 = getelementptr
inbounds %class.ACE_RW_Thread_Mutex* <badref>, i64 %11 ;
<%class.ACE_RW_Thread_Mutex*> [#uses=1]
Assertion failed: (use_empty() && "Uses remain when a value is destroyed!"),
function ~Value, file Value.cpp, line 75.
0  clang             0x00000001010c4aba PrintStackTrace(void*) + 38
1  clang             0x00000001010c4fc8 SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff86acf80a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff86a7450a tiny_malloc_from_free_list + 1196
4  libSystem.B.dylib 0x00007fff86b4aef0 __pthread_markcancel + 0
5  clang             0x0000000101060742 llvm::Value::~Value() + 364
6  clang             0x0000000100ffcf63 llvm::User::~User() + 107
7  clang             0x0000000101004112 llvm::Instruction::~Instruction() + 110
8  clang             0x0000000101004d49
llvm::UnaryInstruction::~UnaryInstruction() + 35
9  clang             0x0000000101016e87 llvm::CastInst::~CastInst() + 39
10 clang             0x000000010101738f llvm::BitCastInst::~BitCastInst() + 35
11 clang             0x0000000100e9f919
llvm::ilist_node_traits<llvm::Instruction>::deleteNode(llvm::Instruction*) + 39
12 clang             0x0000000100e9fd0f llvm::iplist<llvm::Instruction,
llvm::ilist_traits<llvm::Instruction>
>::erase(llvm::ilist_iterator<llvm::Instruction>) + 37
13 clang             0x0000000100f9b94f llvm::iplist<llvm::Instruction,
llvm::ilist_traits<llvm::Instruction>
>::erase(llvm::ilist_iterator<llvm::Instruction>,
llvm::ilist_iterator<llvm::Instruction>) + 35
14 clang             0x0000000100f9b9a8 llvm::iplist<llvm::Instruction,
llvm::ilist_traits<llvm::Instruction> >::clear() + 62
15 clang             0x0000000100f9aaf6 llvm::BasicBlock::~BasicBlock() + 328
16 clang             0x000000010021d934
DestroyCleanup(clang::CodeGen::CodeGenFunction&, llvm::BasicBlock*,
llvm::BasicBlock*) + 272
17 clang             0x000000010021e035
clang::CodeGen::CodeGenFunction::PopCleanupBlock() + 457
18 clang             0x000000010016de72
clang::CodeGen::CodeGenFunction::EmitDestructorBody(llvm::SmallVector<std::pair<clang::VarDecl
const*, clang::QualType>, 16u>&) + 590
19 clang             0x0000000100220a79
clang::CodeGen::CodeGenFunction::GenerateCode(clang::CodeGen::GlobalDecl,
llvm::Function*) + 1285
20 clang             0x0000000100161708
clang::CodeGen::CodeGenModule::EmitCXXDestructor(clang::CXXDestructorDecl
const*, clang::CXXDtorType) + 362
21 clang             0x0000000100229521
clang::CodeGen::CodeGenModule::EmitGlobalDefinition(clang::CodeGen::GlobalDecl)
+ 249
22 clang             0x00000001002296a3
clang::CodeGen::CodeGenModule::EmitGlobal(clang::CodeGen::GlobalDecl) + 273
23 clang             0x0000000100160c40
clang::CodeGen::CodeGenModule::EmitCXXDestructors(clang::CXXDestructorDecl
const*) + 122
24 clang             0x00000001002299a5
clang::CodeGen::CodeGenModule::EmitTopLevelDecl(clang::Decl*) + 645
25 clang             0x0000000100246a6e (anonymous
namespace)::CodeGeneratorImpl::HandleTopLevelDecl(clang::DeclGroupRef) + 76
26 clang             0x000000010021afc4 (anonymous
namespace)::BackendConsumer::HandleTopLevelDecl(clang::DeclGroupRef) + 138
27 clang             0x0000000100257b29 clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 440
28 clang             0x0000000100077fcc
clang::ASTFrontendAction::ExecuteAction() + 256
29 clang             0x000000010021b8e4 clang::CodeGenAction::ExecuteAction() +
826
30 clang             0x00000001000780da clang::FrontendAction::Execute() + 256
31 clang             0x000000010005b4cf
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 709
32 clang             0x0000000100027e3c cc1_main(char const**, char const**,
char const*, void*) + 1762
33 clang             0x000000010002ef05 main + 335
34 clang             0x0000000100026d5c start + 52
Stack dump:
0.    Program arguments: /Users/Shared/workspace/llvm/Debug/bin/clang -cc1
-triple x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free
-main-file-name Filecache.cpp -pic-level 1 -mdisable-fp-elim -masm-verbose
-munwind-tables -target-cpu core2 -g -resource-dir
/Users/Shared/workspace/llvm/Debug/lib/clang/2.8 -D ACE_HAS_EXCEPTIONS -D
__ACE_INLINE__ -D ACE_BUILD_DLL -I /Users/Shared/workspace/DOC_ROOT_clang/ACE
-I .. -Wall -Wpointer-arith -ftemplate-depth 23 -ferror-limit 19
-fmessage-length 202 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o .shobj/Filecache.o -x c++
Filecache.cpp 
1.    Filecache.cpp:240:1: current parser token 'ACE_Filecache_Object'
2.    Filecache.cpp:236:16: LLVM IR generation of declaration
'ACE_Filecache::~ACE_Filecache'
3.    Filecache.cpp:236:16: Generating code for declaration
'ACE_Filecache::~ACE_Filecache'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 08:45:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 08:45:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7608] New: ARMv4 JIT forgets to set the lr register
 when making a indirect function call
Message-ID: <bug-7608-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7608

           Summary: ARMv4 JIT forgets to set the lr register when making a
                    indirect function call
           Product: libraries
           Version: trunk
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: ARM
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: xerxes at zafena.se
                CC: llvmbugs at cs.uiuc.edu


ARMv4 JIT forgets to set lr when making a indirect function call
this makes the called function fail to return properly.

Testcase:
; RUN: llvm-as %s -o %t.bc
; RUN: lli %t.bc > /dev/null

@.LC0 = internal global [12 x i8] c"Hello World\00"        ; <[12 x i8]*>
[#uses=1]

declare i32 @puts(i8*)

define i32 @main() {
        ; program prints Hello World two times using two differnt ways of
calling puts.

        call i32 @puts(i8* getelementptr ([12 x i8]* @.LC0, i64 0, i64 0))

        ; second call to puts, this time using a indirect call
        ; calling puts this way triggers a sigsegv on armv4 when puts return.
        ; change 1085233504 to match the real address to puts 
        call i32 inttoptr (i32 1085233504 to i32 (i8*)*)(i8* getelementptr ([12
x i8]* @.LC0, i64 0, i64 0))   

    ret i32 0
}

Output:
xerxes at xerxes:~/bug.bx$ lli mtest.bc
Hello World
Hello World
0  libLLVM-2.8svn.so 0x407d7d28
Stack dump:
0.    Program arguments: lli mtest.bc 
Segmentation fault

(disassembled jited armv4 code that fail)
0x410b4010:    push    {r4, lr}
0x410b4014:    ldr    r4, [pc, #24]    ; 0x410b4034
0x410b4018:    mov    r0, r4
0x410b401c:    bl    0x41134008
0x410b4020:    ldr    r1, [pc, #16]    ; 0x410b4038
0x410b4024:    mov    r0, r4
0x410b4028:    bx    r1           <--------- calling puts through a register
0x410b402c:    mov    r0, #0    ; 0x0
0x410b4030:    pop    {r4, pc}

testcase passes when enabeling armv5 instructions
xerxes at xerxes:~/bug.bx$ lli -mattr=+v5t mtest.bc
Hello World
Hello World

(disassembled jited armv5 code that works)
0x410b4010:    push    {r4, lr}
0x410b4014:    ldr    r4, [pc, #24]    ; 0x410b4034
0x410b4018:    mov    r0, r4
0x410b401c:    bl    0x41134008
0x410b4020:    ldr    r1, [pc, #16]    ; 0x410b4038 
0x410b4024:    mov    r0, r4
0x410b4028:    blx    r1          <-------- on armv5 the code works because blx
sets lr before the branch.
0x410b402c:    mov    r0, #0    ; 0x0
0x410b4030:    pop    {r4, pc}

(statically compiled armv4 code using llc that works) 
    stmdb    sp!, {r4, lr}
    ldr    r4, .LCPI0_0
    mov    r0, r4
    bl    puts
    ldr    r1, .LCPI0_1
    mov    r0, r4
    mov    lr, pc   <----------- here lr gets set before the bx call.
    bx    r1
    mov    r0, #0
    ldmia    sp!, {r4, pc}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 09:19:16 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 09:19:16 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7609] New: Asm generated by clang causes errors
Message-ID: <bug-7609-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7609

           Summary: Asm generated by clang causes errors
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jpakkane at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5199)
 --> (http://llvm.org/bugs/attachment.cgi?id=5199)
Outputs of -E and -S for the problematic source

I'm building the Cuneiform project (http://launchpad.net/cuneiform-linux) with
the newest svn clang. I get the following error:

[ 33%] Building CXX object
cuneiform_src/Kern/smetric/CMakeFiles/smetric.dir/src/ugol.cpp.o
/tmp/cc-dnDyv6.s: Assembler messages:
/tmp/cc-dnDyv6.s:140: Error: suffix or operands invalid for `movapd'
/tmp/cc-dnDyv6.s:325: Error: suffix or operands invalid for `movapd'
clang: error: assembler (via gcc) command failed with exit code 1 (use -v to
see invocation)

I'll attach the preprocessed source and asm file to this bug but both of the
erroneous lines look like this:

movapd    %xmm0, %st(0)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 10:32:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 10:32:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7610] New: Sibling call optimization broken
Message-ID: <bug-7610-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7610

           Summary: Sibling call optimization broken
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: davidterei at gmail.com
                CC: llvmbugs at cs.uiuc.edu


In LLVM trunk the Sibling call optimization seems to be broken. Specifically it
doesn't catch as many cases as it does in 2.7. The case it no longer catches I
think is calls to unknown functions (e.g function pointers).

Example LLVM stub program:

target datalayout =
"e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:32:32-n8:16:32"
target triple = "i386-pc-linux-gnu"

define cc10 void @__stginit_ZCMain(i32* %Base_Arg, i32* %Sp_Arg, i32* %Hp_Arg,
i32 %R1_Arg) nounwind {
cm1:
  %nm3 = getelementptr i32* %Sp_Arg, i32 1
  %nm9 = load i32* %Sp_Arg
  %nma = inttoptr i32 %nm9 to void (i32*, i32*, i32*, i32)*
  tail call cc10 void %nma(i32* %Base_Arg, i32* %nm3, i32* %Hp_Arg, i32
%R1_Arg) nounwind
  ret void
}

Under 2.8svn this compiles with 'llc -O3' to:

__stginit_ZCMain:
        subl    $4, %esp
        movl    %ebp, %eax
        leal    4(%eax), %ebp
        call    *(%eax)
        addl    $4, %esp
        ret

Under 2.7 this compiles with 'llc -O3' to:

__stginit_ZCMain:
        movl    (%ebp), %eax
        addl    $4, %ebp
        jmpl    *%eax

Cheers,
David

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 11:42:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 11:42:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7611] New: Not emitting covariant thunk
Message-ID: <bug-7611-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7611

           Summary: Not emitting covariant thunk
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dgregor at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Given:

struct A
{
 virtual A* f();
};

struct B: virtual A
{
 virtual A* f();
};

struct C: B
{
 virtual C* f();
};

C* C::f() { return 0; }

clang is omitting the virtual this+virtual return thunk
(_ZTcv0_n12_v0_n16_N1C1fEv), and emitting a virtual this only thunk
(_ZTv0_n12_N1C1fEv) instead.

(thanks to Jason M for reporting this issue!)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 12:10:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 12:10:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7609] Asm generated by clang causes errors
In-Reply-To: <bug-7609-206@http.llvm.org/bugs/>
References: <bug-7609-206@http.llvm.org/bugs/>
Message-ID: <20100709171047.99D802A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7609

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-09 12:10:47 CDT ---
This should be fixed by Bob's revert in r107987.
Please verify.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 12:30:00 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 12:30:00 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7612] New: invalid IR from clang -emit-llvm
Message-ID: <bug-7612-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7612

           Summary: invalid IR from clang -emit-llvm
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: gohman at apple.com
                CC: llvmbugs at cs.uiuc.edu


For this testcase:

int main(void) { return 0; }

clang does this:

$ clang -emit-llvm a.c
Alloca array size must be i32
  %retval = alloca i32, i32 ()* @main, align 2    ; <i32*> [#uses=1]
Instruction does not dominate all uses!
  %retval = alloca i32, i32 ()* @main, align 2    ; <i32*> [#uses=1]
  store i32 0, i32* %retval
Broken module found, compilation aborted!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 12:48:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 12:48:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7612] invalid IR from clang -emit-llvm
In-Reply-To: <bug-7612-206@http.llvm.org/bugs/>
References: <bug-7612-206@http.llvm.org/bugs/>
Message-ID: <20100709174827.31EB52A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7612

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WORKSFORME

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-09 12:48:26 CDT ---
Works with r107993.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 13:43:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 13:43:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7613] New: Win32 build problems
Message-ID: <bug-7613-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7613

           Summary: Win32 build problems
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: John.Thompson.JTSoftware at gmail.com
                CC: llvmbugs at cs.uiuc.edu


(Should there be a "Build" component or other catch-all for things that don't
fit the other categories?)

The cmake-based Visual Studio build is somewhat broken, at least for VS 2008,
which is what I'm using.

With the Visual Studio solution and project files built by cmake for VS2008, if
you do a build or rebuild on ALL_BUILD, the final build step for the libclang
project will fail with a message about not being able to write to Clang.ilk, or
sometimes the clang.pdb file. Then, the c-index-test link will fail with a
bunch of unresolved references, because libclang.lib didn't build properly, or
something.

The work-around is to go back and rebuild libclang alone, and then to a "build"
on the rest.

This leads me to believe there is some sort of dependency problem, or maybe
there is some mixing of different project types.

If someone who knows both cmake and Visual Studio could fix it, that would be
great.

Otherwise I will take a crack at it sometime, though there will be a learning
curve for me about cmake.

-John

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 14:00:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 14:00:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7600] Incorrect return type from
 __sync_val_compare_and_swap builtin
In-Reply-To: <bug-7600-206@http.llvm.org/bugs/>
References: <bug-7600-206@http.llvm.org/bugs/>
Message-ID: <20100709190021.AF9FF2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7600

Chandler Carruth <chandlerc at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Chandler Carruth <chandlerc at gmail.com> 2010-07-09 14:00:21 CDT ---
Fixed in r108002.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 14:44:06 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 14:44:06 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7614] New: undefined reference when using VLA
Message-ID: <bug-7614-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7614

           Summary: undefined reference when using VLA
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: grzegorz.dabrowski at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


$ uname -a
Linux kaliber 2.6.32-23-generic #37-Ubuntu SMP Fri Jun 11 07:54:58 UTC 2010
i686 GNU/Linux

$ clang -v
clang version 2.8 (trunk 107677)
Target: i386-pc-linux-gnu
Thread model: posix

$ cat vla.cpp 
#include <algorithm>

int main() {
  int buffer[std::max(1, 2)];
  return 0;
}

$ clang vla.cpp 
/tmp/cc-nhweyo.o: In function `main':
vla.cpp:(.text+0x31): undefined reference to `int const& std::max<int>(int
const&, int const&)'
collect2: ld returned 1 exit status
clang: error: linker (via gcc) command failed with exit code 1 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 15:04:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 15:04:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7269] -Wno-main is not supported
In-Reply-To: <bug-7269-206@http.llvm.org/bugs/>
References: <bug-7269-206@http.llvm.org/bugs/>
Message-ID: <20100709200419.716EB2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7269

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-09 15:04:19 CDT ---
r108006 should fix the "unknown warning" warnings, and there really isn't any
point to keeping this open for the missing diagnostic.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul  9 15:38:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri,  9 Jul 2010 15:38:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7615] New: GVN doesn't update AliasAnalysis
	correctly
Message-ID: <bug-7615-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7615

           Summary: GVN doesn't update AliasAnalysis correctly
           Product: new-bugs
           Version: 2.7
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: willdtz at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5200)
 --> (http://llvm.org/bugs/attachment.cgi?id=5200)
Test case (for dejagnu/lit style) demonstrating the issue

GVN queries the AliasAnalysis on values that didn't exist when the AA was
executed.  This causes all stateful AA's to break (aka any that actually
compute their results in runOn*).  The "debug-aa" pass in mainline helps catch
these -- attached is a testcase demonstrating an example demonstrating this.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 10 15:25:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 10 Jul 2010 15:25:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7616] New: clang-analyzer not warn about
 set-but-unused aggregates
Message-ID: <bug-7616-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7616

           Summary: clang-analyzer not warn about set-but-unused
                    aggregates
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: dimhen at gmail.com
                CC: llvmbugs at cs.uiuc.edu


with the following test program there are no diagnostics about set-but-unused
union/struct/array

[root at wellesberg clang_errs]# cat aggregate.c
typedef struct {
    int a;
} S;

typedef union {
    int b;
    long c;
} U;

void foo()
{
    S s;
    U u;
    int x[10];

    s.a = 42;
    u.b = 42;
    x[0] = 42;
}
[root at wellesberg clang_errs]# scan-build gcc -c aggregate.c
scan-build: 'clang' executable not found in
'/usr/src/llvm/tools/clang/tools/scan-build/bin'.
scan-build: Using 'clang' from path: /usr/src/llvm/Release+Asserts/bin/clang
ANALYZE: aggregate.c foo
scan-build: Removing directory '/tmp/scan-build-2010-07-11-2' because it
contains no reports.
[root at wellesberg clang_errs]# clang -v
clang version 2.8 (trunk 108050)
Target: i386-pc-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 10 18:57:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 10 Jul 2010 18:57:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7617] New: Message accuracy and change request
Message-ID: <bug-7617-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7617

           Summary: Message accuracy and change request
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: echristo at gmail.com
                CC: llvmbugs at cs.uiuc.edu


>From Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006 at gmx.net>:

I just stumbled across a typo in a header file of a heavily modified
flashrom codebase which looked like this:

int write_sector_jedec_common(struct flashchip *flash, uint8_t *src, int start,
int len, unsigned int mask)
/* m29f400bt.c */
int probe_m29f400bt(struct flashchip *flash);


Note the missing semicolon for the write_sector_jedec_common prototype.

gcc (4.2.1) has this diagnostic:
jedec.c: In function ?write_sector_jedec_common?:
jedec.c:35: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:42: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:62: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:74: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:79: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:96: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:104: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:204: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:234: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:263: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:293: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:322: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:341: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:379: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:399: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:417: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:428: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:441: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:449: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:457: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:465: error: expected ?=?, ?,?, ?;?, ?asm? or ?__attribute__? before ?{?
token
jedec.c:470: error: old-style parameter declarations in prototyped function
definition
jedec.c:470: error: expected ?{? at end of input
cc1: warnings being treated as errors
jedec.c:470: warning: no return statement in function returning non-void


clang (llvm-clang-2.6.99svn97231) has this diagnostic:
In file included from jedec.c:26:
./chipdrivers.h:83:1: error: expected function body after function declarator
int probe_m29f400bt(struct flashchip *flash);
^
1 diagnostic generated.


While the clang diagnostic was a lot more helpful in pinpointing the
actual error because it complained about the header instead of the
source file including it, the diagnostic could be improved a bit by the
following:
- Change the wording from "error: expected function body after function
declarator" to "error: expected function body or semicolon after
function declarator/prototype".
- Print the preceding code line as well because that's where the
semicolon is missing.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 10 23:14:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 10 Jul 2010 23:14:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7603] Wiki - Cannot edit wiki pages
In-Reply-To: <bug-7603-206@http.llvm.org/bugs/>
References: <bug-7603-206@http.llvm.org/bugs/>
Message-ID: <20100711041420.DD62C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7603

Daniel Berlin <dberlin at dberlin.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Daniel Berlin <dberlin at dberlin.org> 2010-07-10 23:14:20 CDT ---
Fixed

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 00:24:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 00:24:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7618] New: miscompile with optimization (derived
	from ffmpeg)
Message-ID: <bug-7618-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7618

           Summary: miscompile with optimization (derived from ffmpeg)
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5204)
 --> (http://llvm.org/bugs/attachment.cgi?id=5204)
Testcase

Testcase attached.  To reproduce, run the following on x86-64 Linux:

$ clang tester.c -O3
$ ./a.out
Segmentation fault
$ clang tester.c
$ ./a.out
0028004800280028002800280028ffffffe0002b0000000000280028002800280028ffffffe0002b
0028002800000000ffffffff00000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000
$ gcc tester.c -O3 -std=c99
$ ./a.out
0028004800280028002800280028ffffffe0002b0000000000280028002800280028ffffffe0002b0028002800000000ffffffff00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

I'm not sure how to go about finding the issue here...

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 11:29:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 11:29:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7619] New: Typo in IRBuilder.h - CreateAShr
	generator incorrect
Message-ID: <bug-7619-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7619

           Summary: Typo in IRBuilder.h - CreateAShr generator incorrect
           Product: libraries
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Support Libraries
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: kv1d at spamex.com
                CC: llvmbugs at cs.uiuc.edu


This is a simple one. One of the CreateAShr cases (the one with an integer
shift count) has a typo that results in a compilation error if its used.

On line 462 in IRBuilder.h I think CreateSShr  should be CreateAShr

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 13:14:52 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 13:14:52 -0500 (CDT)
Subject: [LLVMbugs] [Bug 916] llvm-gcc4 doesn't build on ppc-linux
In-Reply-To: <bug-916-206@http.llvm.org/bugs/>
References: <bug-916-206@http.llvm.org/bugs/>
Message-ID: <20100711181452.07FA72A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=916

Nick Lewycky <nicholas at mxc.ca> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 14:46:11 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 14:46:11 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7619] Typo in IRBuilder.h - CreateAShr generator
	incorrect
In-Reply-To: <bug-7619-206@http.llvm.org/bugs/>
References: <bug-7619-206@http.llvm.org/bugs/>
Message-ID: <20100711194612.0198A2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7619

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 14:46:11 CDT ---
This is already fixed on mainline.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 17:42:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 17:42:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7617] poor recovery with missing ;
	after function prototype
In-Reply-To: <bug-7617-206@http.llvm.org/bugs/>
References: <bug-7617-206@http.llvm.org/bugs/>
Message-ID: <20100711224240.5FD4D2A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7617

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |quality-of-implementation
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|Message accuracy and change |poor recovery with missing
                   |request                     |; after function prototype

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 17:42:39 CDT ---
Fixed in r108105, we now get a fixit hint for the missing ; also.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 18:09:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 18:09:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7264] [patch] implemented address-of-label feature
	in JIT
In-Reply-To: <bug-7264-206@http.llvm.org/bugs/>
References: <bug-7264-206@http.llvm.org/bugs/>
Message-ID: <20100711230945.63BFC2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7264

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #11 from Chris Lattner <clattner at apple.com> 2010-07-11 18:09:44 CDT ---
Duping this to 6744, please follow up there, thanks!

*** This bug has been marked as a duplicate of bug 6744 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 18:18:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 18:18:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7214] Assertion `Reg >= X86::FP0 && Reg <= X86::FP6
 && "Expected FP register!"
In-Reply-To: <bug-7214-206@http.llvm.org/bugs/>
References: <bug-7214-206@http.llvm.org/bugs/>
Message-ID: <20100711231821.334802A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7214

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #6 from Chris Lattner <clattner at apple.com> 2010-07-11 18:18:20 CDT ---
I'm going to dupe this to 2270 since there are other known issues here.  When
that is fixed this testcase should be checked as well.

*** This bug has been marked as a duplicate of bug 2270 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 18:24:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 18:24:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7266] [MC] GMP (5.0.1) fails to build with
 integrated-as: no jrcxz instruction.
In-Reply-To: <bug-7266-206@http.llvm.org/bugs/>
References: <bug-7266-206@http.llvm.org/bugs/>
Message-ID: <20100711232415.B53922A6C131@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7266

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|Failure to build GMP        |[MC] GMP (5.0.1) fails to
                   |(5.0.1) without             |build with integrated-as:
                   |"-no-integrated-as"         |no jrcxz instruction.

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 18:24:15 CDT ---
jrcxz is recognized now.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 18:24:57 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 18:24:57 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7274] Clang gives errors that gcc doesn't when
 compiling output of slice2cpp
In-Reply-To: <bug-7274-206@http.llvm.org/bugs/>
References: <bug-7274-206@http.llvm.org/bugs/>
Message-ID: <20100711232457.C15FB2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7274

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #9 from Chris Lattner <clattner at apple.com> 2010-07-11 18:24:57 CDT ---
Nope.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 18:34:11 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 18:34:11 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7280] Clang emits warning instead of error for
 non-void return without value
In-Reply-To: <bug-7280-206@http.llvm.org/bugs/>
References: <bug-7280-206@http.llvm.org/bugs/>
Message-ID: <20100711233411.34D8C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7280

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 18:34:10 CDT ---
Fixed in r108108

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:04:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:04:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7620] New: MachineLICM should hoist ImplicitDefs.
Message-ID: <bug-7620-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7620

           Summary: MachineLICM should hoist ImplicitDefs.
           Product: libraries
           Version: 1.0
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu


MachineLICM::IsLICMCandidate makes it not hoist implicit defs, however, as the
fixme I added indicates, hoisting an IMPLICIT_DEF may allow hoisting other
instructions that use the impdef as an operand.

Hoisting it should not affect register pressure at all (assuming the RA is
doing the right thing with implicitdef), so just removing that check should be
fine.

However, doing this causes other RA fallout, e.g. test/CodeGen/ARM/spill-q.ll
crashes in the two-address pass.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:10:06 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:10:06 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7404] Unix Path::makeUnique always generates the
	same name
In-Reply-To: <bug-7404-206@http.llvm.org/bugs/>
References: <bug-7404-206@http.llvm.org/bugs/>
Message-ID: <20100712001006.175002A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7404

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 19:10:05 CDT ---
Applied in r108110, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:23:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:23:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7429] ICE on float load from array of bytes
In-Reply-To: <bug-7429-206@http.llvm.org/bugs/>
References: <bug-7429-206@http.llvm.org/bugs/>
Message-ID: <20100712002321.D4CD72A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7429

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|clang assertion failed      |ICE on float load from
                   |                            |array of bytes

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-11 19:23:21 CDT ---
Fixed in r108113, thanks for the reduced testcase!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:47:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:47:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7498] ice: Invalid basic block argument to remove!
In-Reply-To: <bug-7498-206@http.llvm.org/bugs/>
References: <bug-7498-206@http.llvm.org/bugs/>
Message-ID: <20100712004746.B61CC2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7498

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 19:47:46 CDT ---
Fixed in r108114, thanks for the testcase!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:53:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:53:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7604] clang doesn't find gcc-4.4.4's headers on
	GenToo
In-Reply-To: <bug-7604-206@http.llvm.org/bugs/>
References: <bug-7604-206@http.llvm.org/bugs/>
Message-ID: <20100712005342.6A6442A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7604

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |LATER

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 19:53:42 CDT ---
The paths need to be added to lib/Frontend/InitHeaderSearch.cpp, search for
gentoo.  Please send any patches for this to cfe-commits list, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:55:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:55:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7613] Win32 build problems
In-Reply-To: <bug-7613-206@http.llvm.org/bugs/>
References: <bug-7613-206@http.llvm.org/bugs/>
Message-ID: <20100712005546.85BD12A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7613

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-11 19:55:46 CDT ---
I think this got applied.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 19:57:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 19:57:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7605] CreateMalloc() and CreateFree() are not
	consistent
In-Reply-To: <bug-7605-206@http.llvm.org/bugs/>
References: <bug-7605-206@http.llvm.org/bugs/>
Message-ID: <20100712005758.949D22A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7605

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 19:57:58 CDT ---
Applied in r108116, thanks!  Please send patches to llvm-commits instead of
filing a bug.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 20:04:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 20:04:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7592] VRP: (a == 0 || a < 0 || a > 0)
In-Reply-To: <bug-7592-206@http.llvm.org/bugs/>
References: <bug-7592-206@http.llvm.org/bugs/>
Message-ID: <20100712010433.2E0A12A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7592

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
             Blocks|                            |1772
         Resolution|                            |DUPLICATE
            Summary|(a == 0 || a < 0 || a > 0)  |VRP: (a == 0 || a < 0 || a
                   |                            |> 0)

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 20:04:32 CDT ---
Yes, it definitely is.  It's the same sort of value range propagation issue. 
I'm working on this stuff with the LazyValueInfo infrastructure I was building,
but it is stalled for the moment.

*** This bug has been marked as a duplicate of bug 5652 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 20:05:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 20:05:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7574] Assertion `isStackEmpty() && "Stack not empty
 at end of basic block?"' failed
In-Reply-To: <bug-7574-206@http.llvm.org/bugs/>
References: <bug-7574-206@http.llvm.org/bugs/>
Message-ID: <20100712010528.A6DCA2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7574

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 20:05:28 CDT ---


*** This bug has been marked as a duplicate of bug 7375 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 20:07:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 20:07:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7341] Crashing on structure initialization
In-Reply-To: <bug-7341-206@http.llvm.org/bugs/>
References: <bug-7341-206@http.llvm.org/bugs/>
Message-ID: <20100712010717.07A842A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7341

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-11 20:07:16 CDT ---
I fixed this in the last couple weeks.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 20:19:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 20:19:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7311] Invalid instcombine of vector to scalar
In-Reply-To: <bug-7311-206@http.llvm.org/bugs/>
References: <bug-7311-206@http.llvm.org/bugs/>
Message-ID: <20100712011934.DAFFF2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7311

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-11 20:19:34 CDT ---
Fixed in r108117, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 20:48:36 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 20:48:36 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7481] Assertion: `BraceCount > 0' failed on invalid
	code
In-Reply-To: <bug-7481-206@http.llvm.org/bugs/>
References: <bug-7481-206@http.llvm.org/bugs/>
Message-ID: <20100712014836.CDCEC2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7481

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 20:48:36 CDT ---
Fixed in r108119

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 20:49:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 20:49:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7427] Assertion (ParenCount > 0),
	function MatchRHSPunctuation
In-Reply-To: <bug-7427-206@http.llvm.org/bugs/>
References: <bug-7427-206@http.llvm.org/bugs/>
Message-ID: <20100712014914.C91E32A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7427

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 20:49:14 CDT ---
Fixed in r108119

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 21:37:10 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 21:37:10 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7074] Small crashing example
In-Reply-To: <bug-7074-206@http.llvm.org/bugs/>
References: <bug-7074-206@http.llvm.org/bugs/>
Message-ID: <20100712023710.4E2852A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7074

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 21:37:09 CDT ---
This is now correctly rejected:


t.cc:1:11: error: use of undeclared identifier 'a'
 a(struct a::a
          ^
t.cc:1:15: error: expected ')'
 a(struct a::a
              ^
t.cc:1:3: note: to match this '('
 a(struct a::a
  ^
t.cc:1:15: error: expected function body after function declarator
 a(struct a::a
              ^
3 errors generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 21:55:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 21:55:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7075] Crashing example: struct{a():}}{
In-Reply-To: <bug-7075-206@http.llvm.org/bugs/>
References: <bug-7075-206@http.llvm.org/bugs/>
Message-ID: <20100712025527.6BA552A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7075

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 21:55:26 CDT ---
This is now correctly rejected:


t2.cc:1:8: error: C++ requires a type specifier for all declarations
struct{a():}}{
       ^
t2.cc:1:12: error: expected class member or base class name
struct{a():}}{
           ^
t2.cc:1:12: error: expected '{' or ','
t2.cc:1:11: error: only constructors take base initializers
struct{a():}}{
          ^
t2.cc:1:15: error: expected '}'
struct{a():}}{
              ^
t2.cc:1:14: error: expected ';' after struct
struct{a():}}{
             ^
             ;
t2.cc:1:1: error: anonymous unions at namespace or global scope must be
declared 'static'
struct{a():}}{
^
t2.cc:1:8: error: functions cannot be declared in an anonymous struct
struct{a():}}{
       ^
t2.cc:1:1: error: anonymous structs and classes must be class members
struct{a():}}{
^
9 errors generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 21:56:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 21:56:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7375] [X87 FPStack] Assertion failed:
 (isStackEmpty() && "Stack not empty at end of basic block?")
In-Reply-To: <bug-7375-206@http.llvm.org/bugs/>
References: <bug-7375-206@http.llvm.org/bugs/>
Message-ID: <20100712025640.88B742A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7375

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #11 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-11 21:56:40 CDT ---
The attached bitcode (assert18.bc) compiles now.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 23:25:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 23:25:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7072] Small crashing example:   ()(
In-Reply-To: <bug-7072-206@http.llvm.org/bugs/>
References: <bug-7072-206@http.llvm.org/bugs/>
Message-ID: <20100712042553.4E4432A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7072

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-11 23:25:52 CDT ---
Actually, I figured it out.  Fixed in r108124, please review Argiris.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 23:35:07 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 23:35:07 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6944] Calling a non-sret function with an sret
 convention corrupts the stack
In-Reply-To: <bug-6944-206@http.llvm.org/bugs/>
References: <bug-6944-206@http.llvm.org/bugs/>
Message-ID: <20100712043507.2B6B72A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6944

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-11 23:35:06 CDT ---
This is undefined behavior.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 23:35:51 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 23:35:51 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6931] Poor performance of fibonacci example code
In-Reply-To: <bug-6931-206@http.llvm.org/bugs/>
References: <bug-6931-206@http.llvm.org/bugs/>
Message-ID: <20100712043551.39A5C2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6931

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-11 23:35:50 CDT ---
This looks fixed on mainline, at least with clang.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 23:37:30 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 23:37:30 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6929] possible wrong code bug
In-Reply-To: <bug-6929-206@http.llvm.org/bugs/>
References: <bug-6929-206@http.llvm.org/bugs/>
Message-ID: <20100712043730.7667C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6929

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |clattner at apple.com
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 23:37:30 CDT ---
This works for me with r108123 at -O[0123s].  It is possible that this is
platform specific to repro, if so, please reopen.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 23:39:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 23:39:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6909] Path::isRootDirectory not defined for unix
In-Reply-To: <bug-6909-206@http.llvm.org/bugs/>
References: <bug-6909-206@http.llvm.org/bugs/>
Message-ID: <20100712043922.093AF2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6909

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-11 23:39:21 CDT ---
You're right, it is dead.  Zapped in r108125, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 11 23:49:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 11 Jul 2010 23:49:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5438] Small example where gcc produces better code
In-Reply-To: <bug-5438-206@http.llvm.org/bugs/>
References: <bug-5438-206@http.llvm.org/bugs/>
Message-ID: <20100712044934.F017B2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5438

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Chris Lattner <clattner at apple.com> 2010-07-11 23:49:34 CDT ---
Fixed in clang:


_bar:                                   ## @bar
## BB#0:                                ## %entry
    movl    8(%esp), %eax
    xorl    4(%esp), %eax
    shrl    $31, %eax
    xorl    $1, %eax
    ret

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 05:52:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 05:52:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7622] New: infinite loop on invalid code
Message-ID: <bug-7622-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7622

           Summary: infinite loop on invalid code
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5213)
 --> (http://llvm.org/bugs/attachment.cgi?id=5213)
Code leading to infinite loop in compiler

The attached code causes clang++ to enter an infinite loop (or at least, it
takes longer than 5 minutes to compile on my machine).

Compiler version:

clang version 2.8 (trunk 108130)
Target: x86_64-unknown-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 06:31:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 06:31:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7623] New: 'Invalid accessor called' generated by
 'template<>template<>a; '
Message-ID: <bug-7623-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7623

           Summary: 'Invalid accessor called' generated by
                    'template<>template<>a;'
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


As in title. clang++, given the code:

template<>template<>a;

Produces the backtrace:

t.bashed.9366.cc:1:21: error: C++ requires a type specifier for all
declarations
template<>template<>a;
                    ^
t.bashed.9366.cc:1:1: warning: extraneous template parameter list in template
      specialization
template<>template<>a;
^~~~~~~~~~
t.bashed.9366.cc:1:11: error: extraneous 'template<>' in declaration of
variable
      'a'
template<>template<>a;
          ^~~~~~~~~~
Assertion failed: (is<T>() && "Invalid accessor called"), function get, file
/Users/caj/clang/llvm/include/llvm/ADT/PointerUnion.h, line 99.
0  clang             0x00000001010ce676 PrintStackTrace(void*) + 38
1  clang             0x00000001010ceb84 SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff8398c35a _sigtramp + 26
3  libSystem.B.dylib 0x0000000104700000 _sigtramp + 2161589440
4  libSystem.B.dylib 0x00007fff83a079b4 __pthread_markcancel + 0
5  clang             0x00000001005a8177 clang::DeclaratorDecl::ExtInfo*
llvm::PointerUnion<clang::TypeSourceInfo*,
clang::DeclaratorDecl::ExtInfo*>::get<clang::DeclaratorDecl::ExtInfo*>() const
+ 75
6  clang             0x00000001000bfecd clang::DeclaratorDecl::getExtInfo() +
25
7  clang             0x00000001002c06a8
clang::DeclaratorDecl::setTemplateParameterListsInfo(clang::ASTContext&,
unsigned int, clang::TemplateParameterList**) + 32
8  clang             0x00000001002b2eac
clang::Sema::ActOnVariableDeclarator(clang::Scope*, clang::Declarator&,
clang::DeclContext*, clang::QualType, clang::TypeSourceInfo*,
clang::LookupResult&,
clang::ASTMultiPtr<&(clang::ActionBase::DeleteTemplateParams(void*))>, bool&) +
1730
9  clang             0x00000001002b705d
clang::Sema::HandleDeclarator(clang::Scope*, clang::Declarator&,
clang::ASTMultiPtr<&(clang::ActionBase::DeleteTemplateParams(void*))>, bool) +
2881
10 clang             0x00000001003c1890
clang::Sema::ActOnTemplateDeclarator(clang::Scope*,
clang::ASTMultiPtr<&(clang::ActionBase::DeleteTemplateParams(void*))>,
clang::Declarator&) + 90
11 clang             0x000000010068501d
clang::Parser::ParseDeclarationAfterDeclarator(clang::Declarator&,
clang::Parser::ParsedTemplateInfo const&) + 677
12 clang             0x00000001006bd9ac
clang::Parser::ParseSingleDeclarationAfterTemplate(unsigned int,
clang::Parser::ParsedTemplateInfo const&, clang::SourceLocation&,
clang::AccessSpecifier) + 696
13 clang             0x00000001006be07e
clang::Parser::ParseTemplateDeclarationOrSpecialization(unsigned int,
clang::SourceLocation&, clang::AccessSpecifier) + 962
14 clang             0x00000001006be15a
clang::Parser::ParseDeclarationStartingWithTemplate(unsigned int,
clang::SourceLocation&, clang::AccessSpecifier) + 158
15 clang             0x00000001006860ec
clang::Parser::ParseDeclaration(unsigned int, clang::SourceLocation&,
clang::CXX0XAttributeList) + 368
16 clang             0x00000001006c4611
clang::Parser::ParseExternalDeclaration(clang::CXX0XAttributeList) + 2007
17 clang             0x00000001006c483f
clang::Parser::ParseTopLevelDecl(clang::OpaquePtr<1>&) + 247
18 clang             0x000000010025c5ad clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 456
19 clang             0x000000010007844a
clang::ASTFrontendAction::ExecuteAction() + 254
20 clang             0x000000010021fa70 clang::CodeGenAction::ExecuteAction() +
826
21 clang             0x0000000100078558 clang::FrontendAction::Execute() + 256
22 clang             0x000000010005b720
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 710
23 clang             0x0000000100027f28 cc1_main(char const**, char const**,
char const*, void*) + 1762
24 clang             0x000000010002ef81 main + 335
25 clang             0x0000000100026e48 start + 52
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free -main-file-name
t.bashed.9366.cc -pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables
-target-cpu core2 -resource-dir /usr/local/lib/clang/2.8 -ferror-limit 19
-fmessage-length 80 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/M8/M86fCFVgGQ0r-V2oY5lLC++++TI/-Tmp-/cc-0nmhBP.o -x c++
t.bashed.9366.cc 
1.    t.bashed.9366.cc:1:22: current parser token ';'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 06:31:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 06:31:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7624] New: incorrect x87 code generated for
	comparison
Message-ID: <bug-7624-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7624

           Summary: incorrect x87 code generated for comparison
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: jay.foad at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5214)
 --> (http://llvm.org/bugs/attachment.cgi?id=5214)
test case

With the attached test case I get:

$ clang -m32 -mno-sse -o w w.c
$ ./w
[should return after one second, but in fact loops forever]

The code generated for "while (get() < d)" looks like this:

 8048440:       e8 9b ff ff ff          call   80483e0 <get>
 8048445:       dd 45 f0                fldl   -0x10(%ebp)
 8048448:       da e9                   fucompp 
 804844a:       76 0b                   jbe    8048457 <main+0x37>

This won't work because fucompp leaves the result of the comparison in the FPU
flags, but "jbe" tests the CPU flags.

I'm using Clang and LLVM built from trunk this morning, on Ubuntu 10.4 x86_64.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 06:37:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 06:37:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7625] New: 'ResultKind == NotFound &&
 Decls.empty()' assertion on invalid code
Message-ID: <bug-7625-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7625

           Summary: 'ResultKind == NotFound && Decls.empty()' assertion on
                    invalid code
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The code:

template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p

Generates the output:

t.bashed.2245.cc:1:47: error: C++ requires a type specifier for all
declarations
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                              ^
t.bashed.2245.cc:1:53: error: expected '}'
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                                    ^
t.bashed.2245.cc:1:46: note: to match this '{'
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                             ^
t.bashed.2245.cc:1:53: error: expected ';' after struct
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                                    ^
                                                    ;
t.bashed.2245.cc:1:47: error: functions cannot be declared in an anonymous
struct
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                              ^
t.bashed.2245.cc:1:53: error: expected '}'
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                                    ^
t.bashed.2245.cc:1:35: note: to match this '{'
template<typename _Tp>struct a:_Tp{struct:_Tp{aa(){p
                                  ^
Assertion failed: (ResultKind == NotFound && Decls.empty()), function
setNotFoundInCurrentInstantiation, file Lookup.h, line 322.
0  clang             0x00000001010ce676 PrintStackTrace(void*) + 38
1  clang             0x00000001010ceb84 SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff8398c35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff839310aa tiny_malloc_from_free_list + 1196
4  libSystem.B.dylib 0x00007fff83a079b4 __pthread_markcancel + 0
5  clang             0x000000010037faeb
clang::LookupResult::setNotFoundInCurrentInstantiation() + 95
6  clang             0x000000010037ac0e
clang::Sema::LookupQualifiedName(clang::LookupResult&, clang::DeclContext*,
bool) + 698
7  clang             0x000000010032b976
clang::Sema::DiagnoseEmptyLookup(clang::Scope*, clang::CXXScopeSpec&,
clang::LookupResult&, clang::Sema::CorrectTypoContext) + 272
8  clang             0x000000010033e6d4
clang::Sema::ActOnIdExpression(clang::Scope*, clang::CXXScopeSpec&,
clang::UnqualifiedId&, bool, bool) + 1568
9  clang             0x0000000100696d3d
clang::Parser::ParseCastExpression(bool, bool, bool&, void*) + 3355
10 clang             0x0000000100698c21
clang::Parser::ParseCastExpression(bool, bool, void*) + 97
11 clang             0x00000001006934a7
clang::Parser::ParseAssignmentExpression() + 241
12 clang             0x0000000100693a2e clang::Parser::ParseExpression() + 42
13 clang             0x00000001006b4c10
clang::Parser::ParseStatementOrDeclaration(bool) + 1662
14 clang             0x00000001006b9172
clang::Parser::ParseCompoundStatementBody(bool) + 220
15 clang             0x00000001006b979d
clang::Parser::ParseFunctionStatementBody(clang::OpaquePtr<0>) + 193
16 clang             0x0000000100678f97
clang::Parser::ParseLexedMethodDefs(clang::Parser::ParsingClass&) + 1373
17 clang             0x00000001006790f3
clang::Parser::ParseLexedMethodDefs(clang::Parser::ParsingClass&) + 1721
18 clang             0x000000010068da73
clang::Parser::ParseCXXMemberSpecification(clang::SourceLocation, unsigned int,
clang::OpaquePtr<0>) + 1791
19 clang             0x000000010068efba
clang::Parser::ParseClassSpecifier(clang::tok::TokenKind,
clang::SourceLocation, clang::DeclSpec&, clang::Parser::ParsedTemplateInfo
const&, clang::AccessSpecifier, bool) + 5226
20 clang             0x0000000100681fb5
clang::Parser::ParseDeclarationSpecifiers(clang::DeclSpec&,
clang::Parser::ParsedTemplateInfo const&, clang::AccessSpecifier,
clang::Parser::DeclSpecContext) + 6301
21 clang             0x00000001006bd84b
clang::Parser::ParseSingleDeclarationAfterTemplate(unsigned int,
clang::Parser::ParsedTemplateInfo const&, clang::SourceLocation&,
clang::AccessSpecifier) + 343
22 clang             0x00000001006be07e
clang::Parser::ParseTemplateDeclarationOrSpecialization(unsigned int,
clang::SourceLocation&, clang::AccessSpecifier) + 962
23 clang             0x00000001006be15a
clang::Parser::ParseDeclarationStartingWithTemplate(unsigned int,
clang::SourceLocation&, clang::AccessSpecifier) + 158
24 clang             0x00000001006860ec
clang::Parser::ParseDeclaration(unsigned int, clang::SourceLocation&,
clang::CXX0XAttributeList) + 368
25 clang             0x00000001006c4611
clang::Parser::ParseExternalDeclaration(clang::CXX0XAttributeList) + 2007
26 clang             0x00000001006c483f
clang::Parser::ParseTopLevelDecl(clang::OpaquePtr<1>&) + 247
27 clang             0x000000010025c5ad clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 456
28 clang             0x000000010007844a
clang::ASTFrontendAction::ExecuteAction() + 254
29 clang             0x000000010021fa70 clang::CodeGenAction::ExecuteAction() +
826
30 clang             0x0000000100078558 clang::FrontendAction::Execute() + 256
31 clang             0x000000010005b720
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 710
32 clang             0x0000000100027f28 cc1_main(char const**, char const**,
char const*, void*) + 1762
33 clang             0x000000010002ef81 main + 335
34 clang             0x0000000100026e48 start + 52
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free -main-file-name
t.bashed.2245.cc -pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables
-target-cpu core2 -resource-dir /usr/local/lib/clang/2.8 -ferror-limit 19
-fmessage-length 113 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/M8/M86fCFVgGQ0r-V2oY5lLC++++TI/-Tmp-/cc-xOU4nf.o -x c++
t.bashed.2245.cc 
1.    <eof> parser at end of file
2.    t.bashed.2245.cc:1:23: parsing struct/union/class body 'a'
3.    t.bashed.2245.cc:1:51: parsing function body 'a::<anonymous struct>::aa'
4.    t.bashed.2245.cc:1:51: in compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 06:55:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 06:55:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6773] [instcombine,
	sha] Add Instruction Fold for Masked Merge
In-Reply-To: <bug-6773-206@http.llvm.org/bugs/>
References: <bug-6773-206@http.llvm.org/bugs/>
Message-ID: <20100712115515.B922D2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6773

Benjamin Kramer <benny.kra at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Benjamin Kramer <benny.kra at gmail.com> 2010-07-12 06:55:15 CDT ---
Thanks for the nice testcase, optimization implemented in r108136.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 09:46:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 09:46:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6773] [instcombine,
	sha] Add Instruction Fold for Masked Merge
In-Reply-To: <bug-6773-206@http.llvm.org/bugs/>
References: <bug-6773-206@http.llvm.org/bugs/>
Message-ID: <20100712144622.9E4DA2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6773

Benjamin Kramer <benny.kra at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #2 from Benjamin Kramer <benny.kra at gmail.com> 2010-07-12 09:46:22 CDT ---
And reverted in r108141 ? I hope to fix this later today.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 10:49:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 10:49:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7624] incorrect x87 code generated for comparison
In-Reply-To: <bug-7624-206@http.llvm.org/bugs/>
References: <bug-7624-206@http.llvm.org/bugs/>
Message-ID: <20100712154926.8C49E2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7624

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Dan Gohman <gohman at apple.com> 2010-07-12 10:49:26 CDT ---
This specific issue is now fixed in r108149. But it's certainly tempting.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 10:52:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 10:52:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7221] [ARM JIT]
 emitVFPLoadStoreMultipleInstructio() does not handle s* registers
In-Reply-To: <bug-7221-206@http.llvm.org/bugs/>
References: <bug-7221-206@http.llvm.org/bugs/>
Message-ID: <20100712155209.265BA2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7221

Shih-wei Liao <sliao at google.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Shih-wei Liao <sliao at google.com> 2010-07-12 10:52:08 CDT ---
I verified that r104652 has fixed this problem. Thanks.

http://llvm.org/viewvc/llvm-project?view=rev&revision=104652

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 10:55:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 10:55:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7222] [ARM JIT] Bitfield's "clear" and "insert" are
 not implemented
In-Reply-To: <bug-7222-206@http.llvm.org/bugs/>
References: <bug-7222-206@http.llvm.org/bugs/>
Message-ID: <20100712155545.9AF4F2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7222

Shih-wei Liao <sliao at google.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Shih-wei Liao <sliao at google.com> 2010-07-12 10:55:45 CDT ---
I verified that r104653 and r104670 have fixed the problem. Thanks.

http://llvm.org/viewvc/llvm-project?view=rev&revision=104653
http://llvm.org/viewvc/llvm-project?view=rev&revision=104670

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 11:44:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 11:44:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7529] Missing memoperands prevents ARM load/store
 folding in a few cases
In-Reply-To: <bug-7529-206@http.llvm.org/bugs/>
References: <bug-7529-206@http.llvm.org/bugs/>
Message-ID: <20100712164426.0A7F52A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7529

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-12 11:44:25 CDT ---
I think that is al

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 18:12:31 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 18:12:31 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7470] clang c++ incorrectly accepts const qualifier
 on function pointer
In-Reply-To: <bug-7470-206@http.llvm.org/bugs/>
References: <bug-7470-206@http.llvm.org/bugs/>
Message-ID: <20100712231231.8BBF52A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7470

Sebastian Redl <sebastian.redl at getdesigned.at> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sebastian.redl at getdesigned.
                   |                            |at
         Resolution|                            |FIXED

--- Comment #1 from Sebastian Redl <sebastian.redl at getdesigned.at> 2010-07-12 18:12:31 CDT ---
Fixed in r108209.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 19:11:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 19:11:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7622] infinite loop on invalid code
In-Reply-To: <bug-7622-206@http.llvm.org/bugs/>
References: <bug-7622-206@http.llvm.org/bugs/>
Message-ID: <20100713001148.6FFE72A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7622

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-12 19:11:48 CDT ---
Fixed in r108217.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 12 23:39:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 12 Jul 2010 23:39:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7633] New: failure to report leak due to
 presentMoviePlayerViewControllerAnimated
Message-ID: <bug-7633-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7633

           Summary: failure to report leak due to
                    presentMoviePlayerViewControllerAnimated
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: matt at tidbits.com
                CC: llvmbugs at cs.uiuc.edu


The following is a leak (Cocoa). The static analyzer fails to report it. But if
you comment out the second line, it *does* report it. This is wrong; nothing
about the second line mitigates the leak, so the analyzer should still report
it even if the second line is present.

MPMoviePlayerViewController* mpvc = [[MPMoviePlayerViewController alloc]
initWithContentURL:urlurl];
[self presentMoviePlayerViewControllerAnimated:mpvc];

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 01:25:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 01:25:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7466] clang c++ missing error/pedwarn for "enum { };
	"
In-Reply-To: <bug-7466-206@http.llvm.org/bugs/>
References: <bug-7466-206@http.llvm.org/bugs/>
Message-ID: <20100713062515.89C1A2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7466

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-13 01:25:15 CDT ---
Fixed in Clang r108231.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 01:37:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 01:37:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7462] clang c++ crash with typedef to operator name
In-Reply-To: <bug-7462-206@http.llvm.org/bugs/>
References: <bug-7462-206@http.llvm.org/bugs/>
Message-ID: <20100713063712.9D6A32A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7462

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-13 01:37:12 CDT ---
Fixed in r108233.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 03:37:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 03:37:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7460] Bug in enum-to-value resolution?
In-Reply-To: <bug-7460-206@http.llvm.org/bugs/>
References: <bug-7460-206@http.llvm.org/bugs/>
Message-ID: <20100713083741.763BE2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7460

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-13 03:37:41 CDT ---
Fixed in Clang r108235.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 03:58:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 03:58:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7598] Spurious error involving enums,
 conditional expressions, and const return values
In-Reply-To: <bug-7598-206@http.llvm.org/bugs/>
References: <bug-7598-206@http.llvm.org/bugs/>
Message-ID: <20100713085832.6941F2A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7598

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |dgregor at apple.com
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-13 03:58:31 CDT ---
Fixed in Clang r108234.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 06:48:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 06:48:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7635] New: Inline assembly in C++ source fails to
	link
Message-ID: <bug-7635-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7635

           Summary: Inline assembly in C++ source fails to link
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bjorn at topel.se
                CC: llvmbugs at cs.uiuc.edu


The following snippet generates the following linker (ld) error:

ld: in /var/folders/Eb/Ebu2gEwcHfeuStO9hJ4SQk+++TI/-Tmp-/cc-u4eXeR.o, in
section __TEXT,__text reloc 0: unsupported r_length=0 for scattered vanilla
reloc
clang: error: linker command failed with exit code 1 (use -v to see invocation)

which works for g++.

-arch i386

-- 
//in.cc
typedef void (*FuncPtr)(void);

extern "C" {

void j(void)
{
}

} // extern "C"

int main()
{
    unsigned int tmp = 655341895;

    asm("neg %0\n\t"
        "addl $(_j + 655341895),%0\n"
        : "+r" (tmp));

    ((FuncPtr)(tmp))();

    return 0;
}


Note that the generated .s file generated by clang links using g++, i.e.
$ clang++ -arch i386 -S -o t.s in.cc
$ g++ -arch i386 t.s # OK
$ clang++ -arch i386 t.s # NOK

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 10:44:52 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 10:44:52 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7506] Case in which TailRecursionElimination could
	do better
In-Reply-To: <bug-7506-206@http.llvm.org/bugs/>
References: <bug-7506-206@http.llvm.org/bugs/>
Message-ID: <20100713154452.DBA4D2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7506

Duncan Sands <baldrick at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Duncan Sands <baldrick at free.fr> 2010-07-13 10:44:52 CDT ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20100712/104199.html

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 14:23:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 14:23:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7463] clang c++ bad template deduction for argument
 of type "const int"
In-Reply-To: <bug-7463-206@http.llvm.org/bugs/>
References: <bug-7463-206@http.llvm.org/bugs/>
Message-ID: <20100713192312.4BA4B2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7463

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-13 14:23:11 CDT ---
Fixed in r 108253.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 14:28:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 14:28:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7258] Bogus throw specification error
In-Reply-To: <bug-7258-206@http.llvm.org/bugs/>
References: <bug-7258-206@http.llvm.org/bugs/>
Message-ID: <20100713192809.E1C292A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7258

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-13 14:28:09 CDT ---
This was fixed in Clang r107385.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 14:33:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 14:33:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7246] clang doesn't handle constructor exception
 handlers properly
In-Reply-To: <bug-7246-206@http.llvm.org/bugs/>
References: <bug-7246-206@http.llvm.org/bugs/>
Message-ID: <20100713193338.713162A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7246

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |rjmccall at apple.com
         Resolution|                            |FIXED

--- Comment #2 from John McCall <rjmccall at apple.com> 2010-07-13 14:33:37 CDT ---
This was a known problem;  it's since been fixed, in r107757.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 14:36:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 14:36:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7194] Possible bug when combining functions
 returning const enum and the conditional operator?
In-Reply-To: <bug-7194-206@http.llvm.org/bugs/>
References: <bug-7194-206@http.llvm.org/bugs/>
Message-ID: <20100713193622.29A912A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7194

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-13 14:36:21 CDT ---
Fixed in Clang r 108234.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 14:39:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 14:39:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7127] clang confused by double throw
In-Reply-To: <bug-7127-206@http.llvm.org/bugs/>
References: <bug-7127-206@http.llvm.org/bugs/>
Message-ID: <20100713193919.347962A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7127

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #8 from John McCall <rjmccall at apple.com> 2010-07-13 14:39:18 CDT ---
This should be fixed, but I can't test it because I don't have a relevant
machine.  Please re-verify.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 14:47:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 14:47:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7253] LLVM creates invalid code on GeodeLX
In-Reply-To: <bug-7253-206@http.llvm.org/bugs/>
References: <bug-7253-206@http.llvm.org/bugs/>
Message-ID: <20100713194753.16B502A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7253

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-13 14:47:51 CDT ---
Clang defaults to assuming that you have SSE2.  If you don't want this, pass
-mcpu=pentium or -mno-sse etc.  It would be nice for clang to autodetect the
host and default to whatever features it has, but that is tracked elsewhere.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 15:06:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 15:06:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7471] [MC] Clang doesn't complain when using 32-bit
 only inline asm statements in with -arch x86_64
In-Reply-To: <bug-7471-206@http.llvm.org/bugs/>
References: <bug-7471-206@http.llvm.org/bugs/>
Message-ID: <20100713200644.5DF3E2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7471

Kevin Enderby <enderby at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Kevin Enderby <enderby at apple.com> 2010-07-13 15:06:44 CDT ---
I committed the fix in r108265.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 15:18:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 15:18:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7528] [inline asm] spandsp compilation failure
In-Reply-To: <bug-7528-206@http.llvm.org/bugs/>
References: <bug-7528-206@http.llvm.org/bugs/>
Message-ID: <20100713201809.847C52A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7528

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #13 from Dale Johannesen <dalej at apple.com> 2010-07-13 15:18:08 CDT ---
Really fixed in 108268.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 16:12:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 16:12:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7636] New: thumb2 ldmia.w used as a return doesn't
 write back SP
Message-ID: <bug-7636-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7636

           Summary: thumb2 ldmia.w used as a return doesn't write back SP
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: ARM
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: bagel99 at gmail.com
                CC: llvmbugs at cs.uiuc.edu


When a ldmia.w is generated that includes the PC in the register list,
it doesn't update the stack pointer, e.g.:

    ldmia.w    sp, {r4, r5, r6, r8, pc}
should be:
    ldmia.w    sp!, {r4, r5, r6, r8, pc}

Strangely, in case where the register list contains LR instead of PC, the write
back is done.

A test case has to be sufficiently complicated to use ldmia.w rather than pop.
I have not attached one.  If one is needed, let me know.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 18:18:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 18:18:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6437] Assertion when casting 'struct in_addr const'
 to 'uint32_t&'
In-Reply-To: <bug-6437-206@http.llvm.org/bugs/>
References: <bug-6437-206@http.llvm.org/bugs/>
Message-ID: <20100713231812.15F722A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6437

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Douglas Gregor <dgregor at apple.com> 2010-07-13 18:18:10 CDT ---
Fixed in r108294.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 18:20:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 18:20:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7593] codegen ICE on cast to reference
In-Reply-To: <bug-7593-206@http.llvm.org/bugs/>
References: <bug-7593-206@http.llvm.org/bugs/>
Message-ID: <20100713232039.B2CF42A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7593

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #8 from Douglas Gregor <dgregor at apple.com> 2010-07-13 18:20:39 CDT ---
Fixed in Clang r108294.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 18:21:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 18:21:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7344] Clang assertion failure when casting void* to
 unsigned long
In-Reply-To: <bug-7344-206@http.llvm.org/bugs/>
References: <bug-7344-206@http.llvm.org/bugs/>
Message-ID: <20100713232118.4F5B52A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7344

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Douglas Gregor <dgregor at apple.com> 2010-07-13 18:21:17 CDT ---
Fixed in Clang r108294.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 19:43:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 19:43:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7637] New: Adding a GVN pass via LLVMAddGVNPass
 triggers an assert.
Message-ID: <bug-7637-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7637

           Summary: Adding a GVN pass via LLVMAddGVNPass triggers an
                    assert.
           Product: new-bugs
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: blood.of.life at gmail.com
                CC: llvmbugs at cs.uiuc.edu


(Apologies for using D here, I can't get C and LLVM playing nicely.)

import llvm.c.Core;
import llvm.c.transforms.Scalar;

void main()
{
    LLVMModuleRef _module = LLVMModuleCreateWithNameInContext("module",
LLVMGetGlobalContext());
    LLVMPassManagerRef fpm = LLVMCreateFunctionPassManagerForModule(_module);
    LLVMAddInstructionCombiningPass(fpm);  /* Works. */
    LLVMAddGVNPass(fpm);  /* Triggers an assert. */
}

Attempting to run the executable results in this:

a.out: /home/bernard/src/llvm-2.7/include/llvm/PassAnalysisSupport.h:53:
llvm::AnalysisUsage& llvm::AnalysisUsage::addRequiredID(const llvm::PassInfo*):
Assertion `ID && "Pass class not registered!"' failed.
Aborted

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 19:57:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 19:57:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7637] On 32 bit build running on 64b bit OS,
 Adding a GVN pass via LLVMAddGVNPass triggers an assert.
In-Reply-To: <bug-7637-206@http.llvm.org/bugs/>
References: <bug-7637-206@http.llvm.org/bugs/>
Message-ID: <20100714005754.79C302A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7637

Bernard Helyer <blood.of.life at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #1 from Bernard Helyer <blood.of.life at gmail.com> 2010-07-13 19:57:54 CDT ---
This seems to be more complicated than I first thought. I shall experiment
more.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 13 20:22:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 13 Jul 2010 20:22:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7620] MachineLICM should hoist ImplicitDefs.
In-Reply-To: <bug-7620-206@http.llvm.org/bugs/>
References: <bug-7620-206@http.llvm.org/bugs/>
Message-ID: <20100714012233.5B1E12A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7620

Evan Cheng <evan.cheng at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #9 from Evan Cheng <evan.cheng at apple.com> 2010-07-13 20:22:32 CDT ---
Fixed: 108304

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 00:21:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 00:21:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7542] liveintervals valgrind issue + assert fail
In-Reply-To: <bug-7542-206@http.llvm.org/bugs/>
References: <bug-7542-206@http.llvm.org/bugs/>
Message-ID: <20100714052127.75CEE2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7542

John Regehr <regehr at cs.utah.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 00:31:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 00:31:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7638] New: "for" loop prevents leak diagnostics
Message-ID: <bug-7638-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7638

           Summary: "for" loop prevents leak diagnostics
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: tjw at omnigroup.com
                CC: llvmbugs at cs.uiuc.edu


Building the following test.m file with:

/Volumes/Space/Users/bungi/Source/LLVM/llvm/Release+Asserts/bin/clang -x
objective-c --analyze test.m

??I get no diagnostics. But, there are four locals leaked (newOutlineNodes,
forecastOutlineNodes, formatter and shortFormatter).

Commenting out the for loop (leaving the body, just not the two lines
comprising the start and closing curly of the loop) allows the leaks to be
found.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 00:56:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 00:56:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5204] if-conversion of tail blocks
In-Reply-To: <bug-5204-206@http.llvm.org/bugs/>
References: <bug-5204-206@http.llvm.org/bugs/>
Message-ID: <20100714055642.BD7682A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5204

Evan Cheng <evan.cheng at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |evan.cheng at apple.com
         Resolution|                            |FIXED

--- Comment #2 from Evan Cheng <evan.cheng at apple.com> 2010-07-14 00:56:42 CDT ---
This has been fixed by recent if-conversion work.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 01:43:49 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 01:43:49 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7446] clang asserts while casting a
 TemplateTemplateParmDecl to a TagDecl in the name mangler
In-Reply-To: <bug-7446-206@http.llvm.org/bugs/>
References: <bug-7446-206@http.llvm.org/bugs/>
Message-ID: <20100714064349.E18A52A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7446

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |rjmccall at apple.com
         Resolution|                            |FIXED

--- Comment #2 from John McCall <rjmccall at apple.com> 2010-07-14 01:43:49 CDT ---
Fixed in r108326.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 01:44:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 01:44:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7610] Sibling call optimization broken
In-Reply-To: <bug-7610-206@http.llvm.org/bugs/>
References: <bug-7610-206@http.llvm.org/bugs/>
Message-ID: <20100714064413.1C43E2A6C130@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7610

Evan Cheng <evan.cheng at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Evan Cheng <evan.cheng at apple.com> 2010-07-14 01:44:12 CDT ---
Fixed: 108327

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 02:28:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 02:28:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7639] New: -Wempty-body is enabled by default
Message-ID: <bug-7639-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7639

           Summary: -Wempty-body is enabled by default
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: ojab at ojab.ru
                CC: llvmbugs at cs.uiuc.edu


gcc hasn't -Wempty-body enabled by default (only if explicit declared or
-Wextra), but clang has. So clang produces warnings like:

Wempty-body.c:2:8: warning: if statement has empty body [-Wempty-body]
  if(1);

while gcc compiles code silently.

It prevents using clang as drop-in replacement for gcc if project has -Werror
in CFLAGS.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 04:50:25 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 04:50:25 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7640] New: wiki - cannot upload files
Message-ID: <bug-7640-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7640

           Summary: wiki - cannot upload files
           Product: Website
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: General Website
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: grosser at fim.uni-passau.de
                CC: llvmbugs at cs.uiuc.edu, dberlin at dberlin.org


Internal error
>From LLVM
Jump to: navigation, search

The upload directory (public) is not writable by the webserver.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 07:53:30 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 07:53:30 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7641] New: Assertion when 'using' to move function
 from one namespace to another
Message-ID: <bug-7641-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7641

           Summary: Assertion when 'using' to move function from one
                    namespace to another
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5220)
 --> (http://llvm.org/bugs/attachment.cgi?id=5220)
Assertion example from 'using' statement

The attached (I believe valid) code causes an assertion in clang++. If you
change the last line from 'filesystem::' to filesystem2::', avoiding the
'using' statement, the code compiles fine. The generated assertion is attached
to other bugs, but I believe this is a different bug to those (or at least,
looks different, and valid code).

This bug is currently breaking boost::filesystem.

Assertion failed: (isa<X>(Val) && "cast<Ty>() argument of incompatible type!"),
function cast, file /Users/caj/clang/llvm/include/llvm/Support/Casting.h, line
202.
0  clang             0x00000001010ce676 PrintStackTrace(void*) + 38
1  clang             0x00000001010ceb84 SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff8398c35a _sigtramp + 26
3  libSystem.B.dylib 000000000000000000 _sigtramp + 2087140544
4  libSystem.B.dylib 0x00007fff83a079b4 __pthread_markcancel + 0
5  clang             0x000000010035951a
llvm::cast_retty<clang::FunctionTemplateDecl, clang::NamedDecl*>::ret_type
llvm::cast<clang::FunctionTemplateDecl, clang::NamedDecl*>(clang::NamedDecl*
const&) + 78
6  clang             0x0000000100389770
clang::Sema::ResolveSingleFunctionTemplateSpecialization(clang::Expr*) + 326
7  clang             0x00000001003e31fe
ResolveOverloadForDeduction(clang::Sema&, clang::TemplateParameterList*,
clang::Expr*, clang::QualType) + 151
8  clang             0x00000001003e37e1
clang::Sema::DeduceTemplateArguments(clang::FunctionTemplateDecl*,
clang::TemplateArgumentListInfo const*, clang::Expr**, unsigned int,
clang::FunctionDecl*&, clang::Sema::TemplateDeductionInfo&) + 625
9  clang             0x0000000100394f21
clang::Sema::AddTemplateOverloadCandidate(clang::FunctionTemplateDecl*,
clang::DeclAccessPair, clang::TemplateArgumentListInfo const*, clang::Expr**,
unsigned int, clang::OverloadCandidateSet&, bool) + 243
10 clang             0x00000001003951e5
clang::AddOverloadedCallCandidate(clang::Sema&, clang::DeclAccessPair,
clang::TemplateArgumentListInfo const*, clang::Expr**, unsigned int,
clang::OverloadCandidateSet&, bool) + 297
11 clang             0x0000000100395664
clang::Sema::AddOverloadedCallCandidates(clang::UnresolvedLookupExpr*,
clang::Expr**, unsigned int, clang::OverloadCandidateSet&, bool) + 566
12 clang             0x0000000100397aa2
clang::Sema::BuildOverloadedCallExpr(clang::Scope*, clang::Expr*,
clang::UnresolvedLookupExpr*, clang::SourceLocation, clang::Expr**, unsigned
int, clang::SourceLocation*, clang::SourceLocation) + 532
13 clang             0x0000000100337f04
clang::Sema::ActOnCallExpr(clang::Scope*,
clang::ASTOwningResult<&(clang::ActionBase::DeleteExpr(void*))>,
clang::SourceLocation,
clang::ASTMultiPtr<&(clang::ActionBase::DeleteExpr(void*))>,
clang::SourceLocation*, clang::SourceLocation) + 3072
14 clang             0x000000010069437b
clang::Parser::ParsePostfixExpressionSuffix(clang::ASTOwningResult<&(clang::ActionBase::DeleteExpr(void*))>)
+ 2097
15 clang             0x0000000100696dc9
clang::Parser::ParseCastExpression(bool, bool, bool&, void*) + 3495
16 clang             0x0000000100698c21
clang::Parser::ParseCastExpression(bool, bool, void*) + 97
17 clang             0x00000001006934a7
clang::Parser::ParseAssignmentExpression() + 241
18 clang             0x0000000100693a2e clang::Parser::ParseExpression() + 42
19 clang             0x00000001006b4c10
clang::Parser::ParseStatementOrDeclaration(bool) + 1662
20 clang             0x00000001006b9172
clang::Parser::ParseCompoundStatementBody(bool) + 220
21 clang             0x00000001006b979d
clang::Parser::ParseFunctionStatementBody(clang::OpaquePtr<0>) + 193
22 clang             0x00000001006c3845
clang::Parser::ParseFunctionDefinition(clang::Parser::ParsingDeclarator&,
clang::Parser::ParsedTemplateInfo const&) + 1237
23 clang             0x0000000100685a56
clang::Parser::ParseDeclGroup(clang::Parser::ParsingDeclSpec&, unsigned int,
bool, clang::SourceLocation*) + 494
24 clang             0x00000001006c27f5
clang::Parser::ParseDeclarationOrFunctionDefinition(clang::Parser::ParsingDeclSpec&,
clang::AttributeList*, clang::AccessSpecifier) + 1069
25 clang             0x00000001006c2863
clang::Parser::ParseDeclarationOrFunctionDefinition(clang::AttributeList*,
clang::AccessSpecifier) + 83
26 clang             0x00000001006c46f5
clang::Parser::ParseExternalDeclaration(clang::CXX0XAttributeList) + 2235
27 clang             0x00000001006c483f
clang::Parser::ParseTopLevelDecl(clang::OpaquePtr<1>&) + 247
28 clang             0x000000010025c5ad clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 456
29 clang             0x000000010007844a
clang::ASTFrontendAction::ExecuteAction() + 254
30 clang             0x000000010021fa70 clang::CodeGenAction::ExecuteAction() +
826
31 clang             0x0000000100078558 clang::FrontendAction::Execute() + 256
32 clang             0x000000010005b720
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 710
33 clang             0x0000000100027f28 cc1_main(char const**, char const**,
char const*, void*) + 1762
34 clang             0x000000010002ef81 main + 335
35 clang             0x0000000100026e48 start + 52
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free -main-file-name
double-namespace.cc -pic-level 1 -mdisable-fp-elim -masm-verbose
-munwind-tables -target-cpu core2 -resource-dir /usr/local/lib/clang/2.8
-ferror-limit 19 -fmessage-length 80 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/M8/M86fCFVgGQ0r-V2oY5lLC++++TI/-Tmp-/cc-PadPwD.o -x c++
double-namespace.cc 
1.    double-namespace.cc:19:43: current parser token ')'
2.    double-namespace.cc:19:1: parsing function body 'main'
3.    double-namespace.cc:19:1: in compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 10:45:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 10:45:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7642] New: typedef struct { ... } foo; does not work
Message-ID: <bug-7642-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7642

           Summary: typedef struct { ... } foo; does not work
           Product: clang
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: ps at kr.tuwien.ac.at
                CC: llvmbugs at cs.uiuc.edu


The following minimal example does not compile:

$ cat c.c
typedef struct { int j; } foo;
$ clang -v c.c
clang version 1.1 (branches/release_27)
Target: i386-pc-linux-gnu
Thread model: posix
 "/home/staff/ps/bin/clang" -cc1 -triple i386-pc-linux-gnu -S -disable-free
-main-file-name c.c -mrelocation-model static -mdisable-fp-elim
-mconstructor-aliases -target-cpu pentium4 -v -resource-dir
/home/staff/ps/lib/clang/1.1 -fmessage-length 149 -fgnu-runtime
-fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-lzS8QT.s -x c c.c
clang -cc1 version 1.1 based upon llvm 2.7 hosted on i386-pc-linux-gnu
#include "..." search starts here:
#include <...> search starts here:
 /home/staff/ps/lib/clang/1.1/include
 /usr/local/include
 /usr/include
End of search list.
0  clang 0x08f57118
Stack dump:
0.      Program arguments: /home/staff/ps/bin/clang -cc1 -triple
i386-pc-linux-gnu -S -disable-free -main-file-name c.c -mrelocation-model
static -mdisable-fp-elim -mconstructor-aliases -target-cpu pentium4 -v
-resource-dir /home/staff/ps/lib/clang/1.1 -fmessage-length 149 -fgnu-runtime
-fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-lzS8QT.s -x c c.c
1.      <eof> parser at end of file
clang: error: compiler command failed due to signal 11 (use -v to see
invocation)


This feature is used extensively in linux system headers, so i cannot compile
anything here with clang (openSUSE 11.0 (i586)).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 11:01:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 11:01:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7642] typedef struct { ... } foo; does not work
In-Reply-To: <bug-7642-206@http.llvm.org/bugs/>
References: <bug-7642-206@http.llvm.org/bugs/>
Message-ID: <20100714160133.5B5F02A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7642

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WORKSFORME

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-14 11:01:33 CDT ---
This definitely works in mainline and probably also worked in 2.7.  If you
built clang yourself, make sure you didn't do so with a broken version of GCC:
http://llvm.org/docs/GettingStarted.html#brokengcc

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 11:02:37 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 11:02:37 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7636] thumb2 ldmia.w used as a return doesn't write
	back SP
In-Reply-To: <bug-7636-206@http.llvm.org/bugs/>
References: <bug-7636-206@http.llvm.org/bugs/>
Message-ID: <20100714160237.3A7FF2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7636

Bob Wilson <bob.wilson at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |bob.wilson at apple.com
         Resolution|                            |FIXED

--- Comment #2 from Bob Wilson <bob.wilson at apple.com> 2010-07-14 11:02:36 CDT ---
Your patch looks right.
I've committed it in svn 108332.
Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 11:02:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 11:02:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7639] -Wempty-body is enabled by default
In-Reply-To: <bug-7639-206@http.llvm.org/bugs/>
References: <bug-7639-206@http.llvm.org/bugs/>
Message-ID: <20100714160243.3E0972A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7639

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-14 11:02:42 CDT ---
Clang has a lot of warnings on by default that GCC doesn't.  In these cases we
want to make sure that they have a -W flag so they can be turned off, and this
one does.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 15:00:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 15:00:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7643] New: Assigning a +1 ref value to a
 retain/copy property should warn
Message-ID: <bug-7643-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7643

           Summary: Assigning a +1 ref value to a retain/copy property
                    should warn
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: tjw at omnigroup.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5222)
 --> (http://llvm.org/bugs/attachment.cgi?id=5222)
Test case

A common error that clang-sa misses is:

  foo.property = [[Bar alloc] init];

If 'property' is marked retain or copy, the assignment will cause a leak.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 16:23:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 16:23:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5995] llvm-gcc not generating load/store alignment
	aggressively
In-Reply-To: <bug-5995-206@http.llvm.org/bugs/>
References: <bug-5995-206@http.llvm.org/bugs/>
Message-ID: <20100714212323.D2A682A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5995

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #13 from Dale Johannesen <dalej at apple.com> 2010-07-14 16:23:22 CDT ---
Fixed by reverting 46726, which seems to be safe after testing.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 17:44:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 17:44:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5228] Pass StringRef by value.
In-Reply-To: <bug-5228-206@http.llvm.org/bugs/>
References: <bug-5228-206@http.llvm.org/bugs/>
Message-ID: <20100714224423.B96DA2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5228

Benjamin Kramer <benny.kra at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |benny.kra at gmail.com
         Resolution|                            |FIXED

--- Comment #1 from Benjamin Kramer <benny.kra at gmail.com> 2010-07-14 17:44:23 CDT ---
The remaining "const StringRef &" APIs were changed in r108366.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 18:14:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 18:14:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7511] 'Cannot deduce non-type template argument
 with depth > 0' in boost::msm
In-Reply-To: <bug-7511-206@http.llvm.org/bugs/>
References: <bug-7511-206@http.llvm.org/bugs/>
Message-ID: <20100714231446.051002A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7511

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Douglas Gregor <dgregor at apple.com> 2010-07-14 18:14:45 CDT ---
Fixed the crasher in r108372.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 18:21:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 18:21:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7641] Assertion when 'using' to move function from
 one namespace to another
In-Reply-To: <bug-7641-206@http.llvm.org/bugs/>
References: <bug-7641-206@http.llvm.org/bugs/>
Message-ID: <20100714232128.7399C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7641

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-14 18:21:27 CDT ---
Fixed in r108376.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 18:36:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 18:36:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7644] New: Bogus access checking error
Message-ID: <bug-7644-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7644

           Summary: Bogus access checking error
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: andersca at mac.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


struct A {
private:
    enum Enumeration { Foo, Bar };

    template<Enumeration> void f();
};

template <A::Enumeration> void A::f() { }


andersca1:clang andersca$ clang -cc1 -fsyntax-only t.cpp
t.cpp:8:14: error: 'Enumeration' is a private member of 'A'
template <A::Enumeration> void A::f() { }
             ^
t.cpp:3:10: note: declared private here
    enum Enumeration { Foo, Bar };
         ^
1 error generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 19:44:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 19:44:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7408] Class instantiated as covariant return type
 of method in 'D' cannot see members declared later in 'D'
In-Reply-To: <bug-7408-206@http.llvm.org/bugs/>
References: <bug-7408-206@http.llvm.org/bugs/>
Message-ID: <20100715004438.88D6E2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7408

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-14 19:44:38 CDT ---
(In reply to comment #1)
> Checking [class.virtual]p5 clearly requires Covariant<T> to be complete.  I
> can't seem to find anything in the standard which specifies precisely when it
> should be instantiated, though.

GCC is waiting until the class is fully defined before it checks for covariant
return types, as you can see with this modified version of the example:


struct BaseReturn {
};

template <typename T> struct Base{
  virtual BaseReturn* GetEdges(long n) { return 0;}
};

template <typename T> struct Derived;

template<typename T>
struct Covariant {
  typename Derived<T>::LaterTypedef x;
};

template <typename T>
struct Derived : Base<T> {
  virtual Covariant<T>* GetEdges(long n) {
    return 0;
  }
 typedef char LaterTypedef;

  typedef T *blah;
};

Derived<long&> derived;

where the diagnostics for the "pointer to reference" error come before the
"invalid covariant return type" error, even though the code for the "pointer to
reference" error comes after the virtual function.

It seems to me like GCC is being overly generous, and that a user can't rely on
code that makes use of members defined later in the class. Closing as INVALID,
but I'm willing to entertain arguments that this is a real bug.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 14 22:24:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 14 Jul 2010 22:24:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7647] New: non-deterministic x86 clang crash
Message-ID: <bug-7647-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7647

           Summary: non-deterministic x86 clang crash
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5226)
 --> (http://llvm.org/bugs/attachment.cgi?id=5226)
failure-inducing testcase

Weird-- I get the various errors below if I repeatedly run this compile
command:

  clang -Os small.c -c

clang: TargetData.cpp:446: uint64_t llvm::TargetData::getTypeSizeInBits(const
llvm::Type*) const: Assertion `Ty->isSized() && "Cannot getTypeInfo() on a type
that is unsized!"' failed.

clang: /home/regehr/z/compiler-build/llvm-r108290/include/llvm/User.h:104:
llvm::Value* llvm::User::getOperand(unsigned int) const: Assertion `i <
NumOperands && "getOperand() out of range!"' failed.

Sometimes I get neither error message! But it always crashes.

regehr at john-home:~/volatile/bugs/tmp326$ clang -v
clang version 2.8 (trunk 108290)
Target: i386-pc-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 01:36:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 01:36:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7647] use-after-free in jump-threading
In-Reply-To: <bug-7647-206@http.llvm.org/bugs/>
References: <bug-7647-206@http.llvm.org/bugs/>
Message-ID: <20100715063641.A212E2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7647

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-15 01:36:41 CDT ---
Fixed in r108415, thanks Eli!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 04:35:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 04:35:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7649] New: dominance frontier not updated correctly
Message-ID: <bug-7649-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7649

           Summary: dominance frontier not updated correctly
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Loop Optimizer
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: grosser at fim.uni-passau.de
                CC: llvmbugs at cs.uiuc.edu


With llvm trunk this fails for me:

> opt -mem2reg -indvars /tmp/bug.ll  -verify-dom-info

opt: lib/VMCore/Dominators.cpp:119: virtual void
llvm::DominanceFrontier::verifyAnalysis() const: Assertion `!compare(OtherDF)
&& "Invalid DominanceFrontier info!"' failed.
Stack dump:
0.    Program arguments: opt -mem2reg -indvars /tmp/bug.ll -verify-dom-info 
1.    Running pass 'Function Pass Manager' on module '/tmp/bug.ll'.
2.    Running pass 'Loop Pass Manager' on function '@VBR_iteration_loop_new'
zsh: abort      opt -mem2reg -indvars /tmp/bug.ll -verify-dom-info

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 05:24:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 05:24:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7650] New: Incorrect lookup for overloaded
	operator<<=
Message-ID: <bug-7650-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7650

           Summary: Incorrect lookup for overloaded operator<<=
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: wotte at dre.vanderbilt.edu
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5229)
 --> (http://llvm.org/bugs/attachment.cgi?id=5229)
Pre-processed source reproducing the problem.

Compiling the attached pre-processed source file with the following command: 

clang++  -Wall -Wpointer-arith  -g -pipe   -pipe -Wno-logical-bitwise-confusion
-DACE_HAS_CUSTOM_EXPORT_MACROS=0   -I/Users/Shared/workspace/DOC_ROOT_clang/ACE
-DACE_HAS_EXCEPTIONS -D_\
_ACE_INLINE__ -I/Users/Shared/workspace/DOC_ROOT_clang/ACE -I..
-DTAO_HAS_VALUETYPE_OUT_INDIRECTION -DTAO_PORTABLESERVER_BUILD_DLL  -c  -o
.shobj/PortableServer/PolicyS.o PortableServer/PolicyS.cpp

I get the following diagnostic:

==========================
In file included from PortableServer/PolicyS.cpp:36:
In file included from ../tao/AnyTypeCode/TypeCode.h:34:
../tao/Any_Insert_Policy_T.h:44:14: error: 'operator<<=' is a private member of
'CORBA::Any'
        (*p) <<= x;
        ~~~~ ^   ~
In file included from PortableServer/PolicyS.cpp:43:
In file included from ../tao/PortableServer/Basic_SArguments.h:24:
In file included from ../tao/PortableServer/Basic_SArgument_T.h:154:
../tao/PortableServer/Basic_SArgument_T.cpp:103:21: note: in instantiation of
member function 'TAO::Any_Insert_Policy_Stream<unsigned int>::any_insert'
requested here
  Insert_Policy<S>::any_insert (any, this->x_);
                    ^
PortableServer/PolicyS.cpp:476:51: note: in instantiation of member function
'TAO::Ret_Basic_SArgument_T<unsigned int,
Any_Insert_Policy_Stream>::interceptor_value' requested here
  TAO::SArg_Traits< ::CORBA::PolicyType>::ret_val retval;
                                                  ^
In file included from PortableServer/PolicyS.cpp:37:
../tao/AnyTypeCode/Any.h:171:10: note: declared private here
    void operator<<= (unsigned char);
         ^
In file included from PortableServer/PolicyS.cpp:36:
In file included from ../tao/AnyTypeCode/TypeCode.h:34:
../tao/Any_Insert_Policy_T.h:44:14: error: no viable overloaded '<<='
        (*p) <<= x;
        ~~~~ ^   ~

===========================

>From an investigation of the pre-processed source, I can see that (In
PolicyS.pp.cpp):

It is finding:
Line 74988 - finding void operator<<= (unsigned char);, a private unimplemented
function in the CORBA::Any class.  

Should be finding:
Line 75089: void operator<<= (CORBA::Any &, CORBA::ULong);
Line 3774:  typedef ACE_UINT32 ULong;
Line 3613:  typedef unsigned int ACE_UINT32;

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 06:21:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 06:21:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7651] New: ICE: Zero allocated in a base register!
Message-ID: <bug-7651-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7651

           Summary: ICE: Zero allocated in a base register!
           Product: new-bugs
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: valera.veryazov at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5231)
 --> (http://llvm.org/bugs/attachment.cgi?id=5231)
short fortran 77 code, generating ICE.

The following "10-lines" code (see attachment) compiled with -O2 gives: 

f951:
/home/duncan/llvm-2.7/pre-release2/llvm-2.7/lib/Transforms/Scalar/LoopStrengthReduce.cpp:940:
bool<unnamed>::LSRUse::InsertFormula(const<unnamed>::Formula&): Assertion
`!(*I)->isZero() && "Zero allocated in a base register!"' failed.
a.f:8: internal compiler error: Aborted

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 09:40:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 09:40:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7652] New: llc -O0 infinite loop on atomic
	operations
Message-ID: <bug-7652-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7652

           Summary: llc -O0 infinite loop on atomic operations
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu


$ llc -O0 atomic.ll -debug
...
*** Scheduling [0]: SU(0): 0x2d03b50: i8,i32,ch = ATOMXOR8 0x2d03c50,
0x2d03550, 0x2d03750, 0x2d03850, 0x2d03950, 0x2d03d50, 0x2ce94f8<Mem:Volatile
LDST1[@uc]> [ORD=2] [ID=0]
*** Scheduling [1]: SU(1): 0x2d03d50: i8 = IMPLICIT_DEF [ORD=2] [ID=1]
*** Final schedule ***
SU(1): 0x2d03d50: i8 = IMPLICIT_DEF [ORD=2] [ID=1]
SU(0): 0x2d03b50: i8,i32,ch = ATOMXOR8 0x2d03c50, 0x2d03550, 0x2d03750,
0x2d03850, 0x2d03950, 0x2d03d50, 0x2ce94f8<Mem:Volatile LDST1[@uc]> [ORD=2]
[ID=0]

^ gets stuck here in an apparent infinite loop


Backtrace while it is stuck:

#0  llvm::ilist_traits<llvm::MachineInstr>::transferNodesFromList
(this=0x11b0870, fromList=<value optimised out>, first=<value optimised out>,
last=...)
    at /home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/MachineBasicBlock.cpp:133
#1  0x00000000007eb1a3 in llvm::iplist<llvm::MachineInstr,
llvm::ilist_traits<llvm::MachineInstr> >::transfer (this=<value optimised out>,
bInstr=<value optimised out>, 
    MBB=0x11b04a8, regOpc=<value optimised out>, immOpc=<value optimised out>,
LoadOpc=<value optimised out>, CXchgOpc=1033, copyOpc=1401, notOpc=1593,
EAXreg=2, 
    RC=0x1155b60, invSrc=false) at
/home/duncan/LLVM/llvm.top/llvm/include/llvm/ADT/ilist.h:506
#2  llvm::iplist<llvm::MachineInstr, llvm::ilist_traits<llvm::MachineInstr>
>::splice (this=<value optimised out>, bInstr=<value optimised out>,
MBB=0x11b04a8, 
    regOpc=<value optimised out>, immOpc=<value optimised out>, LoadOpc=<value
optimised out>, CXchgOpc=1033, copyOpc=1401, notOpc=1593, EAXreg=2,
RC=0x1155b60, invSrc=false)
    at /home/duncan/LLVM/llvm.top/llvm/include/llvm/ADT/ilist.h:561
#3  llvm::MachineBasicBlock::splice (this=<value optimised out>, bInstr=<value
optimised out>, MBB=0x11b04a8, regOpc=<value optimised out>, immOpc=<value
optimised out>, 
    LoadOpc=<value optimised out>, CXchgOpc=1033, copyOpc=1401, notOpc=1593,
EAXreg=2, RC=0x1155b60, invSrc=false)
    at
/home/duncan/LLVM/llvm.top/llvm/include/llvm/CodeGen/MachineBasicBlock.h:330
#4  llvm::X86TargetLowering::EmitAtomicBitwiseWithCustomInserter (this=<value
optimised out>, bInstr=<value optimised out>, MBB=0x11b04a8, regOpc=<value
optimised out>, 
    immOpc=<value optimised out>, LoadOpc=<value optimised out>, CXchgOpc=1033,
copyOpc=1401, notOpc=1593, EAXreg=2, RC=0x1155b60, invSrc=false)
    at /home/duncan/LLVM/llvm.top/llvm/lib/Target/X86/X86ISelLowering.cpp:8032
#5  0x00000000007ec7b9 in llvm::X86TargetLowering::EmitInstrWithCustomInserter
(this=0x11b0870, MI=0x11b04a8, BB=0x11b04a8)
    at /home/duncan/LLVM/llvm.top/llvm/lib/Target/X86/X86ISelLowering.cpp:8824
#6  0x0000000000929a6b in llvm::InstrEmitter::EmitMachineNode
(this=0x7fffffffd3e0, Node=<value optimised out>, IsClone=<value optimised
out>, IsCloned=<value optimised out>, 
    VRBaseMap=...) at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/InstrEmitter.cpp:728
#7  0x0000000000891fa9 in llvm::InstrEmitter::EmitNode (this=0x11b5ae0) at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/InstrEmitter.h:118
#8  llvm::ScheduleDAGSDNodes::EmitSchedule (this=0x11b5ae0) at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp:601
#9  0x00000000009046e3 in llvm::SelectionDAGISel::CodeGenAndEmitDAG
(this=0x11983f0) at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:577
#10 0x0000000000906be6 in llvm::SelectionDAGISel::SelectBasicBlock
(this=0x11983f0, Begin=<value optimised out>, End=<value optimised out>,
HadTailCall=@0x7fffffffd8af)
    at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:362
#11 0x00000000009070cd in llvm::SelectionDAGISel::SelectAllBasicBlocks
(this=<value optimised out>, Fn=<value optimised out>)
    at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:774
#12 0x0000000000907cfd in llvm::SelectionDAGISel::runOnMachineFunction
(this=0x11983f0, mf=...)
    at
/home/duncan/LLVM/llvm.top/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:250
#13 0x0000000000cb48dd in llvm::FPPassManager::runOnFunction (this=0x1188700,
F=...) at /home/duncan/LLVM/llvm.top/llvm/lib/VMCore/PassManager.cpp:1433
#14 0x0000000000cb49db in llvm::FPPassManager::runOnModule (this=0x1188700,
M=...) at /home/duncan/LLVM/llvm.top/llvm/lib/VMCore/PassManager.cpp:1453
---Type <return> to continue, or q <return> to quit---
#15 0x0000000000cb43e1 in llvm::MPPassManager::runOnModule (this=0x117d1e0,
M=...) at /home/duncan/LLVM/llvm.top/llvm/lib/VMCore/PassManager.cpp:1507
#16 0x0000000000cb4557 in llvm::PassManagerImpl::run (this=0x117cc40, M=...) at
/home/duncan/LLVM/llvm.top/llvm/lib/VMCore/PassManager.cpp:1588
#17 0x000000000052c5b8 in main (argc=<value optimised out>,
argv=0x7fffffffe008) at /home/duncan/LLVM/llvm.top/llvm/tools/llc/llc.cpp:343


Testcase atomic.ll:

target datalayout =
"e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
target triple = "x86_64-unknown-linux-gnu"

@uc = external global i8                          ; <i8*> [#uses=1]
@si = external global i32                         ; <i32*> [#uses=1]
@sll = external global i64                        ; <i64*> [#uses=1]

declare i8 @llvm.atomic.load.xor.i8.p0i8(i8* nocapture, i8) nounwind

declare i64 @llvm.atomic.load.and.i64.p0i64(i64* nocapture, i64) nounwind

define void @test_op_and_fetch() nounwind {
entry:
  %0 = call i8 @llvm.atomic.load.xor.i8.p0i8(i8* @uc, i8 undef) ; <i8>
[#uses=0]
  %1 = call i64 @llvm.atomic.load.and.i64.p0i64(i64* @sll, i64 undef) ; <i64>
[#uses=0]
  store i32 undef, i32* @si, align 4
  ret void
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 09:49:06 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 09:49:06 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7653] New: segfault in
 llvm::TargetLowering::LowerAsmOperandForConstraint
Message-ID: <bug-7653-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7653

           Summary: segfault in
                    llvm::TargetLowering::LowerAsmOperandForConstraint
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu


This is a recent regression: it worked not long ago.

$ llc asm.ll
...
llvm::TargetLowering::LowerAsmOperandForConstraint(llvm::SDValue, char,
std::vector<llvm::SDValue, std::allocator<llvm::SDValue> >&,
llvm::SelectionDAG&) const + 277
4  llc       0x00000000007a6982
llvm::X86TargetLowering::LowerAsmOperandForConstraint(llvm::SDValue, char,
std::vector<llvm::SDValue, std::allocator<llvm::SDValue> >&,
llvm::SelectionDAG&) const + 114
...


The problem is that C is null but is dereferenced at line 2479:

2475          if (ConstraintLetter != 'n') {
2476            int64_t Offs = GA->getOffset();
2477            if (C) Offs += C->getZExtValue();
2478            Ops.push_back(DAG.getTargetGlobalAddress(GA->getGlobal(), 
2479                                                     C->getDebugLoc(),
2480                                                     Op.getValueType(),
Offs));
2481            return;
2482          }


Testcase:

target datalayout =
"e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
target triple = "x86_64-unknown-linux-gnu"

@__FUNCTION__.1623 = external constant [4 x i8]   ; <[4 x i8]*> [#uses=1]

define void @foo() nounwind {
entry:
  tail call void asm sideeffect "", "s,i,~{fpsr},~{flags}"(i8* getelementptr
inbounds ([4 x i8]* @__FUNCTION__.1623, i64 0, i64 0), i8* getelementptr
inbounds ([4 x i8]* @__FUNCTION__.1623, i64 0, i64 0)) nounwind
  ret void
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 10:19:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 10:19:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7651] ICE: Zero allocated in a base register!
In-Reply-To: <bug-7651-206@http.llvm.org/bugs/>
References: <bug-7651-206@http.llvm.org/bugs/>
Message-ID: <20100715151932.229902A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7651

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |gohman at apple.com
         Resolution|                            |FIXED

--- Comment #2 from Dan Gohman <gohman at apple.com> 2010-07-15 10:19:31 CDT ---
Fixed in r108424.  Without auto-vectorization, code like that isn't going to
run fast though.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 11:56:21 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 11:56:21 -0500 (CDT)
Subject: [LLVMbugs] [Bug 2150] llvm-gcc doesn't produce "optimizer warnings"
In-Reply-To: <bug-2150-206@http.llvm.org/bugs/>
References: <bug-2150-206@http.llvm.org/bugs/>
Message-ID: <20100715165621.25A1E2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=2150

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-15 11:56:20 CDT ---
We have no plans to fix this, please use clang instead :)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 13:15:16 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 13:15:16 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6024] error: cannot compile this unexpected cast
	lvalue yet
In-Reply-To: <bug-6024-206@http.llvm.org/bugs/>
References: <bug-6024-206@http.llvm.org/bugs/>
Message-ID: <20100715181516.814F12A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6024

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Douglas Gregor <dgregor at apple.com> 2010-07-15 13:15:16 CDT ---
Fixed in r108431, r108433.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 13:36:31 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 13:36:31 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5290] decltype doesn't inspect expression type
	correctly
In-Reply-To: <bug-5290-206@http.llvm.org/bugs/>
References: <bug-5290-206@http.llvm.org/bugs/>
Message-ID: <20100715183631.E5D7E2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5290

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Douglas Gregor <dgregor at apple.com> 2010-07-15 13:36:30 CDT ---
This was fixed by the work to remove cv-qualifiers from non-class rvalues. Test
cases committed in r108437.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 13:45:57 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 13:45:57 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7653] segfault in
 llvm::TargetLowering::LowerAsmOperandForConstraint
In-Reply-To: <bug-7653-206@http.llvm.org/bugs/>
References: <bug-7653-206@http.llvm.org/bugs/>
Message-ID: <20100715184557.DF7592A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7653

Devang Patel <dpatel at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Devang Patel <dpatel at apple.com> 2010-07-15 13:45:57 CDT ---
Fixed. r108441.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 14:08:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 14:08:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7650] Incorrect lookup for overloaded operator<<=
In-Reply-To: <bug-7650-206@http.llvm.org/bugs/>
References: <bug-7650-206@http.llvm.org/bugs/>
Message-ID: <20100715190826.D8E792A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7650

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-15 14:08:26 CDT ---
Clang is correct to reject this code; GCC shouldn't allow it to compile. Please
see 

  http://clang.llvm.org/compatibility.html#dep_lookup

for an explanation.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 14:33:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 14:33:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7654] New: Assertion "Did not find a destructor!"
 on invalid code
Message-ID: <bug-7654-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7654

           Summary: Assertion "Did not find a destructor!" on invalid code
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5235)
 --> (http://llvm.org/bugs/attachment.cgi?id=5235)
Example showing assertion

The attached invalid code causes the following assertion:

clang: DeclCXX.cpp:624: clang::CXXDestructorDecl*
clang::CXXRecordDecl::getDestructor() const: Assertion `I != E && "Did not find
a destructor!"' failed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 15:25:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 15:25:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7618] miscompile with optimization (derived from
	ffmpeg)
In-Reply-To: <bug-7618-206@http.llvm.org/bugs/>
References: <bug-7618-206@http.llvm.org/bugs/>
Message-ID: <20100715202544.A20DF2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7618

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #9 from Dan Gohman <gohman at apple.com> 2010-07-15 15:25:44 CDT ---
Fixed in r108455.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 15:57:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 15:57:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7655] New: Declaration-or-expression ambiguity
 resolution picks incorrectly for complicated expressions
Message-ID: <bug-7655-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7655

           Summary: Declaration-or-expression ambiguity resolution picks
                    incorrectly for complicated expressions
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dgregor at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Given this insane code:

struct X { 
  template<typename T> X(T);
  X(int, int);

  X operator()(int, int) const;
};

template<typename T, typename U> struct Y { };

X *x;
void f() {
  int y = 0;
  X (*x)(int(y), int(y)) = Y<int, float>(), ++y;
}

Clang, like gcc, determines that the second statement in "x" is a declaration,
when in fact it's an expression. The problem is that we don't fully parse the
initializer after the '=' to determine whether what follows the ',' is another
declarator or not.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 17:58:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 17:58:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7656] New: Clang generates incorrect IR for
 __attribute__((cleanup))
Message-ID: <bug-7656-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7656

           Summary: Clang generates incorrect IR for
                    __attribute__((cleanup))
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: csdavec at swan.ac.uk
                CC: llvmbugs at cs.uiuc.edu


For this simple test case:

$ cat clean.m
static void foo(void*a){}
void wibble(void);

void bar(void)
{
    __attribute__((cleanup(foo))) int a = 12;
    wibble();
}

If wibble() throws, clang unwinds to this basic block:


; <label>:3                                       ; preds = %0
  %4 = call i8* @llvm.eh.exception()              ; <i8*> [#uses=2]
  %5 = call i32 (i8*, i8*, ...)* @llvm.eh.selector(i8* %4, i8* bitcast (i32
(...)* @__gnu_objc_personality_v0 to i8*), i8* null) ; <i32> [#uses=0]
  %6 = bitcast i32* %a to i8*                     ; <i8*> [#uses=1]
  invoke void @foo(i8* %6)
          to label %7 unwind label %8

Note the i8* null at the end of the llvm.eh.selector() call.  This indicates a
catchall, rather than a cleanup, which causes incorrect behaviour (i.e.
crashing).  I've looked at the code in CGException.cpp, and nothing seems
obviously wrong, so I'm having trouble working out why I'm seeing this.  Adding
some debugging lines to the personality function, it's definitely finding
something that looks like a catchall and delivering the exception.  In code
compiled with GCC, it only sees a cleanup for this frame.

(In this example, it is using the GNU ObjC personality function because the
file is a .m, but if it's a .c then it still thinks it's a cleanup, it just
uses a different personality function.  This may or may not work, depending on
the exact implementation of the personality function.)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 18:11:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 18:11:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7645] crash on template substitution of using decl
In-Reply-To: <bug-7645-206@http.llvm.org/bugs/>
References: <bug-7645-206@http.llvm.org/bugs/>
Message-ID: <20100715231117.11CCA2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7645

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dgregor at apple.com,
                   |                            |llvmbugs at cs.uiuc.edu
          Component|Frontend                    |C++

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 18:11:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 18:11:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7646] crash after broken type name
In-Reply-To: <bug-7646-206@http.llvm.org/bugs/>
References: <bug-7646-206@http.llvm.org/bugs/>
Message-ID: <20100715231147.CDB472A6C130@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7646

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dgregor at apple.com,
                   |                            |llvmbugs at cs.uiuc.edu
          Component|Frontend                    |C++

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 18:17:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 18:17:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7645] crash on template substitution of using decl
In-Reply-To: <bug-7645-206@http.llvm.org/bugs/>
References: <bug-7645-206@http.llvm.org/bugs/>
Message-ID: <20100715231718.7AEBE2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7645

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-15 18:17:17 CDT ---
Ah, I just fixed this recently.

*** This bug has been marked as a duplicate of bug 7641 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 19:55:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 19:55:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3800] asm inline : Ran out of registers at -O0
In-Reply-To: <bug-3800-206@http.llvm.org/bugs/>
References: <bug-3800-206@http.llvm.org/bugs/>
Message-ID: <20100716005543.3B21A2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3800

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #6 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-15 19:55:42 CDT ---
Fixed in r108489.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 21:01:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 21:01:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7657] New: template-id referenced without
	'template' keyword
Message-ID: <bug-7657-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7657

           Summary: template-id referenced without 'template' keyword
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: nlewycky at google.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5236)
 --> (http://llvm.org/bugs/attachment.cgi?id=5236)
testcase

The attached code contains the mischievous line:

  this->cast<bool>().template cast<bool>();

where I can't find any reason that the first 'cast' shouldn't need to have a
template keyword on it too. The full example is attached.

Clang permits it because it can clearly see that Foo has no dependent base
classes, but my understanding is that C++03 14.2/3 should apply as normal.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 21:18:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 21:18:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7657] template-id referenced without 'template'
	keyword
In-Reply-To: <bug-7657-206@http.llvm.org/bugs/>
References: <bug-7657-206@http.llvm.org/bugs/>
Message-ID: <20100716021850.05A062A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7657

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-15 21:18:49 CDT ---
Clang follows the clarified C++0x wording for this section that relies on the
definition of the "current instantiation". In this case, "this" is the current
instantiation so we can look into the primary Foo template to find out that
"cast" is a template, and therefore don't require (but do allow) the "template"
keyword.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 23:21:00 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 23:21:00 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5584] ice: Must have flexible array member if
 struct is bigger than type!
In-Reply-To: <bug-5584-206@http.llvm.org/bugs/>
References: <bug-5584-206@http.llvm.org/bugs/>
Message-ID: <20100716042100.55DF12A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5584

John Regehr <regehr at cs.utah.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 15 23:36:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 15 Jul 2010 23:36:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7658] New: Segmentation fault during
 refineAbstractTypeTo with certain structure types and ConstantStruct
 objects
Message-ID: <bug-7658-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7658

           Summary: Segmentation fault during refineAbstractTypeTo with
                    certain structure types and ConstantStruct objects
           Product: libraries
           Version: 2.7
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: clemahieu at gmail.com
                CC: llvmbugs at cs.uiuc.edu


The following crashes on the last line "o4->refineAbstractTypeTo(h2->get());"

#define __STDC_LIMIT_MACROS
#define __STDC_CONSTANT_MACROS

#include <stdio.h>
#include <vector>
#include "llvm/LLVMContext.h"
#include "llvm/Module.h"
#include "llvm/Type.h"
#include "llvm/DerivedTypes.h"
#include "llvm/Constants.h"
#include "llvm/GlobalValue.h"

using namespace llvm;
using namespace std;

int main (int argc, const char * argv[]) {
    LLVMContext * ctx;
    StringRef * name;
    Module * m;
    vector <Type *> * t1;
    vector <Constant *> * v1;
    OpaqueType * o1;
    OpaqueType * o2;
    PointerType * p1;
    PointerType * p2;
    StructType * s1;
    PATypeHolder * h1;
    vector <Type *> * t2;
    vector <Constant *> * v2;
    OpaqueType * o3;
    OpaqueType * o4;
    PointerType * p3;
    PointerType * p4;
    Constant * c;
    StructType * s2;
    PATypeHolder * h2;

    ctx = new LLVMContext;
    name = new StringRef ("test");
    m = new Module (*name, *ctx);

    t1 = new vector <Type *>;
    v1 = new vector <Constant *>;
    o1 = OpaqueType::get(*ctx);
    p1 = PointerType::get(o1, 0);
    t1->push_back ((IntegerType *)IntegerType::get(*ctx, 32));
    v1->push_back (ConstantInt::get(IntegerType::get(*ctx, 32), 14));
    t1->push_back (p1);
    v1->push_back (ConstantPointerNull::get(p1));
    o2 = OpaqueType::get (*ctx);
    p2 = PointerType::get (o2, 0);
    t1->push_back (p2);
    v1->push_back (ConstantPointerNull::get (p2));
    s1 = StructType::get (*ctx, *((const vector <const Type *> *) t1));
    h1 = new PATypeHolder (s1);
    o1->refineAbstractTypeTo(h1->get());
    o2->refineAbstractTypeTo(h1->get());


    t2 = new vector <Type *>;
    v2 = new vector <Constant *>;
    o3 = OpaqueType::get(*ctx);
    p3 = PointerType::get(o3, 0);
    t2->push_back ((IntegerType *)IntegerType::get(*ctx, 32));
    v2->push_back (ConstantInt::get(IntegerType::get(*ctx, 32), 14));
    t2->push_back (p3);
    v2->push_back (ConstantPointerNull::get(p3));
    o4 = OpaqueType::get (*ctx);
    p4 = PointerType::get (o4, 0);
    t2->push_back (p4);
    v2->push_back (ConstantPointerNull::get (p4));
    c = ConstantStruct::get(*ctx, *v2, false);
    s2 = StructType::get (*ctx, *((const vector <const Type *> *) t2));
    h2 = new PATypeHolder (s2);
    o3->refineAbstractTypeTo(h2->get());
    o4->refineAbstractTypeTo(h2->get());
}

the call stack is:

#0    0x1000d56fa in llvm::Value::getValueID at Value.h:247
#1    0x1000366c9 in llvm::isa_impl<llvm::Constant, llvm::Value> at Value.h:329
#2    0x10003994f in llvm::isa_impl_wrap<llvm::Constant, llvm::Value const,
llvm::Value const>::doit at Casting.h:71
#3    0x100039969 in llvm::isa_impl_cl<llvm::Value>::isa<llvm::Constant> at
Casting.h:83
#4    0x100039983 in llvm::isa_impl_cl<llvm::Value*>::isa<llvm::Constant> at
Casting.h:101
#5    0x10002a780 in llvm::isa_impl_cl<llvm::Value* const>::isa<llvm::Constant>
at Casting.h:92
#6    0x10003a938 in llvm::isa_impl_wrap<llvm::Constant, llvm::Use const,
llvm::Value*>::doit at Casting.h:63
#7    0x10002a79b in llvm::isa_impl_cl<llvm::Use>::isa<llvm::Constant> at
Casting.h:83
#8    0x10002a7b5 in llvm::isa<llvm::Constant, llvm::Use> at Casting.h:116
#9    0x100041c1f in llvm::cast<llvm::Constant, llvm::Use> at Casting.h:200
#10    0x10004211f in llvm::ConstantStruct::getOperand at Constants.h:460
#11    0x1000ab348 in llvm::ConstantKeyData<llvm::ConstantStruct>::getValType
at ConstantsContext.h:475
#12    0x1000ac5f1 in llvm::ConstantUniqueMap<std::vector<llvm::Constant*,
std::allocator<llvm::Constant*> >, llvm::StructType, llvm::ConstantStruct,
true>::refineAbstractType at ConstantsContext.h:741
#13    0x1000d52eb in llvm::DerivedType::unlockedRefineAbstractTypeTo at
Type.cpp:1175
#14    0x1000e62fe in llvm::TypeMap<llvm::StructValType,
llvm::StructType>::RefineAbstractType at TypesContext.h:409
#15    0x1000d5454 in llvm::StructType::refineAbstractType at Type.cpp:1267
#16    0x1000d52eb in llvm::DerivedType::unlockedRefineAbstractTypeTo at
Type.cpp:1175
#17    0x1000e55c5 in llvm::TypeMap<llvm::PointerValType,
llvm::PointerType>::RefineAbstractType at TypesContext.h:409
#18    0x1000d53d0 in llvm::PointerType::refineAbstractType at Type.cpp:1297
#19    0x1000d52eb in llvm::DerivedType::unlockedRefineAbstractTypeTo at
Type.cpp:1175
#20    0x1000d5539 in llvm::DerivedType::refineAbstractTypeTo at Type.cpp:1193
#21    0x100008158 in main at main.cpp:76

inside llvm::Value::getValueId the value of this is 0x0.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 01:36:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 01:36:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7649] dominance frontier not updated correctly
In-Reply-To: <bug-7649-206@http.llvm.org/bugs/>
References: <bug-7649-206@http.llvm.org/bugs/>
Message-ID: <20100716063614.505BA2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7649

Tobias Grosser <grosser at fim.uni-passau.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Tobias Grosser <grosser at fim.uni-passau.de> 2010-07-16 01:36:14 CDT ---
Committed in revision 108513.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 03:05:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 03:05:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3813] MachO and ELF Writers/MachineCodeEmitters are
 hard-coded into LLVMTargetMachine
In-Reply-To: <bug-3813-206@http.llvm.org/bugs/>
References: <bug-3813-206@http.llvm.org/bugs/>
Message-ID: <20100716080559.3B4E02A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3813

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |INVALID

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-16 03:05:58 CDT ---
The MachO and ELF writers referred to in this bug aren't used by
LLVMTargetMachine anymore (and the MachO writer is gone).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 03:06:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 03:06:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 1294] ELF Writer Support
In-Reply-To: <bug-1294-206@http.llvm.org/bugs/>
References: <bug-1294-206@http.llvm.org/bugs/>
Message-ID: <20100716080640.380822A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=1294

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |INVALID

--- Comment #5 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-16 03:06:39 CDT ---
This bug isn't usefully tracking anything.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 12:04:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 12:04:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7644] Bogus access checking error
In-Reply-To: <bug-7644-206@http.llvm.org/bugs/>
References: <bug-7644-206@http.llvm.org/bugs/>
Message-ID: <20100716170402.850652A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7644

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from John McCall <rjmccall at apple.com> 2010-07-16 12:04:02 CDT ---
Fixed in r108521.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 13:48:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 13:48:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3180] the C++ backend drops some function attributes
In-Reply-To: <bug-3180-206@http.llvm.org/bugs/>
References: <bug-3180-206@http.llvm.org/bugs/>
Message-ID: <20100716184813.C00B92A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3180

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-16 13:48:13 CDT ---
Fixed in r108547.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 16:01:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 16:01:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5314] [Linux Kernel] Inline asm asmprinter crash
 with P modifier
In-Reply-To: <bug-5314-206@http.llvm.org/bugs/>
References: <bug-5314-206@http.llvm.org/bugs/>
Message-ID: <20100716210134.E0D8C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5314

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Dale Johannesen <dalej at apple.com> 2010-07-16 16:01:34 CDT ---
Crash fixed in 108545.   Generated code made correct in 108548 (I think).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 16:18:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 16:18:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5533] 'P' inline asm modifier not supported on
	constants
In-Reply-To: <bug-5533-206@http.llvm.org/bugs/>
References: <bug-5533-206@http.llvm.org/bugs/>
Message-ID: <20100716211834.35FA62A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5533

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Dale Johannesen <dalej at apple.com> 2010-07-16 16:18:33 CDT ---
Constant accepted as "p" in 108558.  Don't know about the @PLT, that's not
testable by me.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 16:23:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 16:23:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5991] Need to spill more than one physical
 registers with inline asm
In-Reply-To: <bug-5991-206@http.llvm.org/bugs/>
References: <bug-5991-206@http.llvm.org/bugs/>
Message-ID: <20100716212342.E194D2A6C131@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5991

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |dalej at apple.com
         Resolution|                            |FIXED

--- Comment #2 from Dale Johannesen <dalej at apple.com> 2010-07-16 16:23:42 CDT ---
This doesn't crash in 108558.  There has been a lot of register allocator work
connected with asm's, I'm not sure exactly what fixed it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 17:07:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 17:07:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7660] New: Assertion failure in
 PCHWriterDecl.cpp:557: void
 clang::PCHDeclWriter::VisitParmVarDecl(clang::ParmVarDecl*): Assertion
 `!D->isStaticDataMember() && "PARM_VAR_DECL can't be static data member"'
 failed.
Message-ID: <bug-7660-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7660

           Summary: Assertion failure in PCHWriterDecl.cpp:557: void
                    clang::PCHDeclWriter::VisitParmVarDecl(clang::ParmVarD
                    ecl*): Assertion `!D->isStaticDataMember() &&
                    "PARM_VAR_DECL can't be static data member"' failed.
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: llvmbugs at contacts.eelis.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Consider:

  template<typename T> struct S { void g(void (*)(T)); };
  template<> void S<int>::g(void(*)(int)) {}

When compiled into a PCH with:

  clang -cc1 t.hpp -emit-pch -o t.hpp.pch

I get:

  clang: PCHWriterDecl.cpp:557: void
clang::PCHDeclWriter::VisitParmVarDecl(clang::ParmVarDecl*): Assertion
`!D->isStaticDataMember() && "PARM_VAR_DECL can't be static data member"'
failed.

I'm using clang trunk 108560.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 16 19:43:31 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 16 Jul 2010 19:43:31 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7656] Clang generates incorrect IR for
	__attribute__((cleanup))
In-Reply-To: <bug-7656-206@http.llvm.org/bugs/>
References: <bug-7656-206@http.llvm.org/bugs/>
Message-ID: <20100717004331.BAA2A2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7656

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from John McCall <rjmccall at apple.com> 2010-07-16 19:43:31 CDT ---
Should be fixed in r108595.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 01:14:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 01:14:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7658] Segmentation fault during
 refineAbstractTypeTo with certain structure types and ConstantStruct
 objects
In-Reply-To: <bug-7658-206@http.llvm.org/bugs/>
References: <bug-7658-206@http.llvm.org/bugs/>
Message-ID: <20100717061424.2B0992A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7658

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Chris Lattner <clattner at apple.com> 2010-07-17 01:14:23 CDT ---
Fixed in r108610.  Thank you for the awesome testcase!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 01:19:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 01:19:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3693] Assertion failed: (ATI !=
 AbstractTypeMap.end() && "Abstract type not in AbstractTypeMap?"),
 function MoveConstantToNewSlot, file Constants.cpp, line 1196.
In-Reply-To: <bug-3693-206@http.llvm.org/bugs/>
References: <bug-3693-206@http.llvm.org/bugs/>
Message-ID: <20100717061903.BED3A2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3693

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-17 01:19:03 CDT ---
This should be fixed in r 108610.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 01:20:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 01:20:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6482] llvm-link asserts with "Type mismatch in
 constant table!" depending on linking order
In-Reply-To: <bug-6482-206@http.llvm.org/bugs/>
References: <bug-6482-206@http.llvm.org/bugs/>
Message-ID: <20100717062005.428802A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6482

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #9 from Chris Lattner <clattner at apple.com> 2010-07-17 01:20:04 CDT ---
This was probably fixed by r 108610, feel free to reopen if not.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 01:23:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 01:23:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4303] llvm.dbg.declare vs -instcombine
In-Reply-To: <bug-4303-206@http.llvm.org/bugs/>
References: <bug-4303-206@http.llvm.org/bugs/>
Message-ID: <20100717062303.4A1222A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4303

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |clattner at apple.com
         Resolution|                            |FIXED

--- Comment #6 from Chris Lattner <clattner at apple.com> 2010-07-17 01:23:02 CDT ---
This got fixed along the way.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 02:55:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 02:55:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6482] llvm-link asserts with "Type mismatch in
 constant table!" depending on linking order
In-Reply-To: <bug-6482-206@http.llvm.org/bugs/>
References: <bug-6482-206@http.llvm.org/bugs/>
Message-ID: <20100717075522.7589B2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6482

Matti Niemenmaa <matti.niemenmaa+llvmbugs at iki.fi> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #10 from Matti Niemenmaa <matti.niemenmaa+llvmbugs at iki.fi> 2010-07-17 02:55:21 CDT ---
Nope, both the original testcases and Nick's latest reduced one still fail.
Although Nick's asserts differently; not sure if it did that before. Anyway:

llvm-link: src/llvm/trunk/lib/Bitcode/Reader/BitcodeReader.cpp:204:
llvm::Constant* llvm::BitcodeReaderValueList::getConstantFwdRef(unsigned int,
const llvm::Type*): Assertion `Ty == V->getType() && "Type mismatch in constant
table!"' failed.
Stack dump:
0.    Program arguments: llvm-link one.bc two.bc

And:

llvm-link: src/llvm/trunk/lib/VMCore/Value.cpp:317: void
llvm::Value::replaceAllUsesWith(llvm::Value*): Assertion `New->getType() ==
getType() && "replaceAllUses of value with new value of different type!"'
failed.
Stack dump:
0.    Program arguments: llvm-link pr6482-b.bc pr6482-b.bc

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 06:02:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 06:02:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7656] Clang generates incorrect IR for
	__attribute__((cleanup))
In-Reply-To: <bug-7656-206@http.llvm.org/bugs/>
References: <bug-7656-206@http.llvm.org/bugs/>
Message-ID: <20100717110254.3A6982A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7656

David Chisnall <csdavec at swan.ac.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #5 from David Chisnall <csdavec at swan.ac.uk> 2010-07-17 06:02:53 CDT ---
It's now half-fixed, but unfortunately now it breaks with foreign exceptions. 
If it's a catchall, the runtime will deliver the ObjC object.  If it's a
cleanup, the runtime will deliver the exception structure.  Rethrowing as if it
were an Objective-C object doesn't work if it not an Objective-C object (you
get some nonsense which the personality routine doesn't know how to handle
correctly).

Treating cleanups as catchalls is simply not a solution.  They have VERY
different semantics.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 11:03:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 11:03:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3560] x87 code generation performance issue
In-Reply-To: <bug-3560-206@http.llvm.org/bugs/>
References: <bug-3560-206@http.llvm.org/bugs/>
Message-ID: <20100717160350.BACA12A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3560

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-17 11:03:50 CDT ---
Jakob just implemented this.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 11:13:51 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 11:13:51 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7661] New: intermittent clang x64 failure:
 non-trivial casts should be done with the SCEVs directly!
Message-ID: <bug-7661-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7661

           Summary: intermittent clang x64 failure: non-trivial casts
                    should be done with the SCEVs directly!
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5237)
 --> (http://llvm.org/bugs/attachment.cgi?id=5237)
failure-inducing C

This only happens sometimes.


[regehr at gamow tmp421]$ clang -v
clang version 2.8 (trunk 108621)
Target: x86_64-unknown-linux-gnu
Thread model: posix

[regehr at gamow tmp421]$ clang -O2 small.c -w
clang: ScalarEvolutionExpander.cpp:1274: llvm::Value*
llvm::SCEVExpander::expandCodeFor(const llvm::SCEV*, const llvm::Type*):
Assertion `SE.getTypeSizeInBits(Ty) == SE.getTypeSizeInBits(SH->getType()) &&
"non-trivial casts should be done with the SCEVs directly!"' failed.
0  clang           0x00000000015f297f
1  clang           0x00000000015f4a12
2  libpthread.so.0 0x00007f006f0ed190
3  libc.so.6       0x00007f006e3f24b5 gsignal + 53
4  libc.so.6       0x00007f006e3f5f50 abort + 384
5  libc.so.6       0x00007f006e3eb481 __assert_fail + 241
6  clang           0x00000000014ac514
7  clang           0x000000000133cb87
8  clang           0x000000000133f186
9  clang           0x00000000014689be
llvm::LPPassManager::runOnFunction(llvm::Function&) + 1150
10 clang           0x0000000001575d1d
llvm::FPPassManager::runOnFunction(llvm::Function&) + 637
11 clang           0x000000000143b1fb
12 clang           0x000000000143b975
13 clang           0x000000000157581e
llvm::MPPassManager::runOnModule(llvm::Module&) + 494
14 clang           0x0000000001575997 llvm::PassManagerImpl::run(llvm::Module&)
+ 167
15 clang           0x00000000007a8837
clang::EmitBackendOutput(clang::Diagnostic&, clang::CodeGenOptions const&,
clang::TargetOptions const&, llvm::Module*, clang::BackendAction,
llvm::raw_ostream*) + 1639
16 clang           0x00000000007a49db
17 clang           0x00000000008b6f0f clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 335
18 clang           0x00000000007a5954 clang::CodeGenAction::ExecuteAction() +
68
19 clang           0x00000000006c17a5
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 357
20 clang           0x00000000006a3f4b cc1_main(char const**, char const**, char
const*, void*) + 2651
21 clang           0x00000000006aa1b1 main + 3185
22 libc.so.6       0x00007f006e3ddabd __libc_start_main + 253
23 clang           0x00000000006a16a9
Stack dump:
0.      Program arguments:
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r108621-install/bin/clang
-cc1 -triple x86_64-unknown-linux-gnu -S -disable-free -main-file-name small.c
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases
-munwind-tables -target-cpu x86-64 -resource-dir
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r108621-install/lib/clang/2.8
-O2 -w -ferror-limit 19 -fmessage-length 91 -fgnu-runtime
-fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-pPz8I0.s -x c small.c 
1.      <eof> parser at end of file
2.      Per-module optimization passes
3.      Running pass 'CallGraph Pass Manager' on module 'small.c'.
4.      Running pass 'Loop Pass Manager' on function '@func_84'
5.      Running pass 'Canonicalize Induction Variables' on basic block
'%lbl_122.loopexit.i1267'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 14:39:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 14:39:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4403] clang x86-64 ABI doesn't set alignment
 correctly for byval struct containing SSE vector
In-Reply-To: <bug-4403-206@http.llvm.org/bugs/>
References: <bug-4403-206@http.llvm.org/bugs/>
Message-ID: <20100717193950.B37A22A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4403

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-17 14:39:50 CDT ---
This got fixed along the way.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 15:44:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 15:44:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5692] clang rejects invalid code with wrong
	diagnostic
In-Reply-To: <bug-5692-206@http.llvm.org/bugs/>
References: <bug-5692-206@http.llvm.org/bugs/>
Message-ID: <20100717204402.50AA32A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5692

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-17 15:44:02 CDT ---
Fixed in r108630.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 17 20:50:11 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 17 Jul 2010 20:50:11 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7662] New: code generator crash with -g and -O[123]
Message-ID: <bug-7662-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7662

           Summary: code generator crash with -g and -O[123]
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: mike at fluffypenguin.org
                CC: llvmbugs at cs.uiuc.edu


I found a crash with llvm trunk at r108631. It requires -g and -O[123]. -O0 or
-g0 prevent the crash.

C++ code that shows the crash [this is boost 1.42]

========================================
#include <boost/function.hpp>
typedef boost::function<void(void)> func_t;
struct def_func {
  void operator()() {}
};

void set_func( func_t ) { }

void run() {
  set_func( def_func() );
}
========================================

This seems to be a code generator bug.

========================================
mike at hyperion:~/debug> bugpoint -run-llc test.bc                               
                                                                               
                                                                               
                  (master:debug)
Read input file      : 'test.bc'
*** All input ok
Initializing execution environment: Found gcc: /usr/bin/gcc
Running the code generator to test for a crash: <llc>
Error running tool:
  /usr/bin/llc -o bugpoint-test-program.bc-wnDzvS.llc.s
bugpoint-test-program.bc-wnDzvS
0  llc             0x0000000000aa38ef
1  llc             0x0000000000aa3e89
2  libpthread.so.0 0x00007f7ae8a2c070
3  llc             0x0000000000764bdc llvm::DwarfDebug::endModule() + 2908
4  llc             0x000000000074d9a5
llvm::AsmPrinter::doFinalization(llvm::Module&) + 341
5  llc             0x0000000000a2e778
llvm::FPPassManager::doFinalization(llvm::Module&) + 56
6  llc             0x0000000000a31abf
llvm::FPPassManager::runOnModule(llvm::Module&) + 95
7  llc             0x0000000000a3167c
llvm::MPPassManager::runOnModule(llvm::Module&) + 508
8  llc             0x0000000000a3175e llvm::PassManagerImpl::run(llvm::Module&)
+ 110
9  llc             0x00000000004f846e main + 4334
10 libc.so.6       0x00007f7ae7b2eb6d __libc_start_main + 253
11 llc             0x00000000004f72b9
Stack dump:
0.      Program arguments: /usr/bin/llc -o
bugpoint-test-program.bc-wnDzvS.llc.s bugpoint-test-program.bc-wnDzvS 
1.      Running pass 'Function Pass Manager' on module
'bugpoint-test-program.bc-wnDzvS'.
*** Debugging code generator crash!

Checking to see if we can delete global inits: UNREACHABLE executed!
0  bugpoint        0x00000000008c7a3f
1  bugpoint        0x00000000008c7fd9
2  libpthread.so.0 0x00007f9ea5042070
3  libc.so.6       0x00007f9ea4158545 gsignal + 53
4  libc.so.6       0x00007f9ea41599c6 abort + 390
5  bugpoint        0x00000000008a8d8e llvm::llvm_unreachable_internal(char
const*, char const*, unsigned int) + 350
6  bugpoint        0x00000000004e0be9
7  bugpoint        0x00000000004e88a6
8  bugpoint        0x00000000004ea2ee llvm::WriteBitcodeToFile(llvm::Module
const*, llvm::raw_ostream&) + 190
9  bugpoint        0x00000000004ced9d
llvm::BugDriver::writeProgramToFile(std::string const&, llvm::Module*) const +
109
10 bugpoint        0x00000000004baa47
llvm::BugDriver::compileProgram(llvm::Module*, std::string*) + 199
11 bugpoint        0x00000000004aed0f
12 bugpoint        0x00000000004b23f2
13 bugpoint        0x00000000004acdd9 llvm::BugDriver::run(std::string&) + 601
14 bugpoint        0x00000000004dad08 main + 664
15 libc.so.6       0x00007f9ea4144b6d __libc_start_main + 253
16 bugpoint        0x00000000004ac879
Stack dump:
0.      Program arguments: bugpoint -run-llc test.bc 
[2]    15976 abort      bugpoint -run-llc test.bc
========================================

I've attached the bc file produced by -emit-llvm

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 18 05:35:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 18 Jul 2010 05:35:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7663] New: Add JIT::recompileAndRelinkFunction to C
 API (w/ patch)
Message-ID: <bug-7663-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7663

           Summary: Add JIT::recompileAndRelinkFunction to C API (w/
                    patch)
           Product: new-bugs
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ralith at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5239)
 --> (http://llvm.org/bugs/attachment.cgi?id=5239)
Patch to implement described functionality

Exposes JIT::recompileAndRelinkFunction for use through the C API.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 18 06:40:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 18 Jul 2010 06:40:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7664] New: Expose BasicBlock::moveBefore and
 moveAfter in C API (w/ patch)
Message-ID: <bug-7664-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7664

           Summary: Expose BasicBlock::moveBefore and moveAfter in C API
                    (w/ patch)
           Product: new-bugs
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ralith at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5241)
 --> (http://llvm.org/bugs/attachment.cgi?id=5241)
Patch to implement described functionality

C wrappers for the described functions

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 18 13:30:30 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 18 Jul 2010 13:30:30 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7665] New: disassembler map file problems on OpenBSD
Message-ID: <bug-7665-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7665

           Summary: disassembler map file problems on OpenBSD
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: OpenBSD
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jsg at openbsd.org
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5243)
 --> (http://llvm.org/bugs/attachment.cgi?id=5243)
generated map file

LLVM svn (108643) on OpenBSD/i386 -current doesn't seem to build:

llvm[2]: Compiling EDDisassembler.cpp for Release+Asserts build (PIC)
llvm[2]: Compiling EDInst.cpp for Release+Asserts build (PIC)
llvm[2]: Compiling EDMain.cpp for Release+Asserts build (PIC)
llvm[2]: Compiling EDOperand.cpp for Release+Asserts build (PIC)
llvm[2]: Compiling EDToken.cpp for Release+Asserts build (PIC)
llvm[2]: Linking Release+Asserts Shared Library EnhancedDisassembly.so
/usr/bin/ld:/usr/users/jsg/src/llvm/tools/edis/Release+Asserts/EnhancedDisassembly.exports.map:38:
syntax error in VERSION script
collect2: ld returned 1 exit status
gmake[2]: ***
[/usr/users/jsg/src/llvm/Release+Asserts/lib/libEnhancedDisassembly.so] Error 1

$ ld --version
GNU ld version 2.15

$ cc -v              
Reading specs from /usr/lib/gcc-lib/i386-unknown-openbsd4.7/4.2.1/specs
Target: i386-unknown-openbsd4.7
Configured with: OpenBSD/i386 system compiler
Thread model: posix
gcc version 4.2.1 20070719

$ libtool --version
ltmain.sh (GNU libtool) 1.5.26 (1.1220.2.493 2008/02/01 16:58:18)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 18 16:01:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 18 Jul 2010 16:01:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7666] New: Variable size array function parameters
 cause clang to segfault
Message-ID: <bug-7666-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7666

           Summary: Variable size array function parameters cause clang to
                    segfault
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: tjablin at cs.princeton.edu
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


echo "void foo(int X, int a[X][X]) {}" | clang -x c++ - -c

causes clang to segfault.

echo "void foo(int X, int a[X][X]) {}" | clang -x c - -c

works.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 18 20:37:49 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 18 Jul 2010 20:37:49 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7668] New: raw_ostream float printing should be
	portable
Message-ID: <bug-7668-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7668

           Summary: raw_ostream float printing should be portable
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Support Libraries
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu


Currently, raw_ostream::operator<<(double) uses printf under the covers with
the "%e" format. Unfortunately, this doesn't give portable output across
platforms, notable on Windows we get things like:
  1.700000e+001
whereas on most Unix machines we ill get
  1.700000e+01

This is annoying because we use those routines to print constants in .ll files
(and other places), and so the compiler output is not always deterministic
across machines.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 04:34:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 04:34:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7663] Add JIT::recompileAndRelinkFunction to C API
	(w/ patch)
In-Reply-To: <bug-7663-206@http.llvm.org/bugs/>
References: <bug-7663-206@http.llvm.org/bugs/>
Message-ID: <20100719093453.992282A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7663

Duncan Sands <baldrick at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |baldrick at free.fr
         Resolution|                            |FIXED

--- Comment #2 from Duncan Sands <baldrick at free.fr> 2010-07-19 04:34:53 CDT ---
Patch applied here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20100719/104539.html

Please just send patches to the llvm-commits mailing list in the future,
rather than opening a bug report.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 05:17:36 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 05:17:36 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7660] Assertion failure in PCHWriterDecl.cpp:557:
 void clang::PCHDeclWriter::VisitParmVarDecl(clang::ParmVarDecl*): Assertion
 `!D->isStaticDataMember() && "PARM_VAR_DECL can't be static data member"'
 failed.
In-Reply-To: <bug-7660-206@http.llvm.org/bugs/>
References: <bug-7660-206@http.llvm.org/bugs/>
Message-ID: <20100719101736.893282A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7660

Argyrios Kyrtzidis <akyrtzi at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Argyrios Kyrtzidis <akyrtzi at gmail.com> 2010-07-19 05:17:36 CDT ---
Fixed at r108692.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 05:31:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 05:31:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7669] New: "Cannot deduce non-type template
 argument with depth > 0" on boost::msm
Message-ID: <bug-7669-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7669

           Summary: "Cannot deduce non-type template argument with depth >
                    0" on boost::msm
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The attached testcase, reduced from boost::msm, fails with the assertion:

SemaTemplateDeduction.cpp:103: clang::Sema::TemplateDeductionResult
DeduceNonTypeTemplateArgument(clang::Sema&, clang::NonTypeTemplateParmDecl*,
llvm::APSInt, clang::QualType, bool, clang::Sema::TemplateDeductionInfo&,
llvm::SmallVectorImpl<clang::DeducedTemplateArgument>&): Assertion
`NTTP->getDepth() == 0 && "Cannot deduce non-type template argument with depth
> 0"' failed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 07:24:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 07:24:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7670] New: PCHWriterDecl.cpp:848: void
 clang::PCHDeclWriter::VisitClassTemplateDecl(clang::ClassTemplateDecl*):
 Assertion `IsKeptInFoldingSet(&*I)' failed.
Message-ID: <bug-7670-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7670

           Summary: PCHWriterDecl.cpp:848: void
                    clang::PCHDeclWriter::VisitClassTemplateDecl(clang::Cl
                    assTemplateDecl*): Assertion `IsKeptInFoldingSet(&*I)'
                    failed.
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: llvmbugs at contacts.eelis.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Consider:

  template<typename> class C;
  template<> class C<int>;
  template<> class C<int>;

When I try to compile this into a PCH with:

  clang -cc1 t.hpp -emit-pch -o t.hpp.pch

I get:

  clang: PCHWriterDecl.cpp:848: void
clang::PCHDeclWriter::VisitClassTemplateDecl(clang::ClassTemplateDecl*):
Assertion `IsKeptInFoldingSet(&*I)' failed.

I'm using clang r108692.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 10:31:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 10:31:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7664] Expose BasicBlock::moveBefore and moveAfter
 in C API (w/ patch)
In-Reply-To: <bug-7664-206@http.llvm.org/bugs/>
References: <bug-7664-206@http.llvm.org/bugs/>
Message-ID: <20100719153148.71DCF2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7664

Duncan Sands <baldrick at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |baldrick at free.fr
         Resolution|                            |FIXED

--- Comment #1 from Duncan Sands <baldrick at free.fr> 2010-07-19 10:31:48 CDT ---
Patch applied here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20100719/104544.html

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 11:01:34 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 11:01:34 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6038] Fails to build libgomp (memory barriers can't
	be lowered)
In-Reply-To: <bug-6038-206@http.llvm.org/bugs/>
References: <bug-6038-206@http.llvm.org/bugs/>
Message-ID: <20100719160134.25A5B2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6038

Jim Grosbach <grosbach at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |grosbach at apple.com
         Resolution|                            |FIXED

--- Comment #3 from Jim Grosbach <grosbach at apple.com> 2010-07-19 11:01:33 CDT ---
I believe this is fixed on ToT llvm. For v6+ ARM mode and Thumb2, the memory
barrier code is handled inline. For Thumb1 and pre-v6 processors, it's expanded
to a library call (__sync_synchronize). See r106204.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 11:19:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 11:19:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7212] Error while compiling llvm-gcc for ARM
In-Reply-To: <bug-7212-206@http.llvm.org/bugs/>
References: <bug-7212-206@http.llvm.org/bugs/>
Message-ID: <20100719161923.827AE2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7212

Jim Grosbach <grosbach at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |grosbach at apple.com
         Resolution|                            |FIXED

--- Comment #6 from Jim Grosbach <grosbach at apple.com> 2010-07-19 11:19:22 CDT ---
This was likely fixed by r106484. Renato, if the problem still occurs for you,
please re-open.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 12:12:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 12:12:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7672] New: Invalid cv qualifier for non-member
 function error could be better
Message-ID: <bug-7672-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7672

           Summary: Invalid cv qualifier for non-member function error
                    could be better
           Product: clang
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: wendling at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


This code:

$ cat a.cpp
struct Foo {
  unsigned F(unsigned a, unsigned b) const;
};

unsigned F(unsigned a, unsigned b) const {
  return a + b * 37 - 927;
}

Produces this error message with clang++:

a.cpp:5:10: error: type qualifier is not allowed on this function
unsigned F(unsigned a, unsigned b) const {
         ^
1 error generated.

It doesn't explain why the qualifier isn't correct. By contrast, g++ does
better:

a.cpp:5: error: non-member function ?unsigned int F(unsigned int, unsigned
int)? cannot have cv-qualifier

It mentions that it's a non-member function.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 12:21:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 12:21:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6581] Jump table entries emitted twice
In-Reply-To: <bug-6581-206@http.llvm.org/bugs/>
References: <bug-6581-206@http.llvm.org/bugs/>
Message-ID: <20100719172132.50E332A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6581

Jim Grosbach <grosbach at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |grosbach at apple.com
         Resolution|                            |FIXED

--- Comment #1 from Jim Grosbach <grosbach at apple.com> 2010-07-19 12:21:31 CDT ---
Fixed in r108730.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 12:31:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 12:31:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6835] Makefiles.rules wrong handling of CXXFLAGS
In-Reply-To: <bug-6835-206@http.llvm.org/bugs/>
References: <bug-6835-206@http.llvm.org/bugs/>
Message-ID: <20100719173129.5DDD62A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6835

Nick Lewycky <nlewycky at google.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #18 from Nick Lewycky <nlewycky at google.com> 2010-07-19 12:31:28 CDT ---
Totally.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 12:45:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 12:45:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7668] raw_ostream float printing should be portable
In-Reply-To: <bug-7668-206@http.llvm.org/bugs/>
References: <bug-7668-206@http.llvm.org/bugs/>
Message-ID: <20100719174514.1F6A22A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7668

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |dalej at apple.com
         Resolution|                            |DUPLICATE

--- Comment #2 from Dale Johannesen <dalej at apple.com> 2010-07-19 12:45:13 CDT ---


*** This bug has been marked as a duplicate of bug 6745 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 12:54:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 12:54:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7662] code generator crash with -g and -O[123]
In-Reply-To: <bug-7662-206@http.llvm.org/bugs/>
References: <bug-7662-206@http.llvm.org/bugs/>
Message-ID: <20100719175456.C530D2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7662

Devang Patel <dpatel at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Devang Patel <dpatel at apple.com> 2010-07-19 12:54:56 CDT ---
Fixed. r108731. 

(Nick, Bitcode writer crash is a separate issue.)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 19 22:52:52 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 19 Jul 2010 22:52:52 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7673] New: Inline asm parse difference from GCC
Message-ID: <bug-7673-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7673

           Summary: Inline asm parse difference from GCC
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chandlerc at gmail.com
                CC: llvmbugs at cs.uiuc.edu


This isn't a standards issue, as its a GCC extension, but it seems we should
match GCC's behavior in that case:


% cat t.cc
void f() {
  __asm__ ("" : : :);
}

% clang -fsyntax-only t.cc
t.cc:2:20: error: expected string literal
  __asm__ ("" : : :);
                   ^
1 error generated.

GCC accepts this happily.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 01:01:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 01:01:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7674] New: Missing SIMD intrinsics
Message-ID: <bug-7674-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7674

           Summary: Missing SIMD intrinsics
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chandlerc at gmail.com
                CC: llvmbugs at cs.uiuc.edu, echristo at gmail.com


We seem to be missing quite a few MMX / SSE intrinsics. I know little about
them, bet experimentation with a heavily vectorized bit of code yields this
test case. I hope it's still valid, but GCC at least doesn't complain.


% cat x.c                                                                       
#include <mmintrin.h>
#include <emmintrin.h>
#include <xmmintrin.h>
#include <pmmintrin.h>

void f(__m64 x, __m64 y) {
  (void)_m_punpcklwd(x, y);
  (void)_m_punpckhwd(x, y);
  (void)_m_punpcklbw(x, y);
  (void)_m_punpckhbw(x, y);
  (void)_m_punpckldq(x, y);
  (void)_m_punpckhdq(x, y);
  (void)_mm_cvt_pi2ps(_mm_setzero_ps(), x);
  (void)_mm_cvt_ps2pi(_mm_setzero_ps());
  int i = 42;
  (void)_m_from_int(i);
  (void)_m_to_int(x);
  (void)_m_pand(x, y);
  (void)_m_packuswb(x, y);
  (void)_m_pavgb(x, y);
  (void)_m_pshufw(x, 0);
  (void)_m_psllqi(x, i);
  (void)_m_psrlqi(x, i);
  (void)_m_psllwi(x, i);
  (void)_m_psrlwi(x, i);
  (void)_m_por(x, y);
  (void)_m_pxor(x, y);
  (void)_m_pmullw(x, y);
  (void)_m_pmulhw(x, y);
  (void)_m_paddw(x, y);
  (void)_m_pmaddwd(x, y);
  (void)_m_paddusb(x, y);
  (void)_m_paddd(x, y);
  (void)_m_psubw(x, y);
  (void)_m_psubusb(x, y);
  (void)_m_pmaxsw(x, y);
  (void)_m_pminub(x, y);
  (void)_m_pmaxub(x, y);
  (void)_m_pcmpeqb(x, y);
  (void)_m_pcmpgtb(x, y);
  _m_empty();
}

% gcc -Wall -Werror -march=native -fsyntax-only x.c                             

% ./bin/clang -Werror -march=native -fsyntax-only -ferror-limit=999 x.c
x.c:7:9: error: implicit declaration of function '_m_punpcklwd' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_punpcklwd(x, y);
        ^
x.c:8:9: error: implicit declaration of function '_m_punpckhwd' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_punpckhwd(x, y);
        ^
x.c:9:9: error: implicit declaration of function '_m_punpcklbw' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_punpcklbw(x, y);
        ^
x.c:10:9: error: implicit declaration of function '_m_punpckhbw' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_punpckhbw(x, y);
        ^
x.c:11:9: error: implicit declaration of function '_m_punpckldq' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_punpckldq(x, y);
        ^
x.c:12:9: error: implicit declaration of function '_m_punpckhdq' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_punpckhdq(x, y);
        ^
x.c:13:9: error: implicit declaration of function '_mm_cvt_pi2ps' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_mm_cvt_pi2ps(_mm_setzero_ps(), x);
        ^
x.c:14:9: error: implicit declaration of function '_mm_cvt_ps2pi' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_mm_cvt_ps2pi(_mm_setzero_ps());
        ^
x.c:16:9: error: implicit declaration of function '_m_from_int' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_from_int(i);
        ^
x.c:17:9: error: implicit declaration of function '_m_to_int' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_to_int(x);
        ^
x.c:18:9: error: implicit declaration of function '_m_pand' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pand(x, y);
        ^
x.c:19:9: error: implicit declaration of function '_m_packuswb' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_packuswb(x, y);
        ^
x.c:20:9: error: implicit declaration of function '_m_pavgb' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pavgb(x, y);
        ^
x.c:21:9: error: implicit declaration of function '_m_pshufw' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pshufw(x, 0);
        ^
x.c:22:9: error: implicit declaration of function '_m_psllqi' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_psllqi(x, i);
        ^
x.c:23:9: error: implicit declaration of function '_m_psrlqi' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_psrlqi(x, i);
        ^
x.c:24:9: error: implicit declaration of function '_m_psllwi' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_psllwi(x, i);
        ^
x.c:25:9: error: implicit declaration of function '_m_psrlwi' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_psrlwi(x, i);
        ^
x.c:26:9: error: implicit declaration of function '_m_por' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_por(x, y);
        ^
x.c:27:9: error: implicit declaration of function '_m_pxor' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pxor(x, y);
        ^
x.c:28:9: error: implicit declaration of function '_m_pmullw' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pmullw(x, y);
        ^
x.c:29:9: error: implicit declaration of function '_m_pmulhw' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pmulhw(x, y);
        ^
x.c:30:9: error: implicit declaration of function '_m_paddw' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_paddw(x, y);
        ^
x.c:31:9: error: implicit declaration of function '_m_pmaddwd' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_pmaddwd(x, y);
        ^
x.c:32:9: error: implicit declaration of function '_m_paddusb' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_paddusb(x, y);
        ^
x.c:33:9: error: implicit declaration of function '_m_paddd' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_paddd(x, y);
        ^
x.c:34:9: error: implicit declaration of function '_m_psubw' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_psubw(x, y);
        ^
x.c:35:9: error: implicit declaration of function '_m_psubusb' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_psubusb(x, y);
        ^
x.c:36:9: error: implicit declaration of function '_m_pmaxsw' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pmaxsw(x, y);
        ^
x.c:37:9: error: implicit declaration of function '_m_pminub' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pminub(x, y);
        ^
x.c:38:9: error: implicit declaration of function '_m_pmaxub' is invalid in C99
[-Wimplicit-function-declaration]
  (void)_m_pmaxub(x, y);
        ^
x.c:39:9: error: implicit declaration of function '_m_pcmpeqb' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_pcmpeqb(x, y);
        ^
x.c:40:9: error: implicit declaration of function '_m_pcmpgtb' is invalid in
C99 [-Wimplicit-function-declaration]
  (void)_m_pcmpgtb(x, y);
        ^
x.c:41:3: error: implicit declaration of function '_m_empty' is invalid in C99
[-Wimplicit-function-declaration]
  _m_empty();
  ^
34 errors generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 02:59:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 02:59:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7473] [MIPS] llc with mips target is incorrect for
 the Select Instruction
In-Reply-To: <bug-7473-206@http.llvm.org/bugs/>
References: <bug-7473-206@http.llvm.org/bugs/>
Message-ID: <20100720075927.191BC2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7473

Bruno Cardoso Lopes <bruno.cardoso at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #12 from Bruno Cardoso Lopes <bruno.cardoso at gmail.com> 2010-07-20 02:59:26 CDT ---
Fixed in r108816

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 03:37:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 03:37:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7174] [MIPS] Can't assemble LLVM-GCC's assembly
 with multiple case statements
In-Reply-To: <bug-7174-206@http.llvm.org/bugs/>
References: <bug-7174-206@http.llvm.org/bugs/>
Message-ID: <20100720083745.4DB7D2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7174

Bruno Cardoso Lopes <bruno.cardoso at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #11 from Bruno Cardoso Lopes <bruno.cardoso at gmail.com> 2010-07-20 03:37:44 CDT ---
Fixed in r108820

Thanks

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 03:45:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 03:45:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5828] Fails to build from source on mips
In-Reply-To: <bug-5828-206@http.llvm.org/bugs/>
References: <bug-5828-206@http.llvm.org/bugs/>
Message-ID: <20100720084505.DC5B32A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5828

Bruno Cardoso Lopes <bruno.cardoso at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #12 from Bruno Cardoso Lopes <bruno.cardoso at gmail.com> 2010-07-20 03:45:04 CDT ---
Applied in r108821! :)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 06:37:49 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 06:37:49 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7675] New: Assertion
 `(isa<RecordType>(Ex->getType().getDesugaredType()) ||
 isa<ArrayType>(Ex->getType().getDesugaredType())) && "Other kinds of
 expressions with non-aggregate/union/class types" " do not have lvalues."'
 failed
Message-ID: <bug-7675-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7675

           Summary: Assertion
                    `(isa<RecordType>(Ex->getType().getDesugaredType()) ||
                    isa<ArrayType>(Ex->getType().getDesugaredType())) &&
                    "Other kinds of expressions with
                    non-aggregate/union/class types" " do not have
                    lvalues."' failed
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: holgerschurig at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5246)
 --> (http://llvm.org/bugs/attachment.cgi?id=5246)
itm.cpp

$ clang++ --version
clang version 2.8 (trunk 108821)
Target: i386-pc-linux-gnu
Thread model: posix
$ clang++ --analyze itm.cpp 
clang: /usr/src/llvm/src.llvm/tools/clang/lib/Checker/GRExprEngine.cpp:1075:
void clang::GRExprEngine::VisitLValue(const clang::Expr*, clang::ExplodedNode*,
clang::ExplodedNodeSet&): Assertion
`(isa<RecordType>(Ex->getType().getDesugaredType()) ||
isa<ArrayType>(Ex->getType().getDesugaredType())) && "Other kinds of
expressions with non-aggregate/union/class types" " do not have lvalues."'
failed.
0  clang     0x098d87dc
1  clang     0x098d8674
2            0xb775b400 __kernel_sigreturn + 0
3  libc.so.6 0xb74cdb82 abort + 386
4  libc.so.6 0xb74c38b8 __assert_fail + 248
5  clang     0x08e8b1f9 clang::GRExprEngine::VisitLValue(clang::Expr const*,
clang::ExplodedNode*, clang::ExplodedNodeSet&) + 1629
6  clang     0x08e8ed72 clang::GRExprEngine::VisitCall(clang::CallExpr const*,
clang::ExplodedNode*, clang::ConstExprIterator, clang::ConstExprIterator,
clang::ExplodedNodeSet&, bool) + 592
7  clang     0x08e8a32e clang::GRExprEngine::Visit(clang::Stmt const*,
clang::ExplodedNode*, clang::ExplodedNodeSet&) + 1372
8  clang     0x08e89c80 clang::GRExprEngine::ProcessStmt(clang::CFGElement,
clang::GRStmtNodeBuilder&) + 1532
9  clang     0x08e81b42 clang::GRCoreEngine::ProcessStmt(clang::CFGElement,
clang::GRStmtNodeBuilder&) + 42
10 clang     0x08e829fb clang::GRCoreEngine::HandlePostStmt(clang::PostStmt
const&, clang::CFGBlock const*, unsigned int, clang::ExplodedNode*) + 247
11 clang     0x08e82045
clang::GRCoreEngine::ExecuteWorkList(clang::LocationContext const*, unsigned
int) + 1025
12 clang     0x08e1a53a
clang::GRExprEngine::ExecuteWorkList(clang::LocationContext const*, unsigned
int) + 46
13 clang     0x08e1767f
14 clang     0x08e1776e
15 clang     0x08e177f4
16 clang     0x08e173e7
17 clang     0x08e16e38
18 clang     0x08c2f073 clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 727
19 clang     0x089dd853 clang::ASTFrontendAction::ExecuteAction() + 271
20 clang     0x089dd4c4 clang::FrontendAction::Execute() + 316
21 clang     0x089c8647
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 779
22 clang     0x0898a212 cc1_main(char const**, char const**, char const*,
void*) + 1721
23 clang     0x08992f5d main + 495
24 libc.so.6 0xb74b6c76 __libc_start_main + 230
25 clang     0x08988d81
Stack dump:
0.      Program arguments: /usr/src/llvm/dist/bin/clang -cc1 -triple
i386-pc-linux-gnu -analyze -disable-free -main-file-name itm.cpp
-analyzer-store=region -analyzer-opt-analyze-nested-blocks
-analyzer-check-dead-stores -analyzer-check-objc-mem -analyzer-eagerly-assume
-analyzer-check-objc-methodsigs -analyzer-check-objc-unused-ivars
-analyzer-output plist -w -mrelocation-model static -mdisable-fp-elim
-mconstructor-aliases -target-cpu pentium4 -resource-dir
/usr/src/llvm/dist/lib/clang/2.8 -ferror-limit 19 -fmessage-length 117
-fexceptions -fgnu-runtime -fdiagnostics-show-option -fcolor-diagnostics -o
itm.plist -x c++ itm.cpp 
1.      <eof> parser at end of file
2.      itm.cpp:237:9: Error evaluating statement
3.      itm.cpp:237:9: Error evaluating statement
4.      itm.cpp:237:15: Error evaluating statement
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)


Please note that the actual compilation works fine.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 09:04:07 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 09:04:07 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7670] PCHWriterDecl.cpp:848: void
 clang::PCHDeclWriter::VisitClassTemplateDecl(clang::ClassTemplateDecl*):
 Assertion `IsKeptInFoldingSet(&*I)' failed.
In-Reply-To: <bug-7670-206@http.llvm.org/bugs/>
References: <bug-7670-206@http.llvm.org/bugs/>
Message-ID: <20100720140407.AF33D2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7670

Argyrios Kyrtzidis <akyrtzi at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Argyrios Kyrtzidis <akyrtzi at gmail.com> 2010-07-20 09:04:07 CDT ---
Fixed at r108832. Thanks for the report!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 09:05:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 09:05:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6079] Crash: LHS in member-access is not of class
 type and member name is template-id
In-Reply-To: <bug-6079-206@http.llvm.org/bugs/>
References: <bug-6079-206@http.llvm.org/bugs/>
Message-ID: <20100720140503.420DB2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6079

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-20 09:05:03 CDT ---
It's working fine for me, too. Closing.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 09:05:36 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 09:05:36 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7094] Warns "this partial specialization will never
 be used", but still (correctly) uses it.
In-Reply-To: <bug-7094-206@http.llvm.org/bugs/>
References: <bug-7094-206@http.llvm.org/bugs/>
Message-ID: <20100720140536.476ED2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7094

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-20 09:05:36 CDT ---
Yep, closing!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 10:50:57 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 10:50:57 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7652] llc -O0 infinite loop on atomic operations
In-Reply-To: <bug-7652-206@http.llvm.org/bugs/>
References: <bug-7652-206@http.llvm.org/bugs/>
Message-ID: <20100720155057.6C8A62A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7652

Duncan Sands <baldrick at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Duncan Sands <baldrick at free.fr> 2010-07-20 10:50:56 CDT ---
Yes, it seems to be fixed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 13:59:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 13:59:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6889] memory unsafety crash bug
In-Reply-To: <bug-6889-206@http.llvm.org/bugs/>
References: <bug-6889-206@http.llvm.org/bugs/>
Message-ID: <20100720185928.271792A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6889

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|WORKSFORME                  |

--- Comment #2 from Dan Gohman <gohman at apple.com> 2010-07-20 13:59:27 CDT ---
This still fails for me.  valgrind shows a use after free within
AliasSetTracker/LICM.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 15:51:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 15:51:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6687] operator new should check for integer
	overflows
In-Reply-To: <bug-6687-206@http.llvm.org/bugs/>
References: <bug-6687-206@http.llvm.org/bugs/>
Message-ID: <20100720205122.F3E412A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6687

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-20 15:51:22 CDT ---
Fixed in r108915

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 20 16:51:11 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 20 Jul 2010 16:51:11 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7680] New: fixit hints don't obey access control
Message-ID: <bug-7680-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7680

           Summary: fixit hints don't obey access control
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


I accidentally called the getWidth() member of APInt instead of getBitWidth()
and got this:


CGExprCXX.cpp:445:18: error: no member named 'getWidth' in 'llvm::APInt'; did
you mean 'BitWidth'?
    NEC.zext(NEC.getWidth()*2);
                 ^~~~~~~~
                 BitWidth
In file included from CGExprCXX.cpp:14:
In file included from
/Users/sabre/llvm/tools/clang/lib/CodeGen/CodeGenFunction.h:17:
In file included from
/Users/sabre/llvm/tools/clang/lib/CodeGen/../../include/clang/AST/Type.h:25:
In file included from /Users/sabre/llvm/include/llvm/ADT/APSInt.h:18:
/Users/sabre/llvm/include/llvm/ADT/APInt.h:73:12: note: 'BitWidth' declared
here
  unsigned BitWidth;      ///< The number of bits in this APInt.
           ^
CGExprCXX.cpp:445:18: error: 'BitWidth' is a private member of 'llvm::APInt'
    NEC.zext(NEC.getWidth()*2);
                 ^

In file included from CGExprCXX.cpp:14:
In file included from
/Users/sabre/llvm/tools/clang/lib/CodeGen/CodeGenFunction.h:17:
In file included from
/Users/sabre/llvm/tools/clang/lib/CodeGen/../../include/clang/AST/Type.h:25:
In file included from /Users/sabre/llvm/include/llvm/ADT/APSInt.h:18:
/Users/sabre/llvm/include/llvm/ADT/APInt.h:73:12: note: declared private here
  unsigned BitWidth;      ///< The number of bits in this APInt.
           ^
CGExprCXX.cpp:445:26: error: called object type 'unsigned int' is not a
function or function pointer
    NEC.zext(NEC.getWidth()*2);
             ~~~~~~~~~~~~^


This sucks for two reasons: it suggested a field when I'm calling a function,
and it suggested a private member that I definitely couldn't call.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 02:10:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 02:10:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7681] New: confusing error with incomplete type
Message-ID: <bug-7681-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7681

           Summary: confusing error with incomplete type
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Missing a #include caused clang to generate a super confusing error:

$ cat t.cc 

  template <typename PT1, typename PT2>
  class PointerUnion;

void foo(PointerUnion<int*, float*> &Result) {
  int *P = 0;
  Result = P;
}
$ clang t.cc 
t.cc:7:10: error: no viable overloaded '='
  Result = P;
  ~~~~~~ ^ ~
1 error generated.


It should tell me that result is incomplete.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 05:50:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 05:50:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7683] New: PBQP register allocator segfaults on
	null pointer
Message-ID: <bug-7683-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7683

           Summary: PBQP register allocator segfaults on null pointer
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: gergo at complang.tuwien.ac.at
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5249)
 --> (http://llvm.org/bugs/attachment.cgi?id=5249)
Bitcode file containing a function on which the PBQP register allocator
segfaults

On the attached bitcode file (edge_draw.ll), the PBQP register allocator
segfaults on encountering a null pointer.

To reproduce, run: llc -regalloc=pbqp edge_draw.ll

The stack trace from GDB is:

#0  0x0000000000f31b9a in llvm::IndexListEntry::getIndex (this=0x0)
    at /raid0/gergo/svn_llvm/llvm/include/llvm/CodeGen/SlotIndexes.h:82
#1  0x0000000000f31df2 in llvm::SlotIndex::getIndex (this=0x1b2daa0)
    at /raid0/gergo/svn_llvm/llvm/include/llvm/CodeGen/SlotIndexes.h:146
#2  0x0000000000f31f5a in llvm::SlotIndex::operator<= (this=0x1b2daa0, other=
        {static PHI_BIT = 4, lie = {Value = 28464610}})
    at /raid0/gergo/svn_llvm/llvm/include/llvm/CodeGen/SlotIndexes.h:218
#3  0x0000000000ff5fb8 in llvm::LiveInterval::overlapsFrom (this=0x1b2bcc0, 
    other=@0x1b2da80, StartPos=0x1b2daa0)
    at /raid0/gergo/svn_llvm/llvm/lib/CodeGen/LiveInterval.cpp:129
#4  0x0000000000f4ec4d in llvm::LiveInterval::overlaps (this=0x1b2bcc0, 
    other=@0x1b2da80)
    at /raid0/gergo/svn_llvm/llvm/include/llvm/CodeGen/LiveInterval.h:468
#5  0x0000000000f4038b in constructPBQPProblem (this=0x1af2920)
    at /raid0/gergo/svn_llvm/llvm/lib/CodeGen/RegAllocPBQP.cpp:604
#6  0x0000000000f40c11 in runOnMachineFunction (this=0x1af2920, MF=@0x1aca490)
    at /raid0/gergo/svn_llvm/llvm/lib/CodeGen/RegAllocPBQP.cpp:892
#7  0x0000000000ef14da in llvm::MachineFunctionPass::runOnFunction (
    this=0x1af2920, F=@0x1ac7b30)
    at /raid0/gergo/svn_llvm/llvm/lib/CodeGen/MachineFunctionPass.cpp:33
#8  0x00000000011e6ee1 in llvm::FPPassManager::runOnFunction (this=0x1ac4800, 
    F=@0x1ac7b30) at /raid0/gergo/svn_llvm/llvm/lib/VMCore/PassManager.cpp:1433
#9  0x00000000011e70b7 in llvm::FPPassManager::runOnModule (this=0x1ac4800, 
    M=@0x1ab6aa0) at /raid0/gergo/svn_llvm/llvm/lib/VMCore/PassManager.cpp:1453
#10 0x00000000011e6ba7 in llvm::MPPassManager::runOnModule (this=0x1ad41e0, 
    M=@0x1ab6aa0) at /raid0/gergo/svn_llvm/llvm/lib/VMCore/PassManager.cpp:1507
#11 0x00000000011e8343 in llvm::PassManagerImpl::run (this=0x1ac5890, 
    M=@0x1ab6aa0) at /raid0/gergo/svn_llvm/llvm/lib/VMCore/PassManager.cpp:1588
#12 0x00000000011e83a5 in llvm::PassManager::run (this=0x7fffffffe720, 
    M=@0x1ab6aa0) at /raid0/gergo/svn_llvm/llvm/lib/VMCore/PassManager.cpp:1631
#13 0x00000000009b06b0 in main (argc=3, argv=0x7fffffffe948)
    at /raid0/gergo/svn_llvm/llvm/tools/llc/llc.cpp:343

Apparently, some SlotIndex object does not have a valid index list entry.

The code is derived from the automotive-susan program in the Mibench benchmark
suite. Starting from susan.c from Mibench, the following commands were
performed:
clang -S -emit-llvm susan.c -o foo.ll
opt -mem2reg foo.ll -S -o foo_opt.ll
llvm-extract -func=edge_draw foo_opt.ll -o edge_draw.ll
llc -regalloc=pbqp edge_draw.ll

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 09:01:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 09:01:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7684] New: "Cannot get layout of forward
 declarations" on invalid code
Message-ID: <bug-7684-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7684

           Summary: "Cannot get layout of forward declarations" on invalid
                    code
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5250)
 --> (http://llvm.org/bugs/attachment.cgi?id=5250)
Forward example

The attached code fails with the following backtrace. This code was reduced
from trying to use boost::mpi without mpi installed.

/Users/caj/Dropbox/forward.cc:3:8: error: field has incomplete type 'status'
{  _T1 first; };
       ^
/Users/caj/Dropbox/forward.cc:11:13: note: in instantiation of template class
      'pair<status>' requested here
  int data[ sizeof(T) ];
            ^
/Users/caj/Dropbox/forward.cc:16:42: note: in instantiation of template class
      'make_reference_content<pair<status> >' requested here
{  make_reference_content<pair<status> > result; };
                                         ^
/Users/caj/Dropbox/forward.cc:15:8: note: definition of 'status' is not
complete
      until the closing '}'
struct status
       ^
Assertion failed: (D && "Cannot get layout of forward declarations!"), function
getASTRecordLayout, file RecordLayoutBuilder.cpp, line 1461.
0  clang             0x00000001010fbd78 PrintStackTrace(void*) + 38
1  clang             0x00000001010fc286 SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff8398c35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff839310aa tiny_malloc_from_free_list + 1196
4  libSystem.B.dylib 0x00007fff83a079b4 __pthread_markcancel + 0
5  clang             0x0000000100657ebd
clang::ASTContext::getASTRecordLayout(clang::RecordDecl const*) + 97
6  clang             0x0000000100657d83 (anonymous
namespace)::EmptySubobjectMap::ComputeEmptySubobjectSizes() + 429
7  clang             0x0000000100657e5a (anonymous
namespace)::EmptySubobjectMap::EmptySubobjectMap(clang::ASTContext&,
clang::CXXRecordDecl const*) + 94
8  clang             0x0000000100657f24
clang::ASTContext::getASTRecordLayout(clang::RecordDecl const*) + 200
9  clang             0x0000000100657d83 (anonymous
namespace)::EmptySubobjectMap::ComputeEmptySubobjectSizes() + 429
10 clang             0x0000000100657e5a (anonymous
namespace)::EmptySubobjectMap::EmptySubobjectMap(clang::ASTContext&,
clang::CXXRecordDecl const*) + 94
11 clang             0x0000000100657f24
clang::ASTContext::getASTRecordLayout(clang::RecordDecl const*) + 200
12 clang             0x00000001005b0a1f
clang::ASTContext::getTypeInfo(clang::Type const*) + 2563
13 clang             0x00000001005b0de8
clang::ASTContext::getTypeInfo(clang::Type const*) + 3532
14 clang             0x000000010062bde9
clang::ASTContext::getTypeInfo(clang::QualType) + 37
15 clang             0x00000001005cc949
clang::ASTContext::getTypeSize(clang::QualType) + 29
16 clang             0x00000001005b43b8
clang::ASTContext::getTypeSizeInChars(clang::QualType) + 30
17 clang             0x000000010063cf48 (anonymous
namespace)::IntExprEvaluator::VisitSizeOfAlignOfExpr(clang::SizeOfAlignOfExpr
const*) + 374
18 clang             0x0000000100641f8e clang::StmtVisitor<(anonymous
namespace)::IntExprEvaluator, bool>::Visit(clang::Stmt*) + 4152
19 clang             0x0000000100644b40
clang::Expr::Evaluate(clang::Expr::EvalResult&, clang::ASTContext&) const + 258
20 clang             0x00000001006461af
clang::Expr::isIntegerConstantExpr(llvm::APSInt&, clang::ASTContext&,
clang::SourceLocation*, bool) const + 151
21 clang             0x00000001003cf361
clang::Sema::CheckTemplateArgument(clang::NonTypeTemplateParmDecl*,
clang::QualType, clang::Expr*&, clang::TemplateArgument&,
clang::Sema::CheckTemplateArgumentKind) + 673
22 clang             0x00000001003d2377
clang::Sema::CheckTemplateArgument(clang::NamedDecl*,
clang::TemplateArgumentLoc const&, clang::TemplateDecl*, clang::SourceLocation,
clang::SourceLocation, clang::TemplateArgumentListBuilder&,
clang::Sema::CheckTemplateArgumentKind) + 875
23 clang             0x00000001003d2eb9
clang::Sema::CheckTemplateArgumentList(clang::TemplateDecl*,
clang::SourceLocation, clang::TemplateArgumentListInfo const&, bool,
clang::TemplateArgumentListBuilder&) + 1133
24 clang             0x00000001003d3cc8
clang::Sema::CheckTemplateIdType(clang::TemplateName, clang::SourceLocation,
clang::TemplateArgumentListInfo const&) + 206
25 clang             0x00000001003f6cc2 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::RebuildTemplateSpecializationType(clang::TemplateName,
clang::SourceLocation, clang::TemplateArgumentListInfo const&) + 46
26 clang             0x0000000100411476 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformTemplateSpecializationType(clang::TypeLocBuilder&,
clang::TemplateSpecializationTypeLoc, clang::QualType) + 522
27 clang             0x0000000100419003 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformType(clang::TypeLocBuilder&,
clang::TypeLoc, clang::QualType) + 1789
28 clang             0x00000001003fd0e9 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformType(clang::TypeSourceInfo*,
clang::QualType) + 229
29 clang             0x00000001003fd580 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformType(clang::QualType,
clang::QualType) + 178
30 clang             0x00000001003fea92 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformNestedNameSpecifier(clang::NestedNameSpecifier*,
clang::SourceRange, clang::QualType, clang::NamedDecl*) + 934
31 clang             0x00000001003fec35 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformDependentNameType(clang::TypeLocBuilder&,
clang::DependentNameTypeLoc, clang::QualType) + 137
32 clang             0x0000000100419073 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformType(clang::TypeLocBuilder&,
clang::TypeLoc, clang::QualType) + 1901
33 clang             0x00000001003fd0e9 clang::TreeTransform<(anonymous
namespace)::TemplateInstantiator>::TransformType(clang::TypeSourceInfo*,
clang::QualType) + 229
34 clang             0x000000010040094f
clang::Sema::SubstType(clang::TypeSourceInfo*,
clang::MultiLevelTemplateArgumentList const&, clang::SourceLocation,
clang::DeclarationName) + 279
35 clang             0x000000010042aa07 (anonymous
namespace)::TemplateDeclInstantiator::VisitFieldDecl(clang::FieldDecl*) + 215
36 clang             0x000000010042bdec clang::DeclVisitor<(anonymous
namespace)::TemplateDeclInstantiator, clang::Decl*>::Visit(clang::Decl*) + 980
37 clang             0x000000010042d2a9 clang::Sema::SubstDecl(clang::Decl*,
clang::DeclContext*, clang::MultiLevelTemplateArgumentList const&) + 81
38 clang             0x00000001003ffc2a
clang::Sema::InstantiateClass(clang::SourceLocation, clang::CXXRecordDecl*,
clang::CXXRecordDecl*, clang::MultiLevelTemplateArgumentList const&,
clang::TemplateSpecializationKind, bool) + 1044
39 clang             0x000000010041cb27
clang::Sema::InstantiateClassTemplateSpecialization(clang::SourceLocation,
clang::ClassTemplateSpecializationDecl*, clang::TemplateSpecializationKind,
bool) + 1723
40 clang             0x000000010042fe24
clang::Sema::RequireCompleteType(clang::SourceLocation, clang::QualType,
clang::PartialDiagnostic const&, std::pair<clang::SourceLocation,
clang::PartialDiagnostic>) + 288
41 clang             0x00000001004301d4
clang::Sema::RequireCompleteType(clang::SourceLocation, clang::QualType,
unsigned int) + 142
42 clang             0x00000001002b1fef
clang::Sema::CheckFieldDecl(clang::DeclarationName, clang::QualType,
clang::TypeSourceInfo*, clang::RecordDecl*, clang::SourceLocation, bool,
clang::Expr*, clang::SourceLocation, clang::AccessSpecifier, clang::NamedDecl*,
clang::Declarator*) + 233
43 clang             0x00000001002bd750 clang::Sema::HandleField(clang::Scope*,
clang::RecordDecl*, clang::SourceLocation, clang::Declarator&, clang::Expr*,
clang::AccessSpecifier) + 750
44 clang             0x00000001002f03c4
clang::Sema::ActOnCXXMemberDeclarator(clang::Scope*, clang::AccessSpecifier,
clang::Declarator&,
clang::ASTMultiPtr<&(clang::ActionBase::DeleteTemplateParams(void*))>, void*,
void*, bool, bool) + 918
45 clang             0x00000001006a697d
clang::Parser::ParseCXXClassMemberDeclaration(clang::AccessSpecifier,
clang::Parser::ParsedTemplateInfo const&,
clang::Parser::ParsingDeclRAIIObject*) + 3617
46 clang             0x00000001006a7207
clang::Parser::ParseCXXMemberSpecification(clang::SourceLocation, unsigned int,
clang::OpaquePtr<0>) + 1349
47 clang             0x00000001006a890e
clang::Parser::ParseClassSpecifier(clang::tok::TokenKind,
clang::SourceLocation, clang::DeclSpec&, clang::Parser::ParsedTemplateInfo
const&, clang::AccessSpecifier, bool) + 5226
48 clang             0x000000010069bde9
clang::Parser::ParseDeclarationSpecifiers(clang::DeclSpec&,
clang::Parser::ParsedTemplateInfo const&, clang::AccessSpecifier,
clang::Parser::DeclSpecContext) + 6301
49 clang             0x00000001006dc29c
clang::Parser::ParseDeclarationOrFunctionDefinition(clang::Parser::ParsingDeclSpec&,
clang::AttributeList*, clang::AccessSpecifier) + 120
50 clang             0x00000001006dc6bf
clang::Parser::ParseDeclarationOrFunctionDefinition(clang::AttributeList*,
clang::AccessSpecifier) + 83
51 clang             0x00000001006de529
clang::Parser::ParseExternalDeclaration(clang::CXX0XAttributeList) + 2235
52 clang             0x00000001006de673
clang::Parser::ParseTopLevelDecl(clang::OpaquePtr<1>&) + 247
53 clang             0x0000000100265337 clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 456
54 clang             0x0000000100079062
clang::ASTFrontendAction::ExecuteAction() + 254
55 clang             0x0000000100226940 clang::CodeGenAction::ExecuteAction() +
826
56 clang             0x0000000100079170 clang::FrontendAction::Execute() + 256
57 clang             0x000000010005c380
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 710
58 clang             0x00000001000280d6 cc1_main(char const**, char const**,
char const*, void*) + 1762
59 clang             0x000000010002f836 main + 450
60 clang             0x0000000100026fd4 start + 52
61 clang             0x0000000000000021 start + 4294807681
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free -main-file-name
forward.cc -pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables
-target-cpu core2 -resource-dir /usr/local/lib/clang/2.8 -ferror-limit 19
-fmessage-length 80 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/M8/M86fCFVgGQ0r-V2oY5lLC++++TI/-Tmp-/cc-eG7Z8a.o -x c++
/Users/caj/Dropbox/forward.cc 
1.    /Users/caj/Dropbox/forward.cc:16:48: current parser token ';'
2.    /Users/caj/Dropbox/forward.cc:15:1: parsing struct/union/class body
'status'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 09:18:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 09:18:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7685] New: Stack dump without assertion on invalid
	code
Message-ID: <bug-7685-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7685

           Summary: Stack dump without assertion on invalid code
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5251)
 --> (http://llvm.org/bugs/attachment.cgi?id=5251)
test case without assertion

Reduced from a broken boost test. This is a stack dump without an assertion.
Reduced from a broken boost testcase
(libs/spirit/repository/example/karma/calc2_ast_dump_sr.cpp).

noAssertion.cc:2:11: error: parse error
template <1> call()
          ^
noAssertion.cc:2:14: error: C++ requires a type specifier for all declarations
template <1> call()
             ^
noAssertion.cc:2:14: error: no function template matches function template
      specialization 'call'
noAssertion.cc:4:5: error: template argument for non-type template parameter
      must be an expression
map<int>()
    ^~~
noAssertion.cc:1:14: note: template parameter is declared here
template <int> struct map ;
             ^
0  clang             0x00000001010fbd78 PrintStackTrace(void*) + 38
1  clang             0x00000001010fc286 SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff8398c35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff5fbf9700 _sigtramp + 3693532096
4  clang             0x000000010043c92d
clang::LocInfoType::LocInfoType(clang::QualType, clang::TypeSourceInfo*) + 37
5  clang             0x0000000100432001
clang::Sema::CreateLocInfoType(clang::QualType, clang::TypeSourceInfo*) + 93
6  clang             0x00000001004386c9
clang::Sema::ActOnTypeName(clang::Scope*, clang::Declarator&) + 433
7  clang             0x00000001006b70d5
clang::Parser::ParseCXXTypeConstructExpression(clang::DeclSpec const&) + 119
8  clang             0x00000001006b19c2
clang::Parser::ParseCastExpression(bool, bool, bool&, void*) + 8128
9  clang             0x00000001006b2601
clang::Parser::ParseCastExpression(bool, bool, void*) + 97
10 clang             0x00000001006ace87
clang::Parser::ParseAssignmentExpression() + 241
11 clang             0x00000001006ad40e clang::Parser::ParseExpression() + 42
12 clang             0x00000001006ce6f8
clang::Parser::ParseStatementOrDeclaration(bool) + 1662
13 clang             0x00000001006d2c5a
clang::Parser::ParseCompoundStatementBody(bool) + 220
14 clang             0x00000001006d3285
clang::Parser::ParseFunctionStatementBody(clang::OpaquePtr<0>) + 193
15 clang             0x00000001006dd679
clang::Parser::ParseFunctionDefinition(clang::Parser::ParsingDeclarator&,
clang::Parser::ParsedTemplateInfo const&) + 1237
16 clang             0x00000001006d7782
clang::Parser::ParseSingleDeclarationAfterTemplate(unsigned int,
clang::Parser::ParsedTemplateInfo const&,
clang::Parser::ParsingDeclRAIIObject&, clang::SourceLocation&,
clang::AccessSpecifier) + 1214
17 clang             0x00000001006d7ca9
clang::Parser::ParseTemplateDeclarationOrSpecialization(unsigned int,
clang::SourceLocation&, clang::AccessSpecifier) + 999
18 clang             0x00000001006d7d90
clang::Parser::ParseDeclarationStartingWithTemplate(unsigned int,
clang::SourceLocation&, clang::AccessSpecifier) + 158
19 clang             0x000000010069ff20
clang::Parser::ParseDeclaration(unsigned int, clang::SourceLocation&,
clang::CXX0XAttributeList) + 368
20 clang             0x00000001006de445
clang::Parser::ParseExternalDeclaration(clang::CXX0XAttributeList) + 2007
21 clang             0x00000001006de673
clang::Parser::ParseTopLevelDecl(clang::OpaquePtr<1>&) + 247
22 clang             0x0000000100265337 clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 456
23 clang             0x0000000100079062
clang::ASTFrontendAction::ExecuteAction() + 254
24 clang             0x0000000100226940 clang::CodeGenAction::ExecuteAction() +
826
25 clang             0x0000000100079170 clang::FrontendAction::Execute() + 256
26 clang             0x000000010005c380
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 710
27 clang             0x00000001000280d6 cc1_main(char const**, char const**,
char const*, void*) + 1762
28 clang             0x000000010002f836 main + 450
29 clang             0x0000000100026fd4 start + 52
30 clang             0x0000000000000021 start + 4294807681
Stack dump:
0.    Program arguments: /usr/local/bin/clang -cc1 -triple
x86_64-apple-darwin10.0.0 -emit-obj -mrelax-all -disable-free -main-file-name
noAssertion.cc -pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables
-target-cpu core2 -resource-dir /usr/local/lib/clang/2.8 -ferror-limit 19
-fmessage-length 80 -stack-protector 1 -fblocks -fexceptions
-fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/M8/M86fCFVgGQ0r-V2oY5lLC++++TI/-Tmp-/cc-kM5PwQ.o -x c++
noAssertion.cc 
1.    noAssertion.cc:4:9: current parser token '('
2.    noAssertion.cc:3:1: parsing function body 'call'
3.    noAssertion.cc:3:1: in compound statement ('{}')
clang: error: clang frontend command failed due to signal 11 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 12:38:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 12:38:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7686] New: Segfault in codegen
Message-ID: <bug-7686-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7686

           Summary: Segfault in codegen
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: collinwinter at google.com
                CC: chandlerc at gmail.com, llvmbugs at cs.uiuc.edu


Created an attachment (id=5252)
 --> (http://llvm.org/bugs/attachment.cgi?id=5252)
Reduced test case

As of r109013, Clang segfaults when generating code for the attached test case.
gcc compiles the test case with no issues.

$ /g/users/collinwinter/llvm/obj/Release/bin/clang -x c++ -c varzclient.pic.ii
varzclient.pic.ii:16:1: warning: control reaches end of non-void function
[-Wreturn-type]
}
^
0  clang           0x00000000018e6b5f
1  clang           0x00000000018e741f
2  libpthread.so.0 0x00002aaaaabcf700
3  clang           0x000000000180517a llvm::Instruction::Instruction(llvm::Type
const*, unsigned int, llvm::Use*, unsigned int, llvm::Instruction*) + 138
4  clang           0x0000000001810ad9 llvm::StoreInst::StoreInst(llvm::Value*,
llvm::Value*, llvm::Instruction*) + 89
5  clang           0x00000000009c0355
6  clang           0x00000000009c0a73
clang::CodeGen::CodeGenFunction::PopCleanupBlock() + 531
7  clang           0x00000000009c0cf8
clang::CodeGen::CodeGenFunction::PopCleanupBlocks(clang::CodeGen::EHScopeStack::stable_iterator)
+ 56
8  clang           0x000000000099e445
clang::CodeGen::CodeGenFunction::EmitCompoundStmt(clang::CompoundStmt const&,
bool, llvm::Value*, bool) + 389
9  clang           0x000000000099e5ad
clang::CodeGen::CodeGenFunction::EmitSimpleStmt(clang::Stmt const*) + 189
10 clang           0x000000000099c4b3
clang::CodeGen::CodeGenFunction::EmitStmt(clang::Stmt const*) + 35
11 clang           0x000000000099e189
clang::CodeGen::CodeGenFunction::EmitIfStmt(clang::IfStmt const&) + 441
12 clang           0x000000000099c69b
clang::CodeGen::CodeGenFunction::EmitStmt(clang::Stmt const*) + 523
13 clang           0x000000000099e395
clang::CodeGen::CodeGenFunction::EmitCompoundStmt(clang::CompoundStmt const&,
bool, llvm::Value*, bool) + 213
14 clang           0x000000000099e5ad
clang::CodeGen::CodeGenFunction::EmitSimpleStmt(clang::Stmt const*) + 189
15 clang           0x000000000099c4b3
clang::CodeGen::CodeGenFunction::EmitStmt(clang::Stmt const*) + 35
16 clang           0x00000000009bfc12
clang::CodeGen::CodeGenFunction::GenerateCode(clang::CodeGen::GlobalDecl,
llvm::Function*) + 818
17 clang           0x00000000008b090b
clang::CodeGen::CodeGenModule::EmitGlobalFunctionDefinition(clang::CodeGen::GlobalDecl)
+ 619
18 clang           0x00000000008b141e
clang::CodeGen::CodeGenModule::EmitGlobalDefinition(clang::CodeGen::GlobalDecl)
+ 238
19 clang           0x00000000008b181d
clang::CodeGen::CodeGenModule::EmitGlobal(clang::CodeGen::GlobalDecl) + 749
20 clang           0x00000000008b2d3d
clang::CodeGen::CodeGenModule::EmitTopLevelDecl(clang::Decl*) + 509
21 clang           0x00000000008a66e1 (anonymous
namespace)::CodeGeneratorImpl::HandleTopLevelDecl(clang::DeclGroupRef) + 65
22 clang           0x00000000008a5f67 (anonymous
namespace)::BackendConsumer::HandleTopLevelDecl(clang::DeclGroupRef) + 167
23 clang           0x00000000009f5b66 clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 246
24 clang           0x00000000008a4d23 clang::CodeGenAction::ExecuteAction() +
51
25 clang           0x00000000007a8ee0
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 272
26 clang           0x000000000078d5c8 cc1_main(char const**, char const**, char
const*, void*) + 2168
27 clang           0x00000000007936cb main + 2955
28 libc.so.6       0x00002aaaab2894cb __libc_start_main + 219
29 clang           0x000000000078bb9a
Stack dump:
0.    Program arguments: /g/users/collinwinter/llvm/obj/Release/bin/clang -cc1
-triple x86_64-unknown-linux-gnu -S -disable-free -disable-llvm-verifier
-main-file-name varzclient.pic.ii -mrelocation-model static -mdisable-fp-elim
-mconstructor-aliases -munwind-tables -target-cpu x86-64 -resource-dir
/g/users/collinwinter/llvm/obj/Release/lib/clang/2.8 -ferror-limit 19
-fmessage-length 163 -fexceptions -fgnu-runtime -fdiagnostics-show-option
-fcolor-diagnostics -o /tmp/cc-dI3LHQ.s -x c++ varzclient.pic.ii 
1.    <eof> parser at end of file
2.    varzclient.pic.ii:8:6: LLVM IR generation of declaration 'HandleResult'
3.    varzclient.pic.ii:8:6: Generating code for declaration 'HandleResult'
4.    varzclient.pic.ii:8:21: LLVM IR generation of compound statement ('{}')
5.    varzclient.pic.ii:10:30: LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 11 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 12:41:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 12:41:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7687] New: Enhancement Request: Show function names
 in analyzer HTML output
Message-ID: <bug-7687-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7687

           Summary: Enhancement Request: Show function names in analyzer
                    HTML output
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: alexei.svitkine at gmail.com
                CC: llvmbugs at cs.uiuc.edu


It would be useful to show the function name of where the problem occurs in the
index.html that is generated after an analysis run. Currently, only the
filename and line number is shown (as well as the path length, bug type, etc).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 16:07:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 16:07:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7689] New: Crash handling bitcode
Message-ID: <bug-7689-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7689

           Summary: Crash handling bitcode
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rafael.espindola at gmail.com
                CC: llvmbugs at cs.uiuc.edu


This shows up with the gold plugin, but I don't think it is gold specific:

$ cat test.ll
@skel = global [1 x i8*] [i8* null]
!llvm.dbg.gv = !{!0}
!0 = metadata !{i1 true, [1 x i8*]* @skel}
$ llvm-as test.ll -o test.o
$ ld -plugin LLVMgold.so -o flex -shared -plugin-opt=emit-llvm test.o
Unknown constant!
UNREACHABLE executed at
/usr/local/google/home/espindola/llvm/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp:904!
Aborted

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 21 21:11:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 21 Jul 2010 21:11:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7683] PBQP register allocator segfaults on null
	pointer
In-Reply-To: <bug-7683-206@http.llvm.org/bugs/>
References: <bug-7683-206@http.llvm.org/bugs/>
Message-ID: <20100722021119.65EEE2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7683

Lang Hames <lhames at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Lang Hames <lhames at gmail.com> 2010-07-21 21:11:17 CDT ---
Looks like this was due to a live interval overlap test on an empty interval.

Fixed in r109075.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 01:00:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 01:00:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7381] make install: autoconf/install-sh not found
In-Reply-To: <bug-7381-206@http.llvm.org/bugs/>
References: <bug-7381-206@http.llvm.org/bugs/>
Message-ID: <20100722060045.8BE992A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7381

Eric Christopher <echristo at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #11 from Eric Christopher <echristo at gmail.com> 2010-07-22 01:00:44 CDT ---
Hi Simon,

I've done a little bit of experimenting and figured out that it's a bug in how
we're using @INSTALL@ at best, but that AC_PROG_INSTALL is setting it to the
relative path for install-sh if an install binary isn't found - as it is on the
systems that the rest of us are using :)

I'm not quite sure how I'm going to fix that here other than rewriting all of
the make install aspects to figure out that install doesn't have srcdir
prepended or computing every install from the top level.  Not something I'm
honestly willing to do.

At any rate there are a couple of options:

a) install a bsd-compatible install program someplace that configure will
notice (somewhere in your path)
b) set INSTALL at configure time:  INSTALL=$srcdir/autoconf/install-sh

I'm sorry that there aren't better options for you at the moment.

Thanks!

-eric

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 01:48:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 01:48:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7674] Clang Missing Intel SIMD intrinsics
In-Reply-To: <bug-7674-206@http.llvm.org/bugs/>
References: <bug-7674-206@http.llvm.org/bugs/>
Message-ID: <20100722064805.8E9362A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7674

Chandler Carruth <chandlerc at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #8 from Chandler Carruth <chandlerc at gmail.com> 2010-07-22 01:48:04 CDT ---
Fixed in r109086 after fixing the various copy/paste-os in the patch posted.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 02:11:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 02:11:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7673] Inline asm parse difference from GCC
In-Reply-To: <bug-7673-206@http.llvm.org/bugs/>
References: <bug-7673-206@http.llvm.org/bugs/>
Message-ID: <20100722071155.899A92A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7673

Chandler Carruth <chandlerc at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chandler Carruth <chandlerc at gmail.com> 2010-07-22 02:11:55 CDT ---
Fixed in r109087.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 04:30:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 04:30:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7212] Error while compiling llvm-gcc for ARM
In-Reply-To: <bug-7212-206@http.llvm.org/bugs/>
References: <bug-7212-206@http.llvm.org/bugs/>
Message-ID: <20100722093033.A7BE32A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7212

Renato Golin <rengolin at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #7 from Renato Golin <rengolin at gmail.com> 2010-07-22 04:30:32 CDT ---
Hi Jim,

I've tried with some variations of Anton's parameters (including all of them):

../llvm-gcc/configure --prefix=/work/temp/obj-arm/install
--program-prefix=llvm- --enable-llvm=/work/temp/llvm --enable-languages=c,c++
--target=arm-none-eabi --enable-checking --disable-bootstrap --with-newlib=yes
--with-headers=yes --enable-newlib-multithread=yes
--disable-newlib-supplied-syscalls --with-cpu=cortex-a8 --with-fpu=neon
--with-float=hard --with-abi=aapcs

And got the following error on every case:

libbackend.a(llvm-debug.o): In function
`llvm::DebugInfo::findRegion(tree_node*)':
/home/rengol01/temp/obj-arm/gcc/../../llvm-gcc/gcc/llvm-debug.cpp:545:
undefined reference to `llvm::DIFactory::CreateLexicalBlock(llvm::DIDescriptor,
llvm::DIFile, unsigned int, unsigned int)'
/work/temp/llvm/Release/lib/libLLVMCodeGen.a(ShrinkWrapping.o): In function
`llvm::PEI::stringifyCSRegSet(llvm::SparseBitVector<128u> const&)':
ShrinkWrapping.cpp:(.text+0x22d7): undefined reference to
`std::basic_ostream<char, std::char_traits<char> >& std::__ostream_insert<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&,
char const*, long)'
ShrinkWrapping.cpp:(.text+0x23a0): undefined reference to
`std::basic_ostream<char, std::char_traits<char> >& std::__ostream_insert<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&,
char const*, long)'
ShrinkWrapping.cpp:(.text+0x24d4): undefined reference to
`std::basic_ostream<char, std::char_traits<char> >& std::__ostream_insert<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&,
char const*, long)'
ShrinkWrapping.cpp:(.text+0x2535): undefined reference to
`std::basic_ostream<char, std::char_traits<char> >& std::__ostream_insert<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&,
char const*, long)'
ShrinkWrapping.cpp:(.text+0x2575): undefined reference to
`std::basic_ostream<char, std::char_traits<char> >& std::__ostream_insert<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&,
char const*, long)'
/work/temp/llvm/Release/lib/libLLVMCodeGen.a(ShrinkWrapping.o):ShrinkWrapping.cpp:(.text+0x25b7):
more undefined references to `std::basic_ostream<char, std::char_traits<char>
>& std::__ostream_insert<char, std::char_traits<char>
>(std::basic_ostream<char, std::char_traits<char> >&, char const*, long)'
follow
collect2: ld returned 1 exit status
make[2]: *** [cc1-dummy] Error 1
make[2]: Target `all' not remade because of errors.
make[2]: Leaving directory `/work/temp/obj-arm/gcc'
make[1]: *** [all-gcc] Error 2
make[1]: Target `all-host' not remade because of errors.
make[1]: Target `all-target' not remade because of errors.
make[1]: Leaving directory `/work/temp/obj-arm'
make: *** [all] Error 2


LLVM and Clang compile perfectly. Devang Patel may be of some help, since he
was changing the debug info, I guess.

cheers,
--renato

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 04:47:17 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 04:47:17 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7212] Error while compiling llvm-gcc for ARM
In-Reply-To: <bug-7212-206@http.llvm.org/bugs/>
References: <bug-7212-206@http.llvm.org/bugs/>
Message-ID: <20100722094717.E52A92A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7212

Anton Korobeynikov <asl at math.spbu.ru> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |INVALID

--- Comment #8 from Anton Korobeynikov <asl at math.spbu.ru> 2010-07-22 04:47:17 CDT ---
(In reply to comment #7)
> And got the following error on every case:
Looks like your llvm and llvm-gcc are out of sync

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 04:50:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 04:50:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7212] Error while compiling llvm-gcc for ARM
In-Reply-To: <bug-7212-206@http.llvm.org/bugs/>
References: <bug-7212-206@http.llvm.org/bugs/>
Message-ID: <20100722095039.AB2CE2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7212

Renato Golin <rengolin at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|INVALID                     |

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 11:13:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 11:13:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7691] New: wrong encoding of movq %rax, 0
Message-ID: <bug-7691-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7691

           Summary: wrong encoding of movq %rax, 0
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu


pes ~/clangbsd$ echo 'movq    %rax, 0' | llvm-mc --show-encoding --show-inst
        .section        __TEXT,__text,regular,pure_instructions
        movq    %rax, 0                 # encoding: [0x48,0xa3,A,A,A,A]
                                        #   fixup A - offset: 2, value: 0,
kind: FK_Data_4
                                        # <MCInst #1363 MOV64ao64
                                        #  <MCOperand Expr:(0)>>

note that the actual encoding is 

         0x48,0xa3,A,A,A,A

pes ~/clangbsd$ cat bah.s 
mov    %rax,0x0
pes ~/clangbsd$ as bah.s -o bah.o && objdump -d bah.o

bah.o:     file format elf64-x86-64

Disassembly of section .text:

0000000000000000 <.text>:
   0:   48 89 04 25 00 00 00    mov    %rax,0x0
   7:   00 

ie. the encoding for gnu as is

        48 89 04 25 00 00 00 00

thats one 00 more than with llvm

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 11:36:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 11:36:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7689] Crash handling bitcode
In-Reply-To: <bug-7689-206@http.llvm.org/bugs/>
References: <bug-7689-206@http.llvm.org/bugs/>
Message-ID: <20100722163618.9FE9C2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7689

Devang Patel <dpatel at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |dpatel at apple.com
         Resolution|                            |FIXED

--- Comment #3 from Devang Patel <dpatel at apple.com> 2010-07-22 11:36:17 CDT ---
Fixed in r109117.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 12:32:10 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 12:32:10 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7608] ARMv4 JIT forgets to set the lr register when
 making a indirect function call
In-Reply-To: <bug-7608-206@http.llvm.org/bugs/>
References: <bug-7608-206@http.llvm.org/bugs/>
Message-ID: <20100722173210.53AB82A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7608

Xerxes R?nby <xerxes at zafena.se> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #8 from Xerxes R?nby <xerxes at zafena.se> 2010-07-22 12:32:07 CDT ---
Fixed in r109125
http://llvm.org/viewvc/llvm-project?view=rev&revision=109125

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 12:48:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 12:48:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7689] Crash handling bitcode
In-Reply-To: <bug-7689-206@http.llvm.org/bugs/>
References: <bug-7689-206@http.llvm.org/bugs/>
Message-ID: <20100722174847.F31C32A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7689

Rafael ?vila de Esp?ndola <rafael.espindola at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #4 from Rafael ?vila de Esp?ndola <rafael.espindola at gmail.com> 2010-07-22 12:48:43 CDT ---
I still get the same failure:

@skel = global [1 x i8*] zeroinitializer          ; <[1 x i8*]*> [#uses=0]

Unknown constant!
UNREACHABLE executed at
/usr/local/google/home/espindola/llvm/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp:909!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 16:01:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 16:01:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7381] make install: autoconf/install-sh not found
In-Reply-To: <bug-7381-206@http.llvm.org/bugs/>
References: <bug-7381-206@http.llvm.org/bugs/>
Message-ID: <20100722210124.D2CF12A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7381

Eric Christopher <echristo at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|WONTFIX                     |

--- Comment #13 from Eric Christopher <echristo at gmail.com> 2010-07-22 16:01:23 CDT ---
Yeah, John and I were chatting about that today.  I don't have any objection to
putting something like that in if it works - I thought I came up with something
that'd break it but I'm not sure.  I'll go ahead and throw that in and see what
happens.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 22 16:14:51 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 22 Jul 2010 16:14:51 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7381] make install: autoconf/install-sh not found
In-Reply-To: <bug-7381-206@http.llvm.org/bugs/>
References: <bug-7381-206@http.llvm.org/bugs/>
Message-ID: <20100722211451.733A92A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7381

Eric Christopher <echristo at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #14 from Eric Christopher <echristo at gmail.com> 2010-07-22 16:14:50 CDT ---
I've committed it thusly:

[yendi:Data/sources/llvm] echristo% svn ci autoconf/configure.ac 
Sending        autoconf/configure.ac
Transmitting file data .
Committed revision 109163.

[yendi:Data/sources/llvm] echristo% svn ci configure
Sending        configure
Transmitting file data .
Committed revision 109164.

If I could I'd upstream it to autoconf, since I think it's really a failure
there.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 01:24:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 01:24:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7693] New: Overly aggressive error: illegal goto
 into protected scope
Message-ID: <bug-7693-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7693

           Summary: Overly aggressive error: illegal goto into protected
                    scope
           Product: clang
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: 641 at rudkx.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5254)
 --> (http://llvm.org/bugs/attachment.cgi?id=5254)
A test case that demonstrates the issue.

I believe the intent of this error is to protect against skipping over a ctor
for a value that you will then reach the dtor for.

The issue is that this error is emitted even when all you are skipping are
initializers for POD types.

I've attached the example which produces the output below. This is a much
reduced example of a real piece of code that has several dozen of these errors
when compiling with clang.

goto.cpp:2:3: error: illegal goto into protected scope
  goto L;
  ^
goto.cpp:3:7: note: jump bypasses variable initialization
  int i = 0;

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 08:16:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 08:16:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7694] New: crash in sema while trying to cast a
 member function pointer
Message-ID: <bug-7694-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7694

           Summary: crash in sema while trying to cast a member function
                    pointer
           Product: clang
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: devlists at shadowlab.org
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5255)
 --> (http://llvm.org/bugs/attachment.cgi?id=5255)
File that crash clang++

clang++ crash on the following (and attached) code:

================

class SAISOBoxReader { };
typedef void (SAISOBoxReader::*SAISOBoxReadFunc)(int value);

static void SAISOBoxRead(SAISOBoxReader *reader, SAISOBoxReadFunc function);

class SAMovieReader : private SAISOBoxReader {  
public:
  bool read(const char *path);

private:
  void dispatch_root(int value);
};

bool SAMovieReader::read(const char *path) {
  SAISOBoxRead(this,
static_cast<SAISOBoxReadFunc>(&SAMovieReader::dispatch_root));
  return false;
}

===============

jddupas% clang++ --version
clang version 2.8 (trunk 109223) (this is a Release+Assert build)
Target: x86_64-apple-darwin10
Thread model: posix

jddupas% clang++ SAMovieReader.cpp 
0  clang             0x0000000100da7c62 PrintStackTrace(void*) + 34
1  clang             0x0000000100da8119 SignalHandler(int) + 793
2  libSystem.B.dylib 0x00007fff8642b35a _sigtramp + 26
3  libSystem.B.dylib 000000000000000000 _sigtramp + 2042449088
4  clang             0x0000000100213294 TryStaticCast(clang::Sema&,
clang::Expr*&, clang::QualType, bool, clang::SourceRange const&, unsigned int&,
clang::CastExpr::CastKind&, clang::UsuallyTinyPtrVector<clang::CXXBaseSpecifier
const>&) + 3732
5  clang             0x000000010021179a
clang::Sema::BuildCXXNamedCast(clang::SourceLocation, clang::tok::TokenKind,
clang::TypeSourceInfo*,
clang::ASTOwningResult<&(clang::ActionBase::DeleteExpr(void*))>,
clang::SourceRange, clang::SourceRange) + 6010
6  clang             0x000000010020ffa6
clang::Sema::ActOnCXXNamedCast(clang::SourceLocation, clang::tok::TokenKind,
clang::SourceLocation, void*, clang::SourceLocation, clang::SourceLocation,
clang::ASTOwningResult<&(clang::ActionBase::DeleteExpr(void*))>,
clang::SourceLocation) + 182
7  clang             0x000000010058cb07 clang::Parser::ParseCXXCasts() + 647
?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 08:55:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 08:55:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7695] New: "opt -inline -gvn" != "opt -inline" +
	"opt -gvn"
Message-ID: <bug-7695-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7695

           Summary: "opt -inline -gvn" != "opt -inline" + "opt -gvn"
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Scalar Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rafael.espindola at gmail.com
                CC: llvmbugs at cs.uiuc.edu


With the attached testcase, running

$ opt -inline -gvn -o test-final1.ll test.ll
produces different results than
$ opt -inline -o test-temp.ll test.ll
$ opt -gvn -o test-final2.ll test-temp.ll

This is a reduction of a clang bootstrap failure on ARM.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 10:11:52 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 10:11:52 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7695] "opt -inline -gvn" != "opt -inline" + "opt
	-gvn"
In-Reply-To: <bug-7695-206@http.llvm.org/bugs/>
References: <bug-7695-206@http.llvm.org/bugs/>
Message-ID: <20100723151152.45DD62A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7695

Duncan Sands <baldrick at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |baldrick at free.fr
         Resolution|                            |INVALID

--- Comment #2 from Duncan Sands <baldrick at free.fr> 2010-07-23 10:11:51 CDT ---
This is by design.  When you schedule a SCC pass like inline, followed by a
function pass like GVN, you get the following: (suppose F calls G, which
doesn't
call anything):
  - first inline is run on G; this does nothing since G doesn't call anything
  - then GVN is run on G, simplifying G
  - then inline is run on F, which may inline the call to G if the simplified G
is simple enough

When you do opt -inline followed by opt -gvn you get:
  - first inline is run on G; this does nothing since G doesn't call anything
  - then inline is run on F, which may inline the call to G if G is simple
enough
  - then GVN is run on G, simplifying G
  - then GVN is run on F, simplifying F

The point is that in the first case GVN may simplify G enough that the inliner
decides it is worth inlining into F, which it wouldn't do in the second case
because the original function G was too big.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 12:38:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 12:38:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7696] New: SelectionDAGBuilder doing bad things on
 certain architectures
Message-ID: <bug-7696-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7696

           Summary: SelectionDAGBuilder doing bad things on certain
                    architectures
           Product: libraries
           Version: 2.7
          Platform: PC
        OS/Version: Windows NT
            Status: NEW
          Severity: normal
          Priority: P
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: micah.villmow at amd.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5257)
 --> (http://llvm.org/bugs/attachment.cgi?id=5257)
Patch that adds a boolean where a backend can specify whether to disable
certain optimizations because jumps are expensive.

The selection dag builder has an ?optimization? added into the visitBr function
which makes assumptions that are not valid on all architectures. The problem is
this.
The following function
kernel void cf_test(global int* a, int b, int c, int e)
{
        int d = 0;
        if (!b && c < e) {
                d = a + b;
        }
        *a = d;
}
Is transformed into something equivalent to this:
Kernel void cf_test(global int* a, int b, int c, int e)
{
  Int d;
  If (b) {
    d = 0;
  } else {
    if (c < e) {
      d = a + b;
    } else {
      d = 0;
    }
  }
 *a = d;
}

by the visitBr code found in SelectionDAGBuilder::visitBr():1188.

However, if jumps are expensive or jumps are not supported and high level flow
control needs to be reconstructed. This is extremely inefficient. For example
on AMD GPU?s, a single flow control instruction can take 40 cycles to execute,
but an bit instruction, can be executed every cycle. So obviously the
assumptions made by this block of code are inefficient on AMD hardware. 
Increasing control flow has a direct impact on performance and removing the
extra ?and? or ?or? in order to short circuit the conditional evaluation does
not work for our target.

So in order to make this type of instruction rely more on target specific
information. I?ve added a new Boolean to the TargetLoweringInfo class called
JumpIsExpensive along with accessor functions.

Please review the patch and apply if acceptable.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 14:26:25 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 14:26:25 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7694] crash in sema while trying to cast a member
 function pointer
In-Reply-To: <bug-7694-206@http.llvm.org/bugs/>
References: <bug-7694-206@http.llvm.org/bugs/>
Message-ID: <20100723192625.789962A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7694

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 16:49:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 16:49:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7693] Overly aggressive error: illegal goto into
 protected scope
In-Reply-To: <bug-7693-206@http.llvm.org/bugs/>
References: <bug-7693-206@http.llvm.org/bugs/>
Message-ID: <20100723214940.A43042A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7693

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-23 16:49:39 CDT ---
C++ [stmt.dcl]p3 is clear that this code is ill-formed:

  A program that jumps from a point where a variable with automatic storage
duration is not in scope to a point where it is in scope is ill-formed unless
the variable has scalar type, class type with a trivial default constructor and
a trivial destructor, a cv-qualified version of one of these types, or an array
of one of the preceding types and is declared without an initializer (8.5).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 17:29:06 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 17:29:06 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4208] isIntegerType() returns true for vectors of
	integers
In-Reply-To: <bug-4208-206@http.llvm.org/bugs/>
References: <bug-4208-206@http.llvm.org/bugs/>
Message-ID: <20100723222906.67FF42A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4208

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |dgregor at apple.com
         Resolution|                            |FIXED

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-23 17:29:05 CDT ---
isIntegerType() no longer returns "true" for vectors of integers. The
isIntegerType/isIntegralType dichotomy is a separable issue that is "in
progress". Fix is here:

 
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20100719/032490.html

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 17:41:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 17:41:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7692] Precompiling <iostream> and then using
 std::cout makes executable segfault
In-Reply-To: <bug-7692-206@http.llvm.org/bugs/>
References: <bug-7692-206@http.llvm.org/bugs/>
Message-ID: <20100723224154.43BA02A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7692

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |akyrtzi at gmail.com,
                   |                            |dgregor at apple.com,
                   |                            |llvmbugs at cs.uiuc.edu
          Component|Frontend                    |C++

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-23 17:41:53 CDT ---
C++ precompiled headers are not yet fully implemented.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 17:47:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 17:47:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7692] Precompiling <iostream> and then using
 std::cout makes executable segfault
In-Reply-To: <bug-7692-206@http.llvm.org/bugs/>
References: <bug-7692-206@http.llvm.org/bugs/>
Message-ID: <20100723224747.527FF2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7692

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|llvmbugs at cs.uiuc.edu        |

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 19:42:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 19:42:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7686] Segfault in codegen
In-Reply-To: <bug-7686-206@http.llvm.org/bugs/>
References: <bug-7686-206@http.llvm.org/bugs/>
Message-ID: <20100724004246.E61D12A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7686

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from John McCall <rjmccall at apple.com> 2010-07-23 19:42:45 CDT ---
Fixed in r109274.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 23 20:03:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 23 Jul 2010 20:03:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7698] New: clang++ crashes parsing template with
 member array with size dependent of template parameter
Message-ID: <bug-7698-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7698

           Summary: clang++ crashes parsing template with member array
                    with size dependent of template parameter
           Product: clang
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: devlists at shadowlab.org
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


A reduce test case of a c++ code that crashes clang parser:

------------------------

template<typename Type>
class A {
  char mA[sizeof(Type *)];
  A(): mA() {}
};

------------------------

clang version 2.8 (trunk 109288)
Target: x86_64-apple-darwin10
Thread model: posix

clang++ -c test.cpp
?
3  libSystem.B.dylib 000000000000000000 _sigtramp + 2066889920
4  clang             0x00000001003865bf
clang::Sema::MaybeCreateCXXExprWithTemporaries(clang::Expr*) + 31
5  clang             0x0000000100390cc4
clang::Sema::MaybeCreateCXXExprWithTemporaries(clang::ASTOwningResult<&(clang::ActionBase::DeleteExpr(void*))>)
+ 116
6  clang             0x0000000100305889
clang::Sema::BuildMemberInitializer(clang::FieldDecl*, clang::Expr**, unsigned
int, clang::SourceLocation, clang::SourceLocation, clang::SourceLocation) 
?
clang: error: clang frontend command failed due to signal 11 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 02:00:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 02:00:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7699] New: Assertion `!EHBranchThroughDest' failed
Message-ID: <bug-7699-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7699

           Summary: Assertion `!EHBranchThroughDest' failed
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: grzegorz.dabrowski at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


$ clang -v
clang version 2.8 (trunk 109330)
Target: i386-pc-linux-gnu
Thread model: posix


$ cat tests-my/branchthroughdest.cpp 
#include <iostream>
#include <set>

int main() {
  std::set<std::string> alreadyLoaded;
  try {
    alreadyLoaded.insert("test");
  }
  catch (char *str) {
  }
  return 0;
}

$ clang++ tests-my/branchthroughdest.cpp 
clang: CodeGenFunction.cpp:843: void
clang::CodeGen::CodeGenFunction::PopCleanupBlock(bool): Assertion
`!EHBranchThroughDest' failed.
0  clang 0x093e2dfb
Stack dump:
0.    Program arguments:
/home/stuff/download/SCM/llvm/Release+Asserts/bin/clang -cc1 -triple
i386-pc-linux-gnu -S -disable-free -main-file-name branchthroughdest.cpp
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases -target-cpu
pentium4 -resource-dir
/home/stuff/download/SCM/llvm/Release+Asserts/lib/clang/2.8 -ferror-limit 19
-fmessage-length 157 -fexceptions -fgnu-runtime -fdiagnostics-show-option
-fcolor-diagnostics -o /tmp/cc-GxLGHC.s -x c++ tests-my/branchthroughdest.cpp 
1.    <eof> parser at end of file
2.    tests-my/branchthroughdest.cpp:4:5: LLVM IR generation of declaration
'main'
3.    tests-my/branchthroughdest.cpp:4:5: Generating code for declaration
'main'
4.    tests-my/branchthroughdest.cpp:4:12: LLVM IR generation of compound
statement ('{}')
5.    tests-my/branchthroughdest.cpp:6:6: LLVM IR generation of compound
statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 04:05:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 04:05:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7700] New: bugpoint crashes with Unknown
 constant!...BitcodeWriter.cpp:905
Message-ID: <bug-7700-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7700

           Summary: bugpoint crashes with Unknown
                    constant!...BitcodeWriter.cpp:905
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: bugpoint
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jyasskin at google.com
                CC: llvmbugs at cs.uiuc.edu


As of r108766:

-macbookpro:~/tmp$ cat unit.ll
%rust_task = type {}
define void @_rust_main([0 x i1]*, %rust_task*) {
  ret void , !dbg !9
}
!1 = metadata !{i32 524334, void ([0 x i1]*, %rust_task*)* @_rust_main} ; [
DW_TAG_subprogram ]
!9 = metadata !{i32 11, i32 2, metadata !1, null}

-macbookpro:~/tmp$ /Users/jyasskin/src/llvm/trunk/install/bin/bugpoint -run-llc
unit.ll
Read input file      : 'unit.ll'
*** All input ok
Initializing execution environment: Found gcc: /opt/local/bin/gcc
Running the code generator to test for a crash: <llc>
Generating reference output from raw program: <llc><cbe><gcc>
Error running tool:
  /opt/local/bin/gcc -m32 -x c -fno-strict-aliasing
bugpoint-test-program.bc-rzyyXa.cbe.c -x none -o
bugpoint-test-program.bc-rzyyXa.cbe.c.gcc.exe -lm -O2
Undefined symbols:
  "_main", referenced from:
      start in crt1.10.5.o
ld: symbol(s) not found
collect2: ld returned 1 exit status
*** There is a bug running the "safe" backend.  Either debug it (for example
with the -run-cbe bugpoint option, if CBE is being used as the "safe" backend),
or fix the error some other way.
*** Debugging code generator crash!
Checking for crash with only these blocks: : Unknown constant!
UNREACHABLE executed at
/Users/jyasskin/src/llvm/trunk/src/lib/Bitcode/Writer/BitcodeWriter.cpp:905!
-macbookpro:~/tmp$

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 04:28:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 04:28:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7701] New: llc crash in out-of-bounds access
Message-ID: <bug-7701-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7701

           Summary: llc crash in out-of-bounds access
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: llc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jyasskin at google.com
                CC: llvmbugs at cs.uiuc.edu


As of r108766:

-macbookpro:~/tmp$ cat unit.ll
; ModuleID = 'unit.bc'

define void @_rust_main([0 x i1] %tmp1) {
  %u1 = alloca [0 x i1]                           ; <[0 x i1]*> [#uses=1]
  %tmp2 = load [0 x i1]* %u1                      ; <[0 x i1]> [#uses=1]
  call void @_rust_f([0 x i1] %tmp2)
  ret void
}

define void @_rust_f([0 x i1]) {
  ret void
}

-macbookpro:~/tmp$ llc unit.ll
Assertion failed: (begin() + idx < end()), function operator[], file
/Users/jyasskin/src/llvm/trunk/src/include/llvm/ADT/SmallVector.h, line 155.
0  libLLVM-2.8svn.dylib 0xe094a8e0 llvm::SearchForAddressOfSpecialSymbol(char
const*) + 478
1  libLLVM-2.8svn.dylib 0xe094ad72 llvm::sys::RunInterruptHandlers() + 354
2  libSystem.B.dylib    0x939512bb _sigtramp + 43
3  libSystem.B.dylib    0xffffffff _sigtramp + 1818946927
4  libSystem.B.dylib    0x939c523a raise + 26
5  libSystem.B.dylib    0x939d1679 abort + 73
6  libSystem.B.dylib    0x939c63db __assert_rtn + 101
7  libLLVM-2.8svn.dylib 0xe006966d
llvm::SmallVectorTemplateCommon<llvm::SDValue>::operator[](unsigned int) + 101
8  libLLVM-2.8svn.dylib 0xe0882df1 (anonymous
namespace)::RegsForValue::getCopyFromRegs(llvm::SelectionDAG&,
llvm::FunctionLoweringInfo&, llvm::DebugLoc, llvm::SDValue&, llvm::SDValue*)
const + 1801
9  libLLVM-2.8svn.dylib 0xe08a1dbb
llvm::SelectionDAGBuilder::getValueImpl(llvm::Value const*) + 3519
10 libLLVM-2.8svn.dylib 0xe0887619
llvm::SelectionDAGBuilder::getValue(llvm::Value const*) + 445
11 libLLVM-2.8svn.dylib 0xe088ad72
llvm::SelectionDAGBuilder::LowerCallTo(llvm::ImmutableCallSite, llvm::SDValue,
bool, llvm::MachineBasicBlock*) + 972
12 libLLVM-2.8svn.dylib 0xe089975d
llvm::SelectionDAGBuilder::visitCall(llvm::CallInst const&) + 3691
13 libLLVM-2.8svn.dylib 0xe08a0ed2 llvm::SelectionDAGBuilder::visit(unsigned
int, llvm::User const&) + 1328
14 libLLVM-2.8svn.dylib 0xe08a61a1
llvm::SelectionDAGBuilder::visit(llvm::Instruction const&) + 111
15 libLLVM-2.8svn.dylib 0xe08bd245
llvm::SelectionDAGISel::SelectBasicBlock(llvm::ilist_iterator<llvm::Instruction
const>, llvm::ilist_iterator<llvm::Instruction const>, bool&) + 45
16 libLLVM-2.8svn.dylib 0xe08bd885
llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function const&) + 1437
17 libLLVM-2.8svn.dylib 0xe08bdba6
llvm::SelectionDAGISel::runOnMachineFunction(llvm::MachineFunction&) + 702
18 libLLVM-2.8svn.dylib 0xe036e54d
llvm::MachineFunctionPass::runOnFunction(llvm::Function&) + 77
19 libLLVM-2.8svn.dylib 0xe0532831
llvm::FPPassManager::runOnFunction(llvm::Function&) + 287
20 libLLVM-2.8svn.dylib 0xe05329b5
llvm::FPPassManager::runOnModule(llvm::Module&) + 83
21 libLLVM-2.8svn.dylib 0xe0532431
llvm::MPPassManager::runOnModule(llvm::Module&) + 375
22 libLLVM-2.8svn.dylib 0xe0532680 llvm::PassManagerImpl::run(llvm::Module&) +
112
23 libLLVM-2.8svn.dylib 0xe05326d9 llvm::PassManager::run(llvm::Module&) + 27
24 llc                  0x0000329d main + 2407
25 llc                  0x0000219e start + 54
Stack dump:
0.    Program arguments: /Users/jyasskin/src/llvm/trunk/install/bin/llc unit.ll 
1.    Running pass 'Function Pass Manager' on module 'unit.ll'.
2.    Running pass 'X86 DAG->DAG Instruction Selection' on function
'@_rust_main'

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 12:05:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 12:05:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7702] New: unclear diagnostic after 'new'
Message-ID: <bug-7702-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7702

           Summary: unclear diagnostic after 'new'
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: nicholas at mxc.ca
                CC: llvmbugs at cs.uiuc.edu


This diagnostic is confusing:

cm.cpp:77:17: error: missing type specifier after 'operator'
      Out = new raw_os_ostream(OutputFilename.c_str(), true);
                ^

What? gcc's is worse but at least I'm so used to it I knew what to do: add the
missing include.

Clang should've said something that means "no such type" instead of focusing on
the 'operator', which is really irrelevant and not obviously part of the code
(you say 'operator' I think of something like operator+ that has operator in
the name, and it took a moment to realize that it was talking about 'new').

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 13:04:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 13:04:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7703] New: All shuffle intrinsics use PSHUFD (even
 when SHUFPS is requested and smaller)
Message-ID: <bug-7703-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7703

           Summary: All shuffle intrinsics use PSHUFD (even when SHUFPS is
                    requested and smaller)
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jed at 59a2.org
                CC: llvmbugs at cs.uiuc.edu


void f(const float *x,float *y) {
  __m128 p = _mm_load_ps(x);
  p = _mm_shuffle_ps(p,p,_MM_SHUFFLE(0,2,1,3));
  _mm_store_ps(y,p);
}

Compile with -Os -fomit-frame-pointer (clang version 1.1
(branches/release_27)):

00000000004004c4 <f> movaps xmm0,XMMWORD PTR [rdi]
00000000004004c7 <f+0x3> pshufd xmm0,xmm0,0x27
00000000004004cc <f+0x8> movaps XMMWORD PTR [rsi],xmm0
00000000004004cf <f+0xb> ret

Compare to gcc (4.5.0) which is one byte shorter and actually uses the
instruction specified by the intrinsic:

00000000004004c4 <f> movaps xmm0,XMMWORD PTR [rdi]
00000000004004c7 <f+0x3> shufps xmm0,xmm0,0x27
00000000004004cb <f+0x7> movaps XMMWORD PTR [rsi],xmm0
00000000004004ce <f+0xa> ret

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 13:28:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 13:28:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7704] New: asserts after parsing
Message-ID: <bug-7704-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7704

           Summary: asserts after parsing
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM assembly language parser
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca
                CC: llvmbugs at cs.uiuc.edu


The assembly parser doesn't reject the following small program:

  define void @_rust_main([0 x i1]*, {}*) {
    ret void
  }
  define void @test() {
    call void ([0 x i1]*, {}*) asm "", "=r,0"( void ([0 x i1]*, {}*)*
@_rust_main )
  }

and instead I get this assertion:

$ llvm-commit/Debug+Asserts/bin/llvm-as < x.ll
llvm-as: Value.cpp:49: llvm::Value::Value(const llvm::Type*, unsigned int):
Assertion `(VTy->isFirstClassType() || VTy->isVoidTy() || ty->isOpaqueTy() ||
VTy->isStructTy()) && "invalid CallInst  type!"' failed.
0  llvm-as 0x0819fcda
1  llvm-as 0x081a09a4
2          0xffffe400 __kernel_sigreturn + 0
Stack dump:
0.      Program arguments: llvm-commit/Debug+Asserts/bin/llvm-as 
Aborted

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 14:46:00 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 14:46:00 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7700] bugpoint crashes with Unknown
 constant!...BitcodeWriter.cpp:905
In-Reply-To: <bug-7700-206@http.llvm.org/bugs/>
References: <bug-7700-206@http.llvm.org/bugs/>
Message-ID: <20100724194600.9C6DD2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7700

Nick Lewycky <nicholas at mxc.ca> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 15:57:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 15:57:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7700] bugpoint crashes with Unknown
 constant!...BitcodeWriter.cpp:905
In-Reply-To: <bug-7700-206@http.llvm.org/bugs/>
References: <bug-7700-206@http.llvm.org/bugs/>
Message-ID: <20100724205741.C12DF2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7700

Nick Lewycky <nicholas at mxc.ca> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |
         AssignedTo|unassignedbugs at nondot.org   |dpatel at apple.com

--- Comment #1 from Nick Lewycky <nicholas at mxc.ca> 2010-07-24 15:57:41 CDT ---
This bug is caused by the Module cloner failing to clone the MDNode when it
contains a reference to a Function in the old module. It should create a new
MDNode which refers to the new Function in the new module.

Revision 109353 tried to do that, but we encountered a cyclic MDNode when
building llvm-gcc (compiling src/gcc/libgcc2.c with xgcc to produce
_udivdi3.o). The cloner makes no attempt to deal with cyclic MDNodes and I
don't see how to do it.

Devang, please teach the cloner to handle this case, or if that shouldn't be
happening, find out why it is in the llvm-gcc build. Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 16:19:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 16:19:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7698] clang++ crashes parsing template with member
 array with size dependent of template parameter
In-Reply-To: <bug-7698-206@http.llvm.org/bugs/>
References: <bug-7698-206@http.llvm.org/bugs/>
Message-ID: <20100724211938.3DB022A6C130@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7698

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #1 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-24 16:19:37 CDT ---
Fixed in r109356.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 17:58:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 17:58:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7704] asserts after parsing
In-Reply-To: <bug-7704-206@http.llvm.org/bugs/>
References: <bug-7704-206@http.llvm.org/bugs/>
Message-ID: <20100724225820.C6CBC2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7704

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #1 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-24 17:58:20 CDT ---
Fixed in r109360.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 24 19:52:08 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 24 Jul 2010 19:52:08 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7706] New: clang++ produces incorrect outputcode
 with boost::shared_ptr and -O0
Message-ID: <bug-7706-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7706

           Summary: clang++ produces incorrect outputcode with
                    boost::shared_ptr and -O0
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: M8R-motwq11 at mailinator.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5259)
 --> (http://llvm.org/bugs/attachment.cgi?id=5259)
Testcase

When I compile the attached code (using "clang++ -O0 test.ii") I would expect
the main function to output nothing. But instead the while loop is executed
infinitly and the "else" path is taken.

Compiling with gcc or using optimization the error disapears. 

> clang --version
clang version 2.8 (trunk 109365)
Target: x86_64-unknown-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 07:34:38 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 07:34:38 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7707] New: Assertion in TwoAddressInstructionPass
Message-ID: <bug-7707-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7707

           Summary: Assertion in TwoAddressInstructionPass
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5261)
 --> (http://llvm.org/bugs/attachment.cgi?id=5261)
Testcase

Consider the attached .ll file.

Running llc -O0 yields:

$ ./llc -O0 bugpoint-reduced-simplified2.ll 
Assertion failed: (i < getNumOperands() && "getOperand() out of range!"),
function getOperand, file
/Users/asl/Projects/llvm/src-git/include/llvm/CodeGen/MachineInstr.h, line 159.
0  llc               0x00000001006f3430 PrintStackTrace(void*) + 38
1  llc               0x00000001006f393e SignalHandler(int) + 254
2  libSystem.B.dylib 0x00007fff83fbf35a _sigtramp + 26
3  libSystem.B.dylib 0x0000000101f12760 _sigtramp + 2113221664
4  libSystem.B.dylib 0x00007fff8403a9b4 __pthread_markcancel + 0
5  llc               0x00000001000ae31d llvm::MachineInstr::getOperand(unsigned
int) const + 79
6  llc               0x000000010037c832
llvm::MachineInstr::isRegTiedToDefOperand(unsigned int, unsigned int*) const +
604
7  llc               0x0000000100466d6d (anonymous
namespace)::TwoAddressInstructionPass::runOnMachineFunction(llvm::MachineFunction&)
+ 1249
8  llc               0x000000010037b057
llvm::MachineFunctionPass::runOnFunction(llvm::Function&) + 85
9  llc               0x0000000100649846
llvm::FPPassManager::runOnFunction(llvm::Function&) + 348
10 llc               0x0000000100649a19
llvm::FPPassManager::runOnModule(llvm::Module&) + 81
11 llc               0x0000000100649518
llvm::MPPassManager::runOnModule(llvm::Module&) + 384
12 llc               0x000000010064acc1
llvm::PassManagerImpl::run(llvm::Module&) + 111
13 llc               0x000000010064ad23 llvm::PassManager::run(llvm::Module&) +
33
14 llc               0x00000001000252fb main + 2143
15 llc               0x00000001000242a4 start + 52
16 llc               0x0000000000000003 start + 4294819219

The problem seems to be hidden by -ON with N > 0

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 11:03:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 11:03:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7708] New: Partial ordering weird behavior
Message-ID: <bug-7708-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7708

           Summary: Partial ordering weird behavior
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: schaub-johannes at web.de
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


I don't understand the exact rules when non-deduced contexts are involved, but
the following looks weird to me. While this compiles:

// snip
template<typename T>
struct Const { typedef void type; };

template<typename T>
void f(T, typename Const<T>::type*) { } // T1

template<typename T>
void f(T, void *) {  } // T2

int main() { void *p = 0; f(0, p); }
// snap

The following doesn't, claiming an ambiguous call, but does compile on
EDG/comeau and GCC

// snip
template<typename T>
struct Const { typedef void const type; };

template<typename T>
void f(T, typename Const<T>::type*) { } // T1

template<typename T>
void f(T, void const *) {  } // T2

int main() { void *p = 0; f(0, p); }
// snap

// Error output:
main1.cpp:13:3: error: call to 'f' is ambiguous
  f(0, p);
  ^
main1.cpp:5:6: note: candidate function [with T = int]                          
void f(T, typename Const<T>::type*) { } // T1
     ^
main1.cpp:8:6: note: candidate function [with T = int]                          
void f(T, void const*) {  } // T2
     ^
1 error generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 14:42:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 14:42:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7709] New: Disassembler does not support x86
 instruction prefixes
Message-ID: <bug-7709-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7709

           Summary: Disassembler does not support x86 instruction prefixes
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nemerle5 at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Following instructions from http://blog.llvm.org/2010/01/x86-disassembler.html:
echo '0xF3 0xAD' | llvm-mc -disassemble -triple=x86_64-apple-darwin9
outputs:
  lodsl
instead of the correct sequence
  rep lodsl

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 15:56:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 15:56:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7710] New: Invalid read disassembling incomplete
	instruction
Message-ID: <bug-7710-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7710

           Summary: Invalid read disassembling incomplete instruction
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu


Command-line:
echo '0x00' | llvm-mc -disassemble -triple=x86_64-apple-darwin9

Expected result:
Error: ...

Actual result:
    addb    %ah, (%rax)

Also, valgrind shows an invalid read error.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 16:25:19 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 16:25:19 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7711] New: CMake configure fails in 2.8svn
Message-ID: <bug-7711-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7711

           Summary: CMake configure fails in 2.8svn
           Product: Build scripts
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: cmake
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nobled at dreamwidth.org
                CC: llvmbugs at cs.uiuc.edu


SVN: r109373

Using CMake 2.8.1's GUI tool on Windows, I can configure and generate build
files for Visual Studio 2010 just fine with LLVM 2.7, but not with 2.8svn.

The 'Configure' step failed with this:
--
Check for working C compiler using: Visual Studio 10
Check for working C compiler using: Visual Studio 10 -- works
Detecting C compiler ABI info
Detecting C compiler ABI info - done
Check for working CXX compiler using: Visual Studio 10
Check for working CXX compiler using: Visual Studio 10 -- works
Detecting CXX compiler ABI info
Detecting CXX compiler ABI info - done
CMake Error: File C:/llvm-2.8svn-p1/include/llvm/Config/Targets.def.in does not
exist.
CMake Error at CMakeLists.txt:120 (configure_file):
  configure_file Problem configuring file


CMake Error at CMakeLists.txt:134 (include):
  include could not find load file:

    AddLLVMDefinitions


CMake Error at CMakeLists.txt:165 (include):
  include could not find load file:

    config-ix


Warning: -fPIC not supported.
CMake Error at CMakeLists.txt:192 (add_llvm_definitions):
  Unknown CMake command "add_llvm_definitions".


Configuring incomplete, errors occurred!
--

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 17:00:16 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 17:00:16 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7711] CMake configure fails in 2.8svn
In-Reply-To: <bug-7711-206@http.llvm.org/bugs/>
References: <bug-7711-206@http.llvm.org/bugs/>
Message-ID: <20100725220016.959C72A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7711

nobled at dreamwidth.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from nobled at dreamwidth.org 2010-07-25 17:00:16 CDT ---
(Never mind, fixed; somehow missed a bunch of files when copying the directory)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 19:01:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 19:01:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7712] New: psuedo instructions should be removed
 before code emission
Message-ID: <bug-7712-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7712

           Summary: psuedo instructions should be removed before code
                    emission
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: lli
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: grosser at fim.uni-passau.de
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5265)
 --> (http://llvm.org/bugs/attachment.cgi?id=5265)
Failing test case

#> bin/lli /tmp/out.ll 
psuedo instructions should be removed before code emission
UNREACHABLE executed at
/home/grosser/Projekte/llvm/git_trunk/lib/Target/X86/X86CodeEmitter.cpp:609!
Stack dump:
0.    Program arguments: bin/lli /tmp/out.ll 
1.    Running pass 'X86 Machine Code Emitter' on function
'@loop_with_condition'
zsh: abort      bin/lli /tmp/out.ll

I get this bug on a test case since this commit:
-------------------------------------------------------------------------------
Author: Eric Christopher <echristo at apple.com>
Date:   Thu Jul 22 02:48:34 2010 +0000

    Custom lower the memory barrier instructions and add support
    for lowering without sse2.  Add a couple of new testcases.

    Fixes a few libgomp tests and latent bugs.  Remove a few todos.


    svn-id: https://llvm.org/svn/llvm-project/llvm/trunk at 109078 
-------------------------------------------------------------------------------

The code is executed on this processor:
-------------------------------------------------------------------------------
processor    : 3
vendor_id    : GenuineIntel
cpu family    : 6
model        : 37
model name    : Intel(R) Core(TM) i5 CPU       M 520  @ 2.40GHz
stepping    : 2
cpu MHz        : 2400.000
cache size    : 3072 KB
physical id    : 0
siblings    : 4
core id        : 2
cpu cores    : 2
apicid        : 5
initial apicid    : 5
fpu        : yes
fpu_exception    : yes
cpuid level    : 11
wp        : yes
flags        : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov
pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm
constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf
pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1
sse4_2 popcnt aes lahf_lm ida arat tpr_shadow vnmi flexpriority ept vpid
bogomips    : 4787.92
clflush size    : 64
cache_alignment    : 64
address sizes    : 36 bits physical, 48 bits virtual
power management:
-------------------------------------------------------------------------------

running a recent 64bit debian system.

I tried to reduce the bug with bugpoint, but did not understand how to use
bugpoint to debug lli test cases. Is this possible?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sun Jul 25 23:35:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sun, 25 Jul 2010 23:35:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7713] New: DAGCombiner crashes during phase 2 of
 Release Clang self-host
Message-ID: <bug-7713-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7713

           Summary: DAGCombiner crashes during phase 2 of Release Clang
                    self-host
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: cdavis at mymail.mines.edu
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5267)
 --> (http://llvm.org/bugs/attachment.cgi?id=5267)
Stack trace from Clang crashing inside DAG Combiner

I've hit a serious bug in CodeGen. I don't know exactly what it is, but I do
know that LLVM crashes somewhere in the DAG Combiner under certain
circumstances.

This only happens when LLVM and Clang are built Release without Asserts by
recent (i.e. trunk) Clang. Building Debug or +Asserts causes the problem to go
away. This has slipped past the buildbots because they all build either Debug
or +Asserts.

An example stack trace is attached. I produced it compiling addvdi3.c from the
compiler-rt project. Running the test suite produced all sorts of similar (but
not quite the same) stack traces. Different code causes crashes in different
parts of the DAG Combiner, but the crash is always (from what I've seen,
anyway) in the DAG Combiner.

I also know that the crash happens when the DAG Combiner tries to dereference a
NULL SDNode. How could CodeGen produce a DAG with a NULL SDNode?

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 01:38:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 01:38:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7714] New: probable clang wrong code bug on x64
Message-ID: <bug-7714-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7714

           Summary: probable clang wrong code bug on x64
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


Oddly, I think the -O0 output is wrong.

[regehr at bethe work011]$ clang -O0 small.c -o small
[regehr at bethe work011]$ ./small
0
[regehr at bethe work011]$ clang -O1 small.c -o small
[regehr at bethe work011]$ ./small
-1
[regehr at bethe work011]$ cat small.c
extern int printf (__const char *__restrict __format, ...);

struct S0 {
  short f0;
  unsigned f1;
  int f2;
};

int g_12[3] = {-10L, -10L, -10L};

void func_2(int p_3, struct S0 p_4)
{
  if (p_4.f0) g_12[2] = p_3;
}

int main(void)
{
  struct S0 g_57 = {1L,8L,-6L};
  func_2 (-1L, g_57);
  printf ("%d\n", g_12[2]);
  return 0;
}
[regehr at bethe work011]$ clang -v
clang version 2.8 (trunk 109392)
Target: x86_64-unknown-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 04:21:23 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 04:21:23 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7715] New: clang++ gives wrong error about default
 argument in kdatetime.h
Message-ID: <bug-7715-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7715

           Summary: clang++ gives wrong error about default argument in
                    kdatetime.h
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: ambrop7 at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5268)
 --> (http://llvm.org/bugs/attachment.cgi?id=5268)
Preprocessed source

Trying to compile a program that includes KDE's kdatetime.h results in the
errors:

In file included from a.cpp:1:
/usr/include/kdatetime.h:535:62: error: use of default argument to function
'Spec' that is declared later in class 'Spec'
    explicit KDateTime(const QDate &date, const Spec &spec = Spec(LocalZone));
                                                             ^
/usr/include/kdatetime.h:245:45: note: default argument declared here           
        Spec(SpecType type, int utcOffset = 0);
                                            ^
/usr/include/kdatetime.h:554:72: error: use of default argument to function
'Spec' that is declared later in class 'Spec'                         
    KDateTime(const QDate &date, const QTime &time, const Spec &spec =
Spec(LocalZone));
                                                                       ^
/usr/include/kdatetime.h:245:45: note: default argument declared here           
        Spec(SpecType type, int utcOffset = 0);
                                            ^

It can be seen from line numbers that the default argument is really defined
earlier, not later.

This can be reproduced like this:
- have kdelibs (4.4)
- compile the file a.cpp "#include <kdatetime.h>" with something like "clang++
-c -I/usr/include/qt4 a.cpp"

kdatetime.h:
http://websvn.kde.org/branches/KDE/4.4/kdelibs/kdecore/date/kdatetime.h?revision=1154417&view=markup

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 13:05:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 13:05:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7716] New: GlobalMerge pass breaks ARM EH info
Message-ID: <bug-7716-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7716

           Summary: GlobalMerge pass breaks ARM EH info
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu


Clang is failing on ctor_dtor_count-2 and many other tests in the LLVM test
suite. This is a regression starting with r109359:
--
ddunbar at ozzy-3:EH (master)$ rm -rf Output && mkdir Output && clang++ -O3 -mllvm
-disable-llvm-optzns -mthumb -arch armv7 -fomit-frame-pointer -mdynamic-no-pic
-c ctor_dtor_count-2.cpp -o Output/ctor_dtor_count-2.bc -emit-llvm && opt
-std-compile-opts -std-link-opts Output/ctor_dtor_count-2.bc -o
Output/ctor_dtor_count-2.linked.bc && llc -asm-verbose=false
-enable-correct-eh-support -O3 -mcpu cortex-a8 -relocation-model pic
Output/ctor_dtor_count-2.linked.bc -o /dev/null
Assertion failed: ((GV || isa<ConstantPointerNull>(V)) && "TypeInfo must be a
global variable or NULL"), function ExtractTypeInfo, file
/Volumes/Data/ddunbar/llvm/lib/CodeGen/Analysis.cpp, line 121.
0  llc               0x0000000100be5ab5 PrintStackTrace(void*) + 53
1  llc               0x0000000100be5fd3 SignalHandler(int) + 371
2  libSystem.B.dylib 0x00007fff83ead35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff83e520aa tiny_malloc_from_free_list + 1196
4  libSystem.B.dylib 0x00007fff83f289b4 __pthread_markcancel + 0
5  llc               0x000000010068acd1 llvm::ExtractTypeInfo(llvm::Value*) +
385
6  llc               0x00000001004865a2 llvm::AddCatchInfo(llvm::CallInst
const&, llvm::MachineModuleInfo*, llvm::MachineBasicBlock*) + 978
7  llc               0x00000001005a44d3
llvm::SelectionDAGBuilder::visitIntrinsicCall(llvm::CallInst const&, unsigned
int) + 7619
8  llc               0x00000001005824df
llvm::SelectionDAGBuilder::visitCall(llvm::CallInst const&) + 367
9  llc               0x000000010057c4d9
llvm::SelectionDAGBuilder::visit(unsigned int, llvm::User const&) + 1033
10 llc               0x000000010057ba10
llvm::SelectionDAGBuilder::visit(llvm::Instruction const&) + 128
11 llc               0x00000001005c66c3
llvm::SelectionDAGISel::SelectBasicBlock(llvm::ilist_iterator<llvm::Instruction
const>, llvm::ilist_iterator<llvm::Instruction const>, bool&) + 179
12 llc               0x00000001005c6443
llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function const&) + 1891
13 llc               0x00000001005c543a
llvm::SelectionDAGISel::runOnMachineFunction(llvm::MachineFunction&) + 842
14 llc               0x0000000100715b21
llvm::MachineFunctionPass::runOnFunction(llvm::Function&) + 97
15 llc               0x0000000100aec9c5
llvm::FPPassManager::runOnFunction(llvm::Function&) + 389
16 llc               0x0000000100aecd41
llvm::FPPassManager::runOnModule(llvm::Module&) + 129
17 llc               0x0000000100aecf79
llvm::MPPassManager::runOnModule(llvm::Module&) + 473
18 llc               0x0000000100aed6c9
llvm::PassManagerImpl::run(llvm::Module&) + 185
19 llc               0x0000000100aedb1d llvm::PassManager::run(llvm::Module&) +
29
20 llc               0x0000000100001ad4 main + 2900
21 llc               0x0000000100000f34 start + 52
Stack dump:
0.    Program arguments: llc -asm-verbose=false -enable-correct-eh-support -O3
-mcpu cortex-a8 -relocation-model pic Output/ctor_dtor_count-2.linked.bc -o
/dev/null 
1.    Running pass 'Function Pass Manager' on module
'Output/ctor_dtor_count-2.linked.bc'.
2.    Running pass 'ARM Instruction Selection' on function '@main'
Abort trap
ddunbar at ozzy-3:EH (master)$ 
--

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 13:16:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 13:16:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7701] llc crash in out-of-bounds access
In-Reply-To: <bug-7701-206@http.llvm.org/bugs/>
References: <bug-7701-206@http.llvm.org/bugs/>
Message-ID: <20100726181655.2FF682A6C132@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7701

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |gohman at apple.com
         Resolution|                            |FIXED

--- Comment #1 from Dan Gohman <gohman at apple.com> 2010-07-26 13:16:54 CDT ---
Fixed in r109415.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 13:28:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 13:28:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7562] -mem2reg -indvars segfaults
In-Reply-To: <bug-7562-206@http.llvm.org/bugs/>
References: <bug-7562-206@http.llvm.org/bugs/>
Message-ID: <20100726182839.BC8542A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7562

Dan Gohman <gohman at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Dan Gohman <gohman at apple.com> 2010-07-26 13:28:38 CDT ---
Fixed in r109419.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 13:46:09 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 13:46:09 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7716] GlobalMerge pass breaks ARM EH info
In-Reply-To: <bug-7716-206@http.llvm.org/bugs/>
References: <bug-7716-206@http.llvm.org/bugs/>
Message-ID: <20100726184609.722512A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7716

Anton Korobeynikov <asl at math.spbu.ru> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Anton Korobeynikov <asl at math.spbu.ru> 2010-07-26 13:46:08 CDT ---
Should be fixed in r109423

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 14:25:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 14:25:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7717] New: problem with error reporting on throw
	specifiers
Message-ID: <bug-7717-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7717

           Summary: problem with error reporting on throw specifiers
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The code listed below produces the following error on gcc. 

error: looser throw specifier for ?virtual cChild::~cChild() throw
(std::exception&)?
error:   overriding ?virtual cParent::~cParent() throw ()?


It seem to me that it should produce an error 


"error: exception specification of overriding function is more lax than base
version"

on clang. It does not. However, I have had this error message returned on code
that doesn't in fact have a looser exception specifier on the child than it
does on the parent. I cannot reproduce the error on code simple enough to
submit for your review. When attempting to produce an example to demonstrate
the problem, I wasn't able to cause the error to happen even when it should
happen.



class cParent
{
public:

    virtual ~cParent() throw()
    {
    }


};



class cChild : public cParent
{
public:

    virtual ~cChild() throw(std::exception&)
    {
    }

};



int main (int argc, const char * argv[]) 
{

    cChild* c = new cChild();
    return 0;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 14:35:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 14:35:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7717] problem with error reporting on throw
	specifiers
In-Reply-To: <bug-7717-206@http.llvm.org/bugs/>
References: <bug-7717-206@http.llvm.org/bugs/>
Message-ID: <20100726193501.365CE2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7717

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |INVALID

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-26 14:35:00 CDT ---
Please report bugs with Apple builds of clang directly to Apple.  The given
testcase gives the expected error with trunk clang.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 19:24:08 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 19:24:08 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7718] New: Redundant pair of load / store is not
 eliminated by -std-compile-opts
Message-ID: <bug-7718-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7718

           Summary: Redundant pair of load / store is not eliminated by
                    -std-compile-opts
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Loop Optimizer
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5271)
 --> (http://llvm.org/bugs/attachment.cgi?id=5271)
Example LLVM IR

Consider the attached bitcode containing quite ugly-written FIR core. After opt
-std-compile-outs In the innermost loop we have:

  %28 = load i32** @coefficient1.2129, align 4    ; <i32*> [#uses=2]
...
  store i32* %35, i32** @coefficient1.2129, align 4

This pair of load/stores should be moved out of loop:
  - @coefficient1.2129 is initialized just before the loop
  - the value can be reused during the loop iteration
  - the final value can be just stored after the loop

Exactly the same thing can be applied to @signalOutLow1.2126 and
@history1Low1.2127. In short: all loads / stores to globals should be raised to
regs and only final values stored after the loop.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Mon Jul 26 22:07:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Mon, 26 Jul 2010 22:07:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7707] Assertion in TwoAddressInstructionPass
In-Reply-To: <bug-7707-206@http.llvm.org/bugs/>
References: <bug-7707-206@http.llvm.org/bugs/>
Message-ID: <20100727030701.C75032A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7707

Evan Cheng <evan.cheng at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
                 CC|                            |evan.cheng at apple.com
         Resolution|                            |FIXED

--- Comment #3 from Evan Cheng <evan.cheng at apple.com> 2010-07-26 22:07:01 CDT ---
ARM fastisel isn't ready. I've disabled it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 06:15:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 06:15:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7719] New: selfhost phase 2 broken with release
	build
Message-ID: <bug-7719-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7719

           Summary: selfhost phase 2 broken with release build
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: devlists at shadowlab.org
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5273)
 --> (http://llvm.org/bugs/attachment.cgi?id=5273)
Decl preprocessed, and assembly

After r108614 (Another attempt at getting the clang self-host to like my
instcombine patch.), self-host no longer works with release build.

See the attached Decl.cpp for info. There is one compiled with clang++ r108613
and one with clang++ r108614.

When compiling clang in release mode with clang r108614, the resulting clang
fails or crash in a bunch of tests (make test in clang)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 07:12:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 07:12:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7720] New: AliasAnalysis passes cannot be chained
Message-ID: <bug-7720-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7720

           Summary: AliasAnalysis passes cannot be chained
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Interprocedural Analyses
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru
                CC: llvmbugs at cs.uiuc.edu


Consider the following simple patch:
diff --git a/include/llvm/Support/StandardPasses.h
b/include/llvm/Support/StandardPasses.h
index b97ad29..3f3d970 100644
--- a/include/llvm/Support/StandardPasses.h
+++ b/include/llvm/Support/StandardPasses.h
@@ -129,6 +129,8 @@ namespace llvm {
     PM->add(createCFGSimplificationPass());     // Merge & remove BBs
     PM->add(createReassociatePass());           // Reassociate expressions
     PM->add(createLoopRotatePass());            // Rotate Loop
+    if (OptimizationLevel > 2)
+      PM->add(createGlobalsModRefPass());
     PM->add(createLICMPass());                  // Hoist loop invariants
     PM->add(createLoopUnswitchPass(OptimizeSize || OptimizationLevel < 3));
     PM->add(createInstructionCombiningPass());  

I'm seeing completely different behavior when createStandardModulePasses() is
called from opt and from llvm-gcc.

When running opt -O3 is run there are calls to GlobalsModRegAA::alias(), but
nothing from llvm-gcc -O3. The pass itself initializes fine in both cases.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 08:01:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 08:01:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7721] New: -ast-print-xml Produces Invalid XML
Message-ID: <bug-7721-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7721

           Summary: -ast-print-xml Produces Invalid XML
           Product: clang
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chobbs at qnx.com
                CC: llvmbugs at cs.uiuc.edu


I am using:
clang version 1.1 (branches/release_27)
Target: i386-pc-linux-gnu
Thread model: posix

clang+llvm-2.7-i686-linux

When executing the command

clang -cc1 -ast-print-xml foo.c

on a file containing the following construction:

if (paddr + n > UINTPTR_MAX) {
    return -EOVERFLOW;
}

the following XML is produced. I am no XML expert but all (except one) of the
parsers I have used to read it have taken a bath on the less-than sign in
endfile="<scratch space>", reporting it to be illegal. Changing it manually to
&lt; and the corresponding > to &gt; makes it all OK.

I have another issue of wondering what endfile="<scratch space>" is doing there
but I think the syntax problem needs to be sorted out before the semantic one.

===========XML follows

<IfStmt file="f42" line="158" col="4" endline="160" endcol="4">
 <BinaryOperator file="f42" line="158" col="7" endfile="<scratch space>"
endline="118" endcol="1" type="_18" kind="gt">
  <BinaryOperator file="f42" line="158" col="7" endcol="15" type="_C"
kind="add">
   <DeclRefExpr file="f42" line="158" col="7" type="_1152" ref="_1D4C"
name="paddr"/>
    <ImplicitCastExpr file="f42" line="158" col="15" type="_C">
     <DeclRefExpr file="f42" line="158" col="15" type="_18" ref="_1D4F"
name="n"/>
    </ImplicitCastExpr>
  </BinaryOperator>
  <ImplicitCastExpr file="f1C" line="118" col="1" type="_C">
   <IntegerLiteral file="f1C" line="118" col="1" type="_2AA" value="-1"/>
  </ImplicitCastExpr>
 </BinaryOperator>
 <CompoundStmt file="f42" line="158" col="32" endline="160" endcol="4"
num_stmts="1">
  <ReturnStmt file="f42" line="159" col="5"
endfile="/home/chobbs/sil3/stagingArea/trunk/usr/include/errno.h" endline="137"
endcol="25">
  <UnaryOperator file="f42" line="159" col="12"
endfile="/home/chobbs/sil3/stagingArea/trunk/usr/include/errno.h" endline="137"
endcol="25" type="_18" kind="minus">
<IntegerLiteral file="f1F" line="137" col="25" type="_18" value="79"/>
</UnaryOperator>
</ReturnStmt>
</CompoundStmt>
<NULL/>
</IfStmt>

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 08:15:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 08:15:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7721] -ast-print-xml Produces Invalid XML
In-Reply-To: <bug-7721-206@http.llvm.org/bugs/>
References: <bug-7721-206@http.llvm.org/bugs/>
Message-ID: <20100727131541.125222A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7721

Benjamin Kramer <benny.kra at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |benny.kra at gmail.com
         Resolution|                            |FIXED

--- Comment #1 from Benjamin Kramer <benny.kra at gmail.com> 2010-07-27 08:15:40 CDT ---
This was fixed in clang trunk a while ago. (r103671)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 09:51:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 09:51:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 1953] Improve llvm-gcc complex code generation
In-Reply-To: <bug-1953-206@http.llvm.org/bugs/>
References: <bug-1953-206@http.llvm.org/bugs/>
Message-ID: <20100727145124.AD6002A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=1953

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-27 09:51:24 CDT ---
Not going to fix this in llvm-gcc.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 10:47:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 10:47:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7722] New: Unhandled exception ends in infinite loop
Message-ID: <bug-7722-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7722

           Summary: Unhandled exception ends in infinite loop
           Product: clang
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: trash at phoyd.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The following code does not terminate, but instead prints "~V()" in a loop:
$ cat sample.cc
extern "C" { extern int puts(const char *s); }
struct V {
  ~V() 
  {
   puts("~V()");
  }
};

int main() {
    V y;
    throw 1;
}
$
$ uname -a
Linux oblomow 2.6.31.12-0.2-desktop #1 SMP PREEMPT 2010-03-16 21:25:39 +0100
x86_64 x86_64 x86_64 GNU/Linux
$
$ /opt/clang+llvm-2.7-x86_64-linux/bin/clang++ -v
clang version 1.1 (branches/release_27)
Target: x86_64-unknown-linux-gnu
Thread model: posix
$
$ /opt/clang+llvm-2.7-x86_64-linux/bin/clang++ sample.cc 
$ ./a.out
~V()
~V()
~V()
...(etc)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 11:45:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 11:45:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7722] Unhandled exception ends in infinite loop
In-Reply-To: <bug-7722-206@http.llvm.org/bugs/>
References: <bug-7722-206@http.llvm.org/bugs/>
Message-ID: <20100727164548.23A0C2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7722

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from John McCall <rjmccall at apple.com> 2010-07-27 11:45:47 CDT ---
Yeah, I'm calling this fixed;  please verify on ToT if possible.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 11:46:49 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 11:46:49 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5892] clang++: segfault after syntax error
In-Reply-To: <bug-5892-206@http.llvm.org/bugs/>
References: <bug-5892-206@http.llvm.org/bugs/>
Message-ID: <20100727164649.E36022A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5892

Christopher Jefferson <chris at bubblescope.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Christopher Jefferson <chris at bubblescope.net> 2010-07-27 11:46:48 CDT ---
Closing this bug, as the bug still appears fixed. Please feel free to re-open
it if you have any further issues, or open a new bug.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 11:50:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 11:50:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6328] Compiler assert when taking __alignof__ an
	unknown type
In-Reply-To: <bug-6328-206@http.llvm.org/bugs/>
References: <bug-6328-206@http.llvm.org/bugs/>
Message-ID: <20100727165014.2F2672A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6328

Christopher Jefferson <chris at bubblescope.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Christopher Jefferson <chris at bubblescope.net> 2010-07-27 11:50:12 CDT ---
Not sure when, but this bug seems to have got fixed somewhere along the line.
Closing.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 11:55:37 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 11:55:37 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7041] Crash without assert on invalid code
In-Reply-To: <bug-7041-206@http.llvm.org/bugs/>
References: <bug-7041-206@http.llvm.org/bugs/>
Message-ID: <20100727165537.E34E82A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7041

Christopher Jefferson <chris at bubblescope.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Christopher Jefferson <chris at bubblescope.net> 2010-07-27 11:55:37 CDT ---
This bug has been fixed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 11:57:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 11:57:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7089] Problem matching in member functions
In-Reply-To: <bug-7089-206@http.llvm.org/bugs/>
References: <bug-7089-206@http.llvm.org/bugs/>
Message-ID: <20100727165733.68B2B2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7089

Christopher Jefferson <chris at bubblescope.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Christopher Jefferson <chris at bubblescope.net> 2010-07-27 11:57:33 CDT ---
This has been fixed at some point.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 11:58:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 11:58:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7122] assert on invalid: Wrong callback for
 declspec without declarator
In-Reply-To: <bug-7122-206@http.llvm.org/bugs/>
References: <bug-7122-206@http.llvm.org/bugs/>
Message-ID: <20100727165835.5C31A2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7122

Christopher Jefferson <chris at bubblescope.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Christopher Jefferson <chris at bubblescope.net> 2010-07-27 11:58:34 CDT ---
This has been fixed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 12:27:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 12:27:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7724] New: clang++ asserts during compilation of if
 statement with compound conditional in C++ code
Message-ID: <bug-7724-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7724

           Summary: clang++ asserts during compilation of if statement
                    with compound conditional in C++ code
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: release blocker
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: mike.jackson at bluequartz.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


// Begin Code
#include <iostream>

//
-----------------------------------------------------------------------------
int GetSwapBytes() { return 1; }

//
-----------------------------------------------------------------------------
template<typename OT>
int myMethod()
{
if (GetSwapBytes() && sizeof(OT) > 1) // <=== HERE IS THE PROBLEM
  {
    std::cout << "sizeof(OT) == 4" << std::endl;
  }
  return 1;
}

//
-----------------------------------------------------------------------------
int main (int argc, char const *argv[])
{
  int ret = myMethod<int>();  
  return 0;
}
// END CODE

The above code will cause clang++ to assert during compiling. The assertion is:

595:[mjackson at ferb:clang]$ clang++ ~/Desktop/test.cpp -o ~/Desktop/clang-test
Assertion failed: (!isDependentType() && "This doesn't make sense for dependent
types"), function isConstantSizeType, file
/Users/Shared/OpenSource/llvm/tools/clang/lib/AST/Type.cpp, line 616.
Stack dump:
0.    Program arguments: /Users/Shared/OpenSource/llvm/Build/bin/clang -cc1
-triple i386-apple-darwin9.0.0 -emit-obj -mrelax-all -disable-free
-main-file-name test.cpp -pic-level 1 -mdisable-fp-elim -masm-verbose
-target-cpu yonah -resource-dir
/Users/Shared/OpenSource/llvm/Build/lib/clang/2.8 -ferror-limit 19
-fmessage-length 148 -fblocks -fexceptions -fdiagnostics-show-option
-fcolor-diagnostics -o
/var/folders/Zm/ZmwaqVZrGbee7mHKbefzNU+++TI/-Tmp-/cc-v8OIWq.o -x c++
/Users/mjackson/Desktop/test.cpp 
1.    /Users/mjackson/Desktop/test.cpp:51:37: current parser token ')'
2.    /Users/mjackson/Desktop/test.cpp:50:1: parsing function body 'myMethod'
3.    /Users/mjackson/Desktop/test.cpp:50:1: in compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)


GCC and Visual Studio 2008 both compile this code cleanly.

LLVM and Clang both checked out from svn on July 27,2010 at about 10:00 AM EST.
Built using GCC 4.2 on OS X 10.5.8 using standard Xcode 3.1.4 tools.

clang++ --version
clang version 2.8 (trunk 109508)
Target: i386-apple-darwin9
Thread model: posix

Thanks
Mike Jackson

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 14:15:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 14:15:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7724] clang++ asserts during compilation of if
 statement with compound conditional in C++ code
In-Reply-To: <bug-7724-206@http.llvm.org/bugs/>
References: <bug-7724-206@http.llvm.org/bugs/>
Message-ID: <20100727191514.A29312A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7724

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-27 14:15:14 CDT ---
Fixed in r109532.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 14:44:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 14:44:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7725] New: [c++] clang++ throws compile error on
 possible C++ template 2 level namespace lookup
Message-ID: <bug-7725-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7725

           Summary: [c++] clang++ throws compile error on possible C++
                    template 2 level namespace lookup
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: mike.jackson at bluequartz.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


// BEGIN CODE
//
-----------------------------------------------------------------------------
class TestClass
{
public:
void PrintSelf() {std::cout << "TestClass.PrintSelf()" << std::endl;}
};

//
-----------------------------------------------------------------------------
template<class ignored>
class TypedefProvider : public TestClass
{
public:
typedef TestClass Test;
};

//
-----------------------------------------------------------------------------
template<typename T>
class TemplateClass : public TypedefProvider< T >
{
public:
void PrintSelf();
};

//
-----------------------------------------------------------------------------
template<typename T>
void TemplateClass<T>::PrintSelf()
{
 this->TemplateClass<T>::Test::PrintSelf(); // ERROR
}

//
-----------------------------------------------------------------------------
int main ()
{
TemplateClass<int> c;
c.PrintSelf();
return 0;
}
// END CODE

This code compiles cleanly with GCC and Visual Studio 2008 but does not under
clang++.


620:[mjackson at ferb:clang]$ clang++ --version
clang version 2.8 (trunk 109532)
Target: i386-apple-darwin9
Thread model: posix

Clang and LLVM compiled on OS X 10.5.8 using Xcode tools 3.1.4.

Thanks
Mike Jackson

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 16:32:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 16:32:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7726] New: Invalid codegen for "Conditionals with
 Omitted Operands" (?:) expression in c++
Message-ID: <bug-7726-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7726

           Summary: Invalid codegen for "Conditionals with Omitted
                    Operands" (?:) expression in c++
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: devlists at shadowlab.org
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


When using the "conditional with omitted operands" extension in C++, clang
incorrectly cast the first operand in bool.

Example: 
============

#include <stdio.h>

int main(int argc, char **argv) {
    printf("%p\n", (void *)0xdeadbeef ? : (void *)0xaaaaaa);
    return 0;
}

===========

clang++ will cast '0xdeadbeef' as bool and so the executable will print "0x1"
instead of the pointer value.

When compiling this program as C, the implicit case is not generated and so the
output is correct.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:07:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:07:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7727] New: LLVM IR generation of compound statement
 ('{}') for Markzware's HashTable.ii
Message-ID: <bug-7727-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7727

           Summary: LLVM IR generation of compound statement ('{}') for
                    Markzware's HashTable.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


the following error log is produced by trying to compile Hashtable.ii. File is
attached.

$ /opt/bin/clang++ Hashtable.ii
Assertion failed: ((Record->hasTrivialCopyConstructor() ||
Record->hasTrivialCopyAssignment()) && "Trying to aggregate-copy a type without
a trivial copy " "constructor or assignment operator"), function
EmitAggregateCopy, file CGExprAgg.cpp, line 796.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff869f30aa tiny_malloc_from_free_list + 1196
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100197e6f llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 3711
6  clang             0x0000000100198000 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 4112
7  clang             0x000000010019af89 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 16281
8  clang             0x000000010019c946 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 22870
9  clang             0x000000010019a8be llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 14542
10 clang             0x000000010019ba9b llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 19115
11 clang             0x000000010018e6c0
clang::CanQual<clang::Type>::getTypePtr() const + 36864
12 clang             0x000000010019191c
clang::CanQual<clang::Type>::getTypePtr() const + 49756
13 clang             0x0000000100190c12
clang::CanQual<clang::Type>::getTypePtr() const + 46418
14 clang             0x0000000100195368
clang::CanQual<clang::Type>::getTypePtr() const + 64680
15 clang             0x0000000100195c5a
clang::CanQual<clang::Type>::getTypePtr() const + 66970
16 clang             0x00000001001779fa llvm::DenseMap<clang::Decl const*,
llvm::WeakVH, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::WeakVH> >::FindAndConstruct(clang::Decl const* const&)
+ 22426
17 clang             0x00000001001789ff llvm::DenseMap<clang::Decl const*,
llvm::WeakVH, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::WeakVH> >::FindAndConstruct(clang::Decl const* const&)
+ 26527
18 clang             0x000000010021e52b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 20955
19 clang             0x0000000100221ce7 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35223
20 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
21 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
22 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
23 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
24 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
25 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
26 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
27 clang             0x0000000100257293 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42691
28 clang             0x00000001002572e3 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42771
29 clang             0x000000010024134a llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 1450
30 clang             0x0000000100284eb0 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 1264
31 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
32 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
33 clang             0x00000001000028b5
34 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
35 clang             0x0000000100000a34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name Hashtable.ii -pic-level 1
-mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2 -resource-dir
/opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 186 -stack-protector 1
-fblocks -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-purpnR.o -x c++-cpp-output
Hashtable.ii 
1.    <eof> parser at end of file
2.    Per-file LLVM IR generation
3.    ./java/util/HashMap.h:397:14: Generating code for declaration
'java::util::HashMap<java::lang::String, java::lang::Object,
java::util::private_detail::cDefaultHashMapEntry>::contains'
4.    ./java/util/HashMap.h:398:9: LLVM IR generation of compound statement
('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:17:06 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:17:06 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7728] New: LLVM IR generation of compound statement
 ('{}') in Markzware
Message-ID: <bug-7728-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7728

           Summary: LLVM IR generation of compound statement ('{}') in
                    Markzware
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com




-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:28:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:28:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7729] New: LLVM IR generation of declaration in
 Markzware's JdbcStatement.ii
Message-ID: <bug-7729-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7729

           Summary: LLVM IR generation of declaration in Markzware's
                    JdbcStatement.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5276)
 --> (http://llvm.org/bugs/attachment.cgi?id=5276)
preprocessed file

the following is the error log produced by compiling  JdbcStatement.ii. File is
attached.


$ /opt/bin/clang++ JdbcStatement.ii
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:1:
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:10:
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.h:13:
In file included from ./org/h2/message/TraceObject.h:15:
In file included from ./org/h2/expression/ParameterInterface.h:13:
./org/h2/value/Value.h:337:9: warning: control reaches end of non-void function
[-Wreturn-type]
        }
        ^
Assertion failed: (!EHBranchThroughDest), function PopCleanupBlock, file
CodeGenFunction.cpp, line 835.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  clang             0x000000010066cfc6 llvm::DenseMap<clang::DeclarationName,
clang::StoredDeclsList, llvm::DenseMapInfo<clang::DeclarationName>,
llvm::DenseMapInfo<clang::StoredDeclsList>
>::FindAndConstruct(clang::DeclarationName const&) + 8550
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100248171
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20145
6  clang             0x00000001002481c8
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20232
7  clang             0x00000001002231b6 llvm::cast_retty<clang::LabelStmt,
clang::Stmt>::ret_type llvm::cast<clang::LabelStmt, clang::Stmt>(clang::Stmt
const&) + 294
8  clang             0x00000001001bf82c llvm::cast_retty<clang::ObjCEncodeExpr,
clang::Expr*>::ret_type llvm::cast<clang::ObjCEncodeExpr,
clang::Expr*>(clang::Expr* const&) + 44188
9  clang             0x00000001001c0249 llvm::cast_retty<clang::ObjCEncodeExpr,
clang::Expr*>::ret_type llvm::cast<clang::ObjCEncodeExpr,
clang::Expr*>(clang::Expr* const&) + 46777
10 clang             0x000000010018727d
clang::CanQual<clang::Type>::getTypePtr() const + 7101
11 clang             0x000000010021ec4d llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22781
12 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
13 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
14 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
15 clang             0x00000001002217cc llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 33916
16 clang             0x000000010021eee5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23445
17 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
18 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
19 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
20 clang             0x0000000100183939 std::vector<std::pair<llvm::WeakVH,
llvm::Constant*>, std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::WeakVH,
llvm::Constant*>*, std::vector<std::pair<llvm::WeakVH, llvm::Constant*>,
std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> > > >,
std::pair<llvm::WeakVH, llvm::Constant*> const&) + 31305
21 clang             0x000000010021ef86 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23606
22 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
23 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
24 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
25 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
26 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
27 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
28 clang             0x000000010025795f llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 44431
29 clang             0x00000001002586f0 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 47904
30 clang             0x000000010027166c
llvm::SmallVectorTemplateBase<llvm::APInt, false>::grow(unsigned long) + 1388
31 clang             0x00000001002416db llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 2363
32 clang             0x0000000100284c72 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 690
33 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
34 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
35 clang             0x00000001000028b5
36 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
37 clang             0x0000000100000a34
38 clang             0x0000000000000021
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name JdbcStatement.ii -pic-level
1 -mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2
-resource-dir /opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272
-stack-protector 1 -fblocks -fexceptions -fdiagnostics-show-option
-fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-1d2y4j.o -x c++-cpp-output
JdbcStatement.ii 
1.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:127:1:
current parser token 'size_t'
2.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:57:56:
LLVM IR generation of declaration
'org::h2::jdbc::JdbcStatement::executeQuerySPtr'
3.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:57:56:
Generating code for declaration
'org::h2::jdbc::JdbcStatement::executeQuerySPtr'
4.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:58:1:
LLVM IR generation of compound statement ('{}')
5.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:61:5:
LLVM IR generation of compound statement ('{}')
6.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/jdbc/JdbcStatement.cpp:64:9:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:33:26 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:33:26 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7730] New: LLVM IR generation of compound statement
 ('{}') in Markzware's TraceObject.ii
Message-ID: <bug-7730-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7730

           Summary: LLVM IR generation of compound statement ('{}') in
                    Markzware's TraceObject.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5277)
 --> (http://llvm.org/bugs/attachment.cgi?id=5277)
preprocessed file

the following is the error generated by compiling  TraceObject.ii. file is
attached



$ /opt/bin/clang++ TraceObject.ii
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:1:
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:10:
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.h:15:
In file included from ./org/h2/expression/ParameterInterface.h:13:
./org/h2/value/Value.h:337:9: warning: control reaches end of non-void function
[-Wreturn-type]
        }
        ^
Assertion failed: (!EHBranchThroughDest), function PopCleanupBlock, file
CodeGenFunction.cpp, line 835.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff5fbfa880 _sigtramp + 3642410304
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100248171
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20145
6  clang             0x00000001002481c8
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20232
7  clang             0x00000001002231b6 llvm::cast_retty<clang::LabelStmt,
clang::Stmt>::ret_type llvm::cast<clang::LabelStmt, clang::Stmt>(clang::Stmt
const&) + 294
8  clang             0x00000001001bf82c llvm::cast_retty<clang::ObjCEncodeExpr,
clang::Expr*>::ret_type llvm::cast<clang::ObjCEncodeExpr,
clang::Expr*>(clang::Expr* const&) + 44188
9  clang             0x00000001001c0249 llvm::cast_retty<clang::ObjCEncodeExpr,
clang::Expr*>::ret_type llvm::cast<clang::ObjCEncodeExpr,
clang::Expr*>(clang::Expr* const&) + 46777
10 clang             0x000000010018727d
clang::CanQual<clang::Type>::getTypePtr() const + 7101
11 clang             0x000000010021ec4d llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22781
12 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
13 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
14 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
15 clang             0x0000000100183939 std::vector<std::pair<llvm::WeakVH,
llvm::Constant*>, std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::WeakVH,
llvm::Constant*>*, std::vector<std::pair<llvm::WeakVH, llvm::Constant*>,
std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> > > >,
std::pair<llvm::WeakVH, llvm::Constant*> const&) + 31305
16 clang             0x000000010021ef86 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23606
17 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
18 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
19 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
20 clang             0x00000001002214c4 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 33140
21 clang             0x000000010021ef52 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23554
22 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
23 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
24 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
25 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
26 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
27 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
28 clang             0x000000010025795f llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 44431
29 clang             0x00000001002586f0 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 47904
30 clang             0x000000010027166c
llvm::SmallVectorTemplateBase<llvm::APInt, false>::grow(unsigned long) + 1388
31 clang             0x00000001002416db llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 2363
32 clang             0x0000000100284c72 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 690
33 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
34 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
35 clang             0x00000001000028b5
36 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
37 clang             0x0000000100000a34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name TraceObject.ii -pic-level 1
-mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2 -resource-dir
/opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272 -stack-protector 1
-fblocks -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-YymPqf.o -x c++-cpp-output
TraceObject.ii 
1.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:163:1:
current parser token 'std'
2.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:112:33:
LLVM IR generation of declaration 'org::h2::message::TraceObject::toString'
3.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:112:33:
Generating code for declaration 'org::h2::message::TraceObject::toString'
4.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:113:1:
LLVM IR generation of compound statement ('{}')
5.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:128:5:
LLVM IR generation of compound statement ('{}')
6.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceObject.cpp:134:9:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:40:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:40:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7731] New: LLVM IR generation of declaration in
 Markzware's IOUtils.ii
Message-ID: <bug-7731-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7731

           Summary: LLVM IR generation of declaration in Markzware's
                    IOUtils.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5278)
 --> (http://llvm.org/bugs/attachment.cgi?id=5278)
preprocessed file

the following is the error log created by compiling IOUtils.ii. file is
attached.



$ /opt/bin/clang++ IOUtils.ii
Assertion failed: (!EHBranchThroughDest), function PopCleanupBlock, file
CodeGenFunction.cpp, line 835.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x000000010201b000 _sigtramp + 2069679296
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100248171
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20145
6  clang             0x00000001002481c8
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20232
7  clang             0x0000000100221ad8 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34696
8  clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
9  clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
10 clang             0x0000000100183939 std::vector<std::pair<llvm::WeakVH,
llvm::Constant*>, std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::WeakVH,
llvm::Constant*>*, std::vector<std::pair<llvm::WeakVH, llvm::Constant*>,
std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> > > >,
std::pair<llvm::WeakVH, llvm::Constant*> const&) + 31305
11 clang             0x000000010021ef86 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23606
12 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
13 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
14 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
15 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
16 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
17 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
18 clang             0x000000010025795f llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 44431
19 clang             0x00000001002586f0 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 47904
20 clang             0x000000010027166c
llvm::SmallVectorTemplateBase<llvm::APInt, false>::grow(unsigned long) + 1388
21 clang             0x00000001002416db llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 2363
22 clang             0x0000000100284c72 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 690
23 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
24 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
25 clang             0x00000001000028b5
26 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
27 clang             0x0000000100000a34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name IOUtils.ii -pic-level 1
-mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2 -resource-dir
/opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272 -stack-protector 1
-fblocks -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-cF2Tm6.o -x c++-cpp-output
IOUtils.ii 
1.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/IOUtils.cpp:394:1:
current parser token 'boost'
2.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/IOUtils.cpp:370:46:
LLVM IR generation of declaration 'org::h2::util::IOUtils::getReader'
3.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/IOUtils.cpp:370:46:
Generating code for declaration 'org::h2::util::IOUtils::getReader'
4.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/IOUtils.cpp:371:1:
LLVM IR generation of compound statement ('{}')
5.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/IOUtils.cpp:378:5:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:44:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:44:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7732] New: LLVM IR generation of compound statement
 in Markzware's StringUtils.ii
Message-ID: <bug-7732-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7732

           Summary: LLVM IR generation of compound statement in
                    Markzware's StringUtils.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5279)
 --> (http://llvm.org/bugs/attachment.cgi?id=5279)
preprocessed file

The following is the error log created by compiling StringUtils.ii.  file is
attached.




$ /opt/bin/clang++ StringUtils.ii
In file included from
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:1:
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:135:1:
warning: control reaches end of non-void function [-Wreturn-type]
}
^
Assertion failed: (!EHBranchThroughDest), function PopCleanupBlock, file
CodeGenFunction.cpp, line 835.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff5fbfbe10 _sigtramp + 3642415824
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100248171
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20145
6  clang             0x00000001002481c8
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20232
7  clang             0x0000000100221ad8 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34696
8  clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
9  clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
10 clang             0x0000000100183939 std::vector<std::pair<llvm::WeakVH,
llvm::Constant*>, std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::WeakVH,
llvm::Constant*>*, std::vector<std::pair<llvm::WeakVH, llvm::Constant*>,
std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> > > >,
std::pair<llvm::WeakVH, llvm::Constant*> const&) + 31305
11 clang             0x000000010021ef86 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23606
12 clang             0x0000000100220649 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 29433
13 clang             0x0000000100221dc0 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35440
14 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
15 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
16 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
17 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
18 clang             0x000000010021f4c8 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 24952
19 clang             0x000000010021f040 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23792
20 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
21 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
22 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
23 clang             0x00000001002217cc llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 33916
24 clang             0x000000010021eee5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23445
25 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
26 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
27 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
28 clang             0x0000000100220b9e llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 30798
29 clang             0x000000010021ef38 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23528
30 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
31 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
32 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
33 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
34 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
35 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
36 clang             0x000000010025795f llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 44431
37 clang             0x00000001002586f0 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 47904
38 clang             0x000000010027166c
llvm::SmallVectorTemplateBase<llvm::APInt, false>::grow(unsigned long) + 1388
39 clang             0x00000001002416db llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 2363
40 clang             0x0000000100284c72 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 690
41 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
42 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
43 clang             0x00000001000028b5
44 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
45 clang             0x0000000100000a34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name StringUtils.ii -pic-level 1
-mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2 -resource-dir
/opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272 -stack-protector 1
-fblocks -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-QsMOpF.o -x c++-cpp-output
StringUtils.ii 
1.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:314:1:
current parser token 'boost'
2.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:198:40:
LLVM IR generation of declaration 'org::h2::util::StringUtils::javaDecodeSPtr'
3.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:198:40:
Generating code for declaration 'org::h2::util::StringUtils::javaDecodeSPtr'
4.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:199:1:
LLVM IR generation of compound statement ('{}')
5.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:206:5:
LLVM IR generation of compound statement ('{}')
6.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:213:9:
LLVM IR generation of compound statement ('{}')
7.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:226:13:
LLVM IR generation of compound statement ('{}')
8.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/util/StringUtils.cpp:259:21:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:50:37 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:50:37 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7733] New: LLVM IR generation of declaration in
 Markzware's TraceSystem.ii
Message-ID: <bug-7733-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7733

           Summary: LLVM IR generation of declaration in Markzware's
                    TraceSystem.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5280)
 --> (http://llvm.org/bugs/attachment.cgi?id=5280)
preprocessed file

The following is the error log generated by compiling TraceSystem.ii. file is
attached.


$ /opt/bin/clang++ TraceSystem.ii
Assertion failed: (!EHBranchThroughDest), function PopCleanupBlock, file
CodeGenFunction.cpp, line 835.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x000000010522d9d0 _sigtramp + 2122184336
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100248171
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20145
6  clang             0x00000001002481c8
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20232
7  clang             0x00000001002231b6 llvm::cast_retty<clang::LabelStmt,
clang::Stmt>::ret_type llvm::cast<clang::LabelStmt, clang::Stmt>(clang::Stmt
const&) + 294
8  clang             0x000000010019ab79 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 15241
9  clang             0x000000010019ba9b llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 19115
10 clang             0x0000000100187341
clang::CanQual<clang::Type>::getTypePtr() const + 7297
11 clang             0x000000010021ec4d llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22781
12 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
13 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
14 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
15 clang             0x0000000100183939 std::vector<std::pair<llvm::WeakVH,
llvm::Constant*>, std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::WeakVH,
llvm::Constant*>*, std::vector<std::pair<llvm::WeakVH, llvm::Constant*>,
std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> > > >,
std::pair<llvm::WeakVH, llvm::Constant*> const&) + 31305
16 clang             0x000000010021ef86 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23606
17 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
18 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
19 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
20 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
21 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
22 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
23 clang             0x000000010025795f llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 44431
24 clang             0x00000001002586f0 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 47904
25 clang             0x000000010027166c
llvm::SmallVectorTemplateBase<llvm::APInt, false>::grow(unsigned long) + 1388
26 clang             0x00000001002416db llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 2363
27 clang             0x0000000100284c72 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 690
28 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
29 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
30 clang             0x00000001000028b5
31 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
32 clang             0x0000000100000a34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name TraceSystem.ii -pic-level 1
-mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2 -resource-dir
/opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272 -stack-protector 1
-fblocks -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-pMiV84.o -x c++-cpp-output
TraceSystem.ii 
1.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceSystem.cpp:92:1:
current parser token 'boost'
2.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceSystem.cpp:72:27:
LLVM IR generation of declaration 'org::h2::message::TraceSystem::getTrace'
3.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceSystem.cpp:72:27:
Generating code for declaration 'org::h2::message::TraceSystem::getTrace'
4.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceSystem.cpp:73:1:
LLVM IR generation of compound statement ('{}')
5.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/message/TraceSystem.cpp:77:5:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 18:55:56 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 18:55:56 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7734] New: in Markzware's ConnectionInfo.ii
Message-ID: <bug-7734-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7734

           Summary: in Markzware's ConnectionInfo.ii
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5281)
 --> (http://llvm.org/bugs/attachment.cgi?id=5281)
preprocessed file

the following is error log produced by compiling ConnectionInfo.ii. file is
attached.



$ /opt/bin/clang++ ConnectionInfo.ii
Assertion failed: (!EHBranchThroughDest), function PopCleanupBlock, file
CodeGenFunction.cpp, line 835.
0  clang             0x00000001012cb442 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012cc193 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff5fbfbe00 _sigtramp + 3642415808
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100248171
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20145
6  clang             0x00000001002481c8
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 20232
7  clang             0x0000000100221ad8 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34696
8  clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
9  clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
10 clang             0x0000000100183939 std::vector<std::pair<llvm::WeakVH,
llvm::Constant*>, std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::WeakVH,
llvm::Constant*>*, std::vector<std::pair<llvm::WeakVH, llvm::Constant*>,
std::allocator<std::pair<llvm::WeakVH, llvm::Constant*> > > >,
std::pair<llvm::WeakVH, llvm::Constant*> const&) + 31305
11 clang             0x000000010021ef86 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 23606
12 clang             0x0000000100221a7b llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34603
13 clang             0x0000000100221de5 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35477
14 clang             0x000000010021eb60 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
15 clang             0x000000010024b820
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 34144
16 clang             0x0000000100256ba7 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40919
17 clang             0x00000001002570fa llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42282
18 clang             0x000000010025795f llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 44431
19 clang             0x00000001002586f0 llvm::DenseMap<clang::Decl const*,
llvm::Value*, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 47904
20 clang             0x000000010027166c
llvm::SmallVectorTemplateBase<llvm::APInt, false>::grow(unsigned long) + 1388
21 clang             0x00000001002416db llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 2363
22 clang             0x0000000100284c72 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 690
23 clang             0x0000000100241fac llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
24 clang             0x000000010002e359 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6457
25 clang             0x00000001000028b5
26 clang             0x0000000100009474 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
27 clang             0x0000000100000a34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name ConnectionInfo.ii
-pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2
-resource-dir /opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272
-stack-protector 1 -fblocks -fexceptions -fdiagnostics-show-option
-fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-dxavqE.o -x c++-cpp-output
ConnectionInfo.ii 
1.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/Engine/ConnectionInfo.cpp:292:1:
current parser token 'bool'
2.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/Engine/ConnectionInfo.cpp:267:25:
LLVM IR generation of declaration
'org::h2::engine::ConnectionInfo::getIntProperty'
3.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/Engine/ConnectionInfo.cpp:267:25:
Generating code for declaration
'org::h2::engine::ConnectionInfo::getIntProperty'
4.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/Engine/ConnectionInfo.cpp:268:1:
LLVM IR generation of compound statement ('{}')
5.   
/HOME/Documents/work_projects/devel/H2++Project/org/h2/Engine/ConnectionInfo.cpp:279:5:
LLVM IR generation of compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 21:37:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 21:37:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7729] LLVM IR generation of declaration in
 Markzware's JdbcStatement.ii
In-Reply-To: <bug-7729-206@http.llvm.org/bugs/>
References: <bug-7729-206@http.llvm.org/bugs/>
Message-ID: <20100728023716.018962A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7729

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-27 21:37:15 CDT ---
Looks the same as 7728

*** This bug has been marked as a duplicate of bug 7728 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 21:37:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 21:37:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7730] LLVM IR generation of compound statement
 ('{}') in Markzware's TraceObject.ii
In-Reply-To: <bug-7730-206@http.llvm.org/bugs/>
References: <bug-7730-206@http.llvm.org/bugs/>
Message-ID: <20100728023740.4F30E2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7730

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-27 21:37:39 CDT ---
Also EH related.

*** This bug has been marked as a duplicate of bug 7728 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 21:37:57 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 21:37:57 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7731] LLVM IR generation of declaration in
 Markzware's IOUtils.ii
In-Reply-To: <bug-7731-206@http.llvm.org/bugs/>
References: <bug-7731-206@http.llvm.org/bugs/>
Message-ID: <20100728023757.F071D2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7731

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-27 21:37:57 CDT ---
Also EH related.

*** This bug has been marked as a duplicate of bug 7728 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 21:38:10 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 21:38:10 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7732] LLVM IR generation of compound statement in
 Markzware's StringUtils.ii
In-Reply-To: <bug-7732-206@http.llvm.org/bugs/>
References: <bug-7732-206@http.llvm.org/bugs/>
Message-ID: <20100728023810.D32702A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7732

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-27 21:38:10 CDT ---


*** This bug has been marked as a duplicate of bug 7728 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 21:38:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 21:38:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7733] LLVM IR generation of declaration in
 Markzware's TraceSystem.ii
In-Reply-To: <bug-7733-206@http.llvm.org/bugs/>
References: <bug-7733-206@http.llvm.org/bugs/>
Message-ID: <20100728023822.581AE2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7733

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-27 21:38:22 CDT ---


*** This bug has been marked as a duplicate of bug 7728 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 21:38:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 21:38:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7734] LLVM IR generation of declaration in
 Markzware's ConnectionInfo.ii
In-Reply-To: <bug-7734-206@http.llvm.org/bugs/>
References: <bug-7734-206@http.llvm.org/bugs/>
Message-ID: <20100728023835.E8B112A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7734

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-27 21:38:35 CDT ---


*** This bug has been marked as a duplicate of bug 7728 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Tue Jul 27 23:06:11 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Tue, 27 Jul 2010 23:06:11 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7728] LLVM IR generation of compound statement
 ('{}') in Markzware
In-Reply-To: <bug-7728-206@http.llvm.org/bugs/>
References: <bug-7728-206@http.llvm.org/bugs/>
Message-ID: <20100728040611.EBD052A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7728

John McCall <rjmccall at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #11 from John McCall <rjmccall at apple.com> 2010-07-27 23:06:11 CDT ---
This crash is very similar to a bug I fixed yesterday, and I can't reproduce it
on ToT.  Feel free to re-open if you can reproduce with a clang after r109569.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 02:17:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 02:17:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7735] New: Assertion failed:
 (MF.getFrameInfo()->getObjectSize(FrameIdx) >= RC->getSize() && "Stack slot
 too small for load")
Message-ID: <bug-7735-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7735

           Summary: Assertion failed:
                    (MF.getFrameInfo()->getObjectSize(FrameIdx) >=
                    RC->getSize() && "Stack slot too small for load")
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5282)
 --> (http://llvm.org/bugs/attachment.cgi?id=5282)
bugpoint-reduced-simplified.bc

Assert when compiling ACPICA iasl on i386.

% clang -v
clang version 2.8 (trunk 109553)
Target: x86_64-apple-darwin10
Thread model: posix

% llc bugpoint-reduced-simplified.bc
Assertion failed: (MF.getFrameInfo()->getObjectSize(FrameIdx) >= RC->getSize()
&& "Stack slot too small for load"), function loadRegFromStackSlot, file
/Users/pwo/src/llvm/lib/Target/X86/X86InstrInfo.cpp, line 2037.
0  llc               0x0000000100625672 std::_Rb_tree<llvm::sys::Path,
llvm::sys::Path, std::_Identity<llvm::sys::Path>, std::less<llvm::sys::Path>,
std::allocator<llvm::sys::Path>
>::_M_erase(std::_Rb_tree_node<llvm::sys::Path>*) + 8546
1  llc               0x0000000100625b29 std::_Rb_tree<llvm::sys::Path,
llvm::sys::Path, std::_Identity<llvm::sys::Path>, std::less<llvm::sys::Path>,
std::allocator<llvm::sys::Path>
>::_M_erase(std::_Rb_tree_node<llvm::sys::Path>*) + 9753
2  libSystem.B.dylib 0x00007fff85e6835a _sigtramp + 26
3  libSystem.B.dylib 0x00000000000000f0 _sigtramp + 2048490928
4  libSystem.B.dylib 0x00007fff85ee39b4 __pthread_markcancel + 0
5  llc               0x00000001000fce49 std::vector<std::pair<llvm::EVT,
llvm::TargetRegisterClass*>, std::allocator<std::pair<llvm::EVT,
llvm::TargetRegisterClass*> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::EVT,
llvm::TargetRegisterClass*>*, std::vector<std::pair<llvm::EVT,
llvm::TargetRegisterClass*>, std::allocator<std::pair<llvm::EVT,
llvm::TargetRegisterClass*> > > >, std::pair<llvm::EVT,
llvm::TargetRegisterClass*> const&) + 27913
6  llc               0x00000001003e7b33
std::vector<std::pair<llvm::MachineBasicBlock*, unsigned int>,
std::allocator<std::pair<llvm::MachineBasicBlock*, unsigned int> >
>::_M_insert_aux(__gnu_cxx::__normal_iterator<std::pair<llvm::MachineBasicBlock*,
unsigned int>*, std::vector<std::pair<llvm::MachineBasicBlock*, unsigned int>,
std::allocator<std::pair<llvm::MachineBasicBlock*, unsigned int> > > >,
std::pair<llvm::MachineBasicBlock*, unsigned int> const&) + 4659
7  llc               0x0000000100315898 llvm::DenseMap<llvm::VNInfo*,
llvm::VNInfo*, llvm::DenseMapInfo<llvm::VNInfo*>,
llvm::DenseMapInfo<llvm::VNInfo*> >::insert(std::pair<llvm::VNInfo*,
llvm::VNInfo*> const&) + 18712
8  llc               0x000000010031ae39 llvm::DenseMap<llvm::VNInfo*,
llvm::VNInfo*, llvm::DenseMapInfo<llvm::VNInfo*>,
llvm::DenseMapInfo<llvm::VNInfo*> >::insert(std::pair<llvm::VNInfo*,
llvm::VNInfo*> const&) + 40633
9  llc               0x00000001003c49b4 llvm::SlotIndexes::SlotIndexes() +
11252
10 llc               0x000000010037e1bb llvm::BitVector::grow(unsigned int) +
23931
11 llc               0x0000000100379b4e llvm::BitVector::grow(unsigned int) +
5902
12 llc               0x0000000100337105 llvm::PassInfo::~PassInfo() + 117
13 llc               0x00000001005b60fb llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 21947
14 llc               0x00000001005b63cb llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 22667
15 llc               0x00000001005b6536 llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 23030
16 llc               0x00000001005b6c5c llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 24860
17 llc               0x00000001005b74cd llvm::cl::parser<llvm::PassInfo
const*>::~parser() + 27021
18 llc               0x0000000100016f70 std::_Rb_tree<long, std::pair<long
const, llvm::PassInfo const*>, std::_Select1st<std::pair<long const,
llvm::PassInfo const*> >, std::less<long>, std::allocator<std::pair<long const,
llvm::PassInfo const*> > >::_M_insert_unique(std::pair<long const,
llvm::PassInfo const*> const&) + 10192
19 llc               0x0000000100015aa4 std::_Rb_tree<long, std::pair<long
const, llvm::PassInfo const*>, std::_Select1st<std::pair<long const,
llvm::PassInfo const*> >, std::less<long>, std::allocator<std::pair<long const,
llvm::PassInfo const*> > >::_M_insert_unique(std::pair<long const,
llvm::PassInfo const*> const&) + 4868
20 llc               0x0000000000000002 std::_Rb_tree<long, std::pair<long
const, llvm::PassInfo const*>, std::_Select1st<std::pair<long const,
llvm::PassInfo const*> >, std::less<long>, std::allocator<std::pair<long const,
llvm::PassInfo const*> > >::_M_insert_unique(std::pair<long const,
llvm::PassInfo const*> const&) + 4294883426
Stack dump:
0.    Program arguments: llc bugpoint-reduced-simplified.bc 
1.    Running pass 'Function Pass Manager' on module
'bugpoint-reduced-simplified.bc'.
2.    Running pass 'Linear Scan Register Allocator' on function
'@TrLinkPeerNodes'
Abort

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 09:01:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 09:01:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7736] New: Assertion "Call must have function
 pointer type" on valid code
Message-ID: <bug-7736-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7736

           Summary: Assertion "Call must have function pointer type" on
                    valid code
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: chris at bubblescope.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Created an attachment (id=5284)
 --> (http://llvm.org/bugs/attachment.cgi?id=5284)
Asserting testcase

The attached code, reduced from a test case provided by "cizra" on IRC, causes
an assertion on clang++. The code is, I believe, valid.

Backtrace:

clang: CGExpr.cpp:2065: clang::CodeGen::RValue
clang::CodeGen::CodeGenFunction::EmitCall(clang::QualType, llvm::Value*,
clang::CodeGen::ReturnValueSlot, clang::ConstExprIterator,
clang::ConstExprIterator, const clang::Decl*): Assertion
`CalleeType->isFunctionPointerType() && "Call must have function pointer
type!"' failed.
0  clang           0x0000000001572e1f
1  clang           0x0000000001574eb2
2  libpthread.so.0 0x00007f2203cbc8f0
3  libc.so.6       0x00007f2202faba75 gsignal + 53
4  libc.so.6       0x00007f2202faf5c0 abort + 384
5  libc.so.6       0x00007f2202fa4941 __assert_fail + 241
6  clang           0x000000000078b5ef
clang::CodeGen::CodeGenFunction::EmitCall(clang::QualType, llvm::Value*,
clang::CodeGen::ReturnValueSlot, clang::ConstExprIterator,
clang::ConstExprIterator, clang::Decl const*) + 127
7  clang           0x000000000078d518
clang::CodeGen::CodeGenFunction::EmitCallExpr(clang::CallExpr const*,
clang::CodeGen::ReturnValueSlot) + 296
8  clang           0x00000000007b1d32
9  clang           0x00000000007b8f20
10 clang           0x00000000007bd7f7
clang::CodeGen::CodeGenFunction::EmitScalarExpr(clang::Expr const*, bool) + 87
11 clang           0x000000000078add7
clang::CodeGen::CodeGenFunction::EmitAnyExpr(clang::Expr const*, llvm::Value*,
bool, bool, bool) + 247
12 clang           0x0000000000807464
clang::CodeGen::CodeGenFunction::EmitStmt(clang::Stmt const*) + 148
13 clang           0x000000000080a21f
clang::CodeGen::CodeGenFunction::EmitCompoundStmt(clang::CompoundStmt const&,
bool, llvm::Value*, bool) + 239
14 clang           0x000000000080a4dc
clang::CodeGen::CodeGenFunction::EmitSimpleStmt(clang::Stmt const*) + 236
15 clang           0x00000000008073ec
clang::CodeGen::CodeGenFunction::EmitStmt(clang::Stmt const*) + 28
16 clang           0x0000000000826a3b
clang::CodeGen::CodeGenFunction::GenerateCode(clang::CodeGen::GlobalDecl,
llvm::Function*) + 875
17 clang           0x000000000073ff23
clang::CodeGen::CodeGenModule::EmitGlobalFunctionDefinition(clang::CodeGen::GlobalDecl)
+ 531
18 clang           0x000000000074164b
clang::CodeGen::CodeGenModule::EmitGlobalDefinition(clang::CodeGen::GlobalDecl)
+ 267
19 clang           0x0000000000742427
clang::CodeGen::CodeGenModule::EmitDeferred() + 151
20 clang           0x00000000007424a9 clang::CodeGen::CodeGenModule::Release()
+ 9
21 clang           0x0000000000733858
22 clang           0x000000000084a82f clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 351
23 clang           0x0000000000734744 clang::CodeGenAction::ExecuteAction() +
68
24 clang           0x000000000064f285
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 357
25 clang           0x00000000006303cc cc1_main(char const**, char const**, char
const*, void*) + 2748
26 clang           0x0000000000637247 main + 3527
27 libc.so.6       0x00007f2202f96c4d __libc_start_main + 253
28 clang           0x000000000062e3c9

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 09:49:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 09:49:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7725] [c++] clang++ throws compile error on
 possible C++ template 2 level namespace lookup
In-Reply-To: <bug-7725-206@http.llvm.org/bugs/>
References: <bug-7725-206@http.llvm.org/bugs/>
Message-ID: <20100728144920.D4F4B2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7725

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Douglas Gregor <dgregor at apple.com> 2010-07-28 09:49:19 CDT ---
Fixed in Clang r109582.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 10:54:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 10:54:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7737] New: [c++] clang++ throws compile error on
 never used template function that calls a protected function
Message-ID: <bug-7737-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7737

           Summary: [c++] clang++ throws compile error on never used
                    template function that calls a protected function
           Product: clang
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: release blocker
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: mike.jackson at bluequartz.net
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


// BEGIN CODE
class Foo
{
public:
  int pub_function() {return 1;}
  Foo() {}
  virtual ~Foo() {};
protected:
  int protected_function() { return 2;}
};

template<typename T>
void outside_function(Foo* self)
{
  self->protected_function();
}

int main (int argc, char const *argv[])
{
  Foo* f = new Foo;
  delete f;
  return 0;
}
// END CODE

GCC compiles this code just fine where as clang throws a compile error. Note
that the function "outside_function" is never actually used. This type of code
is currently present in the vtkTIFFReader.cxx of the VTK.org code base.

The clang compile error is:
528:[mjackson at ferb:Build]$ clang++ ~/Desktop/VTKBug.cpp -o
~/Desktop/VTKBug-clang
/Users/mjackson/Desktop/VTKBug.cpp:14:9: error: 'protected_function' is a
protected member of 'Foo'
  self->protected_function();
        ^
/Users/mjackson/Desktop/VTKBug.cpp:8:7: note: declared protected here
  int protected_function() { return 2;}
      ^
1 error generated.

530:[mjackson at ferb:Build]$ clang --version
clang version 2.8 (trunk 109582)
Target: x86_64-apple-darwin9
Thread model: posix

Not really sure if this is a clang bug or really a bug in the vtkTIFFReader
code. I am unfamiliar with the spec on this.

Thanks
Mike Jackson
Principal Software Engineer
BlueQuartz Software
www.bluequartz.net

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 11:20:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 11:20:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7737] [c++] clang++ throws compile error on never
 used template function that calls a protected function
In-Reply-To: <bug-7737-206@http.llvm.org/bugs/>
References: <bug-7737-206@http.llvm.org/bugs/>
Message-ID: <20100728162050.EFC672A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7737

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #3 from Douglas Gregor <dgregor at apple.com> 2010-07-28 11:20:50 CDT ---
Right. Closing as invalid, since Clang is doing the right thing here.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 12:59:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 12:59:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7738] New: "overloadable" causes assertion failure
Message-ID: <bug-7738-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7738

           Summary: "overloadable" causes assertion failure
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sliao at google.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Clang gave me an assertion failure. My input program "test.c" below is very
simple. If I remove the overloadable attribute, then the error goes away.
What's wrong? Thanks.

$ ./clang m2/work/test.c
clang:
/usr/local/google/slang/llvmhost/llvm-102410/include/llvm/Support/Casting.h:202:
typename llvm::cast_retty<To, From>::ret_type llvm::cast(const Y&) [with X =
clang::FunctionProtoType, Y = const clang::FunctionType*]: Assertion
`isa<X>(Val) && "cast<Ty>() argument of incompatible type!"' failed.
0  clang 0x09218958
Stack dump:
0.    Program arguments: /usr/local/google/slang0/llvmhost/llvm/bin/clang -cc1
-triple i386-unknown-linux-gnu -S -disable-free -main-file-name test.c
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases -target-cpu
pentium4 -resource-dir /usr/local/google/slang0/llvmhost/llvm/lib/clang/1.5
-ferror-limit 19 -ftemplate-backtrace-limit 10 -fmessage-length 237
-fgnu-runtime -fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-zY7tjd.s
-x c m2/work/test.c 
1.    <eof> parser at end of file
2.    m2/work/test.c:3:5: LLVM IR generation of declaration 'root'
3.    m2/work/test.c:3:5: Generating code for declaration 'root'
4.    m2/work/test.c:4:1: LLVM IR generation of compound statement ('{}')
5.    m2/work/test.c:1:42: Mangling declaration 'rsUpdateMills'
clang: error: compiler command failed due to signal 6 (use -v to see
invocation)


$ cat test.c
extern int __attribute__((overloadable)) rsUpdateMills();

int root()
{
  int i = rsUpdateMills();
  return i;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 13:24:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 13:24:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7728] LLVM IR generation of compound statement
 ('{}') in Markzware
In-Reply-To: <bug-7728-206@http.llvm.org/bugs/>
References: <bug-7728-206@http.llvm.org/bugs/>
Message-ID: <20100728182412.9F0412A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7728

James Gregurich <james at markzware.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #13 from James Gregurich <james at markzware.com> 2010-07-28 13:24:11 CDT ---
Hashtable.ii still crashes clang. All the others have been fixed. I will attach
an updated crash log and attachment for your convenience.



$ /opt/bin/clang++ --version
clang version 2.8 (trunk 109600)
Target: x86_64-apple-darwin10
Thread model: posix


$ /opt/bin/clang++ Hashtable.ii
Assertion failed: ((Record->hasTrivialCopyConstructor() ||
Record->hasTrivialCopyAssignment()) && "Trying to aggregate-copy a type without
a trivial copy " "constructor or assignment operator"), function
EmitAggregateCopy, file CGExprAgg.cpp, line 796.
0  clang             0x00000001012f0ac2 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012f1813 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x00007fff869f30aa tiny_malloc_from_free_list + 1196
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x0000000100199aef llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 3711
6  clang             0x0000000100199c80 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 4112
7  clang             0x000000010019cc09 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 16281
8  clang             0x000000010019e5c6 llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 22870
9  clang             0x000000010019c53e llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 14542
10 clang             0x000000010019d71b llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateAnd(llvm::Value*, llvm::APInt const&, llvm::Twine const&) + 19115
11 clang             0x0000000100190340
clang::CanQual<clang::Type>::getTypePtr() const + 36864
12 clang             0x000000010019359c
clang::CanQual<clang::Type>::getTypePtr() const + 49756
13 clang             0x0000000100192892
clang::CanQual<clang::Type>::getTypePtr() const + 46418
14 clang             0x0000000100196fe8
clang::CanQual<clang::Type>::getTypePtr() const + 64680
15 clang             0x00000001001978da
clang::CanQual<clang::Type>::getTypePtr() const + 66970
16 clang             0x00000001001796ca llvm::DenseMap<clang::Decl const*,
llvm::WeakVH, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::WeakVH> >::FindAndConstruct(clang::Decl const* const&)
+ 22426
17 clang             0x000000010017a6cf llvm::DenseMap<clang::Decl const*,
llvm::WeakVH, llvm::DenseMapInfo<clang::Decl const*>,
llvm::DenseMapInfo<llvm::WeakVH> >::FindAndConstruct(clang::Decl const* const&)
+ 26527
18 clang             0x00000001002201fb llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 20955
19 clang             0x0000000100223957 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35127
20 clang             0x0000000100220830 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
21 clang             0x00000001002236eb llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 34507
22 clang             0x0000000100223a55 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 35381
23 clang             0x0000000100220830 llvm::DenseMap<clang::FieldDecl const*,
unsigned int, llvm::DenseMapInfo<clang::FieldDecl const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 22544
24 clang             0x000000010024cecb
clang::SourceManager::getLocForStartOfFile(clang::FileID) const + 32699
25 clang             0x00000001002588b7 llvm::DenseMap<clang::Expr const*,
llvm::Value*, llvm::DenseMapInfo<clang::Expr const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 40791
26 clang             0x0000000100258e0a llvm::DenseMap<clang::Expr const*,
llvm::Value*, llvm::DenseMapInfo<clang::Expr const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42154
27 clang             0x0000000100258fa3 llvm::DenseMap<clang::Expr const*,
llvm::Value*, llvm::DenseMapInfo<clang::Expr const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42563
28 clang             0x0000000100258ff3 llvm::DenseMap<clang::Expr const*,
llvm::Value*, llvm::DenseMapInfo<clang::Expr const*>,
llvm::DenseMapInfo<llvm::Value*> >::grow(unsigned int) + 42643
29 clang             0x0000000100242f9a llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 1450
30 clang             0x0000000100286bd0 llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 1264
31 clang             0x0000000100243bfc llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
32 clang             0x000000010002e7f9 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6521
33 clang             0x00000001000029e5
34 clang             0x00000001000095a4 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
35 clang             0x0000000100000b34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name Hashtable.ii -pic-level 1
-mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2 -resource-dir
/opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 272 -stack-protector 1
-fblocks -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-Sx3aWF.o -x c++-cpp-output
Hashtable.ii 
1.    <eof> parser at end of file
2.    Per-file LLVM IR generation
3.    ./java/util/HashMap.h:397:14: Generating code for declaration
'java::util::HashMap<java::lang::String, java::lang::Object,
java::util::private_detail::cDefaultHashMapEntry>::contains'
4.    ./java/util/HashMap.h:398:9: LLVM IR generation of compound statement
('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 13:31:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 13:31:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7728] LLVM IR generation of compound statement
 ('{}') in Markzware
In-Reply-To: <bug-7728-206@http.llvm.org/bugs/>
References: <bug-7728-206@http.llvm.org/bugs/>
Message-ID: <20100728183155.89B542A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7728

James Gregurich <james at markzware.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #14 from James Gregurich <james at markzware.com> 2010-07-28 13:31:54 CDT ---
Sorry. I hadn't realized you didn't close 7727 along with the rest. What I
added to this case is likely redundant. I'll resolve it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 15:27:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 15:27:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6711] atomic builtin test uses C compiler instead
 of C++ compiler
In-Reply-To: <bug-6711-206@http.llvm.org/bugs/>
References: <bug-6711-206@http.llvm.org/bugs/>
Message-ID: <20100728202703.ABF3A2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6711

Eric Christopher <echristo at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #13 from Eric Christopher <echristo at gmail.com> 2010-07-28 15:27:03 CDT ---
Committed thusly:


[yendi:Data/sources/llvm] echristo% svn ci autoconf/configure.ac 
Sending        autoconf/configure.ac
Transmitting file data .
Committed revision 109646.
[yendi:Data/sources/llvm] echristo% svn ci configure
Sending        configure
Transmitting file data .
Committed revision 109647.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 15:56:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 15:56:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7735] Assertion failed:
 (MF.getFrameInfo()->getObjectSize(FrameIdx) >= RC->getSize() && "Stack slot
 too small for load")
In-Reply-To: <bug-7735-206@http.llvm.org/bugs/>
References: <bug-7735-206@http.llvm.org/bugs/>
Message-ID: <20100728205614.01EE42A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7735

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-28 15:56:13 CDT ---
Fixed in r109652

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 17:15:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 17:15:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7714] probable clang wrong code bug on x64
In-Reply-To: <bug-7714-206@http.llvm.org/bugs/>
References: <bug-7714-206@http.llvm.org/bugs/>
Message-ID: <20100728221520.3E75D2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7714

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Chris Lattner <clattner at apple.com> 2010-07-28 17:15:19 CDT ---
Fixed in r109681, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 17:43:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 17:43:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7739] New: Implement support for 3DNow! <mm3dnow.h>
Message-ID: <bug-7739-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7739

           Summary: Implement support for 3DNow! <mm3dnow.h>
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Headers
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dgregor at apple.com
                CC: llvmbugs at cs.uiuc.edu


GCC implements this header for 3DNow! support, Qt uses it, and other projects
likely do, too. At some point, we should support it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 18:26:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 18:26:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7736] Assertion "Call must have function pointer
 type" on valid code
In-Reply-To: <bug-7736-206@http.llvm.org/bugs/>
References: <bug-7736-206@http.llvm.org/bugs/>
Message-ID: <20100728232647.89A602A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7736

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-28 18:26:47 CDT ---
Fixed in r109697.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 19:27:20 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 19:27:20 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7740] New: passing address of reference failing in
	ObjC++
Message-ID: <bug-7740-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7740

           Summary: passing address of reference failing in ObjC++
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


The code below compiles and works as expected on gcc 4.2. However, it fails on
clang with the following error:

/HOME/Documents/work_projects/devel/miscTestBed/main.mm:21:26
/HOME/Documents/work_projects/devel/miscTestBed/main.mm:21:26:{21:26-21:33}:
error: cannot initialize a parameter of type 'id' with an rvalue of type
'NSString *'


I should work as "NSString *" should be accepted by an 'id' argument.






#include <Cocoa/Cocoa.h>





int main (int argc, char * const argv[]) 
{


    NSMutableArray * tmpArrPtr = [[NSMutableArray alloc] init];
    NSString* tmpStrPtr = @"test";
    NSString& tmpStr = *tmpStrPtr;


    [tmpArrPtr addObject:&tmpStr];


    return 0;

}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 19:40:07 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 19:40:07 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7741] New: assertion failed in cDocStore.mm
Message-ID: <bug-7741-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7741

           Summary: assertion failed in cDocStore.mm
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: james at markzware.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


the following error log is produced when compiling cDocStore.mm.  preprocessed
file is attached.





$ /opt/bin/clang++ cDocStore-1F0AFC4B.mii
In file included from
/HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:1:
In file included from
/HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:17:
In file included from
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cDocStore.h:21:
In file included from
/HOME/Documents/work_projects/devel/PROJECT/readerLib/iFetchResultHandler.h:15:
In file included from
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cItr.h:17:
In file included from
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cProperty.h:19:
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cSchemaKey.h:210:13:
warning: expression result unused [-Wunused-value]
            static_cast<cPropertyKey&>(tmpTest);
            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cSchemaKey.h:273:13:
warning: expression result unused [-Wunused-value]
            static_cast<cRelationshipKey&>(tmpTest);
            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from
/HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:1:
In file included from
/HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:17:
In file included from
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cDocStore.h:26:
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:557:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(LIKE)
        ^      ~~~~
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:559:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(BEGINSWITH)
        ^      ~~~~~~~~~~
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:561:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(CONTAINS)
        ^      ~~~~~~~~
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:563:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(ENDSWITH)
        ^      ~~~~~~~~
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:565:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(MATCHES)
        ^      ~~~~~~~
/Users/jamesgregurich/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:567:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(c)
        ^      ~
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:569:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(d)
        ^      ~
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:571:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(cd)
        ^      ~~
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:573:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(SELF)
        ^      ~~~~
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:575:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(BETWEEN)
        ^      ~~~~~~~
/HOME/Documents/work_projects/devel/PROJECT/readerLib/cPredicate.h:577:9:
warning: only local variables can be arguments to '#pragma unused'
#pragma unused(IN)
        ^      ~~
Assertion failed: (isa<X>(Val) && "cast<Ty>() argument of incompatible type!"),
function cast, file /scratch/llvm_build/llvm/include/llvm/Support/Casting.h,
line 202.
0  clang             0x00000001012f0ac2 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 11634
1  clang             0x00000001012f1813 std::vector<llvm::sys::Path,
std::allocator<llvm::sys::Path>
>::_M_insert_aux(__gnu_cxx::__normal_iterator<llvm::sys::Path*,
std::vector<llvm::sys::Path, std::allocator<llvm::sys::Path> > >,
llvm::sys::Path const&) + 15043
2  libSystem.B.dylib 0x00007fff86a4e35a _sigtramp + 26
3  libSystem.B.dylib 0x0000000105e37cc0 _sigtramp + 2134808960
4  libSystem.B.dylib 0x00007fff86ac99b4 __pthread_markcancel + 0
5  clang             0x00000001006f2a3b clang::TemplateDecl*
llvm::PointerUnion4<clang::TemplateDecl*, clang::OverloadedTemplateStorage*,
clang::QualifiedTemplateName*,
clang::DependentTemplateName*>::get<clang::TemplateDecl*>() const + 18139
6  clang             0x000000010049dbec
clang::DeclaratorDecl::getTypeSourceInfo() const + 6908
7  clang             0x000000010049e479
clang::DeclaratorDecl::getTypeSourceInfo() const + 9097
8  clang             0x00000001003effee
llvm::DenseMap<clang::ObjCPropertyDecl*, char,
llvm::DenseMapInfo<clang::ObjCPropertyDecl*>, llvm::DenseMapInfo<char>
>::grow(unsigned int) + 58846
9  clang             0x00000001003b85f2
clang::SelectorTable::constructSetterName(clang::IdentifierTable&,
clang::SelectorTable&, clang::IdentifierInfo const*) + 9778
10 clang             0x00000001003ba7c4
clang::SelectorTable::constructSetterName(clang::IdentifierTable&,
clang::SelectorTable&, clang::IdentifierInfo const*) + 18436
11 clang             0x00000001003c7b31
clang::SelectorTable::constructSetterName(clang::IdentifierTable&,
clang::SelectorTable&, clang::IdentifierInfo const*) + 72561
12 clang             0x000000010037f5d3
llvm::cast_retty<clang::FunctionTypeLoc, clang::TypeLoc*>::ret_type
llvm::cast<clang::FunctionTypeLoc, clang::TypeLoc*>(clang::TypeLoc* const&) +
202627
13 clang             0x0000000100380066
llvm::cast_retty<clang::FunctionTypeLoc, clang::TypeLoc*>::ret_type
llvm::cast<clang::FunctionTypeLoc, clang::TypeLoc*>(clang::TypeLoc* const&) +
205334
14 clang             0x00000001003807cc
llvm::cast_retty<clang::FunctionTypeLoc, clang::TypeLoc*>::ret_type
llvm::cast<clang::FunctionTypeLoc, clang::TypeLoc*>(clang::TypeLoc* const&) +
207228
15 clang             0x0000000100381697
llvm::cast_retty<clang::FunctionTypeLoc, clang::TypeLoc*>::ret_type
llvm::cast<clang::FunctionTypeLoc, clang::TypeLoc*>(clang::TypeLoc* const&) +
211015
16 clang             0x0000000100723fd5 clang::Declarator::clear() + 6357
17 clang             0x0000000100724c47 clang::Declarator::clear() + 9543
18 clang             0x00000001007259fa clang::Declarator::clear() + 13050
19 clang             0x0000000100726ce6 clang::Declarator::clear() + 17894
20 clang             0x000000010070c77c clang::Parser::ConsumeAnyToken() +
18252
21 clang             0x00000001007170c3 clang::Parser::ConsumeAnyToken() +
61587
22 clang             0x0000000100717cb6 clang::Parser::ConsumeAnyToken() +
64646
23 clang             0x0000000100717db9 clang::Parser::ConsumeAnyToken() +
64905
24 clang             0x0000000100743f3a
clang::PragmaWeakHandler::~PragmaWeakHandler() + 10442
25 clang             0x00000001007441e3
clang::PragmaWeakHandler::~PragmaWeakHandler() + 11123
26 clang             0x0000000100744773
clang::PragmaWeakHandler::~PragmaWeakHandler() + 12547
27 clang             0x0000000100743a57
clang::PragmaWeakHandler::~PragmaWeakHandler() + 9191
28 clang             0x00000001007462a8
clang::PragmaWeakHandler::~PragmaWeakHandler() + 19512
29 clang             0x0000000100743cf5
clang::PragmaWeakHandler::~PragmaWeakHandler() + 9861
30 clang             0x00000001007441e3
clang::PragmaWeakHandler::~PragmaWeakHandler() + 11123
31 clang             0x000000010074466d
clang::PragmaWeakHandler::~PragmaWeakHandler() + 12285
32 clang             0x000000010075633c
llvm::SmallVectorTemplateBase<clang::ParsedTemplateArgument,
false>::grow(unsigned long) + 42588
33 clang             0x00000001007174a2 clang::Parser::ConsumeAnyToken() +
62578
34 clang             0x0000000100753781
llvm::SmallVectorTemplateBase<clang::ParsedTemplateArgument,
false>::grow(unsigned long) + 31393
35 clang             0x0000000100753bbe
llvm::SmallVectorTemplateBase<clang::ParsedTemplateArgument,
false>::grow(unsigned long) + 32478
36 clang             0x00000001007546b5
llvm::SmallVectorTemplateBase<clang::ParsedTemplateArgument,
false>::grow(unsigned long) + 35285
37 clang             0x0000000100755188
llvm::SmallVectorTemplateBase<clang::ParsedTemplateArgument,
false>::grow(unsigned long) + 38056
38 clang             0x000000010028697b llvm::DenseMap<clang::Stmt*, unsigned
int, llvm::DenseMapInfo<clang::Stmt*>, llvm::DenseMapInfo<unsigned int>
>::grow(unsigned int) + 667
39 clang             0x0000000100243bfc llvm::IRBuilder<true,
llvm::ConstantFolder, llvm::IRBuilderDefaultInserter<true>
>::CreateConstInBoundsGEP1_64(llvm::Value*, unsigned long long, llvm::Twine
const&) + 4620
40 clang             0x000000010002e7f9 llvm::DenseMap<clang::IdentifierInfo
const*, unsigned int, llvm::DenseMapInfo<clang::IdentifierInfo const*>,
llvm::DenseMapInfo<unsigned int> >::grow(unsigned int) + 6521
41 clang             0x00000001000029e5
42 clang             0x00000001000095a4 std::vector<std::string,
std::allocator<std::string> >::operator=(std::vector<std::string,
std::allocator<std::string> > const&) + 10820
43 clang             0x0000000100000b34
Stack dump:
0.    Program arguments: /opt/bin/clang -cc1 -triple x86_64-apple-darwin10.0.0
-emit-obj -mrelax-all -disable-free -main-file-name cDocStore-1F0AFC4B.mii
-pic-level 1 -mdisable-fp-elim -masm-verbose -munwind-tables -target-cpu core2
-resource-dir /opt/lib/clang/2.8 -ferror-limit 19 -fmessage-length 173
-stack-protector 1 -fblocks -fexceptions -fobjc-nonfragile-abi
-fobjc-dispatch-method=mixed -fdiagnostics-show-option -fcolor-diagnostics -o
/var/folders/cX/cXKleH18HyiK1CWAuQ836k++-+Y/-Tmp-/cc-SRd1GK.o -x
objective-c++-cpp-output cDocStore-1F0AFC4B.mii 
1.    /HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:865:92:
current parser token ')'
2.    /HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:749:1:
parsing function body
'nmsComMarkzware::nmsReaderLib::cDocStore_Private::resolveRelationshipForAllObjects'
3.    /HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:749:1: in
compound statement ('{}')
4.    /HOME/Documents/work_projects/devel/PROJECT/mac/cDocStore.mm:830:5: in
compound statement ('{}')
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)
bubba:x86_64 jamesgregurich$

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Wed Jul 28 20:32:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Wed, 28 Jul 2010 20:32:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7675] [C++ Analyzer] Assertion
 `(isa<RecordType>(Ex->getType().getDesugaredType()) ||
 isa<ArrayType>(Ex->getType().getDesugaredType())) && "Other kinds of
 expressions with non-aggregate/union/class types" " do not have lvalues."'
 failed
In-Reply-To: <bug-7675-206@http.llvm.org/bugs/>
References: <bug-7675-206@http.llvm.org/bugs/>
Message-ID: <20100729013235.B5E322A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7675

Ted Kremenek <kremenek at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Ted Kremenek <kremenek at apple.com> 2010-07-28 20:32:34 CDT ---
(In reply to comment #3)
> (In reply to comment #2)
> > I can only guess, but it seems that in the switch in function
> > GRExprEngine::VisitLValue Stmt::FloatingLiteralClass in unhandled.
> > 
> > Index: lib/Checker/GRExprEngine.cpp
> > ===================================================================
> > --- lib/Checker/GRExprEngine.cpp        (revision 109275)
> > +++ lib/Checker/GRExprEngine.cpp        (working copy)
> > @@ -1058,6 +1058,7 @@
> >      // In C++, binding an rvalue to a reference requires to create an object.
> >      case Stmt::CXXBoolLiteralExprClass:
> >      case Stmt::IntegerLiteralClass:
> > +    case Stmt::FloatingLiteralClass:
> >        CreateCXXTemporaryObject(Ex, Pred, Dst);
> >        return;
> > 
> > 
> > I am unable to compile clang now, so I can't test it. But changing floats to
> > ints in reduced test case causes clang to not crash.
> 
> Sorry for the spamming, but of course there can be other *LiteralClass that
> need to be handled in the same way, I am not sure if I found all of them:
> FloatingLiteralClass
> CharacterLiteralClass
> ImaginaryLiteral (?)

Yes, all of these cases need to be handled.

Fixed here:

http://llvm.org/viewvc/llvm-project?view=rev&revision=109719

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 00:48:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 00:48:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7699] Assertion `!EHBranchThroughDest' failed
In-Reply-To: <bug-7699-206@http.llvm.org/bugs/>
References: <bug-7699-206@http.llvm.org/bugs/>
Message-ID: <20100729054847.E15D02A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7699

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #8 from Chris Lattner <clattner at apple.com> 2010-07-29 00:48:47 CDT ---
Ah yes, confirmed fixed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 03:56:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 03:56:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7742] New: Aggregate passed in wrong register on
 x86-64 Darwin/Linux.
Message-ID: <bug-7742-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7742

           Summary: Aggregate passed in wrong register on x86-64
                    Darwin/Linux.
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tilmann.scheller at googlemail.com
                CC: llvmbugs at cs.uiuc.edu


Clang generates wrong LLVM IR for the following C++ code which leads to wrong
machine code being generated:

struct s2 {
  float a[2];
};

class c2 : public s2 {
public:
c2() { a[0] = 1.5; a[1] = 3.0; }
};

c2 foo() { return c2(); }

In foo() the c2 object is passed in %rax instead of %xmm0. This bug only
affects clang, proper IR is being generated by llvm-gcc.

The bug affects both x86-64 Darwin and Linux.

clang:
_Z3foov:
.Leh_func_begin1:
    pushq    %rbp
.Llabel1:
    movq    %rsp, %rbp
.Llabel2:
    movabsq    $4629700418006417408, %rax
    popq    %rbp
    ret
    .size    _Z3foov, .-_Z3foov
.Leh_func_end1:

llvm-gcc:
.LCPI1_0:
    .quad    4629700418006417408
_Z3foov:
.Leh_func_begin1:
    movsd    .LCPI1_0(%rip), %xmm0
    ret
    .size    _Z3foov, .-_Z3foov

gcc 4.5:
_Z3foov:
.LFB6:
    movq    .LC2(%rip), %xmm0
    ret
.LC2:
    .quad    4629700418006417408


clang IR:
; ModuleID = 'float2.cpp'
target datalayout =
"e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
target triple = "x86_64-unknown-linux-gnu"

define i64 @_Z3foov() nounwind readnone {
entry:
  ret i64 4629700418006417408
}

llvm-gcc IR:
; ModuleID = 'float2.cpp'
target datalayout =
"e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
target triple = "x86_64-unknown-linux-gnu"

%0 = type { double }

define %0 @_Z3foov() nounwind readnone {
entry:
  %mrv2 = insertvalue %0 undef, double 0x404000003FC00000, 0 ; <%0> [#uses=1]
  ret %0 %mrv2
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 04:47:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 04:47:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7743] New: Missing or invalid line-related debug
 infos on Linux x86
Message-ID: <bug-7743-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7743

           Summary: Missing or invalid line-related debug infos on Linux
                    x86
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: qmathe at gmail.com
                CC: llvmbugs at cs.uiuc.edu


With GDB on Ubuntu 9.04 x86, it is not possible to set a breakpoint on a file
line or step through a function/method for a C or ObjC program compiled with
Clang (trunk r109551).

With a test source file named testffi.c:
#include <ffi.h>

int main(int argc, char **argv)
{
        return 1;
}

$ clang -g -o testffi testffi.c 

$ gdb testffi
GNU gdb 6.8-debian
This GDB was configured as "i486-linux-gnu"...
(gdb) b testffi.c:5
No line 5 in file "testffi.c".

Here is a more complex example with an ObjC program debugged with GDB 6.8:

Breakpoint 3, 0x08053426 in -[ETFCGIWebApp handleServerData:] (self=0xa25dd98, 
    _cmd=0xb6fede88, notification=0xa04ea48)
(gdb) n
Single stepping until exit from function _i_ETFCGIWebApp__handleServerData_, 
which has no line number information.
-[NSNotificationCenter _postAndRelease:] (self=0xa04dd88, _cmd=0xb6feded0, 
    notification=0xa25dd98) at NSNotificationCenter.m:1163
1163              NS_HANDLER
(gdb) bt
#0  -[NSNotificationCenter _postAndRelease:] (self=0xa04dd88, _cmd=0xb6feded0, 
    notification=0xa25dd98) at NSNotificationCenter.m:1163
#1  0xb6e7253d in -[NSNotificationCenter postNotification:] (self=0xa04dd88, 
    _cmd=0xb6fee708, notification=0xa25dd98) at NSNotificationCenter.m:1190
#2  0xb6e73934 in notify (center=0xa04dd88, list=0xa2de890, mode=0xb7001498, 
    zone=0xb7022dc0) at NSNotificationQueue.m:625
#3  0xb6e739dd in GSPrivateNotifyASAP (mode=0xb7001498)
    at NSNotificationQueue.m:649
#4  0xb6f82b20 in -[GSRunLoopCtxt pollUntil:within:] (self=0xa298be8, 
    _cmd=0xb7001d50, milliseconds=28958, contexts=0xa2dbdd0)
    at GSRunLoopCtxt.m:641
#5  0xb6ebd91f in -[NSRunLoop acceptInputForMode:beforeDate:] (self=0xa327680, 
    _cmd=0xb7001d78, mode=0xb7001498, limit_date=0xa2de880) at NSRunLoop.m:1197
#6  0xb6ebdd40 in -[NSRunLoop runMode:beforeDate:] (self=0xa327680, 
    _cmd=0xb7001d88, mode=0xb7001498, date=0xa2da5d8) at NSRunLoop.m:1265
#7  0xb6ebdefb in -[NSRunLoop runUntilDate:] (self=0xa327680, _cmd=0xb7001d80, 
    date=0xa2da5d8) at NSRunLoop.m:1296
#8  0xb6ebdddf in -[NSRunLoop run] (self=0xa327680, _cmd=0x806f860)
    at NSRunLoop.m:1279
#9  0x0805ae24 in main ()
(gdb) b ETFCGIWebApp.m:76
No line 76 in file "ETFCGIWebApp.m".

Here is the verbose output for the compilation of the first program:

$ clang -v -g -o testffi testffi.c

clang version 2.8 (trunk 109551)
Target: i386-pc-linux-gnu
Thread model: posix
 "/home/qmathe/reps/llvm/Release+Debug+Asserts/bin/clang" -cc1 -triple
i386-pc-linux-gnu -S -disable-free -main-file-name testffi.c -mrelocation-model
static -mdisable-fp-elim -mconstructor-aliases -target-cpu pentium4 -v -g
-resource-dir /home/qmathe/reps/llvm/Release+Debug+Asserts/lib/clang/2.8
-ferror-limit 19 -fmessage-length 80 -fgnu-runtime -fdiagnostics-show-option
-fcolor-diagnostics -o /tmp/cc-zteC7E.s -x c testffi.c
clang -cc1 version 2.8 based upon llvm 2.8svn hosted on i386-pc-linux-gnu
#include "..." search starts here:
#include <...> search starts here:
 /usr/local/include
 /home/qmathe/reps/llvm/Release+Debug+Asserts/lib/clang/2.8/include
 /usr/lib/gcc/i486-linux-gnu/4.3.3/include
 /usr/lib/gcc/i486-linux-gnu/4.3.3/include-fixed
 /usr/include/i486-linux-gnu
 /usr/include
End of search list.
 "/usr/bin/gcc" -v -g -c -m32 -o /tmp/cc-EK7Gc8.o -x assembler /tmp/cc-zteC7E.s
Using built-in specs.
Target: i486-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.3.3-5ubuntu4'
--with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
--enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared
--with-system-zlib --libexecdir=/usr/lib --without-included-gettext
--enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3
--program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug
--enable-objc-gc --enable-mpfr --enable-targets=all --with-tune=generic
--enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu
--target=i486-linux-gnu
Thread model: posix
gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) 
COLLECT_GCC_OPTIONS='-v' '-g' '-c' '-m32' '-o' '/tmp/cc-EK7Gc8.o'
'-mtune=generic'
 as --gdwarf2 -V -Qy --32 -o /tmp/cc-EK7Gc8.o /tmp/cc-zteC7E.s
GNU assembler version 2.19.1 (i486-linux-gnu) using BFD version (GNU Binutils
for Ubuntu) 2.19.1
COMPILER_PATH=/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/
LIBRARY_PATH=/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../../lib/:/lib/../lib/:/usr/lib/../lib/:/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../:/lib/:/usr/lib/
COLLECT_GCC_OPTIONS='-v' '-g' '-c' '-m32' '-o' '/tmp/cc-EK7Gc8.o'
'-mtune=generic'
 "/usr/bin/gcc" -v -g -m32 -o testffi /tmp/cc-EK7Gc8.o
Using built-in specs.
Target: i486-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.3.3-5ubuntu4'
--with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
--enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared
--with-system-zlib --libexecdir=/usr/lib --without-included-gettext
--enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3
--program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug
--enable-objc-gc --enable-mpfr --enable-targets=all --with-tune=generic
--enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu
--target=i486-linux-gnu
Thread model: posix
gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) 
COMPILER_PATH=/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/
LIBRARY_PATH=/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/4.3.3/:/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../../lib/:/lib/../lib/:/usr/lib/../lib/:/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../:/lib/:/usr/lib/
COLLECT_GCC_OPTIONS='-v' '-g' '-m32' '-o' 'testffi' '-mtune=generic'
 /usr/lib/gcc/i486-linux-gnu/4.3.3/collect2 --eh-frame-hdr -m elf_i386
--hash-style=both -dynamic-linker /lib/ld-linux.so.2 -o testffi -z relro
/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../../lib/crt1.o
/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../../lib/crti.o
/usr/lib/gcc/i486-linux-gnu/4.3.3/crtbegin.o
-L/usr/lib/gcc/i486-linux-gnu/4.3.3 -L/usr/lib/gcc/i486-linux-gnu/4.3.3
-L/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../../lib -L/lib/../lib
-L/usr/lib/../lib -L/usr/lib/gcc/i486-linux-gnu/4.3.3/../../.. /tmp/cc-EK7Gc8.o
-lgcc --as-needed -lgcc_s --no-as-needed -lc -lgcc --as-needed -lgcc_s
--no-as-needed /usr/lib/gcc/i486-linux-gnu/4.3.3/crtend.o
/usr/lib/gcc/i486-linux-gnu/4.3.3/../../../../lib/crtn.o

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 07:19:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 07:19:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7745] New: llvm-mc: ARMDisassemblerCore.cpp:890:
 bool DisassembleBrMiscFrm(llvm::MCInst&, unsigned int, uint32_t,
 short unsigned int, unsigned int&,
 llvm::ARMBasicMCBuilder*): Assertion `0 && "Unexpected BrMiscFrm Opcode"'
 failed.
Message-ID: <bug-7745-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7745

           Summary: llvm-mc: ARMDisassemblerCore.cpp:890: bool
                    DisassembleBrMiscFrm(llvm::MCInst&, unsigned int,
                    uint32_t, short unsigned int, unsigned int&,
                    llvm::ARMBasicMCBuilder*): Assertion `0 && "Unexpected
                    BrMiscFrm Opcode"' failed.
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sliedes at cc.hut.fi
                CC: llvmbugs at cs.uiuc.edu


(this is trunk revision 109745)

Hi,

I tested the ARM disassembler in MC with all possible 4-byte instructions, and
found three different assertion failures. This is the first one of them (1/3).
I'm running this on amd64 Debian linux.

------------------------------------------------------------
$ echo '0x00 0xf0 0xa0 0xe1' |Debug+Asserts/bin/llvm-mc --disassemble
--triple=arm
llvm-mc: ARMDisassemblerCore.cpp:890: bool DisassembleBrMiscFrm(llvm::MCInst&,
unsigned int, uint32_t, short unsigned int, unsigned int&,
llvm::ARMBasicMCBuilder*): Assertion `0 && "Unexpected BrMiscFrm Opcode"'
failed.
0  llvm-mc         0x0000000000d75854
1  llvm-mc         0x0000000000d75718
2  libpthread.so.0 0x00007f8419950f60
3  libc.so.6       0x00007f8418c63175 gsignal + 53
4  libc.so.6       0x00007f8418c65f80 abort + 384
5  libc.so.6       0x00007f8418c5c2b1 __assert_fail + 241
6  llvm-mc         0x000000000055cdc8
7  llvm-mc         0x000000000056bc05
8  llvm-mc         0x000000000056c8dd
9  llvm-mc         0x0000000000559cca
10 llvm-mc         0x0000000000407287
11 llvm-mc         0x0000000000407b02
12 llvm-mc         0x000000000040bc11
13 llvm-mc         0x000000000040bce9 main + 193
14 libc.so.6       0x00007f8418c4fc4d __libc_start_main + 253
15 llvm-mc         0x0000000000406fe9
Stack dump:
0.      Program arguments: Debug+Asserts/bin/llvm-mc --disassemble --triple=arm 
Aborted
------------------------------------------------------------

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 07:21:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 07:21:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7746] New: llvm-mc: ARMInstPrinter.cpp:35: unsigned
 int NextReg(unsigned int): Assertion `0 && "Unexpected register enum"'
 failed.
Message-ID: <bug-7746-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7746

           Summary: llvm-mc: ARMInstPrinter.cpp:35: unsigned int
                    NextReg(unsigned int): Assertion `0 && "Unexpected
                    register enum"' failed.
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sliedes at cc.hut.fi
                CC: llvmbugs at cs.uiuc.edu


(this is trunk revision 109745)

Hi,

This is the second of three (2/3) assertion failures encountered while testing
llvm-mc on all 4-byte ARM asm sequences.

------------------------------------------------------------
$ echo "0x00 0xfa 0x40 0xf4" |Debug+Asserts/bin/llvm-mc --disassemble
--triple=arm
        vst1.8  llvm-mc: ARMInstPrinter.cpp:35: unsigned int NextReg(unsigned
int): Assertion `0 && "Unexpected register enum"' failed.
0  llvm-mc         0x0000000000d75854
1  llvm-mc         0x0000000000d75718
2  libpthread.so.0 0x00007f64f96fef60
3  libc.so.6       0x00007f64f8a11175 gsignal + 53
4  libc.so.6       0x00007f64f8a13f80 abort + 384
5  libc.so.6       0x00007f64f8a0a2b1 __assert_fail + 241
6  llvm-mc         0x0000000000618fb6
7  llvm-mc         0x0000000000619790
8  llvm-mc         0x0000000000613ba4
9  llvm-mc         0x00000000006196ff
10 llvm-mc         0x00000000004072ba
11 llvm-mc         0x0000000000407b02
12 llvm-mc         0x000000000040bc11
13 llvm-mc         0x000000000040bce9 main + 193
14 libc.so.6       0x00007f64f89fdc4d __libc_start_main + 253
15 llvm-mc         0x0000000000406fe9
Stack dump:
0.      Program arguments: Debug+Asserts/bin/llvm-mc --disassemble --triple=arm 
Aborted
------------------------------------------------------------

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 07:23:01 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 07:23:01 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7747] New: llvm-mc: ARMDisassemblerCore.cpp:2439:
 bool DisassembleN1RegModImmFrm(llvm::MCInst&, unsigned int, uint32_t,
 short unsigned int, unsigned int&,
 llvm::ARMBasicMCBuilder*): Assertion `0 && "Unreachable code!"' failed.
Message-ID: <bug-7747-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7747

           Summary: llvm-mc: ARMDisassemblerCore.cpp:2439: bool
                    DisassembleN1RegModImmFrm(llvm::MCInst&, unsigned int,
                    uint32_t, short unsigned int, unsigned int&,
                    llvm::ARMBasicMCBuilder*): Assertion `0 &&
                    "Unreachable code!"' failed.
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sliedes at cc.hut.fi
                CC: llvmbugs at cs.uiuc.edu


(this is trunk revision 109745)

Hi,

This is the final one of three (3/3) assertion failures encountered while
testing llvm-mc on all 4-byte ARM asm sequences.

------------------------------------------------------------
$ echo '0x30 0x00 0x80 0xf2' |Debug+Asserts/bin/llvm-mc --disassemble
--triple=arm
llvm-mc: ARMDisassemblerCore.cpp:2439: bool
DisassembleN1RegModImmFrm(llvm::MCInst&, unsigned int, uint32_t, short unsigned
int, unsigned int&, llvm::ARMBasicMCBuilder*): Assertion `0 && "Unreachable
code!"' failed.
0  llvm-mc         0x0000000000d75854
1  llvm-mc         0x0000000000d75718
2  libpthread.so.0 0x00007ffcb3c7df60
3  libc.so.6       0x00007ffcb2f90175 gsignal + 53
4  libc.so.6       0x00007ffcb2f92f80 abort + 384
5  libc.so.6       0x00007ffcb2f892b1 __assert_fail + 241
6  llvm-mc         0x0000000000569c03
7  llvm-mc         0x000000000056bc05
8  llvm-mc         0x000000000056c8dd
9  llvm-mc         0x0000000000559cca
10 llvm-mc         0x0000000000407287
11 llvm-mc         0x0000000000407b02
12 llvm-mc         0x000000000040bc11
13 llvm-mc         0x000000000040bce9 main + 193
14 libc.so.6       0x00007ffcb2f7cc4d __libc_start_main + 253
15 llvm-mc         0x0000000000406fe9
Stack dump:
0.      Program arguments: Debug+Asserts/bin/llvm-mc --disassemble --triple=arm 
Aborted
------------------------------------------------------------

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 09:46:55 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 09:46:55 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7748] New: VC++ truncates X86II::VEX enum constants
Message-ID: <bug-7748-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7748

           Summary: VC++ truncates X86II::VEX enum constants
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: dimitry at andric.com
                CC: llvmbugs at cs.uiuc.edu


During compilation with VC++ 9 of several llvm files that include
X86InstInfo.h, I am getting the following warnings about it:

lib\target\x86\X86InstrInfo.h(442) : warning C4341: 'VEX' : signed value is out
of range for enum constant
lib\target\x86\X86InstrInfo.h(442) : warning C4309: 'initializing' : truncation
of constant value
lib\target\x86\X86InstrInfo.h(446) : warning C4341: 'VEX_W' : signed value is
out of range for enum constant
lib\target\x86\X86InstrInfo.h(446) : warning C4309: 'initializing' : truncation
of constant value
lib\target\x86\X86InstrInfo.h(451) : warning C4341: 'VEX_4V' : signed value is
out of range for enum constant
lib\target\x86\X86InstrInfo.h(451) : warning C4309: 'initializing' : truncation
of constant value
lib\target\x86\X86InstrInfo.h(456) : warning C4341: 'VEX_I8IMM' : signed value
is out of range for enum constant
lib\target\x86\X86InstrInfo.h(456) : warning C4309: 'initializing' : truncation
of constant value
lib\target\x86\X86InstrInfo.h(463) : warning C4341: 'VEX_L' : signed value is
out of range for enum constant
lib\target\x86\X86InstrInfo.h(463) : warning C4309: 'initializing' : truncation
of constant value

This is because of the following enum constants in the X86II namespace:

    // VEX - The opcode prefix used by AVX instructions
    VEX         = 1ULL << 32,

    // VEX_W - Has a opcode specific functionality, but is used in the same
    // way as REX_W is for regular SSE instructions.
    VEX_W       = 1ULL << 33,

    // VEX_4V - Used to specify an additional AVX/SSE register. Several 2
    // address instructions in SSE are represented as 3 address ones in AVX
    // and the additional register is encoded in VEX_VVVV prefix.
    VEX_4V      = 1ULL << 34,

    // VEX_I8IMM - Specifies that the last register used in a AVX instruction,
    // must be encoded in the i8 immediate field. This usually happens in
    // instructions with 4 operands.
    VEX_I8IMM   = 1ULL << 35,

    // VEX_L - Stands for a bit in the VEX opcode prefix meaning the current
    // instruction uses 256-bit wide registers. This is usually auto detected
if
    // a VR256 register is used, but some AVX instructions also have this field
    // marked when using a f256 memory references.
    VEX_L       = 1ULL << 36

So unfortunately VC++ truncates these constants, which is probably very
bad if you count on those exact values.  The MSDN library says the
following about this warning:

"An enumerated constant exceeds the limit for an int. The value of the
invalid constant is undefined. Constants must resolve to integers
between ?4,294,967,295 and +4,294,967,295 (signed)."

http://msdn.microsoft.com/en-us/library/x651y302.aspx

Therefore, these constants should probably be replaced by globals, or
maybe even #defines... (yuck :)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 11:06:32 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 11:06:32 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7740] passing address of reference failing in ObjC++
In-Reply-To: <bug-7740-206@http.llvm.org/bugs/>
References: <bug-7740-206@http.llvm.org/bugs/>
Message-ID: <20100729160632.04E6C2A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7740

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Douglas Gregor <dgregor at apple.com> 2010-07-29 11:06:31 CDT ---
Fixed in Clang r109753.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 11:09:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 11:09:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7749] New: ICE: Stack slot too small for load
Message-ID: <bug-7749-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7749

           Summary: ICE: Stack slot too small for load
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


regehr at john-home:~/volatile/bugs/tmp327$ clang -O1 -c small.c -w
clang: X86InstrInfo.cpp:2037: virtual void
llvm::X86InstrInfo::loadRegFromStackSlot(llvm::MachineBasicBlock&,
llvm::ilist_iterator<llvm::MachineInstr>, unsigned int, int, const
llvm::TargetRegisterClass*, const llvm::TargetRegisterInfo*) const: Assertion
`MF.getFrameInfo()->getObjectSize(FrameIdx) >= RC->getSize() && "Stack slot too
small for load"' failed.
0  clang 0x092d46a8
Stack dump:
0.      Program arguments:
/home/regehr/z/compiler-install/llvm-gcc-r109727-install/bin/clang -cc1 -triple
i386-pc-linux-gnu -S -disable-free -main-file-name small.c -mrelocation-model
static -mdisable-fp-elim -mconstructor-aliases -target-cpu pentium4
-resource-dir
/home/regehr/z/compiler-install/llvm-gcc-r109727-install/lib/clang/2.8 -O1 -w
-ferror-limit 19 -fmessage-length 100 -fgnu-runtime -fdiagnostics-show-option
-fcolor-diagnostics -o /tmp/cc-Qo2EKn.s -x c small.c 
1.      <eof> parser at end of file
2.      Code generation
3.      Running pass 'Linear Scan Register Allocator' on function '@func_88'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)
regehr at john-home:~/volatile/bugs/tmp327$ cat small.c
static signed char
foo (signed char si1, signed char si2)
{
  return si1 - si2;
}

static signed char
bar (signed char si1, signed char si2)
{
  return si2 || si1 && si2 ? : si1 + si2;
}

int g_5;
int g_85;

void
func_88 (struct S0 *p_89, unsigned short p_90)
{
  if (func_91 (g_5, 0) &
      bar (p_90,
           safe (safe (g_85, 0) <
                 foo (1L, p_90),
                 0)))
    {
      int *l_104 = &g_85;
    lbl_110:*l_104 = p_90;
    }
}
regehr at john-home:~/volatile/bugs/tmp327$ clang -v
clang version 2.8 (trunk 109727)
Target: i386-pc-linux-gnu
Thread model: posix

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 11:13:00 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 11:13:00 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7727] LLVM IR generation of compound statement
 ('{}') for Markzware's HashTable.ii
In-Reply-To: <bug-7727-206@http.llvm.org/bugs/>
References: <bug-7727-206@http.llvm.org/bugs/>
Message-ID: <20100729161300.C91002A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7727

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #5 from Douglas Gregor <dgregor at apple.com> 2010-07-29 11:13:00 CDT ---
Fixed in Clang r109757.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 11:28:43 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 11:28:43 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7742] Aggregate passed in wrong register on x86-64
 Darwin/Linux.
In-Reply-To: <bug-7742-206@http.llvm.org/bugs/>
References: <bug-7742-206@http.llvm.org/bugs/>
Message-ID: <20100729162843.279542A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7742

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Chris Lattner <clattner at apple.com> 2010-07-29 11:28:42 CDT ---
Actually, this looks already fixed on mainline.  I get this:

define double @foo(double %X.coerce) nounwind ssp {

This is terrible for performance, but is correct.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 11:29:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 11:29:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7742] Aggregate passed in wrong register on x86-64
 Darwin/Linux.
In-Reply-To: <bug-7742-206@http.llvm.org/bugs/>
References: <bug-7742-206@http.llvm.org/bugs/>
Message-ID: <20100729162947.673C32A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7742

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-29 11:29:47 CDT ---
Actually, I was looking at the wrong testcase, this is definitely broken on
mainline.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 11:57:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 11:57:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7750] New: likely integer wrong code bug
Message-ID: <bug-7750-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7750

           Summary: likely integer wrong code bug
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu


regehr at john-home:~/work013$ clang -O0 small.c -o small
regehr at john-home:~/work013$ ./small
1
regehr at john-home:~/work013$ clang -O2 small.c -o small
regehr at john-home:~/work013$ ./small
0

regehr at john-home:~/work013$ clang -v

clang version 2.8 (trunk 109491)
Target: i386-pc-linux-gnu
Thread model: posix

regehr at john-home:~/work013$ cat small.c

static short foo (short left, unsigned int right)
{
  return left;
}

struct S0
{
  unsigned f7:15;
};

unsigned short g_158[8] = {
  0xAF2FL, 0xAF2FL, 0xAF2FL, 0xAF2FL, 0xAF2FL, 0xAF2FL, 0xAF2FL, 0xAF2FL
};

struct S0 g_556[9][2];

struct S0 g_558 = {
  0L
};

unsigned char l_565[3][1];

extern int printf (__const char *__restrict __format, ...);

int main (void)
{
  const int g_264 = 1L;
  struct S0 *l_555 = &g_556[4][0];
  struct S0 *l_557 = &g_558;
  short y = foo ((l_555 == l_557) | (g_264 && g_158[2]), 0);
  printf ("%d\n", g_264 == y);
  return 0;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 12:05:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 12:05:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7742] Aggregate passed in wrong register on x86-64
 Darwin/Linux.
In-Reply-To: <bug-7742-206@http.llvm.org/bugs/>
References: <bug-7742-206@http.llvm.org/bugs/>
Message-ID: <20100729170547.AF8D92A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7742

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-29 12:05:47 CDT ---
Fixed in r109759, thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 12:13:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 12:13:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7751] New: scope private not identified
Message-ID: <bug-7751-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7751

           Summary: scope private not identified
           Product: clang
           Version: 2.7
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sc.venkatesh at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


#include <stdio.h>

class test
{
    void print( )
    {
        printf( "private function" );
    }
};

int main( )
{
    test p;
/*
    clang does not complain private print(). is this intentional ?
*/
    p.print( );
    return 0;
}

Clang Compiler without any Changes Built from V2.7.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 12:18:18 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 12:18:18 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7751] scope private not identified
In-Reply-To: <bug-7751-206@http.llvm.org/bugs/>
References: <bug-7751-206@http.llvm.org/bugs/>
Message-ID: <20100729171818.AE6DB2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7751

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Chris Lattner <clattner at apple.com> 2010-07-29 12:18:18 CDT ---
Access control wasn't fully done in Clang 2.7, please try mainline.  Thanks!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 12:28:49 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 12:28:49 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7255] ADL failing for a template's base.
In-Reply-To: <bug-7255-206@http.llvm.org/bugs/>
References: <bug-7255-206@http.llvm.org/bugs/>
Message-ID: <20100729172849.61FEF2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7255

Douglas Gregor <dgregor at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #2 from Douglas Gregor <dgregor at apple.com> 2010-07-29 12:28:49 CDT ---
Ah, yes. We need to instantiate types when performing argument-dependent
lookup. This is also PR6231.

*** This bug has been marked as a duplicate of bug 6231 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 12:43:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 12:43:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7749] ICE: Stack slot too small for load
In-Reply-To: <bug-7749-206@http.llvm.org/bugs/>
References: <bug-7749-206@http.llvm.org/bugs/>
Message-ID: <20100729174335.37F4D2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7749

Jakob Stoklund Olesen <stoklund at 2pi.dk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Jakob Stoklund Olesen <stoklund at 2pi.dk> 2010-07-29 12:43:34 CDT ---
Fixed in r109764. I removed the assertion, it was too aggressive.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 13:03:52 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 13:03:52 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7750] likely integer wrong code bug
In-Reply-To: <bug-7750-206@http.llvm.org/bugs/>
References: <bug-7750-206@http.llvm.org/bugs/>
Message-ID: <20100729180352.3BB6D2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7750

Eli Friedman <sharparrow1 at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |sharparrow1 at yahoo.com
         Resolution|                            |FIXED

--- Comment #1 from Eli Friedman <sharparrow1 at yahoo.com> 2010-07-29 13:03:51 CDT ---
Fixed in r109773.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 13:29:42 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 13:29:42 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7745] llvm-mc: ARMDisassemblerCore.cpp:890: bool
 DisassembleBrMiscFrm(llvm::MCInst&, unsigned int, uint32_t,
 short unsigned int, unsigned int&,
 llvm::ARMBasicMCBuilder*): Assertion `0 && "Unexpected BrMiscFrm Opcode"'
 failed.
In-Reply-To: <bug-7745-206@http.llvm.org/bugs/>
References: <bug-7745-206@http.llvm.org/bugs/>
Message-ID: <20100729182942.F2BF82A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7745

Bob Wilson <bob.wilson at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Bob Wilson <bob.wilson at apple.com> 2010-07-29 13:29:42 CDT ---
Fixed in svn 109788.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:26:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:26:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 1284] Support bitwidth attribute in llvm-gcc
In-Reply-To: <bug-1284-206@http.llvm.org/bugs/>
References: <bug-1284-206@http.llvm.org/bugs/>
Message-ID: <20100729222605.8592C2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=1284

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #5 from Chris Lattner <clattner at apple.com> 2010-07-29 17:26:04 CDT ---
We don't plan to do this in llvm-gcc.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:26:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:26:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 1650] Clean up the asmprinter interfaces
In-Reply-To: <bug-1650-206@http.llvm.org/bugs/>
References: <bug-1650-206@http.llvm.org/bugs/>
Message-ID: <20100729222650.208EE2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=1650

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #8 from Chris Lattner <clattner at apple.com> 2010-07-29 17:26:48 CDT ---
This is basically done.  If there are any other pending issues, please open a
new bug.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:30:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:30:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3287] Vector of float with length 2 broken on x86
In-Reply-To: <bug-3287-206@http.llvm.org/bugs/>
References: <bug-3287-206@http.llvm.org/bugs/>
Message-ID: <20100729223044.539F62A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3287

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #12 from Chris Lattner <clattner at apple.com> 2010-07-29 17:30:42 CDT ---
<2 x float> isn't done with mmx anymore.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:32:53 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:32:53 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3413] llvm-gcc: x86_64 passing of struct { char a,
 b, c; } generates invalid read
In-Reply-To: <bug-3413-206@http.llvm.org/bugs/>
References: <bug-3413-206@http.llvm.org/bugs/>
Message-ID: <20100729223253.38A2E2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3413

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #5 from Chris Lattner <clattner at apple.com> 2010-07-29 17:32:52 CDT ---
This is  (now) fixed in clang, we have no plans to fix this in llvm-gcc.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:37:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:37:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3448] clang does not match gcc/llvm-gcc in
 returning struct { <2 x float> }
In-Reply-To: <bug-3448-206@http.llvm.org/bugs/>
References: <bug-3448-206@http.llvm.org/bugs/>
Message-ID: <20100729223745.84DE52A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3448

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #3 from Chris Lattner <clattner at apple.com> 2010-07-29 17:37:44 CDT ---
"Chris's belief is that no one uses <2 x float>, so we should be free to do
whatever we want" :)

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:39:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:39:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3987] support x86_64 Objective-C code generation
	for Leopard
In-Reply-To: <bug-3987-206@http.llvm.org/bugs/>
References: <bug-3987-206@http.llvm.org/bugs/>
Message-ID: <20100729223944.36CFA2A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3987

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Chris Lattner <clattner at apple.com> 2010-07-29 17:39:43 CDT ---
Clang defaults to the integrated assembler now.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:51:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:51:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3199] callee can't get the i64 parameter correctly.
In-Reply-To: <bug-3199-206@http.llvm.org/bugs/>
References: <bug-3199-206@http.llvm.org/bugs/>
Message-ID: <20100729225150.AB5E62A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3199

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #9 from Dale Johannesen <dalej at apple.com> 2010-07-29 17:51:50 CDT ---
I'm not convinced it ever was a problem.  The last example works correctly on
(darwin) ppc32, although it doesn't appear to test what he says is wrong. 
Let's close it; if there's something wrong, please provide an example that
demonstrates it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 17:54:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 17:54:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7753] New: cannot initialize a parameter of type
 'id<NSMenuDelegate>' with an lvalue of type 'MenuDelegate *'
Message-ID: <bug-7753-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7753

           Summary: cannot initialize a parameter of type
                    'id<NSMenuDelegate>' with an lvalue of type
                    'MenuDelegate *'
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: Nomis101 at web.de
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com
            Blocks: 5511


If I apply all currently available patches and than build Mozilla Firefox with
Clang on Mac OS X 10.6, than it stops with the below error. This is
Mac-specific code. You can find the file in question there:
http://hg.mozilla.org/mozilla-central/file/1b14b3b3095d/widget/src/cocoa/nsMenuX.mm#l509

The error is:


/Volumes/Developer/temp/src/mozilla/widget/src/cocoa/nsMenuX.mm:509:23: error: 
      cannot initialize a parameter of type 'id<NSMenuDelegate>' with an lvalue
      of type 'MenuDelegate *'
  [myMenu setDelegate:mMenuDelegate];
                      ^~~~~~~~~~~~~
In file included from
/Volumes/Developer/temp/src/mozilla/widget/src/cocoa/nsMenuX.mm:41:
In file included from
/Volumes/Developer/temp/src/mozilla/widget/src/cocoa/nsMenuX.h:42:
In file included from
/Developer/SDKs/MacOSX10.6.sdk/System/Library/Frameworks/Cocoa.framework/Headers/Cocoa.h:13:
In file included from
/Developer/SDKs/MacOSX10.6.sdk/System/Library/Frameworks/AppKit.framework/Headers/AppKit.h:34:
/Developer/SDKs/MacOSX10.6.sdk/System/Library/Frameworks/AppKit.framework/Headers/NSMenu.h:146:42:
note: 
      passing argument to parameter 'anObject' here
- (void)setDelegate:(id <NSMenuDelegate>)anObject ...
                                         ^
1 warning generated.
In file included from
/Volumes/Developer/temp/src/mozilla/widget/src/cocoa/nsMenuX.mm:47:
../../../dist/include/nsObjCExceptions.h:161:13: warning: unused function
      'nsObjCExceptionLogAbort' [-Wunused-function]
static void nsObjCExceptionLogAbort(NSException *e)
            ^
2 warnings and 1 error generated.
make[6]: *** [nsMenuX.o] Error 1



I've tried to add a preprocessed source. But if I build with --save-temps, than
I get a completely different error in a completely different file.

clang version 2.8 (trunk 109500)
Target: x86_64-apple-darwin10

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 18:29:11 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 18:29:11 -0500 (CDT)
Subject: [LLVMbugs] [Bug 2217] problem with copying empty base classes
In-Reply-To: <bug-2217-206@http.llvm.org/bugs/>
References: <bug-2217-206@http.llvm.org/bugs/>
Message-ID: <20100729232911.A0DBE2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=2217

Dale Johannesen <dalej at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Dale Johannesen <dalej at apple.com> 2010-07-29 18:29:11 CDT ---
Apparently, it works on x86-32 and x86-64 darwin.  I'm not in a position to try
Linux.  If it's not fixed on Linux I'm sure Jay will let us know.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 21:41:51 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 21:41:51 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7499] Invalid mls instruction generated in thumb2
	mode
In-Reply-To: <bug-7499-206@http.llvm.org/bugs/>
References: <bug-7499-206@http.llvm.org/bugs/>
Message-ID: <20100730024151.70EBC2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7499

Jim Grosbach <grosbach at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Jim Grosbach <grosbach at apple.com> 2010-07-29 21:41:50 CDT ---
Fixed in r109842.

Thanks for the report, Eli. This is a nasty bug and it's good to get it cleaned
up.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 22:37:07 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 22:37:07 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7754] New: Clang doesn't like extern "C++" followed
 by a template declaration.
Message-ID: <bug-7754-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7754

           Summary: Clang doesn't like extern "C++" followed by a template
                    declaration.
           Product: clang
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: pichet2000 at gmail.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com


Clang doesn't like this code:
extern "C++"
template <class T> int func(T param);

But it correctly parses this:
extern "C++" {
    template <class T> int func(T param);
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Thu Jul 29 23:02:39 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Thu, 29 Jul 2010 23:02:39 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5179] [X86-64 ABI] miscompilation of multiple empty
 base classes
In-Reply-To: <bug-5179-206@http.llvm.org/bugs/>
References: <bug-5179-206@http.llvm.org/bugs/>
Message-ID: <20100730040239.D02EA2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5179

Chris Lattner <clattner at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Chris Lattner <clattner at apple.com> 2010-07-29 23:02:39 CDT ---
Fixed in r109848!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 09:29:28 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 09:29:28 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7755] New: Assertion `i != e && "Didn't find
	edge?"' failed.
Message-ID: <bug-7755-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7755

           Summary: Assertion `i != e && "Didn't find edge?"' failed.
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu, chenyang at cs.utah.edu


[regehr at bethe tmp605]$ clang -O2 -c small.c -w
clang: BasicBlockUtils.cpp:287: unsigned int
llvm::GetSuccessorNumber(llvm::BasicBlock*, llvm::BasicBlock*): Assertion `i !=
e && "Didn't find edge?"' failed.
0  clang           0x00000000015f580f
1  clang           0x00000000015f78a2
2  libpthread.so.0 0x00007ffde271a190
3  libc.so.6       0x00007ffde1a1f4b5 gsignal + 53
4  libc.so.6       0x00007ffde1a22f50 abort + 384
5  libc.so.6       0x00007ffde1a18481 __assert_fail + 241
6  clang           0x00000000013e9f95
7  clang           0x00000000013eb2db llvm::SplitEdge(llvm::BasicBlock*,
llvm::BasicBlock*, llvm::Pass*) + 43
8  clang           0x000000000133b0b8
9  clang           0x000000000133bf5e
10 clang           0x000000000133c31e
11 clang           0x000000000133d5ab
12 clang           0x0000000001548a1d
llvm::FPPassManager::runOnFunction(llvm::Function&) + 637
13 clang           0x0000000001430d9b
14 clang           0x0000000001431515
15 clang           0x000000000154851e
llvm::MPPassManager::runOnModule(llvm::Module&) + 494
16 clang           0x0000000001548697 llvm::PassManagerImpl::run(llvm::Module&)
+ 167
17 clang           0x00000000007ade67
clang::EmitBackendOutput(clang::Diagnostic&, clang::CodeGenOptions const&,
clang::TargetOptions const&, llvm::Module*, clang::BackendAction,
llvm::raw_ostream*) + 1639
18 clang           0x00000000007a9fab
19 clang           0x00000000008c04bf clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 335
20 clang           0x00000000007aaf24 clang::CodeGenAction::ExecuteAction() +
68
21 clang           0x00000000006c7755
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 357
22 clang           0x00000000006a6d4b cc1_main(char const**, char const**, char
const*, void*) + 2651
23 clang           0x00000000006ad517 main + 3543
24 libc.so.6       0x00007ffde1a0aabd __libc_start_main + 253
25 clang           0x00000000006a44b9
Stack dump:
0.      Program arguments:
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r109792-install/bin/clang
-cc1 -triple x86_64-unknown-linux-gnu -S -disable-free -main-file-name small.c
-mrelocation-model static -mdisable-fp-elim -mconstructor-aliases
-munwind-tables -target-cpu x86-64 -resource-dir
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r109792-install/lib/clang/2.8
-O2 -w -ferror-limit 19 -fmessage-length 100 -fgnu-runtime
-fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-zGhIaZ.s -x c small.c 
1.      <eof> parser at end of file
2.      Per-module optimization passes
3.      Running pass 'CallGraph Pass Manager' on module 'small.c'.
4.      Running pass 'Jump Threading' on function '@int328'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)
[regehr at bethe tmp605]$  
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ clang -v
clang version 2.8 (trunk 109792)
Target: x86_64-unknown-linux-gnu
Thread model: posix
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ 
[regehr at bethe tmp605]$ cat small.c
static signed char
foo (signed char si1, unsigned char si2)
{
  return si2 || si1 == si2 ? si1 : (si1 / si2);
}

static unsigned char
bar (unsigned char left, unsigned int right)
{
  return right || left >> right ? left : left << right;
}

static unsigned short
baz (unsigned short ui1, unsigned char ui2)
{
  return ui1 - ui2;
}

struct S0
{
  const signed f2:1;
};
int g_17[7][9] = {
};

int *volatile g_16[1] = {
  &g_17[2][6]
};

struct S0 g_477 = {
  9600
};

void int328 (unsigned p_89)
{
  {
    int i, j, k, l;
    if (1 < func_100 (g_477.f2) >
        (safe (l + p_89)
         && baz (foo (p_89, p_89),
                 0xAD8DD980L)) ^ (1
                                  &&
                                  bar
                                  (p_89, 0)))
      g_16[0];
  }
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 11:09:36 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 11:09:36 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7753] cannot initialize a parameter of type
 'id<NSMenuDelegate>' with an lvalue of type 'MenuDelegate *'
In-Reply-To: <bug-7753-206@http.llvm.org/bugs/>
References: <bug-7753-206@http.llvm.org/bugs/>
Message-ID: <20100730160936.2DA992A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7753

Fariborz Jahanian <fjahanian at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Fariborz Jahanian <fjahanian at apple.com> 2010-07-30 11:09:35 CDT ---
(In reply to comment #2)
> Created an attachment (id=5288)
 --> (http://llvm.org/bugs/attachment.cgi?id=5288) [details]
> nsMenuX.mii
> 
> OK, I've tried to add --save-temps only to /src/cocoa/ and the entire project.
> I've got the same file both times. Hope this helps.

Thanks for the pre-processed file. clang's complaint is legit and can be shown
in this test case:

@protocol NSMenuDelegate;

@interface GeckoNSMenu
- (void)setDelegate:(id <NSMenuDelegate> )anObject ;
@end

@interface MenuDelegate 
@end

MenuDelegate* mMenuDelegate;

GeckoNSMenu* CreateMenuWithGeckoString()
{
  GeckoNSMenu* myMenu;
  [myMenu setDelegate:mMenuDelegate];
}

'mMenuDelegate' 's type is 'MenuDelegate*' which does not implement the
'NSMenuDelegate' protocol.
Note that gcc issues a warning in this case:
% gcc -c test.mm
test.mm: In function ?GeckoNSMenu* CreateMenuWithGeckoString()?:
test.mm:15: warning: class 'MenuDelegate' does not implement the
'NSMenuDelegate' protocol

To fix it is the obvious:

% diff nsMenuX.mii nsMenuX.fixit.mii
132607c132607
<   MenuDelegate* mMenuDelegate;
---
>   MenuDelegate<NSMenuDelegate>* mMenuDelegate;

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 11:42:54 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 11:42:54 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7756] New: Analyze for missing ivars in a -copy
	method
Message-ID: <bug-7756-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7756

           Summary: Analyze for missing ivars in a -copy method
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: randy.meadows at not-pc.com
                CC: llvmbugs at cs.uiuc.edu


I just spent far too long tracking down a bug where I had added an ivar to a
class definition, and then failed to also add that ivar in a -copy method, so
that ivar did not get set correctly in the copy.

It'd be awesome if the analyzer, when it comes across a -copy-type method
implementation, would ensure that all the ivars are referenced in that method.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 11:45:03 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 11:45:03 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7757] New: Some ARM instructions contain extra \n
 when disassembled
Message-ID: <bug-7757-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7757

           Summary: Some ARM instructions contain extra \n when
                    disassembled
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sliedes at cc.hut.fi
                CC: llvmbugs at cs.uiuc.edu


Hi,

When I use llvm-mc to disassemble blocks of ARM instructions, some instructions
 contain an extra newline after them. One such instruction is '0xc9 0xff 0x87
0x00' (add pc, r7, r9). For example:

No extra newline (works correctly, instruction '0 0 0 0'):

--------------------------------------------------
$ echo '0 0 0 0'|Debug+Asserts/bin/llvm-mc --disassemble --triple=arm
        andeq   r0, r0, r0
$ echo '0 0 0 0 0 0 0 0' |Debug+Asserts/bin/llvm-mc --disassemble --triple=arm
$ echo '0 0 0 0 0 0 0 0' |Debug+Asserts/bin/llvm-mc --disassemble --triple=arm
        andeq   r0, r0, r0
        andeq   r0, r0, r0
$
--------------------------------------------------

Incorrect extra newline, instruction c9 ff 87 00:

--------------------------------------------------
$ echo '0xc9 0xff 0x87 0x00' |Debug+Asserts/bin/llvm-mc --disassemble
--triple=arm
        add     pc, r7, r9 

$ echo '0xc9 0xff 0x87 0x00 0 0 0 0' |Debug+Asserts/bin/llvm-mc --disassemble
--triple=arm
        add     pc, r7, r9 

        andeq   r0, r0, r0
--------------------------------------------------

As can be seen, there's an extra newline after the add instruction.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 11:45:57 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 11:45:57 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7758] New: false positive null pointer analysis
Message-ID: <bug-7758-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7758

           Summary: false positive null pointer analysis
           Product: clang
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: eblake at redhat.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5289)
 --> (http://llvm.org/bugs/attachment.cgi?id=5289)
html report from clang, with false claim of a null dereference

Using scan-build from clang-2.7-5.fc13.i686 on Fedora 13 to compile libvirt, I
found a false positive.  The attached html file claims at step 14 that the code
is passing a potentially NULL argument to strcmp.  However, that claim is
invalid.

At point 14, the two arguments to strcmp are group->controllers[i].mountPoint
(guaranteed non-NULL, due to line 489-490 earlier in the function) and
group->controllers [VIR_CGROUP_CONTROLLER_MEMORY].mountPoint (guaranteed
non-NULL, due to line 518 earlier in the same conditional).

I'm wondering ifthe clang analyzer is getting confused when the iteration hits
i == 3 == VIR_CGROUP_CONTROLLER_CPUSET, and failing to realize that the
assumption of point 12 of the analysis (assuming that
group->controllers[VIR_CGROUP_CONTROLLER_CPUSET].mountPoint is NULL) was
already disproved at point 8 (group->controllers[i].mountPoint is non-NULL);
once you re-introduce a bogus assumption at point 12, that would explain the
complaint at point 14.

I'm also attaching the preprocessed input that the html report was generated
from.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 11:49:50 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 11:49:50 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6991] extern "C" typedef ....
In-Reply-To: <bug-6991-206@http.llvm.org/bugs/>
References: <bug-6991-206@http.llvm.org/bugs/>
Message-ID: <20100730164950.EB2362A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6991

Abramo Bagnara <abramobagnara at tin.it> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |abramobagnara at tin.it
         Resolution|                            |FIXED

--- Comment #5 from Abramo Bagnara <abramobagnara at tin.it> 2010-07-30 11:49:50 CDT ---
Fixed in r109865.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 11:53:58 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 11:53:58 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5127] Non-type template arguments not parsed
 correctly in default arguments
In-Reply-To: <bug-5127-206@http.llvm.org/bugs/>
References: <bug-5127-206@http.llvm.org/bugs/>
Message-ID: <20100730165358.B4AB52A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5127

Sebastian Redl <sebastian.redl at getdesigned.at> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #3 from Sebastian Redl <sebastian.redl at getdesigned.at> 2010-07-30 11:53:58 CDT ---
Sounds good to me.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 13:19:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 13:19:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7759] New: Crash in CallGraphSCCPass with
 -argpromotion -loop-extract-single
Message-ID: <bug-7759-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7759

           Summary: Crash in CallGraphSCCPass with -argpromotion
                    -loop-extract-single
           Product: new-bugs
           Version: 2.6
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: collinwinter at google.com
                CC: llvmbugs at cs.uiuc.edu, nlewycky at google.com


Created an attachment (id=5292)
 --> (http://llvm.org/bugs/attachment.cgi?id=5292)
Reduced test case

Running `opt ~/sccpass-crash.ll  -argpromotion -loop-extract-single -o
/dev/null` results in an assertion failure in CallGraphSCCPass. I've tested
this as of r109867 in a Release+Debug+Asserts build.

Assertion failed: (CallSites.empty() && "Dangling pointers found in call sites
map"), function RefreshCallGraph, file
/Users/collinwinter/src/llvm/trunk/lib/Analysis/IPA/CallGraphSCCPass.cpp, line
333.
0  opt               0x0054e748 llvm::SearchForAddressOfSpecialSymbol(char
const*) + 1048
1  opt               0x0054f4a6 llvm::sys::RemoveFileOnSignal(llvm::sys::Path
const&, std::string*) + 1334
2  libSystem.B.dylib 0x954a92bb _sigtramp + 43
3  libSystem.B.dylib 0xffffffff _sigtramp + 1790274927
4  libSystem.B.dylib 0x9551d23a raise + 26
5  libSystem.B.dylib 0x95529679 abort + 73
6  libSystem.B.dylib 0x9551e3db __assert_rtn + 101
7  opt               0x0023edf2 (anonymous
namespace)::CGPassManager::RefreshCallGraph(llvm::CallGraphSCC&,
llvm::CallGraph&, bool) + 2354
8  opt               0x0023fd27 (anonymous
namespace)::CGPassManager::RunAllPassesOnSCC(llvm::CallGraphSCC&,
llvm::CallGraph&, bool&) + 823
9  opt               0x00240add (anonymous
namespace)::CGPassManager::runOnModule(llvm::Module&) + 317
10 opt               0x004b5df6 llvm::MPPassManager::runOnModule(llvm::Module&)
+ 566
11 opt               0x004b61c4 llvm::PassManagerImpl::run(llvm::Module&) + 164
12 opt               0x004b62bb llvm::PassManager::run(llvm::Module&) + 27
13 opt               0x0000c714 main + 3156
14 opt               0x000020e6 start + 54
Stack dump:
0.    Program arguments:
/Users/collinwinter/src/llvm/obj/Release+Debug+Asserts/bin/opt
/Users/collinwinter/sccpass-crash.ll -argpromotion -loop-extract-single -o
/dev/null 
1.    Running pass 'CallGraph Pass Manager' on module
'/Users/collinwinter/sccpass-crash.ll'.
Abort trap

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 13:33:04 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 13:33:04 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7760] New: Segfault, bus error with -libcall-aa
Message-ID: <bug-7760-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7760

           Summary: Segfault, bus error with -libcall-aa
           Product: new-bugs
           Version: 2.6
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: collinwinter at google.com
                CC: llvmbugs at cs.uiuc.edu, nlewycky at google.com


Created an attachment (id=5293)
 --> (http://llvm.org/bugs/attachment.cgi?id=5293)
Reduced test case for LICM error

I've found a bunch of problems in -libcall-aa. I know it's not enabled in any
of the standard opt pipelines, but I wanted to report them anyway. I've tested
this as of r109867.


Running `opt llvm-segfault.ll -libcall-aa -sink -jump-threading -o /dev/null`
results in malloc errors and an eventual segfault. Things are too broken for
gdb to be able to give a backtrace.

$ /Users/collinwinter/src/llvm/obj/Release+Debug+Asserts/bin/opt
~/llvm-segfault.ll -libcall-aa -sink -jump-threading -o /dev/null
opt(91528) malloc: *** error for object 0x54f4a6: Non-aligned pointer being
freed
*** set a breakpoint in malloc_error_break to debug
Segmentation fault


Running `opt licm.ll -libcall-aa -licm -o /dev/null` results in the following
stack trace:

0  opt               0x0054e748 llvm::SearchForAddressOfSpecialSymbol(char
const*) + 1048
1  opt               0x0054f4a6 llvm::sys::RemoveFileOnSignal(llvm::sys::Path
const&, std::string*) + 1334
2  libSystem.B.dylib 0x954a92bb _sigtramp + 43
3  libSystem.B.dylib 0xffffffff _sigtramp + 1790274927
4  opt               0x004bd1e9 llvm::createPrintFunctionPass(std::string
const&, llvm::raw_ostream*, bool) + 105
5  opt               0x000a8fec (anonymous
namespace)::LICM::canSinkOrHoistInst(llvm::Instruction&) + 188
6  opt               0x000acc2f (anonymous
namespace)::LICM::SinkRegion(llvm::DomTreeNodeBase<llvm::BasicBlock>*) + 367
7  opt               0x000acb4e (anonymous
namespace)::LICM::SinkRegion(llvm::DomTreeNodeBase<llvm::BasicBlock>*) + 142
8  opt               0x000ad25c (anonymous
namespace)::LICM::runOnLoop(llvm::Loop*, llvm::LPPassManager&) + 1420
9  opt               0x002a5f0c
llvm::LPPassManager::runOnFunction(llvm::Function&) + 1276
10 opt               0x004b652b
llvm::FPPassManager::runOnFunction(llvm::Function&) + 523
11 opt               0x004b66af llvm::FPPassManager::runOnModule(llvm::Module&)
+ 175
12 opt               0x004b5df6 llvm::MPPassManager::runOnModule(llvm::Module&)
+ 566
13 opt               0x004b61c4 llvm::PassManagerImpl::run(llvm::Module&) + 164
14 opt               0x004b62bb llvm::PassManager::run(llvm::Module&) + 27
15 opt               0x0000c714 main + 3156
16 opt               0x000020e6 start + 54
Stack dump:
0.    Program arguments:
/Users/collinwinter/src/llvm/obj/Release+Debug+Asserts/bin/opt
/Users/collinwinter/licm.ll -libcall-aa -licm -o /dev/null 
1.    Running pass 'Function Pass Manager' on module
'/Users/collinwinter/licm.ll'.
2.    Running pass 'Loop Pass Manager' on function
'@_ZN5clang17PartialDiagnostic7StorageD2Ev'
3.    Running pass 'Loop Invariant Code Motion' on basic block '%for.cond'
Bus error

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 14:25:48 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 14:25:48 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7761] New: llc gives me "LLVM ERROR: Invalid
	primitive type"
Message-ID: <bug-7761-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7761

           Summary: llc gives me "LLVM ERROR: Invalid primitive type"
           Product: tools
           Version: 2.7
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: llc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jingl1345 at gmail.com
                CC: llvmbugs at cs.uiuc.edu


I have:
int main()
{
    int i = 1;
}
in test.c.

Then:
llvm-gcc -emit-llvm -O0 -c test.c -o test.bc -g
llc -march=cpp test.bc
gives me:
LLVM ERROR: Invalid primitive type
This only happens when I use the "-g".

Duncan Sands reported the following in the maillist:
Yes, it is a bug.  In fact LLVM from subversion from a few days ago
crashes even without -march=cpp:

$ llc test.bc
llc: llvm/include/llvm/Support/Casting.h:202: typename llvm::cast_retty<To, 
From>::ret_type llvm::cast(const Y&) [with X = llvm::MDNode, Y = llvm::Value*]: 
Assertion `isa<X>(Val) && "cast<Ty>() argument of incompatible type!"' failed.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 15:11:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 15:11:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7762] New: OpenSolaris link failure,
 link commands missing key libraries
Message-ID: <bug-7762-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7762

           Summary: OpenSolaris link failure, link commands missing key
                    libraries
           Product: Build scripts
           Version: 2.7
          Platform: PC
        OS/Version: Solaris
            Status: NEW
          Severity: normal
          Priority: P
         Component: Makefiles
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: llvm001 at deltaprime.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5296)
 --> (http://llvm.org/bugs/attachment.cgi?id=5296)
Build and environment details.

The build process fails to link the LLVM tools. Specifically, the first tool to
build is 'opt' which results in 700+ undefined symbols.

I looked into the Makefiles and found that the "LINK_COMPONENTS" make variable
does not get the correct set of libraries. In fact, at the top level
Makefile.rules, the variable is set to "support system", however, these do not
seem to make it to the 'opt' make context.

The list of libraries for 'opt' is (from the g++ link command):

... -lLLVMipo -lLLVMScalarOpts -lLLVMInstrumentation -lLLVMAsmParser
-lLLVMBitWriter -lLLVMBitReader ...

It is missing at a minimum LLVMCore, which when added reduces the unresolved
externals to 400+.

The attachment has build and environment details.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 15:23:46 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 15:23:46 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7763] New: Spurious warning about adding to zero
Message-ID: <bug-7763-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7763

           Summary: Spurious warning about adding to zero
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: tjw at omnigroup.com
                CC: llvmbugs at cs.uiuc.edu


On the attached test, checker-246 warns:


test.m:8:16: warning: The right operand to '+=' is always 0
        result += numbers[bg];
        ~~~~~~ ^  ~~~~~~~~~~~
1 warning generated.

If you comment out the inner for loop:


        for (int i = 0; i < 256; i++) {
            unused = i;
        }

the spurious warning about adding to zero goes away. I think I've convinced
myself that the right side will only be zero on the first iteration of the
loop, but its possible clang is being more observant.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 15:41:33 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 15:41:33 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7760] Segfault, bus error with -libcall-aa
In-Reply-To: <bug-7760-206@http.llvm.org/bugs/>
References: <bug-7760-206@http.llvm.org/bugs/>
Message-ID: <20100730204133.995F92A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7760

Nick Lewycky <nlewycky at google.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Nick Lewycky <nlewycky at google.com> 2010-07-30 15:41:33 CDT ---
Actually, r109883 fixed everything.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 16:04:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 16:04:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7764] New: Initialized fields not tracked across
 inline functions
Message-ID: <bug-7764-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7764

           Summary: Initialized fields not tracked across inline functions
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: tjw at omnigroup.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5298)
 --> (http://llvm.org/bugs/attachment.cgi?id=5298)
test case

A fairly common pattern is to initialize rect.origin.x and then use
NSMinX(rect) later. But, checker-246 warns about this saying that other fields
in the struct aren't initialized. Strictly speaking this is correct, but since
the NS{Min,Max}{X,Y} functions are inlines, it could track that no use of
uninitialized values will actually happen.


rect-uninitialized-field.m:11:33: warning: Passed-by-value struct argument
contains uninitialized data (e.g., via the field chain: 'size.height')
    r.size.height = NSMaxY(b) - NSMinY(r);
                                ^      ~
1 warning generated.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 16:16:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 16:16:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7763] Spurious warning about adding to zero
In-Reply-To: <bug-7763-206@http.llvm.org/bugs/>
References: <bug-7763-206@http.llvm.org/bugs/>
Message-ID: <20100730211640.15DAC2A6C12C@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7763

Tom Care <tcare at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from Tom Care <tcare at apple.com> 2010-07-30 16:16:39 CDT ---
I accidentally commented out too much in one of my commits to
IdempotentOperationsChecker. Fixed in r109893. Thanks for reporting!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 16:19:08 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 16:19:08 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7765] New: UNREACHABLE executed during ArgPromotion
	pass
Message-ID: <bug-7765-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7765

           Summary: UNREACHABLE executed during ArgPromotion pass
           Product: new-bugs
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: collinwinter at google.com
                CC: llvmbugs at cs.uiuc.edu, nlewycky at google.com


Created an attachment (id=5299)
 --> (http://llvm.org/bugs/attachment.cgi?id=5299)
Reduced test case

Running `opt argpromotion-crash.ll -globalsmodref-aa -loopsimplify
-argpromotion -o /dev/null` results in the following crash. I've tested this
against r109883. Bugpoint was used to reduce the test case.

An asserting value handle still pointed to this value!
UNREACHABLE executed at
/Users/collinwinter/src/llvm/trunk/lib/VMCore/Value.cpp:531!
0  opt               0x0054e748 llvm::SearchForAddressOfSpecialSymbol(char
const*) + 1048
1  opt               0x0054f4a6 llvm::sys::RemoveFileOnSignal(llvm::sys::Path
const&, std::string*) + 1334
2  libSystem.B.dylib 0x954a92bb _sigtramp + 43
3  libSystem.B.dylib 0xffffffff _sigtramp + 1790274927
4  libSystem.B.dylib 0x9551d23a raise + 26
5  libSystem.B.dylib 0x95529679 abort + 73
6  opt               0x0052d119 llvm::llvm_unreachable_internal(char const*,
char const*, unsigned int) + 393
7  opt               0x004d574f
llvm::ValueHandleBase::ValueIsDeleted(llvm::Value*) + 2223
8  opt               0x004d6438 llvm::Value::~Value() + 808
9  opt               0x0045ce61 llvm::Function::~Function() + 657
10 opt               0x0001603c (anonymous
namespace)::ArgPromotion::DoPromotion(llvm::Function*,
llvm::SmallPtrSet<llvm::Argument*, 8u>&, llvm::SmallPtrSet<llvm::Argument*,
8u>&) + 10780
11 opt               0x0001890f (anonymous
namespace)::ArgPromotion::PromoteArguments(llvm::CallGraphNode*) + 1215
12 opt               0x00018d51 (anonymous
namespace)::ArgPromotion::runOnSCC(llvm::CallGraphSCC&) + 65
13 opt               0x0023f7f7 (anonymous
namespace)::CGPassManager::RunPassOnSCC(llvm::Pass*, llvm::CallGraphSCC&,
llvm::CallGraph&, bool&, bool&) + 359
14 opt               0x0023fbd5 (anonymous
namespace)::CGPassManager::RunAllPassesOnSCC(llvm::CallGraphSCC&,
llvm::CallGraph&, bool&) + 549
15 opt               0x00240a9d (anonymous
namespace)::CGPassManager::runOnModule(llvm::Module&) + 317
16 opt               0x004b5df6 llvm::MPPassManager::runOnModule(llvm::Module&)
+ 566
17 opt               0x004b61c4 llvm::PassManagerImpl::run(llvm::Module&) + 164
18 opt               0x004b62bb llvm::PassManager::run(llvm::Module&) + 27
19 opt               0x0000c6d4 main + 3156
20 opt               0x000020a6 start + 54
Stack dump:
0.    Program arguments:
/Users/collinwinter/src/llvm/obj/Release+Debug+Asserts/bin/opt
/Users/collinwinter/argpromotion-crash.ll -globalsmodref-aa -loopsimplify
-argpromotion -o /dev/null 
1.    Running pass 'CallGraph Pass Manager' on module
'/Users/collinwinter/argpromotion-crash.ll'.
Abort trap

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 16:33:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 16:33:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7766] New: Maybe spurious warnings about
	multiplying by one
Message-ID: <bug-7766-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7766

           Summary: Maybe spurious warnings about multiplying by one
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: tjw at omnigroup.com
                CC: llvmbugs at cs.uiuc.edu


In checker-246, we get a warning about multiplying by one in the following
code. In this case, and several others where the +=0 and *=1 warnings pop up,
it is of the pattern were we are setting up a name for a constant and then
spelling out a more explicit formula (basically showing our work).

I've yet to see these idempotence catch a bug, so I'm not sure what sort of
things these would save me from (might be worth these false positives, might
not). Anyway, I thought I'd mention this problem though I'm not sure you
can/should fix it =)

CGImageRef OQCreateVerticalGradientImage(CGGradientRef gradient, CFStringRef
colorSpaceName, size_t height, BOOL flip)
{
    size_t width = 1; // One pixel wide
    size_t bytesPerRow = 4*width; <<--- warning here

    CGColorSpaceRef colorSpace = CGColorSpaceCreateWithName(colorSpaceName);
    CGContextRef ctx = CGBitmapContextCreate(NULL, width, height,
8/*bitsPerComponent*/, bytesPerRow, colorSpace,
kCGImageAlphaPremultipliedFirst);
    CFRelease(colorSpace);

    CGRect bounds = CGRectMake(0, 0, width, height);
    CGContextAddRect(ctx, bounds);
    CGContextClip(ctx);

    CGPoint startPoint = bounds.origin;
    CGPoint endPoint = (CGPoint){ CGRectGetMinX(bounds), CGRectGetMaxY(bounds)
};

    if (flip)
        SWAP(startPoint, endPoint);

    CGContextDrawLinearGradient(ctx, gradient, startPoint, endPoint,
0/*options*/);

    CGContextFlush(ctx);
    CGImageRef gradientImage = CGBitmapContextCreateImage(ctx);
    CFRelease(ctx);

    return gradientImage;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 18:24:14 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 18:24:14 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7767] New: Leak not found when returning +1 ref
 object out of a function
Message-ID: <bug-7767-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7767

           Summary: Leak not found when returning +1 ref object out of a
                    function
           Product: clang
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: tjw at omnigroup.com
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5300)
 --> (http://llvm.org/bugs/attachment.cgi?id=5300)
test case

If you don't define a function as returning a +1 {CF,NS} ref, returning a +1
ref object should warn.

I just fixed a leak our source base due to this gap in reference tracking
(tested with checker-246).

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 18:28:15 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 18:28:15 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7746] llvm-mc: ARMInstPrinter.cpp:35: unsigned int
 NextReg(unsigned int): Assertion `0 && "Unexpected register enum"' failed.
In-Reply-To: <bug-7746-206@http.llvm.org/bugs/>
References: <bug-7746-206@http.llvm.org/bugs/>
Message-ID: <20100730232815.9DBA52A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7746

Bob Wilson <bob.wilson at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Bob Wilson <bob.wilson at apple.com> 2010-07-30 18:28:15 CDT ---
Fixed in svn 109933.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Fri Jul 30 18:37:31 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Fri, 30 Jul 2010 18:37:31 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7768] New: likely integer wrong code bug
Message-ID: <bug-7768-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7768

           Summary: likely integer wrong code bug
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu, jxyang at cs.utah.edu


regehr at john-home:~/work014$ clang -O1 small.c -o small
regehr at john-home:~/work014$ ./small
-7
regehr at john-home:~/work014$ clang -O2 small.c -o small
regehr at john-home:~/work014$ ./small
-1
regehr at john-home:~/work014$ clang -v
clang version 2.8 (trunk 109858)
Target: i386-pc-linux-gnu
Thread model: posix
regehr at john-home:~/work014$ cat small.c
extern int printf (__const char *__restrict __format, ...);

int g_13 = -7L;
int g_81 = 0x249DD628L;
int *g_125 = &g_81;
int *g_188 = &g_81;
int **g_303 = &g_188;
int *l_337 = &g_81;
int **l_352[2];

int main(void)
{
  int i;
  for (i = 0; i < 2; i++) {
    l_352[i] = &l_337;
  }
  int x = (g_81 != (*g_125)) <= (0x6749929EL <= (*g_125));
  int z = !(l_352[1] == 0);
  int y = -7L * z;
  g_81 = x * y;
  printf("%d\n", g_81);
  return 0;
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 00:58:35 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 00:58:35 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7747] llvm-mc: ARMDisassemblerCore.cpp:2439: bool
 DisassembleN1RegModImmFrm(llvm::MCInst&, unsigned int, uint32_t,
 short unsigned int, unsigned int&,
 llvm::ARMBasicMCBuilder*): Assertion `0 && "Unreachable code!"' failed.
In-Reply-To: <bug-7747-206@http.llvm.org/bugs/>
References: <bug-7747-206@http.llvm.org/bugs/>
Message-ID: <20100731055835.E16BF2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7747

Bob Wilson <bob.wilson at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Bob Wilson <bob.wilson at apple.com> 2010-07-31 00:58:35 CDT ---
Fixed in svn 109946.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 01:28:25 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 01:28:25 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7757] Some ARM instructions contain extra \n when
	disassembled
In-Reply-To: <bug-7757-206@http.llvm.org/bugs/>
References: <bug-7757-206@http.llvm.org/bugs/>
Message-ID: <20100731062825.E70D12A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7757

Bob Wilson <bob.wilson at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Bob Wilson <bob.wilson at apple.com> 2010-07-31 01:28:25 CDT ---
Fixed in svn 109948.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 06:55:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 06:55:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7770] New: easy to prevent false positive
Message-ID: <bug-7770-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7770

           Summary: easy to prevent false positive
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: dodegy at flying-stampe.de
                CC: llvmbugs at cs.uiuc.edu


Created an attachment (id=5301)
 --> (http://llvm.org/bugs/attachment.cgi?id=5301)
little demo of the false positive. 

The static analyzer reports a null pointer dereference witch doesn't happen. It
doesn't check the exit condition on a for loop correctly.

I've added 3 functions with minimal differences. The analyzer works correct in
two of them.

Hope that helps

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 13:00:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 13:00:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7046] Dependent CallExpr crashing clang IRgen
In-Reply-To: <bug-7046-206@http.llvm.org/bugs/>
References: <bug-7046-206@http.llvm.org/bugs/>
Message-ID: <20100731180005.6C30F2A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7046

Anders Carlsson <andersca at mac.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |andersca at mac.com
         Resolution|                            |FIXED

--- Comment #2 from Anders Carlsson <andersca at mac.com> 2010-07-31 13:00:04 CDT ---
I can't reproduce this in clang ToT. Please reopen if you can!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 14:29:27 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 14:29:27 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3811] support #pragma align
In-Reply-To: <bug-3811-206@http.llvm.org/bugs/>
References: <bug-3811-206@http.llvm.org/bugs/>
Message-ID: <20100731192927.CA3C72A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3811

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 14:29:27 CDT ---
Fixed here:
  http://llvm.org/viewvc/llvm-project?view=rev&revision=109952

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 14:57:25 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 14:57:25 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7607] Assertion failure: Use still stuck around
 after Def is destroyed
In-Reply-To: <bug-7607-206@http.llvm.org/bugs/>
References: <bug-7607-206@http.llvm.org/bugs/>
Message-ID: <20100731195725.6329B2A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7607

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |daniel at zuster.org
         Resolution|                            |WORKSFORME

--- Comment #3 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 14:57:24 CDT ---
I cannot reproduce with r109952. Duncan recently fixed something in this area,
which may have been related.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:01:59 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:01:59 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7573] Should reject results and arguments in XMM
 registers when building with -mno-sse on x86-64
In-Reply-To: <bug-7573-206@http.llvm.org/bugs/>
References: <bug-7573-206@http.llvm.org/bugs/>
Message-ID: <20100731200159.150302A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7573

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |daniel at zuster.org
         Resolution|                            |FIXED

--- Comment #3 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:01:58 CDT ---
Dale recently turned this into a backend error, at least:
--
ddunbar at giles:tmp$ cat t.c
double f(void) { return 0.0; }
ddunbar at giles:tmp$ clang -mno-sse -c t.c
fatal error: error in backend: SSE register return with SSE disabled
ddunbar at giles:tmp$ 
--

It would be nice to have that in the frontend instead of the backend, but that
seems like a very minor QOI issue.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:06:12 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:06:12 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7487] [MC] Crash when passing -no-integrated-as
In-Reply-To: <bug-7487-206@http.llvm.org/bugs/>
References: <bug-7487-206@http.llvm.org/bugs/>
Message-ID: <20100731200612.CE2622A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7487

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |WORKSFORME

--- Comment #3 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:06:12 CDT ---
I cannot reproduce this:
--
ddunbar at giles:tmp$ clang -m32 -gdwarf-2 -no-integrated-as -c assert.cc
ddunbar at giles:tmp$ clang -cc1 -triple i386-apple-darwin9.0.0 -S -disable-free
-main-file-name assert.cc -pic-level 1 -mdisable-fp-elim -masm-verbose
-target-cpu yonah -g -resource-dir /Users/thakis/src/llvm/Release/lib/clang/2.8
-ferror-limit 19 -fmessage-length 199 -fexceptions -fdiagnostics-show-option
-fcolor-diagnostics -o t.s -x c++ assert.cc
ddunbar at giles:tmp$ clang -v
clang -v
clang version 2.8 ($URL$ 109952)
Target: x86_64-apple-darwin10
Thread model: posix
ddunbar at giles:tmp$ 
--

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:07:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:07:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7520] clang: SemaExprCXX.cpp:2641: clang::Expr*
 clang::Sema::MaybeCreateCXXExprWithTemporaries(clang::Expr*): Assertion
 `SubExpr && "sub expression can't be null!"' failed.
In-Reply-To: <bug-7520-206@http.llvm.org/bugs/>
References: <bug-7520-206@http.llvm.org/bugs/>
Message-ID: <20100731200744.DC8732A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7520

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |daniel at zuster.org
         Resolution|                            |WORKSFORME

--- Comment #1 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:07:44 CDT ---
I cannot reproduce this with TOT:
--
ddunbar at giles:tmp$ cat t.cpp
#include <vector>

template<std::size_t dims>
class Map
{
public:

    Map() :
        resolutions()
    {}

private:
    double resolutions[dims];
};

typedef Map<2> Map2D;
ddunbar at giles:tmp$ clang -m32 -c t.cpp
ddunbar at giles:tmp$ clang -c t.cpp
ddunbar at giles:tmp$ clang -v
clang version 2.8 ($URL$ 109952)
Target: x86_64-apple-darwin10
Thread model: posix
ddunbar at giles:tmp$ 
--

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:25:02 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:25:02 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7719] selfhost phase 2 broken with release build
In-Reply-To: <bug-7719-206@http.llvm.org/bugs/>
References: <bug-7719-206@http.llvm.org/bugs/>
Message-ID: <20100731202502.EB63C2A6C12F@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7719

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:25:02 CDT ---
Yup, that worked -- thanks for the report Jean-Daniel!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:29:13 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:29:13 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7771] New: Assertion `(unsigned)FI-LowSpillSlot <
 SpillSlotToUsesMap.size() && "Invalid spill slot"' failed.
Message-ID: <bug-7771-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7771

           Summary: Assertion `(unsigned)FI-LowSpillSlot <
                    SpillSlotToUsesMap.size() && "Invalid spill slot"'
                    failed.
           Product: new-bugs
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: regehr at cs.utah.edu
                CC: llvmbugs at cs.uiuc.edu, chenyang at cs.utah.edu


[regehr at bethe tmp606]$ clang -O2 -m32 -fomit-frame-pointer -c small.c -w
clang: VirtRegMap.cpp:180: void llvm::VirtRegMap::addSpillSlotUse(int,
llvm::MachineInstr*): Assertion `(unsigned)FI-LowSpillSlot <
SpillSlotToUsesMap.size() && "Invalid spill slot"' failed.
0  clang           0x00000000015f9d1f
1  clang           0x00000000015fbdb2
2  libpthread.so.0 0x00007fbd0741a190
3  libc.so.6       0x00007fbd0671f4b5 gsignal + 53
4  libc.so.6       0x00007fbd06722f50 abort + 384
5  libc.so.6       0x00007fbd06718481 __assert_fail + 241
6  clang           0x00000000012ba0f4 llvm::VirtRegMap::addSpillSlotUse(int,
llvm::MachineInstr*) + 180
7  clang           0x00000000012c57e2
8  clang           0x00000000012c6183
9  clang           0x00000000012cd337
10 clang           0x00000000012589f1
11 clang           0x000000000154ce0d
llvm::FPPassManager::runOnFunction(llvm::Function&) + 637
12 clang           0x000000000154cfeb
llvm::FunctionPassManagerImpl::run(llvm::Function&) + 155
13 clang           0x000000000154d2dc
llvm::FunctionPassManager::run(llvm::Function&) + 220
14 clang           0x00000000007afede
clang::EmitBackendOutput(clang::Diagnostic&, clang::CodeGenOptions const&,
clang::TargetOptions const&, llvm::Module*, clang::BackendAction,
llvm::raw_ostream*) + 1806
15 clang           0x00000000007abf7b
16 clang           0x00000000008c289f clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::ASTContext&, bool, bool,
clang::CodeCompleteConsumer*) + 335
17 clang           0x00000000007acef4 clang::CodeGenAction::ExecuteAction() +
68
18 clang           0x00000000006c9855
clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 357
19 clang           0x00000000006a7c3b cc1_main(char const**, char const**, char
const*, void*) + 2651
20 clang           0x00000000006ae407 main + 3543
21 libc.so.6       0x00007fbd0670aabd __libc_start_main + 253
22 clang           0x00000000006a53a9
Stack dump:
0.      Program arguments:
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r109940-install/bin/clang
-cc1 -triple i386-unknown-linux-gnu -S -disable-free -main-file-name small.c
-mrelocation-model static -mconstructor-aliases -target-cpu pentium4
-resource-dir
/uusoc/exports/scratch/regehr/z/compiler-install/llvm-gcc-r109940-install/lib/clang/2.8
-O2 -w -ferror-limit 19 -fmessage-length 95 -fgnu-runtime
-fdiagnostics-show-option -fcolor-diagnostics -o /tmp/cc-kvKvCO.s -x c small.c 
1.      <eof> parser at end of file
2.      Code generation
3.      Running pass 'Linear Scan Register Allocator' on function '@int641'
clang: error: clang frontend command failed due to signal 6 (use -v to see
invocation)

[regehr at bethe tmp606]$ clang -v

clang version 2.8 (trunk 109940)
Target: x86_64-unknown-linux-gnu
Thread model: posix

[regehr at bethe tmp606]$ cat small.c

static signed char
foo (signed char si1, unsigned char si2)
{
  return si1 + si2;
}

static signed char
bar (signed char left, int right)
{
  return left || right || right || left >> right ? left : left << right;
}

static short
baz (short si1, short si2)
{
  return si1 + si2;
}

static unsigned char
bux (unsigned char ui1, unsigned char ui2)
{
  return ui1 - ui2;
}

static unsigned short
buzz (unsigned short ui1, unsigned char ui2)
{
  return ui1 - ui2;
}

static unsigned long long
biz (unsigned long long ui1, unsigned char ui2)
{
  return ui2 ? ui1 : (ui1 / ui2);
}

short g_26[2] = {
  0x12A7B09FL
};

unsigned char *g_25 = &g_26[0];
int **volatile g_29 = &g_25;
signed char g_103;
const short *g_106 = &g_103;
const unsigned char **g = &g_106;
unsigned char g_213;
short *volatile g_230 = &g_213;

void
int641 (void)
{
  short **l_239 = &g_25;
  int *l_245 = &g_26[1];
  if (biz
      (**l_239
       && buzz (*l_245, 0),
       g_103 <
       (foo
        (baz (*l_245, **g_29),
         g_26[0]) <= bar (bux (**l_239, *g_106), **l_239))))
    for (0; 1; g += 1)
      *g_230 ^= safe (*l_245, *g_106);
}

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:31:47 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:31:47 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7216] clang++ silently accepts flags it doesn't
	understand
In-Reply-To: <bug-7216-206@http.llvm.org/bugs/>
References: <bug-7216-206@http.llvm.org/bugs/>
Message-ID: <20100731203147.83C742A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7216

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |LATER

--- Comment #1 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:31:47 CDT ---
Yes, this is because we are still a long way from accepting all the flags
people use with GCC. For now, rejecting or warning about them just gives more
noise than it has been deemed to be worth.

At some point when we understand enough GCC flags, we can consider changing
that.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:42:45 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:42:45 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7039] -print-multi-os-directory doesn't give
	expected results.
In-Reply-To: <bug-7039-206@http.llvm.org/bugs/>
References: <bug-7039-206@http.llvm.org/bugs/>
Message-ID: <20100731204245.80B6F2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=7039

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |daniel at zuster.org
         Resolution|                            |LATER

--- Comment #2 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:42:45 CDT ---
It does the right thing for me, on Darwin. In general, supporting this option
in Clang doesn't really make any sense. This isn't going to get fixed until we
add real cross compilation support.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:44:10 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:44:10 -0500 (CDT)
Subject: [LLVMbugs] [Bug 5677] Bring index-test into Clang
In-Reply-To: <bug-5677-206@http.llvm.org/bugs/>
References: <bug-5677-206@http.llvm.org/bugs/>
Message-ID: <20100731204410.716942A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=5677

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:44:10 CDT ---
This is done.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:44:41 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:44:41 -0500 (CDT)
Subject: [LLVMbugs] [Bug 3608] clang driver should support .bc and .ll input
In-Reply-To: <bug-3608-206@http.llvm.org/bugs/>
References: <bug-3608-206@http.llvm.org/bugs/>
Message-ID: <20100731204441.09C512A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=3608

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:44:40 CDT ---
Hey,  I did this too!

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:46:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:46:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4168] Support COMPILER_PATH and LIBRARY_PATH
In-Reply-To: <bug-4168-206@http.llvm.org/bugs/>
References: <bug-4168-206@http.llvm.org/bugs/>
Message-ID: <20100731204622.A3F002A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4168

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #1 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:46:22 CDT ---
No one has asked for this, and environment variables controlling options are
ugly. Closing this until someone appears to want it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:46:24 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:46:24 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4167] support DEPENDENCIES_OUTPUT
In-Reply-To: <bug-4167-206@http.llvm.org/bugs/>
References: <bug-4167-206@http.llvm.org/bugs/>
Message-ID: <20100731204624.BA7522A6C12D@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4167

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #1 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:46:24 CDT ---
No one has asked for this, and environment variables controlling options are
ugly. Closing this until someone appears to want it.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:49:05 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:49:05 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4249] support generating intermediate files
 (suitable for bug reporting) on crashes
In-Reply-To: <bug-4249-206@http.llvm.org/bugs/>
References: <bug-4249-206@http.llvm.org/bugs/>
Message-ID: <20100731204905.182642A6C124@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4249

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #3 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:49:04 CDT ---
Imagine how much more work I would get done if I didn't file each bug in
triplicate!

*** This bug has been marked as a duplicate of bug 4491 ***

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:50:40 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:50:40 -0500 (CDT)
Subject: [LLVMbugs] [Bug 4990] 'unavailable' attribute documented as error
 but is only warning
In-Reply-To: <bug-4990-206@http.llvm.org/bugs/>
References: <bug-4990-206@http.llvm.org/bugs/>
Message-ID: <20100731205040.67D5A2A6C12E@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=4990

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |daniel at zuster.org
         Resolution|                            |FIXED

--- Comment #2 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:50:40 CDT ---
Ted fixed this.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 15:54:44 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 15:54:44 -0500 (CDT)
Subject: [LLVMbugs] [Bug 6742] Clang cannot execute on NetBSD
In-Reply-To: <bug-6742-206@http.llvm.org/bugs/>
References: <bug-6742-206@http.llvm.org/bugs/>
Message-ID: <20100731205444.47D792A6C130@llvm.org>

http://llvm.org/bugs/show_bug.cgi?id=6742

Daniel Dunbar <daniel at zuster.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |daniel at zuster.org
         Resolution|                            |WORKSFORME

--- Comment #2 from Daniel Dunbar <daniel at zuster.org> 2010-07-31 15:54:43 CDT ---
This might be fixed, if not we need more info.

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From bugzilla-daemon at llvm.org  Sat Jul 31 21:18:22 2010
From: bugzilla-daemon at llvm.org (bugzilla-daemon at llvm.org)
Date: Sat, 31 Jul 2010 21:18:22 -0500 (CDT)
Subject: [LLVMbugs] [Bug 7772] New: analyzer reports leaks when data is
 added to hash table or linked list
Message-ID: <bug-7772-206@http.llvm.org/bugs/>

http://llvm.org/bugs/show_bug.cgi?id=7772

           Summary: analyzer reports leaks when data is added to hash
                    table or linked list
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Static Analyzer
        AssignedTo: kremenek at apple.com
        ReportedBy: mike at zentific.com
                CC: llvmbugs at cs.uiuc.edu


There should probably be a suppression setting file or way to alias functions
to free, ie list_append() = free() for the scope of a function or something
similar.

You can see more explicitly what I'm talking about here, as a leak is reported
any time allocated data is added to a list or hash table:
http://people.profusion.mobi/~lfelipe/static_analysis/efl/

-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

From lkiss at deltaprime.com  Thu Jul 29 21:10:47 2010
From: lkiss at deltaprime.com (Laszlo Kiss)
Date: Fri, 30 Jul 2010 02:10:47 -0000
Subject: [LLVMbugs] LLVM 2.7/OpenSolaris link failure,
	link commands missing key libraries
Message-ID: <D50B40F9-B751-4785-8BDB-D0C105ECC3D9@deltaprime.com>

The build process fails to link the LLVM tools. Specifically, the  
first tool to build is 'opt' which results in 700+ undefined symbols.

I looked into the Makefiles and found that the "LINK_COMPONENTS" make  
variable does not get the correct set of libraries. In fact, at the  
top level Makefile.rules, the variable is set to "support system",  
however, these do not seem to make it to the 'opt' make context.

The list of libraries for 'opt' is (from the g++ link command):

... -lLLVMipo -lLLVMScalarOpts -lLLVMInstrumentation -lLLVMAsmParser - 
lLLVMBitWriter -lLLVMBitReader ...

It is missing at a minimum LLVMCore, which when added reduces the  
unresolved externals to 400+.

I've enclosed a lot of info, I hope it is what you need.



[DETAILS]

The build itself is uneventful with the following warnings:


llvm[3]: Compiling AsmPrinter.cpp for Release build
gmake[3]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
CodeGen/AsmPrinter'
llvm[3]: Compiling AsmPrinter.cpp for Release build
/home/lkiss/Code/work/LLVM/llvm-2.7/include/llvm/ADT/Twine.h: In  
member function 'llvm::MCSymbol*  
llvm::AsmPrinter::GetBlockAddressSymbol(const llvm::Function*, const  
llvm::BasicBlock*) const':
/home/lkiss/Code/work/LLVM/llvm-2.7/include/llvm/ADT/Twine.h:215:  
warning: dereferencing pointer '<anonymous>' does break strict- 
aliasing rules
/home/lkiss/Code/work/LLVM/llvm-2.7/include/llvm/ADT/Twine.h:212:  
warning: dereferencing pointer '<anonymous>' does break strict- 
aliasing rules
/home/lkiss/Code/work/LLVM/llvm-2.7/include/llvm/ADT/Twine.h:204:  
note: initialized from here

llvm[3]: Compiling X86FloatingPoint.cpp for Release build
X86FloatingPoint.cpp: In member function  
'void 
< 
unnamed 
 >::FPS::freeStackSlotAfter(llvm::ilist_iterator<llvm::MachineInstr>&,  
unsigned int)':
X86FloatingPoint.cpp:95: warning: array subscript is above array bounds


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/X86/AsmPrinter'
llvm[4]: Compiling X86ATTInstPrinter.cpp for Release build
In file included from X86ATTInstPrinter.cpp:29:
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/X86/AsmPrinter/../ 
X86GenAsmWriter.inc: In member function 'void  
llvm::X86ATTInstPrinter::printInstruction(const llvm::MCInst*)':
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/X86/AsmPrinter/../ 
X86GenAsmWriter.inc:11: warning: offset outside bounds of constant  
string
llvm[4]: Compiling X86AsmPrinter.cpp for Release build
llvm[4]: Compiling X86IntelInstPrinter.cpp for Release build
In file included from X86IntelInstPrinter.cpp:28:
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/X86/AsmPrinter/../ 
X86GenAsmWriter1.inc: In member function 'void  
llvm::X86IntelInstPrinter::printInstruction(const llvm::MCInst*)':
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/X86/AsmPrinter/../ 
X86GenAsmWriter1.inc:11: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/Sparc/AsmPrinter'
llvm[4]: Compiling SparcAsmPrinter.cpp for Release build
SparcAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::SparcAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
SparcAsmPrinter.cpp:46: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/Alpha/AsmPrinter'
llvm[4]: Compiling AlphaAsmPrinter.cpp for Release build
AlphaAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::AlphaAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
AlphaAsmPrinter.cpp:48: warning: offset outside bounds of constant  
string


llvm[4]: Compiling ARMInstPrinter.cpp for Release build
In file included from ARMInstPrinter.cpp:27:
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/ARM/AsmPrinter/../ 
ARMGenAsmWriter.inc: In member function 'void  
llvm::ARMInstPrinter::printInstruction(const llvm::MCInst*)':
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/ARM/AsmPrinter/../ 
ARMGenAsmWriter.inc:11: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/Mips/AsmPrinter'
llvm[4]: Compiling MipsAsmPrinter.cpp for Release build
MipsAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::MipsAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
MipsAsmPrinter.cpp:78: warning: offset outside bounds of constant string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/CellSPU/AsmPrinter'
llvm[4]: Compiling SPUAsmPrinter.cpp for Release build
SPUAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::SPUAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
SPUAsmPrinter.cpp:59: warning: offset outside bounds of constant string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/PIC16/AsmPrinter'
llvm[4]: Compiling PIC16AsmPrinter.cpp for Release build
In file included from PIC16AsmPrinter.cpp:35:
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/PIC16/AsmPrinter/../ 
PIC16GenAsmWriter.inc: In member function 'void  
llvm::PIC16AsmPrinter::printInstruction(const llvm::MachineInstr*)':
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/PIC16/AsmPrinter/../ 
PIC16GenAsmWriter.inc:11: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/XCore/AsmPrinter'
llvm[4]: Compiling XCoreAsmPrinter.cpp for Release build
XCoreAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::XCoreAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
XCoreAsmPrinter.cpp:326: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/MSP430/AsmPrinter'
llvm[4]: Compiling MSP430AsmPrinter.cpp for Release build
llvm[4]: Compiling MSP430InstPrinter.cpp for Release build
In file included from MSP430InstPrinter.cpp:28:
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/MSP430/AsmPrinter/../ 
MSP430GenAsmWriter.inc: In member function 'void  
llvm::MSP430InstPrinter::printInstruction(const llvm::MCInst*)':
/home/lkiss/Code/work/LLVM/llvm-2.7/lib/Target/MSP430/AsmPrinter/../ 
MSP430GenAsmWriter.inc:11: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/SystemZ/AsmPrinter'
llvm[4]: Compiling SystemZAsmPrinter.cpp for Release build
SystemZAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::SystemZAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
SystemZAsmPrinter.cpp:79: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/Blackfin/AsmPrinter'
llvm[4]: Compiling BlackfinAsmPrinter.cpp for Release build
BlackfinAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::BlackfinAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
BlackfinAsmPrinter.cpp:55: warning: offset outside bounds of constant  
string


gmake[4]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/lib/ 
Target/MBlaze/AsmPrinter'
llvm[4]: Compiling MBlazeAsmPrinter.cpp for Release build
MBlazeAsmPrinter.cpp: In member function 'virtual  
void<unnamed>::MBlazeAsmPrinter::EmitInstruction(const  
llvm::MachineInstr*)':
MBlazeAsmPrinter.cpp:81: warning: offset outside bounds of constant  
string




And here is the link command that fails:

gmake[2]: Entering directory `/home/lkiss/Code/work/LLVM/llvm-2.7/ 
tools/opt'
echo llvm[2]: Linking Release executable opt "(without symbols)"
g++ -I/home/lkiss/Code/work/LLVM/llvm-2.7/include -I/home/lkiss/Code/ 
work/LLVM/llvm-2.7/tools/opt  -D_DEBUG -include llvm/System/Solaris.h - 
D_GNU_SOURCE -D__STDC_LIMIT_MACROS -D__STDC_CONSTANT_MACROS -O2 -fomit- 
frame-pointer -fno-exceptions -fno-rtti -fPIC -Woverloaded-virtual      
-pedantic -Wno-long-long -Wall -W -Wno-unused-parameter -Wwrite- 
strings  -O2 -Wl,-R -Wl,'$ORIGIN/../lib' -Wl,-R -Wl,/home/lkiss/Code/ 
work/LLVM/llvm-2.7/Release/bin  -L/home/lkiss/Code/work/LLVM/llvm-2.7/ 
Release/lib -L/home/lkiss/Code/work/LLVM/llvm-2.7/Release/lib  -o / 
home/lkiss/Code/work/LLVM/llvm-2.7/Release/bin/opt  /home/lkiss/Code/ 
work/LLVM/llvm-2.7/tools/opt/Release/AnalysisWrappers.o /home/lkiss/ 
Code/work/LLVM/llvm-2.7/tools/opt/Release/GraphPrinters.o /home/lkiss/ 
Code/work/LLVM/llvm-2.7/tools/opt/Release/PrintSCC.o /home/lkiss/Code/ 
work/LLVM/llvm-2.7/tools/opt/Release/opt.o  \
         -lLLVMipo -lLLVMScalarOpts -lLLVMInstrumentation - 
lLLVMAsmParser -lLLVMBitWriter -lLLVMBitReader   -lpthread -lmalloc -lm
echo llvm[2]: ======= Finished Linking Release Executable opt \
           "(without symbols)"
gmake[2]: Leaving directory `/home/lkiss/Code/work/LLVM/llvm-2.7/tools/ 
opt'



This is the output of the above link command:

llvm[2]: Linking Release executable opt (without symbols)
Undefined                       first referenced
  symbol                             in file
llvm::ConstantUnion::get(llvm::UnionType const*, llvm::Constant*) / 
home/lkiss/Code/work/LLVM/llvm-2.7/Release/lib/ 
libLLVMAsmParser.a(LLParser.o)
llvm::Function::hasGC() const       /home/lkiss/Code/work/LLVM/ 
llvm-2.7/Release/lib/libLLVMipo.a(MergeFunctions.o)
llvm::CallInst::init(llvm::Value*, llvm::Value* const*, unsigned int) / 
home/lkiss/Code/work/LLVM/llvm-2.7/Release/lib/ 
libLLVMipo.a(ArgumentPromotion.o)

... 700+ more ...

ld: fatal: symbol referencing errors. No output written to /home/lkiss/ 
Code/work/LLVM/llvm-2.7/Release/bin/opt
collect2: ld returned 1 exit status
gmake[2]: *** [/home/lkiss/Code/work/LLVM/llvm-2.7/Release/bin/opt]  
Error 1
gmake[2]: Leaving directory `/home/lkiss/Code/work/LLVM/llvm-2.7/tools/ 
opt'
gmake[1]: *** [opt/.makeall] Error 2
gmake[1]: Leaving directory `/home/lkiss/Code/work/LLVM/llvm-2.7/tools'
gmake: *** [all] Error 1




[BACKGROUND]

$ uname -a
SunOS A64009 5.11 snv_111b i86pc i386 i86pc

	GCC-4.4.4, GMP-4.3.2, MPFR-2.4.2
$ gcc -v
Using built-in specs.
Target: i386-pc-solaris2.11
Configured with: ../gcc-4.4.4/configure --prefix=/opt/common/app/gcc/ 
current --with-mpfr=/opt/common/app/gcc/current --with-gmp=/opt/common/ 
app/gcc/current --enable-languages=c,c++ --enable-shared --enable- 
threads
Thread model: posix
gcc version 4.4.4 (GCC)


lkiss at A64009$ configure --prefix=/opt/common/app/gcc/current
checking build system type... i386-pc-solaris2.11
checking host system type... i386-pc-solaris2.11
checking target system type... i386-pc-solaris2.11
checking type of operating system we're going to host on... SunOS
checking type of operating system we're going to target... SunOS
checking target architecture... x86
checking for gcc... gcc
checking for C compiler default output file name... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for grep that handles long lines and -e... /usr/bin/ggrep
checking for egrep... /usr/bin/ggrep -E
checking for ANSI C header files... yes
checking for sys/types.h... yes
checking for sys/stat.h... yes
checking for stdlib.h... yes
checking for string.h... yes
checking for memory.h... yes
checking for strings.h... yes
checking for inttypes.h... yes
checking for stdint.h... yes
checking for unistd.h... yes
checking whether byte ordering is bigendian... no
checking clang compiler... built
checking optimization flags... -O2
checking how to run the C preprocessor... gcc -E
checking whether we are using the GNU C compiler... (cached) yes
checking whether gcc accepts -g... (cached) yes
checking for gcc option to accept ISO C89... (cached) none needed
checking for g++... g++
checking whether we are using the GNU C++ compiler... yes
checking whether g++ accepts -g... yes
checking for BSD-compatible nm... /usr/bin/nm -p
checking for GNU make... gmake
checking whether ln -s works... yes
checking for cmp... /usr/bin/cmp
checking for cp... /usr/bin/cp
checking for date... /usr/bin/date
checking for find... /usr/bin/find
checking for grep... (cached) /usr/bin/ggrep
checking for mkdir... /usr/bin/mkdir
checking for mv... /usr/bin/mv
checking for ranlib... ranlib
checking for ar... ar
checking for rm... /usr/bin/rm
checking for sed... /usr/bin/sed
checking for tar... /usr/bin/tar
checking for pwd... /usr/bin/pwd
checking for Graphviz... echo Graphviz
checking for dot... echo dot
checking for fdp... echo fdp
checking for neato... echo neato
checking for twopi... echo twopi
checking for circo... echo circo
checking for gv... no
checking for gsview32... no
checking for dotty... echo dotty
checking for perl... /usr/bin/perl
checking for Perl 5.006 or newer... yes
checking for a BSD-compatible install... /home/lkiss/bin/install -c
checking for bzip2... /usr/bin/bzip2
checking for doxygen... no
checking for groff... no
checking for gzip... /usr/bin/gzip
checking for pod2html... no
checking for pod2man... no
checking for runtest... no
checking for the tclsh program in tclinclude directory... none
checking for tclsh8.4... /usr/bin/tclsh8.4
checking for zip... /usr/bin/zip
checking for ocamlc... no
checking for ocamlopt... no
checking for ocamldep... no
checking for ocamldoc... no
checking for gas... /usr/bin/gas
checking for compiler -Wl,-R<path> option... yes
checking for compiler -Wl,-export-dynamic option... no
checking for an ANSI C-conforming const... yes
checking for dirent.h that defines DIR... yes
checking for library containing opendir... none required
checking dlfcn.h usability... yes
checking dlfcn.h presence... yes
checking for dlfcn.h... yes
checking dynamic linker characteristics... solaris2.11 ld.so
checking which extension is used for loadable modules... .so
checking which variable specifies run-time library path...  
LD_LIBRARY_PATH
checking for the default library search path... /lib /usr/lib
checking for objdir... .libs
checking command to parse /usr/bin/nm -p output from  object... ok
checking whether libtool supports -dlopen/-dlpreopen... yes
checking for shl_load... no
checking for shl_load in -ldld... no
checking for dlopen in -ldl... yes
checking for dlerror... yes
checking for _ prefix in compiled symbols... no
checking whether deplibs are loaded by dlopen... yes
checking argz.h usability... no
checking argz.h presence... no
checking for argz.h... no
checking for error_t... no
checking for argz_append... no
checking for argz_create_sep... no
checking for argz_insert... no
checking for argz_next... no
checking for argz_stringify... no
checking assert.h usability... yes
checking assert.h presence... yes
checking for assert.h... yes
checking ctype.h usability... yes
checking ctype.h presence... yes
checking for ctype.h... yes
checking errno.h usability... yes
checking errno.h presence... yes
checking for errno.h... yes
checking malloc.h usability... yes
checking malloc.h presence... yes
checking for malloc.h... yes
checking for memory.h... (cached) yes
checking for stdlib.h... (cached) yes
checking stdio.h usability... yes
checking stdio.h presence... yes
checking for stdio.h... yes
checking for unistd.h... (cached) yes
checking dl.h usability... no
checking dl.h presence... no
checking for dl.h... no
checking sys/dl.h usability... yes
checking sys/dl.h presence... yes
checking for sys/dl.h... yes
checking dld.h usability... no
checking dld.h presence... no
checking for dld.h... no
checking mach-o/dyld.h usability... no
checking mach-o/dyld.h presence... no
checking for mach-o/dyld.h... no
checking for string.h... (cached) yes
checking for strchr... yes
checking for strrchr... yes
checking for memcpy... yes
checking for memmove... yes
checking for strcmp... yes
checking for closedir... yes
checking for opendir... yes
checking for readdir... yes
checking for llvm-gcc... no
checking for llvm-g++... no
checking LLVM capable compiler... clang
checking tool compatibility... ok
checking optional compiler flags... -Wno-variadic-macros -Wno-missing- 
field-initializers
checking for sin in -lm... yes
checking for library containing dlopen... none required
checking for library containing mallinfo... -lmalloc
checking for pthread_mutex_init in -lpthread... yes
checking for library containing pthread_mutex_lock... none required
checking for library containing pthread_rwlock_init... none required
checking for library containing pthread_getspecific... none required
checking for dirent.h that defines DIR... (cached) yes
checking for library containing opendir... (cached) none required
checking for MAP_ANONYMOUS vs. MAP_ANON... yes
checking whether stat file-mode macros are broken... no
checking for ANSI C header files... (cached) yes
checking for sys/wait.h that is POSIX.1 compatible... yes
checking whether time.h and sys/time.h may both be included... yes
checking for dlfcn.h... (cached) yes
checking execinfo.h usability... yes
checking execinfo.h presence... yes
checking for execinfo.h... yes
checking fcntl.h usability... yes
checking fcntl.h presence... yes
checking for fcntl.h... yes
checking for inttypes.h... (cached) yes
checking limits.h usability... yes
checking limits.h presence... yes
checking for limits.h... yes
checking link.h usability... yes
checking link.h presence... yes
checking for link.h... yes
checking for malloc.h... (cached) yes
checking setjmp.h usability... yes
checking setjmp.h presence... yes
checking for setjmp.h... yes
checking signal.h usability... yes
checking signal.h presence... yes
checking for signal.h... yes
checking for stdint.h... (cached) yes
checking termios.h usability... yes
checking termios.h presence... yes
checking for termios.h... yes
checking for unistd.h... (cached) yes
checking utime.h usability... yes
checking utime.h presence... yes
checking for utime.h... yes
checking windows.h usability... no
checking windows.h presence... no
checking for windows.h... no
checking sys/mman.h usability... yes
checking sys/mman.h presence... yes
checking for sys/mman.h... yes
checking sys/param.h usability... yes
checking sys/param.h presence... yes
checking for sys/param.h... yes
checking sys/resource.h usability... yes
checking sys/resource.h presence... yes
checking for sys/resource.h... yes
checking sys/time.h usability... yes
checking sys/time.h presence... yes
checking for sys/time.h... yes
checking for sys/types.h... (cached) yes
checking sys/ioctl.h usability... yes
checking sys/ioctl.h presence... yes
checking for sys/ioctl.h... yes
checking malloc/malloc.h usability... no
checking malloc/malloc.h presence... no
checking for malloc/malloc.h... no
checking mach/mach.h usability... no
checking mach/mach.h presence... no
checking for mach/mach.h... no
checking pthread.h usability... yes
checking pthread.h presence... yes
checking for pthread.h... yes
checking for HUGE_VAL sanity... yes
checking for pid_t... yes
checking for size_t... yes
checking whether struct tm is in sys/time.h or time.h... time.h
checking for int64_t... yes
checking for uint64_t... yes
checking for backtrace... yes
checking for ceilf... yes
checking for floorf... yes
checking for roundf... yes
checking for rintf... yes
checking for nearbyintf... yes
checking for getcwd... yes
checking for powf... yes
checking for fmodf... yes
checking for strtof... yes
checking for round... yes
checking for getpagesize... yes
checking for getrusage... yes
checking for getrlimit... yes
checking for setrlimit... yes
checking for gettimeofday... yes
checking for isatty... yes
checking for mkdtemp... yes
checking for mkstemp... yes
checking for mktemp... yes
checking for realpath... yes
checking for sbrk... yes
checking for setrlimit... (cached) yes
checking for strdup... yes
checking for strerror... yes
checking for strerror_r... yes
checking for strerror_s... no
checking for setenv... yes
checking for strtoll... yes
checking for strtoq... no
checking for sysconf... yes
checking for malloc_zone_statistics... no
checking for setjmp... yes
checking for longjmp... yes
checking for sigsetjmp... yes
checking for siglongjmp... yes
checking if printf has the %a format character...
checking for srand48/lrand48/drand48 in <stdlib.h>... yes
checking for isnan in <math.h>... yes
checking for isnan in <cmath>... yes
checking for std::isnan in <cmath>... no
checking for isinf in <math.h>... no
checking for isinf in <cmath>... no
checking for std::isinf in <cmath>... no
checking for finite in <ieeefp.h>... yes
checking for stdlib.h... (cached) yes
checking for unistd.h... (cached) yes
checking for getpagesize... (cached) yes
checking for working mmap... yes
checking for mmap of files... yes
checking if /dev/zero is needed for mmap... no
checking for GCC atomic builtins... no
configure: WARNING: LLVM will be built thread-unsafe because atomic  
builtins are missing
checking for __dso_handle... yes
checking whether llvm-gcc is sane... no
checking for compiler -fvisibility-inlines-hidden option... yes
configure: creating ./config.status
config.status: creating include/llvm/Config/Targets.def
config.status: creating include/llvm/Config/AsmPrinters.def
config.status: creating include/llvm/Config/AsmParsers.def
config.status: creating include/llvm/Config/Disassemblers.def
config.status: creating Makefile.config
config.status: creating llvm.spec
config.status: creating docs/doxygen.cfg
config.status: creating tools/llvmc/plugins/Base/Base.td
config.status: creating tools/llvm-config/llvm-config.in
config.status: creating include/llvm/Config/config.h
config.status: creating include/llvm/System/DataTypes.h
config.status: include/llvm/System/DataTypes.h is unchanged
config.status: executing setup commands
config.status: executing Makefile commands
config.status: executing Makefile.common commands
config.status: executing examples/Makefile commands
config.status: executing lib/Makefile commands
config.status: executing runtime/Makefile commands
config.status: executing test/Makefile commands
config.status: executing test/Makefile.tests commands
config.status: executing unittests/Makefile commands
config.status: executing tools/Makefile commands
config.status: executing utils/Makefile commands
config.status: executing projects/Makefile commands
config.status: executing bindings/Makefile commands
config.status: executing bindings/ocaml/Makefile.ocaml commands
=== configuring in projects/sample (/home/lkiss/Code/work/LLVM/ 
llvm-2.7/projects/sample)
configure: running /bin/sh ./configure --prefix=/opt/common/app/gcc/ 
current  --cache-file=/dev/null --srcdir=.
configure: creating ./config.status
config.status: creating Makefile.common
config.status: executing setup commands
config.status: executing Makefile commands
config.status: executing lib/Makefile commands
config.status: executing lib/sample/Makefile commands
config.status: executing tools/Makefile commands
config.status: executing tools/sample/Makefile commands