http://llvm.org/bugs/show_bug.cgi?id=3534

           Summary: Add paq8p to llvm testsuite
           Product: Test Suite
           Version: trunk
          Platform: PC
               URL: http://www.cs.fit.edu/~mmahoney/compression/paq8p.zip
        OS/Version: Linux
            Status: NEW
          Severity: enhancement
          Priority: P2
         Component: Programs Tests
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2518)
 --> (http://llvm.org/bugs/attachment.cgi?id=2518)
testfile

This file shows interesting runtimes with different optimizations on  a 256k
file generated from /dev/urandom.

(Built with g++ -march=native -DNOASM)

On x86-64:

g++-4.3 -O3: 55.75 s
g++-4.3 -O2: 39.25 s
g++-4.3 -Os: 41.14 s

llvm-g++ -O3: 53.23 s
llvm-g++ -O2: 53.37 s
llvm-g++ -Os: 55.86 s

So llvm's -O3 is slightly better than gcc's, however it can't achieve gcc's
-O2/-Os. I think this may have something to do with inlining.

Example run:
llvm-g++ -O2 -DNOASM paq8p.cpp;
./a.out file

It prints execution time.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 08:00:18 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 08:00:18 -0600
Subject: [LLVMbugs] [Bug 3535] New: const char * assigned to char * in
	Unix/Signals.inc
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3535

           Summary: const char * assigned to char * in Unix/Signals.inc
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: System Library
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: salimma at fedoraproject.org
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2519)
 --> (http://llvm.org/bugs/attachment.cgi?id=2519)
Patch for Unix/Signals.inc

strrchr returns a const char *, but it is assigned to a char * variable twice
in the file. This breaks compiling, at least on gcc-4.4.0 prereleases.

Switching to const char * (instead of coercing to char *) seems to work; the
char array is never modified.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 11:50:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 11:50:08 -0600
Subject: [LLVMbugs] [Bug 3536] New: Assertion failed:
 (BlockScopes.count(Blocks[i]) && " Did not find block in scope map!"),
 function PopCleanupBlock, file CodeGenFunction.cpp, line 651.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3536

           Summary: Assertion failed: (BlockScopes.count(Blocks[i]) && "Did
                    not find block in scope map!"), function
                    PopCleanupBlock, file CodeGenFunction.cpp, line 651.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

I am getting this when compiling the attached

lev /tmp# clang -emit-llvm vm_pageout.c
Assertion failed: (BlockScopes.count(Blocks[i]) && "Did not find block in scope
map!"), function PopCleanupBlock, file CodeGenFunction.cpp, line 651.
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 cs.uiuc.edu  Tue Feb 10 11:52:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 11:52:19 -0600
Subject: [LLVMbugs] [Bug 3537] New: Assert when running the attached file
	through llc
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3537

           Summary: Assert when running the attached file through llc
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: andersca at mac.com
                CC: llvmbugs at cs.uiuc.edu

Starting program: /Users/andersca/llvm/Debug/bin/llc
bugpoint-reduced-simplified.bc
Reading symbols for shared libraries +++. done
Assertion failed: (InReg && "Value not in map!"), function getValue, file
SelectionDAGBuild.cpp, line 952.

Program received signal SIGABRT, Aborted.
0x90157e42 in __kill ()
(gdb) bt
#0  0x90157e42 in __kill ()
#1  0x90157e34 in kill$UNIX2003 ()
#2  0x901ca23a in raise ()
#3  0x901d6679 in abort ()
#4  0x901cb3db in __assert_rtn ()
#5  0x005b8282 in llvm::SelectionDAGLowering::getValue (this=0x1705a30,
V=0x170a7bc) at SelectionDAGBuild.cpp:952
#6  0x005c5a0f in llvm::SelectionDAGLowering::visitLoad (this=0x1705a30,
I=@0x17029bc) at SelectionDAGBuild.cpp:2807
#7  0x005b7344 in llvm::SelectionDAGLowering::visit (this=0x1705a30, Opcode=25,
I=@0x17029bc) at Instruction.def:131
#8  0x005ce508 in llvm::SelectionDAGLowering::visit (this=0x1705a30,
I=@0x17029bc) at SelectionDAGBuild.cpp:819
#9  0x005e3d01 in llvm::SelectionDAGISel::SelectBasicBlock (this=0x17056d0,
LLVMBB=0x17024e0, Begin={>
= {<> = {}, }, NodePtr = 0x17029bc},
End={> = {<> = {}, }, NodePtr = 0x17029f0}) at SelectionDAGISel.cpp:470
#10 0x005e472f in llvm::SelectionDAGISel::SelectAllBasicBlocks (this=0x17056d0,
Fn=@0x17018e0, MF=@0x170aa00, MMI=0x1709240, DW=0x1705f40, TII=@0x18034f0) at
SelectionDAGISel.cpp:856
#11 0x005e4b06 in llvm::SelectionDAGISel::runOnFunction (this=0x17056d0,
Fn=@0x17018e0) at SelectionDAGISel.cpp:327
#12 0x0085f804 in llvm::FPPassManager::runOnFunction (this=0x1703520,
F=@0x17018e0) at PassManager.cpp:1323
#13 0x0085fa5a in llvm::FunctionPassManagerImpl::run (this=0x1703340,
F=@0x17018e0) at PassManager.cpp:1281
#14 0x0085fbe0 in llvm::FunctionPassManager::run (this=0xbffff780,
F=@0x17018e0) at PassManager.cpp:1226
#15 0x000033c5 in main (argc=2, argv=0xbffff828) at llc.cpp:317

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 11:56:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 11:56:24 -0600
Subject: [LLVMbugs] [Bug 3538] New: Cannot yet select: 0x80255a6e0: ch =
 declare 0x802579090, 0x802578da8, 0x802578208
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3538

           Summary: Cannot yet select: 0x80255a6e0: ch = declare
                    0x802579090, 0x802578da8, 0x802578208
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

getting this when compiling the attached

lev /tmp# ccc -c -g msdosfs_conv.c                                     

....

Cannot yet select: 0x80255a6e0: ch = declare 0x802579090, 0x802578da8,
0x802578208

note that the -g is important, I dont know how to reproduce with plain 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 cs.uiuc.edu  Tue Feb 10 12:14:17 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 12:14:17 -0600
Subject: [LLVMbugs] [Bug 3535] const char * assigned to char * in
	Unix/Signals.inc
In-Reply-To: 
Message-ID: <200902101814.n1AIEH1R015697@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3535

Dan Gohman  changed:

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

--- Comment #2 from Dan Gohman   2009-02-10 12:14:17 ---
It's in newer versions of GLIBC. This is now fixed in trunk.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 12:18:41 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 12:18:41 -0600
Subject: [LLVMbugs] [Bug 3441] mixing ANSI C declaration and K&R function is
 sometimes allowed and sometimes not
In-Reply-To: 
Message-ID: <200902101818.n1AIIf7D015877@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3441

Daniel Dunbar  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |daniel at zuster.org
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #6 from Daniel Dunbar   2009-02-10 12:18:40 ---

*** This bug has been marked as a duplicate of bug 2821 ***

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 14:51:35 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 14:51:35 -0600
Subject: [LLVMbugs] [Bug 3539] New: TargetInstrInfo:: AnalyzeBranch can
 modify the MBB being analyzed
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3539

           Summary: TargetInstrInfo::AnalyzeBranch can modify the MBB being
                    analyzed
           Product: libraries
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: evan.cheng at apple.com
                CC: llvmbugs at cs.uiuc.edu

AnalyzeBranch may delete instructions in the MBB if they are after the last
unconditional branch.

This may invalidate iterators or client's data structure (e.g. liveintervals
maps). The current workaround is to only allow this when the caller explicitly
permits so. The only such client is branch folding. 

The current fix is to not allow AnalyzeBranch to delete instructions. Instead,
branch folding should perform this optimization.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 15:13:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 15:13:50 -0600
Subject: [LLVMbugs] [Bug 3540] New: error: 'struct pppoe_tag' may not be
 used as an array element due to flexible array member
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3540

           Summary: error: 'struct pppoe_tag' may not be used as an array
                    element due to flexible array member
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

getting this when compiling the attached

lev /tmp$ clang -emit-llvm pppoed.c 
In file included from /usr/src/libexec/pppoed/pppoed.c:1:
In file included from /usr/src/libexec/pppoed/pppoed.c:41:
/usr/include/netgraph/ng_pppoe.h:213:22: error: 'struct pppoe_tag' may not be
used as an array element due to flexible array member
 struct pppoe_tag tag[];
                     ^
1 diagnostic 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 cs.uiuc.edu  Tue Feb 10 15:25:41 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 15:25:41 -0600
Subject: [LLVMbugs] [Bug 3528] clang/x86_64 miscompiles
 MultiSource/Benchmarks/SciMark2-C/ scimark2
In-Reply-To: 
Message-ID: <200902102125.n1ALPfwQ022342@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3528

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-10 15:25:41 ---
I added some #includes, plz 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 cs.uiuc.edu  Tue Feb 10 15:36:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 15:36:12 -0600
Subject: [LLVMbugs] [Bug 3541] New: improve builtin function handling
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3541

           Summary: improve builtin function handling
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu, daniel at zuster.org,
                    dgregor at apple.com

The X86-64 ABI work exposed a GCC "feature" that clang isn't implementing. 
Basically, if you use "calloc" without a prototype, GCC knows that it returns a
void* and gives it the right prototype.  clang doesn't know this, and so it
defaults to returning int, which is different according to the ABI.

The proposed fix that we discussed is as follows:

1) add these builtins to the builtins.def file.
2) add a new "give me your builtin ID" method to FunctionDecl.  It would return
true if the name is an identifier, if the identifier has a builtin ID, and if
the decl context is the translation unit.  Stuff the like the printf format
checking stuff can be keyed off this instead of using strcmp of the callee
name.
3) when handling a definition (either implicit definition or redefinition)
check to see if we're redefining a builtin.  If so, synthesize the builtin
definition first before doing decl merging.
4) If decl merging encounters an incompatibility, this should be a hard error,
not warning like GCC does.  We don't want downstream clients to have to worry
about mangled builtin decls.
5) someday long in the future, we can add support for -ffree-standing,
-fno-builtin-calloc, etc.  This would just not prepopulate the identifier info
for calloc with a builtin ID.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 15:43:57 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 15:43:57 -0600
Subject: [LLVMbugs] [Bug 3542] New: Implement __attribute__((overloadable))
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3542

           Summary: Implement __attribute__((overloadable))
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu, dgregor at apple.com

We need this for tgmath.h, Doug volunteered to tackle 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 cs.uiuc.edu  Tue Feb 10 15:50:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 15:50:08 -0600
Subject: [LLVMbugs] [Bug 3540] Unsupported GNU extension: use of structures
 with a flexible array initializer in an array, struct, or union
In-Reply-To: 
Message-ID: <200902102150.n1ALo8KZ023255@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3540

Douglas Gregor  changed:

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

--- Comment #3 from Douglas Gregor   2009-02-10 15:50:07 ---
Fixed here:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012076.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 cs.uiuc.edu  Tue Feb 10 15:58:52 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 15:58:52 -0600
Subject: [LLVMbugs] [Bug 3527] clang/x86_64 has no builtin versions of
	libcalls.
In-Reply-To: 
Message-ID: <200902102158.n1ALwqXF023726@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3527

Daniel Dunbar  changed:

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

--- Comment #5 from Daniel Dunbar   2009-02-10 15:58:51 ---

*** This bug has been marked as a duplicate of bug 3541 ***

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 16:08:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 16:08:33 -0600
Subject: [LLVMbugs] [Bug 3523] clang/x86_64 miscompiles
 MultiSource/Benchmarks/McCat/09-vor/ vor
In-Reply-To: 
Message-ID: <200902102208.n1AM8XCx024102@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3523

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-10 16:08:32 ---

*** This bug has been marked as a duplicate of bug 3541 ***

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 16:12:35 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 16:12:35 -0600
Subject: [LLVMbugs] [Bug 3525] clang/x86_64 miscompiles
 MultiSource/Benchmarks/MiBench/ network-dijkstra/network-dijkstra
In-Reply-To: 
Message-ID: <200902102212.n1AMCZXQ024253@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3525

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-10 16:12:35 ---

*** This bug has been marked as a duplicate of bug 3541 ***

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 16:17:49 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 16:17:49 -0600
Subject: [LLVMbugs] [Bug 3531] clang/x86_64 miscompiles
 SingleSource/Benchmarks/Misc/ himenobmtxpa
In-Reply-To: 
Message-ID: <200902102217.n1AMHnsv024551@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3531

Daniel Dunbar  changed:

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

--- Comment #4 from Daniel Dunbar   2009-02-10 16:17:49 ---

*** This bug has been marked as a duplicate of bug 3541 ***

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 17:04:37 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 17:04:37 -0600
Subject: [LLVMbugs] [Bug 3536] Assertion failed:
 (BlockScopes.count(Blocks[i]) && " Did not find block in scope map!"),
 function PopCleanupBlock, file CodeGenFunction.cpp, line 651.
In-Reply-To: 
Message-ID: <200902102304.n1AN4bF2026302@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3536

Anders Carlsson  changed:

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

--- Comment #3 from Anders Carlsson   2009-02-10 17:04:37 ---
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012089.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 cs.uiuc.edu  Tue Feb 10 19:04:15 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 19:04:15 -0600
Subject: [LLVMbugs] [Bug 3543] New: Certain ({ }) (?) expressions crash
	clang / ccc
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3543

           Summary: Certain ({ }) (?) expressions crash clang / ccc
           Product: new-bugs
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: blocker
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: bkr at jumper.org
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2524)
 --> (http://llvm.org/bugs/attachment.cgi?id=2524)
crashes

I tried to narrow this further and further down for 3 hours now - 
this is as good as it gets.

I think it is somehow related to ({ }) expression but I am not sure.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 10 23:24:52 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 10 Feb 2009 23:24:52 -0600
Subject: [LLVMbugs] [Bug 3544] New: Incorrect warning when compiling
	AIUtilities
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3544

           Summary: Incorrect warning when compiling AIUtilities
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: catfish.man at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Steps to reproduce:
1) Check out the Adium source code from svn://svn.adiumx.com/adium/trunk
2) Attempt to compile with clang using the xcode project in
Frameworks/AIUtilities/

Results:
/Users/david/adium/trunk/Frameworks/AIUtilities
Framework/Source/AIStringAdditions.m:155:31: warning: incompatible pointer
types initializing 'NSString *', expected 'NSMutableString *'
        NSMutableString *newString = [NSMutableString string];
                                     ^~~~~~~~~~~~~~~~~~~~~~~~
1 diagnostic generated.

This is incorrect, because the method +string declared on NSString returns an
id. It appears that clang is looking at the method -string declared in
AIStringAdditions.h, which does return an NSString. Commenting that declaration
out makes the warning go away.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 00:01:43 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 00:01:43 -0600
Subject: [LLVMbugs] [Bug 3545] New: Fatal error when assembling RBSplitView
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3545

           Summary: Fatal error when assembling RBSplitView
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: catfish.man at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2526)
 --> (http://llvm.org/bugs/attachment.cgi?id=2526)
The offending file

Steps to reproduce:
1) llc RBSplitView.bc -o RBSplitView.s; as RBSplitView.s

Results:
RBSplitView.s:1816:FATAL:Symbol Llabel357 already defined.

Versions:
llvm/clang r64284
Apple Inc version cctools-698.1~1, GNU assembler version 1.38

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 02:27:18 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 02:27:18 -0600
Subject: [LLVMbugs] [Bug 3495] LocalSpiller:: RewriteMBB does not propagate
 the information about AvailableSpills between MBBs
In-Reply-To: 
Message-ID: <200902110827.n1B8RIrd011924@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3495

Evan Cheng  changed:

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

--- Comment #1 from Evan Cheng   2009-02-11 02:27:17 ---
Fixed.
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073597.html

This eliminated 9 reloads and turned 6 into copies in the test case. It
eliminated over 2000 copies in 403.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 cs.uiuc.edu  Wed Feb 11 04:02:05 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 04:02:05 -0600
Subject: [LLVMbugs] [Bug 3520] llvm-gcc never calls
	SimplifyLibCalls::doInitialization
In-Reply-To: 
Message-ID: <200902111002.n1BA25P4025151@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3520

Duncan Sands  changed:

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

--- Comment #3 from Duncan Sands   2009-02-11 04:02:01 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073599.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 cs.uiuc.edu  Wed Feb 11 08:29:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 08:29:19 -0600
Subject: [LLVMbugs] [Bug 3546] New: Redefinition of typedef doesn' t print
 previously definition in some cases.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3546

           Summary: Redefinition of typedef doesn't print previously
                    definition in some cases.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: minor
          Priority: P2
         Component: Basic
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: skabet at gmail.com
                CC: llvmbugs at cs.uiuc.edu

$ cat test.c
#include "/usr/include/netinet/in.h"
#include "/usr/include/unistd.h"

$ clang test.c
In file included from test.c:2:
/usr/include/unistd.h:238:20: error: typedef redefinition with different types
('__intptr_t' vs 'long')
typedef __intptr_t intptr_t;
                   ^
1 diagnostic generated.

-------------------------

$ cat test.c
#include "/usr/include/unistd.h"
#include "/usr/include/netinet/in.h"

$ clang test.c
In file included from test.c:2:
In file included from /usr/include/netinet/in.h:24:
/usr/local/Headers/stdint.h:78:23: error: typedef redefinition with different
types ('long' vs '__intptr_t')
typedef long          intptr_t;
                      ^
In file included from test.c:1:
/usr/include/unistd.h:238:20: note: previous definition is here
typedef __intptr_t intptr_t;
                   ^
2 diagnostics generated.

-----------------------

The only difference is the order the files are included. I haven't been able to
reduce the testcase yet.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 08:30:48 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 08:30:48 -0600
Subject: [LLVMbugs] [Bug 3547] New: llvm2.5prerelase: configure requires flex
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3547

           Summary: llvm2.5prerelase: configure requires flex
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

If flex is not found, I get this error on Linux:
configure: error: flex not found but required

AFAIK flex&bison are no longer required since lib/AsmParser was rewritten from
scratch.

If so the configure check should be removed on trunk + release branch.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 09:14:35 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 09:14:35 -0600
Subject: [LLVMbugs] [Bug 3548] New: Bus error with -O4 on memory deallocation
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3548

           Summary: Bus error with -O4 on memory deallocation
           Product: tools
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: j at jawed.name
                CC: llvmbugs at cs.uiuc.edu

Experiencing an easily reproducible Bus Error on memory deallocation with the
PHP project on Darwin with the O4 compiler flag (or with -flto).

1) grab the php source (http://snaps.php.net)
2) tar xjvf php5.3[...].tar.bz2
3) cd php5.3[...]
4) ./configure --enable-debug --enable-cli --disable-all --without-sqlite3
5) make && make install && make test

The symbol in question is zend_llist_destroy. I would attach the bugpoint
report however it is too large and given the issue is easy to reproduce it
seems less practical.

I'll leave it to the community to determine the real severity of this 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 cs.uiuc.edu  Wed Feb 11 09:20:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 09:20:19 -0600
Subject: [LLVMbugs] [Bug 3549] New: extremelly slow -O2 compilation
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3549

           Summary: extremelly slow -O2 compilation
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

when compiling the attached test case I am getting these times:

lev /tmp$ time ccc -c parse.c
1.933u 0.081s 0:02.04 98.5%     22993+3509k 0+19io 0pf+0w

lev /tmp$ time ccc -c -g parse.c
2.647u 0.122s 0:02.85 96.8%     22070+3357k 0+30io 0pf+0w

ie. 2-3 seconds for plain build but

lev /tmp$ time ccc -c -O2 parse.c
467.364u 0.611s 8:07.90 95.9%   23927+3528k 0+20io 0pf+0w

thats over 8 minutes!

I dont know if the problem is in clang or llvm but top shows clang binary
running. this is with r64302, during the compilation the clang process took
~100MB

this is on amd64/freebsd

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 09:30:37 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 09:30:37 -0600
Subject: [LLVMbugs] [Bug 3550] New: problem with tail call to function that
 accesses alloca' d memory
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3550

           Summary: problem with tail call to function that accesses
                    alloca'd memory
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Interprocedural Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jay.foad at antixlabs.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2528)
 --> (http://llvm.org/bugs/attachment.cgi?id=2528)
test case

As I understand it, if you have a tail call from function F to function G, G
isn't allowed to access any memory alloca'd in F.

With the attached example, if I do:

llvm-as -o - bug.ll | opt -inline -scalarrepl -tailcallelim | llvm-dis

and then search for "tail.*memcpy", I find a tail call to memcpy which is
definitely reading and writing memory alloca'd in the calling function:

    %vint1 = alloca %"struct.stlpmtx_std::vector >"
    %tmp13.i = getelementptr
%"struct.stlpmtx_std::vector >"* %vint1, i32 0, i32 0
    %tmp14.i = getelementptr
%"struct.stlp_priv::_Vector_base >"* %tmp13.i, i32 0,
i32 2
    %tmp1.i.i.i.i291 = getelementptr
%"struct.stlp_priv::_STLP_alloc_proxy >"*
%tmp14.i, i32 0, i32 0
    %tmp1.i.i.i.i.i.i = getelementptr %"struct.StackAllocator"*
%tmp1.i.i.i.i291, i32 0, i32 0
    %tmp3.i.i.i.i.i.i.i = bitcast %struct.State* %tmp1.i.i.i.i.i.i to i8*
    tail call void @llvm.memcpy.i32(i8* %tmp3.i.i.i.i.i.i.i, i8*
%tmp4.i.i.i.i.i.i.i, i32 32, i32 4) nounwind

(I've only shown the instructions relevant to the memory that's being *written*
by the memcpy, but the memory being *read* also comes from a similar alloca.)

I'm using a recent build of LLVM on Linux/x86, from rev 64210 of head of svn.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 10:41:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 10:41:24 -0600
Subject: [LLVMbugs] [Bug 3551] New: llc asserts on given bitcode
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3551

           Summary: llc asserts on given bitcode
           Product: libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: criswell at uiuc.edu
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2529)
 --> (http://llvm.org/bugs/attachment.cgi?id=2529)
LLVM Assembly of Bitcode file

The llc tool asserts on the attached bitcode file.  The stack trace is:

Cannot yet select: 0x1810398: ch = declare 0x170eae4, 0x181062c, 0x1810524

#0  0x929d2b9e in __kill ()
#1  0x929d2b91 in kill$UNIX2003 ()
#2  0x92a49ec2 in raise ()
#3  0x92a5947f in abort ()
#4  0x00386f0f in (anonymous namespace)::X86DAGToDAGISel::CannotYetSelect
(this=0x170e9a0, N={Node = 0x1810398, ResNo = 0}) at X86GenDAGISel.inc:52907
#5  0x0039b6f1 in (anonymous namespace)::X86DAGToDAGISel::Select_DECLARE
(this=0x170e9a0, N=@0xbfffedf4) at X86GenDAGISel.inc:51187
#6  0x004b32ff in (anonymous namespace)::X86DAGToDAGISel::SelectCode
(this=0x170e9a0, N={Node = 0x1810398, ResNo = 0}) at X86GenDAGISel.inc:51249
#7  0x0031256c in (anonymous namespace)::X86DAGToDAGISel::Select
(this=0x170e9a0, N={Node = 0x1810398, ResNo = 0}) at
/Users/criswell/src/llvm22/lib/Target/X86/X86ISelDAGToDAG.cpp:1588
#8  0x004b7557 in (anonymous namespace)::X86DAGToDAGISel::SelectRoot
(this=0x170e9a0, DAG=@0x170ead0) at DAGISelHeader.h:119
#9  0x00315105 in (anonymous namespace)::X86DAGToDAGISel::InstructionSelect
(this=0x170e9a0) at
/Users/criswell/src/llvm22/lib/Target/X86/X86ISelDAGToDAG.cpp:709
#10 0x005e1e44 in llvm::SelectionDAGISel::CodeGenAndEmitDAG (this=0x170e9a0) at
/Users/criswell/src/llvm22/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:656
#11 0x005e3b17 in llvm::SelectionDAGISel::SelectBasicBlock (this=0x170e9a0,
LLVMBB=0x170bc20, Begin={>
= {<> = {}, }, NodePtr = 0x170bf64},
End={> = {<> = {}, }, NodePtr = 0x170bf90}) at
/Users/criswell/src/llvm22/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:500
#12 0x005e4365 in llvm::SelectionDAGISel::SelectAllBasicBlocks (this=0x170e9a0,
Fn=@0x1707a40, MF=@0x1714200, MMI=0x1712510, DW=0x170f210, TII=@0x18062f0) at
/Users/criswell/src/llvm22/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:856
#13 0x005e473c in llvm::SelectionDAGISel::runOnFunction (this=0x170e9a0,
Fn=@0x1707a40) at
/Users/criswell/src/llvm22/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp:327
#14 0x0086091c in llvm::FPPassManager::runOnFunction (this=0x170c7f0,
F=@0x1707a40) at /Users/criswell/src/llvm22/lib/VMCore/PassManager.cpp:1323
#15 0x00860b72 in llvm::FunctionPassManagerImpl::run (this=0x170c610,
F=@0x1707a40) at /Users/criswell/src/llvm22/lib/VMCore/PassManager.cpp:1281
#16 0x00860cf8 in llvm::FunctionPassManager::run (this=0xbffff5e0,
F=@0x1707a40) at /Users/criswell/src/llvm22/lib/VMCore/PassManager.cpp:1226
#17 0x00002dc5 in main (argc=2, argv=0xbffff694) at
/Users/criswell/src/llvm22/tools/llc/llc.cpp:317

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 13:20:37 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 13:20:37 -0600
Subject: [LLVMbugs] [Bug 3552] New: LLVM-GCC fails to build on MingW (non
	bootstrap).
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3552

           Summary: LLVM-GCC fails to build on MingW (non bootstrap).
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jlerouge at apple.com
                CC: dpatel at apple.com, gohman at apple.com, llvmbugs at cs.uiuc.edu

LLVM-GCC doesn't build on MingW, starting with revision 61207, doing a
non-bootstrap build.

* rev 61207 introduces the following error:

make[4]: Entering directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3/include'
if [ ! -d "./i686-pc-mingw32/bits/stdc++.h.gch" ]; then \
  mkdir -p ./i686-pc-mingw32/bits/stdc++.h.gch; \
fi; \
/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/./gcc/xgcc
-shared-libgcc
-B/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/./gcc
-nostdinc++ -L/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc
-build/i686-pc-mingw32/libstdc++-v3/src
-L/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3/src/.libs
-L/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/winsup/
mingw
-L/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/winsup/w32api/lib
-isystem
/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-gcc4.2-src/winsup/mingw/include
-isystem /c/cygwin/home/jlerouge/buildb
ot/llvm-test-fixed-gcc4.2-src/winsup/w32api/include
-Bc:/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/i686-pc-mingw32/bin/
-Bc:/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/i686-pc-mingw32/lib/
-isystem c:/cygwin/home/jler
ouge/buildbot/llvm-test-fixed-61207/i686-pc-mingw32/include -isystem
c:/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/i686-pc-mingw32/sys-include
-Winvalid-pch -Wno-deprecated -x c++-header -g -O2 
-I/c/cygwin/home/jlerouge/buildb
ot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3/include/i686-pc-mingw32
-I/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3/include
-I/c/cygwin/home/jlerouge/buildbot/llvm-test
-fixed-gcc4.2-src/libstdc++-v3/libsupc++ -O0 -g
/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-gcc4.2-src/libstdc++-v3/include/precompiled/stdc++.h
-o i686-pc-mingw32/bits/stdc++.h.gch/O0g.gch
make[4]: *** [i686-pc-mingw32/bits/stdc++.h.gch/O0g.gch] Error 1
make[4]: Leaving directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3/include'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3'
make[2]: *** [all] Error 2
make[2]: Leaving directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build/i686-pc-mingw32/libstdc++-v3'
make[1]: *** [all-target-libstdc++-v3] Error 2
make[1]: Leaving directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61207/gcc-build'

SVN log:
------------------------------------------------------------------------
r61207 | dpatel | 2008-12-18 11:18:20 -0800 (Thu, 18 Dec 2008) | 3 lines
Changed paths:
   M /llvm-gcc-4.2/trunk/gcc/llvm-debug.cpp

Reuse DIType slot used for foward declaration for real definition.

------------------------------------------------------------------------

* And then, rev 61215 introduces another error, earlier in the build process:

/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/gcc-build/./gcc/xgcc
-B/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/gcc-build/./gcc/
-L/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/gcc-build/i686-pc-mingw32/win
sup/mingw
-L/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/gcc-build/i686-pc-mingw32/winsup/w32api/lib
-isystem
/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-gcc4.2-src/winsup/mingw/include
-isystem /c/cygwin/home/jlerouge/bu
ildbot/llvm-test-fixed-gcc4.2-src/winsup/w32api/include
-Bc:/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/i686-pc-mingw32/bin/
-Bc:/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/i686-pc-mingw32/lib/
-isystem c:/cygwin/home/
jlerouge/buildbot/llvm-test-fixed-61215/i686-pc-mingw32/include -isystem
c:/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/i686-pc-mingw32/sys-include
-c -DHAVE_CONFIG_H -O2 -g -O2  -I.
-I../../../../llvm-test-fixed-gcc4.2-src/libi
berty/../include  -W -Wall -pedantic -Wwrite-strings -Wstrict-prototypes
-Wc++-compat ../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c -o
pex-win32.o
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c: In function
'argv_to_cmdline':
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c:365: warning:
request for implicit conversion from 'void *' to 'char *' not permitted in C++
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c: In function
'find_executable':
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c:440: warning:
request for implicit conversion from 'void *' to 'char *' not permitted in C++
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c: In function
'win32_spawn':
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c:566: warning:
request for implicit conversion from 'void *' to 'char **' not permitted in C++
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c:573: warning:
request for implicit conversion from 'void *' to 'char *' not permitted in C++
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c: At top level:
../../../../llvm-test-fixed-gcc4.2-src/libiberty/pex-win32.c:900: internal
compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
make[2]: *** [pex-win32.o] Error 1
make[2]: Leaving directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/gcc-build/i686-pc-mingw32/libiberty'
make[1]: *** [all-target-libiberty] Error 2
make[1]: Leaving directory
`/c/cygwin/home/jlerouge/buildbot/llvm-test-fixed-61215/gcc-build'
make: *** [all] Error 2

SVN log:
------------------------------------------------------------------------
r61215 | djg | 2008-12-18 14:03:42 -0800 (Thu, 18 Dec 2008) | 3 lines
Changed paths:
   M /llvm/trunk/lib/Target/X86/X86RegisterInfo.cpp

When emitting instructions that define EFLAGS and the EFLAGS value isn't
used, mark the defs as dead.

------------------------------------------------------------------------

I believe those two problems are still there in the current 2.5 branch (as of
rev 64291).

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 15:16:16 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 15:16:16 -0600
Subject: [LLVMbugs] [Bug 3553] New: FPGrowth LLC Regression
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3553

           Summary: FPGrowth LLC Regression
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org
                CC: llvmbugs at cs.uiuc.edu

This is a regression from 2.4 (linux x86). Reduced with bugppoint.

*** The following function is being miscompiled:  main_bb7
You can reproduce the problem with the command line: 
  llc -f bugpoint.test.bc -o bugpoint.test.bc.s
  gcc ./bugpoint.safe.bc.cbe.c.so bugpoint.test.bc.s -o bugpoint.test.bc.exe
-Wl,-R.
  bugpoint.test.bc.exe /home/vadve/shared/benchmarks/fpgrowth/mushroom.dat
The shared object was created with:
  llc -march=c bugpoint.safe.bc -o temporary.c
  gcc -xc temporary.c -O2 -o ./bugpoint.safe.bc.cbe.c.so -fPIC -shared
-fno-strict-aliasing

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 15:32:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 15:32:24 -0600
Subject: [LLVMbugs] [Bug 3554] New: SPEC/CFP2006/447.dealII LLC Regression
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3554

           Summary: SPEC/CFP2006/447.dealII LLC Regression
           Product: new-bugs
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: blocker
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org
                CC: llvmbugs at cs.uiuc.edu

This is a regression from 2.4 (x86 Mac OS X). Reduced with bugpoint:

*** The following function is being miscompiled:  __tcf_72252_entry_2E_ce
You can reproduce the problem with the command line: 
  llc -f bugpoint.test.bc -o bugpoint.test.bc.s
  gcc ./bugpoint.safe.bc.cbe.c.dylib bugpoint.test.bc.s -o bugpoint.test.bc.exe
  bugpoint.test.bc.exe 10
The shared object was created with:
  llc -march=c bugpoint.safe.bc -o temporary.c
  gcc -xc temporary.c -O2 -o ./bugpoint.safe.bc.cbe.c.dylib -fPIC -shared
-fno-strict-aliasing

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 17:43:25 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 17:43:25 -0600
Subject: [LLVMbugs] [Bug 3555] New: altivec.vaddsbs intrinsic doesn't work
	at all
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3555

           Summary: altivec.vaddsbs intrinsic doesn't work at all
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: PowerPC
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nlewycky at google.com
                CC: llvmbugs at cs.uiuc.edu

$ cat vaddsbs.ll
declare <16 x i8> @llvm.ppc.altivec.vaddsbs(<16 x i8> %a, <16 x i8>)

define void @foo(<16 x i8> %x) {
  %foo = call <16 x i8> @llvm.ppc.altivec.vaddsbs(<16 x i8> %x, <16 x i8> %x)
  ret void
}
$ llvm-as vaddsbs.ll
$ llc -march=ppc32 vaddsbs.bc
llc: SelectionDAGBuild.cpp:2925: void
llvm::SelectionDAGLowering::visitTargetIntrinsic(llvm::CallInst&, unsigned
int): Assertion `TLI.isTypeLegal(Op.getValueType()) && "Intrinsic uses a
non-legal type?"' failed.
0   llc       0x000000000110ae70
1   llc       0x000000000110b1c4
2   libc.so.6 0x00007f0049d1d100
3   libc.so.6 0x00007f0049d1d095 gsignal + 53
4   libc.so.6 0x00007f0049d1eaf0 abort + 272
5   libc.so.6 0x00007f0049d162df __assert_fail + 239
6   llc       0x0000000000d7bf9d
llvm::SelectionDAGLowering::visitTargetIntrinsic(llvm::CallInst&, unsigned int)
+ 685
7   llc       0x0000000000d7c7b9
llvm::SelectionDAGLowering::visitIntrinsicCall(llvm::CallInst&, unsigned int) +
145
8   llc       0x0000000000d810d6
llvm::SelectionDAGLowering::visitCall(llvm::CallInst&) + 244
9   llc       0x0000000000d8cb08 llvm::SelectionDAGLowering::visit(unsigned
int, llvm::User&) + 952
10  llc       0x0000000000d8cbfa
llvm::SelectionDAGLowering::visit(llvm::Instruction&) + 44
[..]

The type on the instruction is correct according to the comments in PowerPC.td,
and if it weren't llvm-as would've failed in the verifier. This happens for
march=ppc32 and march=ppc64.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 18:27:49 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 18:27:49 -0600
Subject: [LLVMbugs] [Bug 3556] New: GVN nukes load it shouldn't
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3556

           Summary: GVN nukes load it shouldn't
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Scalar Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nlewycky at google.com
                CC: resistor at mac.com, llvmbugs at cs.uiuc.edu

The testcase shows a function @luaO_str2d that calls strtod and strtoul with
nocapture attributes marked on them.

GVN is removing the load of 'char *endptr' after the call to strtod. It's not
allowed to do that as nobody said it wouldn't write to that pointer.

Here's the beginning of the problem function:

define hidden i32 @luaO_str2d(i8* %s, double* nocapture %result) nounwind {
entry:
        %.loc = alloca i8*              ;  [#uses=2]
        %endptr = alloca i8*, align 8           ;  [#uses=6]
        %0 = call double @strtod(i8* noalias %s, i8** noalias %endptr) nounwind 
        ;  [#uses=1]
        store double %0, double* %result, align 8
        %1 = load i8** %endptr, align 8         ;  [#uses=1]
        %2 = icmp eq i8* %1, %s         ;  [#uses=1]
        br i1 %2, label %entry.bb12_crit_edge, label %bb1

There's no way GVN could remove %1 here, but it does anyways:

$ opt -gvn b.bc -debug
GVN iteration: 0
GVN removed:    %1 = load i8** %endptr, align 8         ;  [#uses=0]
GVN REMOVING NONLOCAL LOAD:     %2 = load i8** %endptr, align 8         ; 
[#uses=1]
GVN removed:    %2 = load i8** %endptr, align 8         ;  [#uses=0]
GVN REMOVING NONLOCAL LOAD:     %5 = load i8** %endptr, align 8         ; 
[#uses=1]
GVN removed:    %.rle1 = phi i8* [ undef, %bb1.bb4_crit_edge ], [ undef,
%bb1.bb4_crit_edge1 ]          ;  [#uses=0]
GVN removed:    %.rle = phi i8* [ undef, %bb4 ], [ undef, %bb1.bb5_crit_edge ]
;  [#uses=0]
GVN removed:    %5 = load i8** %endptr, align 8         ;  [#uses=0]
GVN iteration: 1

and I can't seem to make it obviously happen with a smaller testcase (ie., just
the block of .ll quoted with ret i1 %2).

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 18:37:30 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 18:37:30 -0600
Subject: [LLVMbugs] [Bug 3556] GVN nukes load it shouldn't
In-Reply-To: 
Message-ID: <200902120037.n1C0bUTs025988@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3556

Nick Lewycky  changed:

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

--- Comment #2 from Nick Lewycky   2009-02-11 18:37:30 ---
Yarr... if it writes %s into %endptr, then %s isn't nocapture is 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 cs.uiuc.edu  Wed Feb 11 19:17:15 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 19:17:15 -0600
Subject: [LLVMbugs] [Bug 3557] New: llvm-2.5 Release build fail to compile
 UnitTests/ SignlessTypes/cast2.cpp
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3557

           Summary: llvm-2.5 Release build fail to compile
                    UnitTests/SignlessTypes/cast2.cpp
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Keywords: compile-fail
          Severity: critical
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: dodohack at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2540)
 --> (http://llvm.org/bugs/attachment.cgi?id=2540)
compile by llvm-g++ with -O0

[hwalin at earth SignlessTypes]$ /temp/obj/Release/bin/llc -f cast2.llvm.bc -o
cast2.s
Value still in symbol table! Type = 'i32' Name = 'retval7'
Value still in symbol table! Type = 'opaque' Name = ''
Value still in symbol table! Type = 'i32 *' Name = 'iftmp.1'
Value still in symbol table! Type = 'i32 *' Name = 'retval'
Value still in symbol table! Type = 'i32 *' Name = 'incnum'
Value still in symbol table! Type = 'i8 *' Name = 'inbuff_addr'
llc: /temp/llvm-2.5/lib/VMCore/ValueSymbolTable.cpp:29:
llvm::ValueSymbolTable::~ValueSymbolTable(): Assertion `vmap.empty() && "Values
remain in symbol table!"' failed.
0   llc       0x08a6f177
1   libc.so.6 0x00baff91 abort + 257
2   libc.so.6 0x00ba793e __assert_fail + 238
3   llc       0x08a0ecba

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 19:31:53 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 19:31:53 -0600
Subject: [LLVMbugs] [Bug 3558] New: ASTContext::getIntWidth, getTypeInfo,
 etc should be const
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3558

           Summary: ASTContext::getIntWidth, getTypeInfo, etc should be
                    const
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Keywords: code-cleanup
          Severity: normal
          Priority: P2
         Component: AST
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu

There are several accessors that are "logically" const but not marked as such. 
This causes const_casts in other logically const methods like
getLegacyIntegralTypeEncoding.  We should just mark the data they use as
mutable or avoid calling getCanonicalType in it.

-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 cs.uiuc.edu  Wed Feb 11 20:00:00 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 20:00:00 -0600
Subject: [LLVMbugs] [Bug 3557] llvm-2.5 Release build fail to compile
 UnitTests/SignlessTypes/ cast2.cpp
In-Reply-To: 
Message-ID: <200902120200.n1C200Op028975@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3557

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-11 19:59:59 ---
I believe you are using GCC 4.1.x which is known broken.  Please build llc with
a different compiler.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 11 20:03:01 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 11 Feb 2009 20:03:01 -0600
Subject: [LLVMbugs] [Bug 3555] altivec.vaddsbs intrinsic doesn't work at all
In-Reply-To: 
Message-ID: <200902120203.n1C231pq029089@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3555

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-11 20:03:01 ---
i agree that the error message is pretty gross, but this does work fine:

$ cat t.ll
declare <16 x i8> @llvm.ppc.altivec.vaddsbs(<16 x i8> %a, <16 x i8>)

define <16 x i8>  @foo(<16 x i8> %x) nounwind {
  %foo = call <16 x i8> @llvm.ppc.altivec.vaddsbs(<16 x i8> %x, <16 x i8> %x)
  ret <16 x i8>  %foo
}

$ llvm-as < t.ll | llc  -o - -march=ppc32 -mcpu=g5
...
_foo:
        vaddsbs v2, v2, v2
        blr 

Make sure to pick a CPU with altivec 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 cs.uiuc.edu  Thu Feb 12 00:12:01 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 00:12:01 -0600
Subject: [LLVMbugs] [Bug 3545] Fatal error when assembling RBSplitView
In-Reply-To: 
Message-ID: <200902120612.n1C6C1W4004884@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3545

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-12 00:12:01 ---
This was fixed 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 cs.uiuc.edu  Thu Feb 12 00:21:43 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 00:21:43 -0600
Subject: [LLVMbugs] [Bug 3546] Redefinition of typedef doesn' t print
 previously definition in some cases.
In-Reply-To: 
Message-ID: <200902120621.n1C6LhwV005321@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3546

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-12 00:21:42 ---
This is correct behavior.  When you #include with a full path
"/usr/include/stdio.h", neither clang nor gcc consider the file to be in a
system header directory.  This is an extwarn that is silenced in system
headers, so without that "bit" on the file, the warning gets emitted and mapped
to an 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 cs.uiuc.edu  Thu Feb 12 00:38:59 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 00:38:59 -0600
Subject: [LLVMbugs] [Bug 3543] Certain ({ }) (?) expressions crash clang /
	ccc
In-Reply-To: 
Message-ID: <200902120638.n1C6cxc5006032@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3543

Chris Lattner  changed:

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

--- Comment #2 from Chris Lattner   2009-02-12 00:38:58 ---
I see this:

$  clang  clang_crash_oeh8Rx.mi
/Users/bkr/vc/jumper/tmp/ClangCrash/test.m:6:13: error: expected expression
 int x = ({ __label__ lplabel; static LOGPOINT lplog ....
            ^

This is because we don't support __label__ yet.  If this is not the problem you
are seeing, please reopen.

*** This bug has been marked as a duplicate of bug 3429 ***

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 00:56:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 00:56:33 -0600
Subject: [LLVMbugs] [Bug 3537] Codegen prepare assertion
In-Reply-To: 
Message-ID: <200902120656.n1C6uXkO006741@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3537

Chris Lattner  changed:

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

--- Comment #3 from Chris Lattner   2009-02-12 00:56:32 ---
Fixed, thanks!
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073640.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 cs.uiuc.edu  Thu Feb 12 01:07:55 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 01:07:55 -0600
Subject: [LLVMbugs] [Bug 3550] problem with tail call to function that
 accesses alloca'd memory
In-Reply-To: 
Message-ID: <200902120707.n1C77t5G007173@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3550

Chris Lattner  changed:

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

--- Comment #5 from Chris Lattner   2009-02-12 01:07:55 ---
Whoa, excellent analysis and patch, great catch:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073642.html

Your GVN patch is also great:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073641.html

Thanks Jay!

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 01:44:04 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 01:44:04 -0600
Subject: [LLVMbugs] [Bug 3559] New: llvm-gcc emits debug info for gimple
 temporaries ( non-user vars)
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3559

           Summary: llvm-gcc emits debug info for gimple temporaries (non-
                    user vars)
           Product: tools
           Version: 2.2
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu

I see an llvm.declare for A.5 in the output of:

$ llvm-gcc t.c -emit-llvm -g -o - -S

on this testcase:

void foo(int X) {
  int A[X];
}

Also, I do not see an llvm.declare of "a" itself, the VLA.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 02:00:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 02:00:50 -0600
Subject: [LLVMbugs] [Bug 3560] New: fpmath code generation performance issue
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3560

           Summary: fpmath code generation performance issue
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: j at jawed.name
                CC: llvmbugs at cs.uiuc.edu

Experiencing a 2x slowdown with nested loops in PHP source.

1) grab the php source (http://snaps.php.net)
2) tar xjvf php5.3[...].tar.bz2
3) cd php5.3[...]
4) ./configure --enable-debug --enable-cli --disable-all --without-sqlite3
5) make && make install && make test

ZEND_POST_INC_SPEC_CV_HANDLER is poorly generated, llvm-gcc generates x86 asm
which uses fldl/fstpl vs gcc which does not (and consequently does not
experience the slowdown).

The test in question is nestedloop(12), performs better/on par with gcc with
>=msse2.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 03:45:21 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 03:45:21 -0600
Subject: [LLVMbugs] [Bug 3561] New: Assertion `StackTop > 0 && "Cannot pop
 empty stack!"' failed.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3561

           Summary: Assertion `StackTop > 0 && "Cannot pop empty stack!"'
                    failed.
           Product: new-bugs
           Version: unspecified
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2542)
 --> (http://llvm.org/bugs/attachment.cgi?id=2542)
testcase .ll

This breaks the llvm-gcc build when targeting the i386 cpu.  To reproduce:
$ llc -mcpu=i386 fp_crash.bc
llc: lib/Target/X86/X86FloatingPoint.cpp:553:
void::FPS::popStackAfter(llvm::ilist_iterator&):
Assertion `StackTop > 0 && "Cannot pop empty stack!"' 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 cs.uiuc.edu  Thu Feb 12 07:51:04 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 07:51:04 -0600
Subject: [LLVMbugs] [Bug 3561] Assertion `StackTop > 0 && "Cannot pop empty
 stack!"' failed.
In-Reply-To: 
Message-ID: <200902121351.n1CDp4HL031171@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3561

Duncan Sands  changed:

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

--- Comment #3 from Duncan Sands   2009-02-12 07:51:03 ---
Fixed in record time by Evan here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073655.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 cs.uiuc.edu  Thu Feb 12 10:52:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 10:52:12 -0600
Subject: [LLVMbugs] [Bug 3562] New: clarify error on gcc vla-in-struct
	extension
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3562

           Summary: clarify error on gcc vla-in-struct extension
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang should give a clearer error on code that uses gcc's "vla-in-struct"
extension (is there a name for it?):
--
ddunbar at milton:gcc$ cat t.c
void f0(int n,...) {
  struct S {
    double          a[n];
  }               s;
  double x = s.a[0];
}
ddunbar at milton:gcc$ clang t.c
t.c:3:21: error: fields must have a constant size
    double          a[n];
                    ^
t.c:5:14: error: subscripted value is neither array nor pointer
  double x = s.a[0];
             ^   ~
2 diagnostics generated.
--

Something like "This gcc extension is not supported, and will hopefully never
be. Change your code you loon." It'd be nice not to get the second error as
well.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 11:33:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 11:33:33 -0600
Subject: [LLVMbugs] [Bug 3538] codegen crash on llvm.declare of a VLA
In-Reply-To: 
Message-ID: <200902121733.n1CHXX9G007780@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3538

Chris Lattner  changed:

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

--- Comment #3 from Chris Lattner   2009-02-12 11:33:32 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073661.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 cs.uiuc.edu  Thu Feb 12 12:48:42 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 12:48:42 -0600
Subject: [LLVMbugs] [Bug 3551] llc asserts on given bitcode
In-Reply-To: 
Message-ID: <200902121848.n1CImgRM010739@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3551

Devang Patel  changed:

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

--- Comment #4 from Devang Patel   2009-02-12 12:48:42 ---
Chris,

Your rev. 64382 patch fixed this. 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 cs.uiuc.edu  Thu Feb 12 13:36:34 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 13:36:34 -0600
Subject: [LLVMbugs] [Bug 3563] New: clang stdint. h has incorrect definition
 for intptr_t on x86-32 Linux
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3563

           Summary: clang stdint.h has incorrect definition for intptr_t on
                    x86-32 Linux
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu

Per description: intptr_t is an int on x86-32 Linux, but the clang stdint.h
defines it as long.

I think we need target-specific code to deal with 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 cs.uiuc.edu  Thu Feb 12 15:21:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 15:21:08 -0600
Subject: [LLVMbugs] [Bug 3564] New: __attribute__((__section__(x))) symbols
 generation problems
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3564

           Summary: __attribute__((__section__(x))) symbols generation
                    problems
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

when I compile the attached with clang and gcc I am getting different things:

lev /tmp$ gcc -c dataset.c; nm dataset.o
0000000000000000 r __set_foo_sym_bar.1873
0000000000000000 b bar
0000000000000000 T foo

lev /tmp$ ccc -c dataset.c ; nm dataset.o
ccc: Unknown host 'freebsd', using generic host information.
0000000000000000 b bar
0000000000000000 T foo
0000000000000000 d foo.__set_foo_sym_bar

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 15:50:55 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 15:50:55 -0600
Subject: [LLVMbugs] [Bug 3565] New: MallocBench/gs crashes
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3565

           Summary: MallocBench/gs crashes
           Product: Test Suite
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Nightly Tester
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nlewycky at google.com
                CC: nicholas at mxc.ca, llvmbugs at cs.uiuc.edu

Something is wrong with this test:

llvm-test/MultiSource/Benchmarks/MallocBench/gs$ cat Output/gs.out-nat
Reading ghost.ps... exit 139
RunSafely.sh detected a failure with these command-line arguments:  500 0
/dev/null Output/gs.out-nat Output/gs.native -DNODISPLAY
/usr/local/google/home/nlewycky/llvm/projects/llvm-test/MultiSource/Benchmarks/MallocBench/gs/INPUT/large.ps

Are those command-line arguments right? -DNODISPLAY? Well, at least
INPUT/large.ps does exist. But it doesn't seem to help much:

$ Output/gs.native 
Reading ghost.ps... Segmentation fault
$ Output/gs.native INPUT/large.ps 
Reading ghost.ps... Segmentation fault
$ Output/gs.native -DNODISPLAY INPUT/large.ps 
Reading ghost.ps... Segmentation fault
$ ls ./ghost.ps
ghost.ps

This is on x86_64 Linux with gcc 4.2.4. The LLVM-built code fails the exact
same way.

Rebuilding with debug shows:
Program received signal SIGSEGV, Segmentation fault.
0x000000000041bc47 in zastore (op=0x634d40) at zarray.c:72
72              op[-size] = *op;
(gdb) bt
#0  0x000000000041bc47 in zastore (op=0x634d40) at zarray.c:72
#1  0x0000000000418343 in interp (pref=) at interp.c:345
#2  0x0000000000418870 in interpret (pref=, user_errors=1)
    at interp.c:140
#3  0x0000000000403b51 in run_file (file_name=0x42756d "ghost.ps", 
    user_errors=1) at gs.c:272
#4  0x0000000000403c9e in init2 () at gs.c:244
#5  0x0000000000403f5b in main (argc=1, argv=0x7fff4b16c618) at gs.c:88
(gdb) p size
$1 = 4294967293
(gdb) p op
$2 = (ref *) 0x634d40

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 17:47:40 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 17:47:40 -0600
Subject: [LLVMbugs] [Bug 3566] New: support __attribute__((used)
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3566

           Summary: support __attribute__((used)
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang should support __attribute__((used)
--
ddunbar at ozzy:tmp$ cat t.c
static int a __attribute__((used)) = 52;
ddunbar at ozzy:tmp$ xcc -c t.c
ddunbar at ozzy:tmp$ nm t.o
nm: no name list
--

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 19:18:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 19:18:33 -0600
Subject: [LLVMbugs] [Bug 3543] Certain ({ }) (?) expressions crash clang /
	ccc
In-Reply-To: 
Message-ID: <200902130118.n1D1IXNY025831@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3543

Bjoern Kriews  changed:

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

--- Comment #3 from Bjoern Kriews   2009-02-12 19:18:32 ---
This is a multiple mess up :-)

I messed up by running into another potential problem,
which I just filed under

http://llvm.org/bugs/show_bug.cgi?id=3567

And clang messes up by crashing with exactly the same 
trace regardless if __label__ is present or not.

I will attach another crashes directory (crashed2.zip)
which show this.

Kind regards, Bjoern

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 12 20:58:02 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 12 Feb 2009 20:58:02 -0600
Subject: [LLVMbugs] [Bug 3568] New: clang arbitrary-width integer types
 don't calculate size/ alignment correctly
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3568

           Summary: clang arbitrary-width integer types don't calculate
                    size/alignment correctly
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu

I added support for arbitrary-width integer types in r64434.  The one known bug
in the implementation is that Sema doesn't calculate the size/alignment
consistently with LLVM.  (And for 128-bit integers, it calculates the alignment
to be 128 bits, which as far as I know is not consistent with gcc's TImode.)

LLVM logic for determining the size/alignment:
If there is at least one known integer type whose bitwidth is equal to or
larger than the bitwidth of the given integer type, the alignment is that of
the smallest such type.  Otherwise, the alignment is the alignment of the
largest known integer type.  The size is then the bitwidth plus whatever
padding is necessary to satisfy the alignment.

What makes this non-trivial is that TargetInfo doesn't expose a list of integer
types outside of the target description string, and it's definitely too much
work to parse the target description.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 00:07:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 00:07:12 -0600
Subject: [LLVMbugs] [Bug 3569] New: Assertion failed: (0 && " VAArgInst is
 not yet implemented for x86-64!")
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3569

           Summary: Assertion failed: (0 && "VAArgInst is not yet
                    implemented for x86-64!")
           Product: new-bugs
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jlerouge at apple.com
                CC: llvmbugs at cs.uiuc.edu

I am getting the following assertion when using lli on a random IR that
contains vararg calls:

Assertion failed: (0 && "VAArgInst is not yet implemented for x86-64!"),
function LowerVAARG, file
/Users/julien/work/llvm/test/llvm-src/lib/Target/X86/X86ISelLowering.cpp,
line 5809.

Thanks,
Julien

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 00:37:07 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 00:37:07 -0600
Subject: [LLVMbugs] [Bug 3569] Assertion failed: (0 && " VAArgInst is not
 yet implemented for x86-64!")
In-Reply-To: 
Message-ID: <200902130637.n1D6b75A003485@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3569

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-13 00:37:07 ---
Unfortunately, the va_arg instruction isn't implemented yet for x86-64, and may
not even work in all cases for x86-32.  The front-end is expected to do the
inline expansion for this (insanely complex) operation right now.

*** This bug has been marked as a duplicate of bug 1740 ***

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 01:31:44 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 01:31:44 -0600
Subject: [LLVMbugs] [Bug 3570] New: Mac OS X: Linking to llvm modifies
 behaviour of abort()
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3570

           Summary: Mac OS X: Linking to llvm modifies behaviour of abort()
           Product: libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Support Libraries
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: alexei.svitkine at gmail.com
                CC: llvmbugs at cs.uiuc.edu

The following behavior applies to Mac OS X 10.5.6 with latest updates.

Consider program abort.cpp:

#include 

int main(const int argc, char **argv)
{
        abort();
        return 0;
}

Compile it normally and run:

g++ crash.cpp && ./a.out 

Produces the expected crash behaviour.

Now do the same thing, while linking to llvm:

g++ crash.cpp `llvm-config --ldflags` `llvm-config --libs` && ./a.out

As you can see the abort() behaviour is modified. Before aborting, the process
hangs for about a minute. Finally, (if you have crash reporting enabled for
non-GUI processes), the crash reporter window pops up, but instead of having
the stack trace of my program as expected (and as the case before), it instead
reports that the "application ReportCrash quit unexpectedly".

Linking to llvm libraries should not have such a side-effect on the behaviour
of 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 cs.uiuc.edu  Fri Feb 13 03:59:55 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 03:59:55 -0600
Subject: [LLVMbugs] [Bug 3571] New: [2.5] Qt build failed
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3571

           Summary: [2.5] Qt build failed
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         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=2551)
 --> (http://llvm.org/bugs/attachment.cgi?id=2551)
Testcase

Currently Qt build is failed on 2.5 at least on mingw, which is definitely a
serious regression.

The assertion in codegen is triggered. Same can be seen on TOT:

$ ./llc bugpoint-reduced-simplified.bc 
llc: /home/asl/proj/llvm/src/lib/CodeGen/LiveVariables.cpp:129: void
llvm::LiveVariables::HandleVirtRegUse(unsigned int, llvm::MachineBasicBlock*,
llvm::MachineInstr*): Assertion `MRI->getVRegDef(reg) && "Register use before
def!"' 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 cs.uiuc.edu  Fri Feb 13 06:54:47 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 06:54:47 -0600
Subject: [LLVMbugs] [Bug 3572] New: missing __attribute__((__used__))
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3572

           Summary: missing __attribute__((__used__))
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

clang ignore __used__ attribute which prevents FreeBSD kernel from linking. the
generated scterm-teken.o contains zero symbols and then linker complains about
missing symbols.

lev /tmp$ cc -c -std=c99 scterm-teken.c && nm scterm-teken.o | wc -l
In file included from ../../../dev/syscons/scterm-teken.c:37,
                 from ../../../dev/syscons/scterm-teken.c:2:
../../../sys/systm.h:159: warning: conflicting types for built-in function
'log'
      56

lev /tmp$ ccc -c -std=c99 scterm-teken.c && nm scterm-teken.o | wc -l

... warnings ...

       0

please test the attached

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 07:08:34 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 07:08:34 -0600
Subject: [LLVMbugs] [Bug 3572] missing __attribute__((__used__))
In-Reply-To: 
Message-ID: <200902131308.n1DD8YvC027269@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3572

Anton Korobeynikov  changed:

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

--- Comment #2 from Anton Korobeynikov   2009-02-13 07:08:34 ---

*** This bug has been marked as a duplicate of bug 3566 ***

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 10:40:44 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 10:40:44 -0600
Subject: [LLVMbugs] [Bug 3573] New: lli crash -force-interpreter and stdlib
 function calls
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3573

           Summary: lli crash -force-interpreter and stdlib function calls
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: reynaudd at loria.fr
                CC: llvmbugs at cs.uiuc.edu

The following C program crashes when compiled with -O3 and run with lli
-force-interpreter on my system:

#include 
#include 

int main(int argc, char **argv) {
    int number;
    if(argc<2) {
        printf("Usage: %s \n", argv[0]);
        exit(1);
    }

// other calls such as atof also provoke the crash
    number = atoi(argv[1]);
    printf("hello %d\n", number);
}

Steps to reproduce the crash:

~/test/llvm$ llvm-gcc -emit-llvm -O3 -c hello.c -o hello.bc
~/test/llvm$ lli -force-interpreter hello.bc 42
lli: /tmp/buildd/llvm-2.2/lib/Support/APInt.cpp:109: llvm::APInt::APInt(const
llvm::APInt&): Assertion `BitWidth <= MAX_INT_BITS && "bitwidth too large"'
failed.
lli[0x843d88]
/lib/libc.so.6[0x7f6f3b9daf60]
/lib/libc.so.6(gsignal+0x35)[0x7f6f3b9daed5]
/lib/libc.so.6(abort+0x183)[0x7f6f3b9dc3f3]
/lib/libc.so.6(__assert_fail+0xe9)[0x7f6f3b9d3dc9]
lli(llvm::APInt::APInt(llvm::APInt const&)+0xc3)[0x820773]
lli(llvm::Interpreter::callExternalFunction(llvm::Function*,
std::vector >
const&)+0xc7)[0x5be997]
lli(llvm::Interpreter::callFunction(llvm::Function*,
std::vector >
const&)+0x55f)[0x5b6adf]
lli(llvm::Interpreter::visitCallSite(llvm::CallSite)+0x22e)[0x5b86ce]
lli(llvm::InstVisitor::visit(llvm::Instruction&)+0xa6)[0x5c22b6]
lli(llvm::Interpreter::run()+0x61)[0x5b92a1]
lli(llvm::Interpreter::runFunction(llvm::Function*,
std::vector >
const&)+0xbc)[0x5bf35c]
lli(llvm::ExecutionEngine::runFunctionAsMain(llvm::Function*,
std::vector,
std::allocator >, std::allocator, std::allocator > > > const&, char const*
const*)+0x497)[0x5c89d7]
lli(main+0x21d)[0x4b362d]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7f6f3b9c71a6]
lli(std::ios_base::Init::~Init()+0x41)[0x4b2ec9]
Aborted

The program runs correctly without -force-interpreter, without -emit-llvm and
with -O0.

Configuration info (llvm was obtained with apt-get):

~/test/llvm$ lli --version
Low Level Virtual Machine (http://llvm.org/):
  llvm version 2.2 (Debian 2.2-12)
  Optimized build with assertions.

~/test/llvm$ uname -a
Linux XXX 2.6.26-1-amd64 #1 SMP Sat Jan 10 17:57:00 UTC 2009 x86_64 GNU/Linux

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 11:19:21 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 11:19:21 -0600
Subject: [LLVMbugs] [Bug 3573] lli crash -force-interpreter and stdlib
	function calls
In-Reply-To: 
Message-ID: <200902131719.n1DHJL09004153@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3573

Anton Korobeynikov  changed:

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

--- Comment #1 from Anton Korobeynikov   2009-02-13 11:19:20 ---
2.2 is really ancient version. TOT correctly reports lack of interpreter
support for external function atoi

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 11:23:58 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 11:23:58 -0600
Subject: [LLVMbugs] [Bug 3574] New: missed opt: ptrtoint(inttoptr (a -> b),
	c) -> sext
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3574

           Summary: missed opt: ptrtoint(inttoptr (a -> b), c) -> sext
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

The following snippet should simplify to an sext on x86_64:
--
ddunbar at ozzy:CodeGen$ cat /tmp/t.ll
; ModuleID = ''
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-f80:128:128"
target triple = "x86_64-apple-darwin10.0"
define i64 @f0(i32 %a0) nounwind {
       %t0 = inttoptr i32 %a0 to i8*
       %t1 = ptrtoint i8* %t0 to i64
       ret i64 %t1
}
ddunbar at ozzy:CodeGen$ llvm-as < /tmp/t.ll | opt -instcombine | llvm-dis
; ModuleID = ''
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-f80:128:128"
target triple = "x86_64-apple-darwin10.0"

define i64 @f0(i32 %a0) nounwind {
        %t0 = inttoptr i32 %a0 to i8*           ;  [#uses=1]
        %t1 = ptrtoint i8* %t0 to i64           ;  [#uses=1]
        ret i64 %t1
}
--

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 11:39:18 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 11:39:18 -0600
Subject: [LLVMbugs] [Bug 3542] Implement __attribute__((overloadable))
In-Reply-To: 
Message-ID: <200902131739.n1DHdIq2005083@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3542

Douglas Gregor  changed:

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

--- Comment #4 from Douglas Gregor   2009-02-13 11:39:18 ---
This is now implemented in Clang. Documentation is here:

  http://clang.llvm.org/docs/LanguageExtensions.html#overloading-in-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 cs.uiuc.edu  Fri Feb 13 14:35:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 14:35:08 -0600
Subject: [LLVMbugs] [Bug 3575] New: warn when adding attributes following
	definition
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3575

           Summary: warn when adding attributes following definition
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang should warn about the following code:
--
static void A() {

}
static void A() __attribute__((constructor));
--

The attribute is ignored because it follows the definition. This is a more
systemic problem than just attribute constructor.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 15:05:57 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 15:05:57 -0600
Subject: [LLVMbugs] [Bug 3576] New: llvm-gcc ignores local attribute used
 without -funit-at-a-time
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3576

           Summary: llvm-gcc ignores local attribute used without -funit-at-
                    a-time
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

ddunbar at ozzy:CodeGen$ echo "void f1() { static int l0 __attribute__((used)) =
5222; }" | llvm-gcc -emit-llvm -S -o - -x c -
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-f80:128:128"
target triple = "i386-apple-darwin10.0"

define void @f1() nounwind {
entry:
        br label %return

return:         ; preds = %entry
        ret void
}
ddunbar at ozzy:CodeGen$ echo "void f1() { static int l0 __attribute__((used)) =
5222; }" | llvm-gcc -emit-llvm -S -o - -x c - -funit-at-a-time
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-f80:128:128"
target triple = "i386-apple-darwin10.0"
@l0.1444 = internal global i32 5222             ;  [#uses=1]
@llvm.used = appending global [1 x i8*] [ i8* bitcast (i32* @l0.1444 to i8*) ],
section "llvm.metadata"         ; <[1 x i8*]*> [#uses=0]

define void @f1() nounwind {
entry:
        br label %return

return:         ; preds = %entry
        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 cs.uiuc.edu  Fri Feb 13 15:34:53 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 15:34:53 -0600
Subject: [LLVMbugs] [Bug 3577] New: clang does not detect some compile time
	constants.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3577

           Summary: clang does not detect some compile time constants.
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bolzoni at cs.unipr.it
                CC: llvmbugs at cs.uiuc.edu, bagnara at cs.unipr.it

The following pieces of code are from the gcc test suite.
While gcc compiles without problems, clang does not detect that
the values assigned to the static variables are compile time known.

The last two pieces are really bizarre, but the other two use idioms
actually pretty common in low level C code.

/*920928-1.c*/
struct{int c;}v;
static long i=((char*)&(v.c)-(char*)&v);
/*end*/

/*930326-1.c*/
struct
{
  char a, b, f[3];
} s;
long i = s.f-&s.b;
/*end*/

/*labels-3.c*/
int foo (int a)
{
  static const short ar[] = { &&l1 - &&l1, &&l2 - &&l1 };
  void *p = &&l1 + ar[a];
  goto *p;
 l1:
  return 1;
 l2:
  return 2;
}
/*end*/

/*20001116-1.c */
int x[60];
char *y = ((char*)&(x[2*8 + 2]) - 8);
int z = (&"Foobar"[1] - &"Foobar"[0]);
/*end*/

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 15:57:02 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 15:57:02 -0600
Subject: [LLVMbugs] [Bug 3577] clang does not detect some compile time
	constants.
In-Reply-To: 
Message-ID: <200902132157.n1DLv25B014965@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3577

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-13 15:57:01 ---
Unless this occurs in real world code, we should not care about fixing this
stuff.  Note that:

 int z = (&"Foobar"[1] - &"Foobar"[0]);

is not guaranteed to be evaluatable to a constant for many reasons.  For
example, with -fwritable-strings, the two strings will be distinct.

Yes we want to be compatible with GCC, but there is a line of pragmatics that
we need to draw also.  We're not going to reimplement all of GCC's "fold"
optimizer 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 cs.uiuc.edu  Fri Feb 13 16:06:01 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 16:06:01 -0600
Subject: [LLVMbugs] [Bug 3578] New: ConstantFold doesn't handle vector
	bitcast?
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3578

           Summary: ConstantFold doesn't handle vector bitcast?
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: natebegeman at mac.com
                CC: llvmbugs at cs.uiuc.edu

test/Transform/InstCombine/shufflevec-constant.ll doesn't codegen to a constant
load on x86, where it should.  Rumor has it this is due to ConstantFold not
handling bitcast (<2 x float>  to <1 x double>).

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 16:09:51 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 16:09:51 -0600
Subject: [LLVMbugs] [Bug 3566] support __attribute__((used)
In-Reply-To: 
Message-ID: <200902132209.n1DM9pYS015518@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3566

Daniel Dunbar  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 cs.uiuc.edu  Fri Feb 13 16:13:48 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 16:13:48 -0600
Subject: [LLVMbugs] [Bug 3562] clarify error on gcc vla-in-struct extension
In-Reply-To: 
Message-ID: <200902132213.n1DMDm0c015671@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3562

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-13 16:13:48 ---
Diagnostic message fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012311.html

Bogus second diagnostic fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012309.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 cs.uiuc.edu  Fri Feb 13 16:29:22 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 16:29:22 -0600
Subject: [LLVMbugs] [Bug 3563] clang stdint. h has incorrect definition for
 intptr_t on x86-32 Linux
In-Reply-To: 
Message-ID: <200902132229.n1DMTMt1016395@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3563

Chris Lattner  changed:

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

--- Comment #5 from Chris Lattner   2009-02-13 16:29:22 ---
fixed, plz verify:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012314.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 cs.uiuc.edu  Fri Feb 13 16:30:58 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 16:30:58 -0600
Subject: [LLVMbugs] [Bug 3579] New: different clang -E from gcc -E on this
	file
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3579

           Summary: different clang -E from gcc -E on this file
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

hi.. I when I preprocess this file with clang and gcc I am getting different
results:

lev ~$ gcc -E subr_smp.c > subr_smp.gcc.c
lev ~$ ccc -E subr_smp.c > subr_smp.clang.c
ccc: Unknown host 'freebsd', using generic host information.
lev ~$ ls -l subr_smp.gcc.c subr_smp.clang.c
-rw-r--r--  1 rdivacky  rdivacky  75499 Feb 13 23:26 subr_smp.clang.c
-rw-r--r--  1 rdivacky  rdivacky  75256 Feb 13 23:26 subr_smp.gcc.c
lev ~$ 

this is one of the causes why fbsd kernel does not link (it. uses a macro that
makes  symbol name based on the line of definition - clang has name collisions
while gcc does 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 cs.uiuc.edu  Fri Feb 13 16:58:53 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 16:58:53 -0600
Subject: [LLVMbugs] [Bug 3564] __attribute__((__section__(x))) symbols
 generation problems
In-Reply-To: 
Message-ID: <200902132258.n1DMwrU6017658@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3564

Daniel Dunbar  changed:

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

--- Comment #15 from Daniel Dunbar   2009-02-13 16:58:52 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012321.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 cs.uiuc.edu  Fri Feb 13 19:27:17 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 19:27:17 -0600
Subject: [LLVMbugs] [Bug 3581] New: -strip-debug does not remove all debug
	info.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3581

           Summary: -strip-debug does not remove all debug info.
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Interprocedural Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: dpatel at apple.com
                CC: llvmbugs at cs.uiuc.edu

see attached example.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 20:10:41 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 20:10:41 -0600
Subject: [LLVMbugs] [Bug 3511] clang generates i64 @main(i64, i64),
 which lli doesn't accept
In-Reply-To: 
Message-ID: <200902140210.n1E2AfiH025086@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3511

Daniel Dunbar  changed:

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

--- Comment #2 from Daniel Dunbar   2009-02-13 20:10:40 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012339.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 cs.uiuc.edu  Fri Feb 13 20:11:23 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 20:11:23 -0600
Subject: [LLVMbugs] [Bug 3524] clang/x86_64 miscompiles
 MultiSource/Applications/sgefa/sgefa
In-Reply-To: 
Message-ID: <200902140211.n1E2BN76025125@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3524

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-13 20:11:23 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012339.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 cs.uiuc.edu  Fri Feb 13 20:37:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 20:37:50 -0600
Subject: [LLVMbugs] [Bug 3582] New: fp inline asm assertion failure
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3582

           Summary: fp inline asm assertion failure
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nlewycky at google.com
                CC: rafael.espindola at gmail.com, llvmbugs at cs.uiuc.edu

Created an attachment (id=2559)
 --> (http://llvm.org/bugs/attachment.cgi?id=2559)
testcase

This delta-reduced testcase:

  static int FastIntRound (double x)
  {
    int result;
    __asm__ __volatile__ ("fistpl %0":"=m" (result):"t" (x):"st");
  }
  void foo() {
    FastIntRound(1e6);
    FastIntRound(1e6);
  }

compiles to the attached bitcode file on a 32-bit x86 linux machine, and emits
this ugly assertion failure at codegen time:

$ llc x.bc -f
llc: X86FloatingPoint.cpp:995:
void::FPS::handleSpecialFP(llvm::ilist_iterator&):
Assertion `(StackTop == 1 || StackTop == 2) && "Stack should have one or two
element on it to return!"' failed.
(gdb) bt
#0  0x00007fc3af63c095 in raise () from /lib/libc.so.6
#1  0x00007fc3af63daf0 in abort () from /lib/libc.so.6
#2  0x00007fc3af6352df in __assert_fail () from /lib/libc.so.6
#3  0x0000000000b16cdd in handleSpecialFP (this=0x17a98a0, I=@0x7fffb87417b0)
    at X86FloatingPoint.cpp:994
#4  0x0000000000b17c44 in processBasicBlock (this=0x17a98a0, MF=@0x17b3b50, 
    BB=@0x17b4150) at X86FloatingPoint.cpp:257
#5  0x0000000000b180a9 in runOnMachineFunction (this=0x17a98a0, MF=@0x17b3b50)
    at X86FloatingPoint.cpp:211
#6  0x00000000008763d5 in llvm::MachineFunctionPass::runOnFunction (
    this=0x17a98a0, F=@0x177d170)
    at
/usr/local/google/home/nlewycky/llvm/include/llvm/CodeGen/MachineFunctionPass.h:42
#7  0x0000000001086eb3 in llvm::FPPassManager::runOnFunction (this=0x177ffe0, 
    F=@0x177d170) at PassManager.cpp:1323
#8  0x0000000001087432 in llvm::FunctionPassManagerImpl::run (this=0x177f9f0, 
    F=@0x177d170) at PassManager.cpp:1281
#9  0x000000000108759c in llvm::FunctionPassManager::run (this=0x7fffb8741a70, 
    F=@0x177d170) at PassManager.cpp:1226
#10 0x00000000008104eb in main (argc=3, argv=0x7fffb8741c88) at llc.cpp:317
(gdb) up 3
#3  0x0000000000b16cdd in handleSpecialFP (this=0x17a98a0, I=@0x7fffb87417b0)
    at X86FloatingPoint.cpp:994
994         assert((StackTop == 1 || StackTop == 2)
(gdb) p StackTop
$1 = 0
(gdb) up
#4  0x0000000000b17c44 in processBasicBlock (this=0x17a98a0, MF=@0x17b3b50, 
    BB=@0x17b4150) at X86FloatingPoint.cpp:257
257         case X86II::SpecialFP:  handleSpecialFP(I); break;
(gdb) p (*I).dump()
  FpSET_ST0_80 %FP0, %ST0
$2 = void
(gdb)

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 21:16:44 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 21:16:44 -0600
Subject: [LLVMbugs] [Bug 3582] fp inline asm assertion failure
In-Reply-To: 
Message-ID: <200902140316.n1E3Gijp027105@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3582

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-13 21:16:43 ---
the cross product of inline asm and x87 floating point stack is a pretty
ultimate evil. 

*** This bug has been marked as a duplicate of bug 879 ***

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 13 23:55:31 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 13 Feb 2009 23:55:31 -0600
Subject: [LLVMbugs] [Bug 3543] Certain ({ }) (?) expressions crash clang /
	ccc
In-Reply-To: 
Message-ID: <200902140555.n1E5tVdd031916@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3543

Ted Kremenek  changed:

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

--- Comment #8 from Ted Kremenek   2009-02-13 23:55:30 ---
Fixed:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012348.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 cs.uiuc.edu  Sat Feb 14 00:20:27 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 00:20:27 -0600
Subject: [LLVMbugs] [Bug 3552] LLVM-GCC fails to build on MingW (non
	bootstrap).
In-Reply-To: 
Message-ID: <200902140620.n1E6KR2K032626@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3552

Julien Lerouge  changed:

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

--- Comment #13 from Julien Lerouge   2009-02-14 00:20:27 ---
Yep, the build went through. 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 cs.uiuc.edu  Sat Feb 14 04:37:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 04:37:24 -0600
Subject: [LLVMbugs] [Bug 3583] New: Correctness bug in commit r64314,
 trunk/lib/CodeGen/  SelectionDAG/TargetLowering.cpp
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3583

           Summary: Correctness bug in commit r64314,
                    trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp
           Product: new-bugs
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: steinke-dirk at web.de
                CC: llvmbugs at cs.uiuc.edu

In
   bool ValueHasExactlyOneBitSet(SDValue Val, const SelectionDAG &DAG)
it is no longer safe to shortcut by
  if (Val.getOpcode() == ISD::SHL || Val.getOpcode() == ISD::SRL)
    return ValueHasExactlyOneBitSet(Val.getOperand(0), DAG);

Either the shortcut is dropped completely, or it has to be checked that the
shift-amount is not big enough to shift out the single 1-bit from the first
operand. 

E.g. 0x8000 >> (y&15) has still exactly one bit set, although we don't know
exactly which one is.
But we know the first operands 1-bit position is 15, and the upper limit of the
shift amount is 15.

A more tricky case would be
    (0x8000 >> (y&14)) >> (z&1)
This is still safe, because the 1-bit position of the first operand is between
1 and 15, and the upper limit of the shift amount is 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 cs.uiuc.edu  Sat Feb 14 07:14:59 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 07:14:59 -0600
Subject: [LLVMbugs] [Bug 3584] New: Integer literal related vertex node
 contains arbitrary precision number with the wrong sign .
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3584

           Summary: Integer literal related vertex node contains arbitrary
                    precision number with the wrong sign.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: AST
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bolzoni at cs.unipr.it
                CC: llvmbugs at cs.uiuc.edu, bagnara at cs.unipr.it

Test case:
------->
extern void f(unsigned long long int a);
int g(void) {
  f(9223372036854775808lu);
}
-------<

being `node' a the pointer to the integer literal expression, the code:
------->
  if (node-> getValue(). isStrictlyPositive()) {
    std::cout << "it is positive!";
  }
  else {
    std::cout << "it is not positive!";
  }
-------<
gives `it is not positive' as output.

I tried making a test case using only llvm::APInt, but it seemed working 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 cs.uiuc.edu  Sat Feb 14 10:31:02 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 10:31:02 -0600
Subject: [LLVMbugs] [Bug 3585] New: wrong defines for FreeBSD
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3585

           Summary: wrong defines for FreeBSD
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

please apply the attached patch, it fixes defines for fbsd. adds __ELF__ and
removes bsd

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 14 10:44:15 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 10:44:15 -0600
Subject: [LLVMbugs] [Bug 3585] wrong defines for FreeBSD
In-Reply-To: 
Message-ID: <200902141644.n1EGiFZ8030511@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3585

Anton Korobeynikov  changed:

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

--- Comment #2 from Anton Korobeynikov   2009-02-14 10:44:14 ---
Applied in
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012358.html

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 cs.uiuc.edu  Sat Feb 14 11:20:36 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 11:20:36 -0600
Subject: [LLVMbugs] [Bug 3586] New: cant redefine malloc() prototype
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3586

           Summary: cant redefine malloc() prototype
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

dgregor in r64504 made malloc and bunch of other functions to have a constant
prototype... in a case of malloc it is void*(unsigned long)

in FreeBSD kernel malloc() has different prototype:

void    *malloc(unsigned long size, struct malloc_type *type, int flags)
__malloc_like;

clang produces a hard error when one attempts to redefine malloc, can there be
a flag to turn off the hard error? or something like 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 cs.uiuc.edu  Sat Feb 14 14:54:02 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 14:54:02 -0600
Subject: [LLVMbugs] [Bug 3586] cant redefine malloc() prototype
In-Reply-To: 
Message-ID: <200902142054.n1EKs2pN005697@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3586

Douglas Gregor  changed:

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

--- Comment #2 from Douglas Gregor   2009-02-14 14:54:01 ---
Added the -ffreestanding option that will have us act as a freestanding
implementation, which disables the implicit declarations of builtin library
functions (like malloc). The test case works with this option, then fails (as
it should) when we're acting as a hosted implementation (the default).

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 14 16:57:37 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 16:57:37 -0600
Subject: [LLVMbugs] [Bug 3586] cant redefine malloc() prototype
In-Reply-To: 
Message-ID: <200902142257.n1EMvb9n009892@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3586

Roman Divacky  changed:

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

--- Comment #3 from Roman Divacky   2009-02-14 16:57:37 ---
ccc (the new one) got broken by r64566

lev ROMAN# make
/usr/local/bin/ccc -c -O2 -pipe -fno-strict-aliasing -march=native -std=c99 -g
-Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes
-Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -Wundef
-Wno-pointer-sign -fformat-extensions -nostdinc -I. -I../../..
-I../../../contrib/altq -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS -include
opt_global.h -finline-limit=8000 --param inline-unit-growth=100 --param
large-function-growth=1000 -fno-omit-frame-pointer -mcmodel=kernel
-mno-red-zone -mfpmath=387 -mno-sse -mno-sse2 -mno-sse3 -mno-mmx -mno-3dnow
-msoft-float -fno-asynchronous-unwind-tables -ffreestanding -fstack-protector
../../../amd64/amd64/genassym.c
ccc: Unknown host 'freebsd', using generic host information.
*** Error code 245

it works ok when I remove the -ffreestanding option (works = fails on malloc
being redefined)

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 14 17:40:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 17:40:50 -0600
Subject: [LLVMbugs] [Bug 3584] Integer literal related vertex node contains
 arbitrary precision number with the wrong sign .
In-Reply-To: 
Message-ID: <200902142340.n1ENeoue011498@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3584

Eli Friedman  changed:

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

--- Comment #1 from Eli Friedman   2009-02-14 17:40:49 ---
The method isStrictlyPositive doesn't do what you think it does... it always
assumes the input is signed.  You need to check isSigned() as well.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 14 23:46:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 23:46:14 -0600
Subject: [LLVMbugs] [Bug 3526] clang/x86_64 miscompiles
 MultiSource/Benchmarks/MiBench/ telecomm-adpcm/telecomm-adpcm
In-Reply-To: 
Message-ID: <200902150546.n1F5kEjk023627@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3526

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-14 23:46:13 ---
This was fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012339.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 cs.uiuc.edu  Sat Feb 14 23:46:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 23:46:19 -0600
Subject: [LLVMbugs] [Bug 3529] clang/x86_64 miscompiles
 MultiSource/Benchmarks/mediabench/adpcm /rawcaudio/rawcaudio
In-Reply-To: 
Message-ID: <200902150546.n1F5kJvM023639@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3529

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-14 23:46:19 ---
This was fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012339.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 cs.uiuc.edu  Sat Feb 14 23:49:03 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 14 Feb 2009 23:49:03 -0600
Subject: [LLVMbugs] [Bug 3530] clang/x86_64 miscompiles
 SingleSource/Benchmarks/McGill/exptree
In-Reply-To: 
Message-ID: <200902150549.n1F5n3wK023732@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3530

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-14 23:49:03 ---
This was fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012339.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 cs.uiuc.edu  Sun Feb 15 00:06:37 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 00:06:37 -0600
Subject: [LLVMbugs] [Bug 3571] [2.5] Qt build failed
In-Reply-To: 
Message-ID: <200902150606.n1F66bdr024343@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3571

Evan Cheng  changed:

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

--- Comment #5 from Evan Cheng   2009-02-15 00:06:36 ---
Fixed.
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073766.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 cs.uiuc.edu  Sun Feb 15 00:18:55 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 00:18:55 -0600
Subject: [LLVMbugs] [Bug 3588] New: merging with K&R declaration mismatch
	with gcc
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3588

           Summary: merging with K&R declaration mismatch with gcc
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang & gcc differ in how they treat the following code:
--
void g0(int, int);
void g0();
//void g0(int);

void f1() {
  g0(1, 2, 3);
}
--

clang merges the declarations so that g0 is g0() after the second declaration;
and therefore allows the commented out redefinition, and the call. gcc errors
on both.

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 15 02:36:51 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 02:36:51 -0600
Subject: [LLVMbugs] [Bug 3522] Missing register save when only used in
 landing pad ( llc -fast ok)
In-Reply-To: 
Message-ID: <200902150836.n1F8apRr029516@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3522

Evan Cheng  changed:

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

--- Comment #2 from Evan Cheng   2009-02-15 02:36:50 ---
Fixed.
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073767.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 cs.uiuc.edu  Sun Feb 15 10:22:20 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 10:22:20 -0600
Subject: [LLVMbugs] [Bug 3512] [2.5] segfault with dynamic alloca+call
	combination
In-Reply-To: 
Message-ID: <200902151622.n1FGMKXK013614@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3512

Tomas Lindquist Olsen  changed:

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

--- Comment #7 from Tomas Lindquist Olsen   2009-02-15 10:22:19 ---
this problem indeed no longer exists when using the latest release_25 branch.
ty whoever fixed it :P

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 15 12:38:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 12:38:29 -0600
Subject: [LLVMbugs] [Bug 3586] cant redefine malloc() prototype
In-Reply-To: 
Message-ID: <200902151838.n1FIcTVZ017748@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3586

Roman Divacky  changed:

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

--- Comment #6 from Roman Divacky   2009-02-15 12:38:28 ---
yes, it works now... sorry for the wrong diagnostics

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 15 13:40:28 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 13:40:28 -0600
Subject: [LLVMbugs] [Bug 3589] New: clang with -O2 generates some strange
	"puts" call
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3589

           Summary: clang with -O2 generates some strange "puts" call
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

when compiling the attached with -O2 there is a strange call to "puts"
generated which is not generated at -O0

lev /tmp$ /usr/local/bin/ccc -S -O2 -ffreestanding imgact_shell.c && grep puts
imgact_shell.s
ccc: Unknown host 'freebsd', using generic host information.
        call    puts
lev /tmp$ /usr/local/bin/ccc -S -ffreestanding imgact_shell.c && grep puts
imgact_shell.s
ccc: Unknown host 'freebsd', using generic host information.
lev /tmp$ 

there is no such thing as "puts" in the source:

lev /tmp$ grep -c puts imgact_shell.c 
0

there is similar case with "putchar" but I dont have a test case for that. this
is likely coming from llvm optimizations or something, feel free to reassign

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 15 14:00:43 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 14:00:43 -0600
Subject: [LLVMbugs] [Bug 3589] clang with -O2 generates some strange "puts"
	call
In-Reply-To: 
Message-ID: <200902152000.n1FK0hNZ020235@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3589

Daniel Dunbar  changed:

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

--- Comment #3 from Daniel Dunbar   2009-02-15 14:00:43 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012401.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 cs.uiuc.edu  Sun Feb 15 15:07:23 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 15:07:23 -0600
Subject: [LLVMbugs] [Bug 3579] different clang -E from gcc -E on this file
In-Reply-To: 
Message-ID: <200902152107.n1FL7N2d022115@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3579

Chris Lattner  changed:

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

--- Comment #11 from Chris Lattner   2009-02-15 15:07:23 ---
Fixed:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012403.html
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090209/012404.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 cs.uiuc.edu  Sun Feb 15 15:47:05 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 15:47:05 -0600
Subject: [LLVMbugs] [Bug 3590] New: basicaa/aa/memcpyopt bug causing
 llvm-gcc build failure
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3590

           Summary: basicaa/aa/memcpyopt bug causing llvm-gcc build failure
           Product: libraries
           Version: 1.0
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Global Analyses
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2563)
 --> (http://llvm.org/bugs/attachment.cgi?id=2563)
testcase, crashes with llvm-as < t2.ll | llc

this crashes, and is preventing llvm-gcc from building:

$ opt bugpoint-reduced-function.bc -memcpyopt

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 15 16:05:54 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 16:05:54 -0600
Subject: [LLVMbugs] [Bug 3590] basicaa/aa/memcpyopt bug causing llvm-gcc
	build failure
In-Reply-To: 
Message-ID: <200902152205.n1FM5sE4024191@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3590

Chris Lattner  changed:

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

--- Comment #3 from Chris Lattner   2009-02-15 16:05:54 ---
I apparently had stale .o files or something, clean rebuild 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 cs.uiuc.edu  Sun Feb 15 18:01:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 15 Feb 2009 18:01:24 -0600
Subject: [LLVMbugs] [Bug 3583] Correctness bug in commit r64314,
 trunk/lib/CodeGen/SelectionDAG  /TargetLowering.cpp
In-Reply-To: 
Message-ID: <200902160001.n1G01O7G028148@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3583

Dan Gohman  changed:

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

--- Comment #1 from Dan Gohman   2009-02-15 18:01:23 ---
Thanks for spotting this! I've commited a fix here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090209/073806.html

I didn't address the tricky cases you describe; this code is
currently used primarily to help enable use of the BT
instruction on x86, and this only requires handling simple
cases. If someone finds a need for handling trickier cases,
the code can be extended.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 03:27:43 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 03:27:43 -0600
Subject: [LLVMbugs] [Bug 3592] New: GCC warning is clang error,
 implicitly defined functions and different signature in the source
 file.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3592

           Summary: GCC warning is clang error, implicitly defined functions
                    and different signature in the source file.
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bolzoni at cs.unipr.it
                CC: llvmbugs at cs.uiuc.edu, bagnara at cs.unipr.it

Here a test case:
------>
static void* malloc(int);
static void* malloc(int size) {
  return ((void*)0); /*do not use heap in this file*/
}
------<
until I do not include stdlib.h we have:
$ gcc -c mall.c
mall.c:2: warning: conflicting types for built-in function ???malloc???
$ clang mall.c 
mall.c:2:14: error: conflicting types for 'malloc'
static void* malloc(int);
             ^
mall.c:2:14: note: 'malloc' was implicitly declared here with type 'void
*(unsigned long)'
2 diagnostics generated.
$ 

While clang has this behaviour since long time, it seems sometime after r64376
the signature of malloc changed to void*(*)(unsigned long) and the linux-2.4.37
kernel can not compile any more for this 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 cs.uiuc.edu  Mon Feb 16 11:20:45 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 11:20:45 -0600
Subject: [LLVMbugs] [Bug 3593] New: LLVM optimizes infinite loop to
 return^W^W^W^W^Wruns infinite loops in less than 6 seconds ?
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3593

           Summary: LLVM optimizes infinite loop to return^W^W^W^W^Wruns
                    infinite loops in less than 6 seconds?
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: penguinofthegods at gmail.com
                CC: llvmbugs at cs.uiuc.edu

>From the web demo:

void test(void) { test(); }

Output from LLVM disassembler

; ModuleID = '/tmp/webcompile/_8177_0.bc'
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"
target triple = "i386-pc-linux-gnu"

define void @test() nounwind readnone {
entry:
        ret void
}

Not sure where to put this report, sorry.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 11:59:00 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 11:59:00 -0600
Subject: [LLVMbugs] [Bug 2892] crash on redefinition of builtin
In-Reply-To: 
Message-ID: <200902161759.n1GHx0oX010197@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2892

Douglas Gregor  changed:

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

--- Comment #4 from Douglas Gregor   2009-02-16 11:59:00 ---
Fixed here:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012435.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 cs.uiuc.edu  Mon Feb 16 12:01:04 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 12:01:04 -0600
Subject: [LLVMbugs] [Bug 3541] improve builtin function handling
In-Reply-To: 
Message-ID: <200902161801.n1GI14NU010370@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3541

Douglas Gregor  changed:

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

--- Comment #16 from Douglas Gregor   2009-02-16 12:01:04 ---
This is now implemented in Clang, with -ffreestanding support to turn off the
builtin C library 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 cs.uiuc.edu  Mon Feb 16 12:21:07 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 12:21:07 -0600
Subject: [LLVMbugs] [Bug 3588] merging with K&R declaration mismatch with gcc
In-Reply-To: 
Message-ID: <200902161821.n1GIL7I0011166@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3588

Douglas Gregor  changed:

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

--- Comment #1 from Douglas Gregor   2009-02-16 12:21:06 ---
Fixed here:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012437.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 cs.uiuc.edu  Mon Feb 16 12:29:35 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 12:29:35 -0600
Subject: [LLVMbugs] [Bug 2719] [sema] missing warning on useless typedef
In-Reply-To: 
Message-ID: <200902161829.n1GITZEn011485@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2719

Douglas Gregor  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dgregor at apple.com
             Status|NEW                         |RESOLVED
         Resolution|                            |WORKSFORME

--- Comment #1 from Douglas Gregor   2009-02-16 12:29:34 ---
We now get a warning for this:

/Users/dgregor/foo.c:1:1: error: declaration does not declare anything
typedef struct S {
^~~~~~~

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 12:36:49 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 12:36:49 -0600
Subject: [LLVMbugs] [Bug 3592] GCC warning is clang error,
 implicitly defined functions and different signature in the source
 file.
In-Reply-To: 
Message-ID: <200902161836.n1GIanPR011721@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3592

Douglas Gregor  changed:

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

--- Comment #1 from Douglas Gregor   2009-02-16 12:36:48 ---
This has been fixed in r64639. We now:

  - Warn when a builtin is redeclared with an incompatible signature
  - Error when a builtin is defined (if used as a hosted implementation)
  - Suppress the generation of builtins when the option -ffreestanding is
provided

I assume that the Linux kernel is building with -ffreestanding?

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 12:38:10 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 12:38:10 -0600
Subject: [LLVMbugs] [Bug 2586] clang is less permissive than gcc in
 accepting function redefinitions
In-Reply-To: 
Message-ID: <200902161838.n1GIcAbe011771@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2586

Chris Lattner  changed:

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

--- Comment #2 from Chris Lattner   2009-02-16 12:38:10 ---
i agree.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 13:29:13 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 13:29:13 -0600
Subject: [LLVMbugs] [Bug 3230] sched creates crazy register pressure +
	spilling
In-Reply-To: 
Message-ID: <200902161929.n1GJTDJC014222@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3230

Evan Cheng  changed:

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

--- Comment #1 from Evan Cheng   2009-02-16 13:29:13 ---
This is fixed recently.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 13:35:13 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 13:35:13 -0600
Subject: [LLVMbugs] [Bug 2825] cannot codegen __builtin_va_arg
In-Reply-To: 
Message-ID: <200902161935.n1GJZDS8014563@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2825

Daniel Dunbar  changed:

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

--- Comment #7 from Daniel Dunbar   2009-02-16 13:35:13 ---
va_arg on x86_64 is 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 cs.uiuc.edu  Mon Feb 16 14:12:35 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 14:12:35 -0600
Subject: [LLVMbugs] [Bug 3594] New: clang: ccc -fPIC/-DPIC not working on
	x86-64
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3594

           Summary: clang: ccc -fPIC/-DPIC not working on x86-64
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Now that va-arg has been fixed on x86-64,  I've tested clang and got a linker
error, here's a reduced testcase:

$ ccc -fPIC -DPIC -c foo.c -o foo.o
$ ccc -shared foo.o -o foo.so
ccc: Unknown host 'linux', using generic host information.
/usr/bin/ld: foo.o: relocation R_X86_64_32S against `a' can not be used when
making a shared object; recompile with -fPIC
foo.o: could not read symbols: Bad value
collect2: ld returned 1 exit status

That error message is what you usually get with gcc if you DON'T use PIC.
But I used PIC with ccc, and I still get the error.

/* foo.c */
int a;
int foo()
{
    return 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 cs.uiuc.edu  Mon Feb 16 14:26:34 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 14:26:34 -0600
Subject: [LLVMbugs] [Bug 3595] New: instcombine miscompilation
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3595

           Summary: instcombine miscompilation
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2565)
 --> (http://llvm.org/bugs/attachment.cgi?id=2565)
bugpoint-passinput.bc

Using SVN r64664 of clang to compile ClamAV, I get 1 error in 'make check' of
ClamAV:

99%: Checks: 235, Failures: 1, Errors: 0
../../../clamav-devel/unit_tests/check_str.c:61:F:cli_unescape:test_unescape_hex:0:
cli_unescape truncated
FAIL: check_clamav

Bugpoint has reduced this to be an instcombine bug.

Here's bugpoint's output:
*** Found miscompiling pass: -instcombine
Emitted bitcode to 'bugpoint-passinput.bc'

*** You can reproduce the problem with: opt bugpoint-passinput.bc -instcombine
Checking to see if the program is misoptimized when these functions are run
through the pass: cli_unescape cli_hex2int output_utf8 main
  Optimizing functions being tested: done.
  Checking to see if the merged program executes correctly: 
yup.

*** The following functions are being miscompiled:  cli_unescape cli_hex2int
output_utf8 main
Extracted a loop from the breaking portion of the program.
  Testing after loop extraction:
  Optimizing functions being tested: done.
  Checking to see if the merged program executes correctly: 
yup.
*** Loop extraction masked the problem.  Undoing.
Checking to see if the program is misoptimized when all blocks are extracted.
  Optimizing functions being tested: done.
  Checking to see if the merged program executes correctly: 
yup.
Checking to see if the program is misoptimized when all but these 29 blocks are
extracted: entry for.cond for.body if.then15 lor.lhs.false lor.lhs.false29
lor.lhs.false38 lor.lhs.false47 lor.lhs.false56 if.then65 ...
  Optimizing functions being tested: done.
  Checking to see if the merged program executes correctly: 
yup.
Outputting reduced bitcode files which expose the problem:
  Non-optimized portion: Emitted bitcode to 'bugpoint-tonotoptimize.bc'
  Portion that is input to optimizer: Emitted bitcode to
'bugpoint-tooptimize.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 cs.uiuc.edu  Mon Feb 16 14:31:56 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 14:31:56 -0600
Subject: [LLVMbugs] [Bug 3596] New: doesn't notice that program Aborts
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3596

           Summary: doesn't notice that program Aborts
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu
        Depends on: 3595

Created an attachment (id=2570)
 --> (http://llvm.org/bugs/attachment.cgi?id=2570)
x.bc

+++ This bug was initially created as a clone of Bug #3595 +++

While trying to reduce a testcase for PR3595, I stumbled upon a bugpoint bug.

If I don't print anything in my testcase, but only return 0 / abort(), then
bugpoint says that the failing program matches the reference output!

bugpoint  -simplifycfg -domtree -domfrontier -mem2reg -instcombine -simplifycfg
-simplify-libcalls -instcombine -jump-threading -simplifycfg -domtree
-domfrontier -scalarrepl -instcombine -break-crit-edges -condprop -tailcallelim
-simplifycfg -reassociate -domtree -loops -loopsimplify -domfrontier
-scalar-evolution -lcssa -loop-rotate -licm -lcssa -loop-unswitch
-scalar-evolution -lcssa -loop-index-split -instcombine -scalar-evolution
-lcssa -indvars -scalar-evolution -lcssa -loop-deletion -instcombine -memdep
-gvn -memdep -memcpyopt -sccp -instcombine -break-crit-edges -condprop -domtree
-memdep -dse -adce -simplifycfg -preverify -domtree -verify x.bc
Read input file      : 'x.bc'
*** All input ok
Running selected passes on program to test for crash: Success!
Initializing execution environment: Found gcc: /usr/lib/ccache/gcc
Running the code generator to test for a crash: 
Generating reference output from raw program: Reference
output is: bugpoint.reference.out

*** Checking the code generator...

*** Debugging miscompilation!
Checking to see if '' compile correctly:  yup.
Checking to see if '-simplifycfg -domtree -domfrontier -mem2reg -instcombine
-simplifycfg -simplify-libcalls -instcombine -jump-threading -simplifycfg
-domtree -domfrontier -scalarrepl -instcombine -break-crit-edges -condprop
-tailcallelim -simplifycfg -reassociate -domtree -loops -loopsimplify
-domfrontier -scalar-evolution -lcssa -loop-rotate -licm -lcssa -loop-unswitch
-scalar-evolution -lcssa -loop-index-split -instcombine -scalar-evolution
-lcssa -indvars -scalar-evolution -lcssa -loop-deletion -instcombine -memdep
-gvn -memdep -memcpyopt -sccp -instcombine -break-crit-edges -condprop -domtree
-memdep -dse -adce -simplifycfg -preverify -domtree -verify' compile correctly:
 yup.
*** Optimized program matches reference output!  No problem detected...
bugpoint can't help you with your problem!

Once I added a printf("OK"); to the testcase, bugpoint was able to reduce 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 cs.uiuc.edu  Mon Feb 16 16:36:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 16:36:29 -0600
Subject: [LLVMbugs] [Bug 3597] New: linux 2.4.37 compilation fails in linux.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3597

           Summary: linux 2.4.37 compilation fails in linux.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: parser
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bolzoni at cs.unipr.it
                CC: llvmbugs at cs.uiuc.edu

in the file ia32_ioctl.c a clang assertions fail:

$ gcc -w -c ia32_ioctl.i
$ clang -w ia32_ioctl.i 
clang: Builtins.cpp:71: std::string
clang::Builtin::Context::getHeaderName(unsigned int) const: Assertion `NameEnd
&& "Missing ':' after header name"' failed.
0   clang     0x0000000000e03321
1   libc.so.6 0x00007f5bd7a37150
2   libc.so.6 0x00007f5bd7a370c5 gsignal + 53
3   libc.so.6 0x00007f5bd7a385e3 abort + 387
4   libc.so.6 0x00007f5bd7a300e9 __assert_fail + 233
5   clang     0x00000000008bfe0c
6   clang     0x000000000083189d
clang::Sema::LazilyCreateBuiltin(clang::IdentifierInfo*, unsigned int,
clang::Scope*, bool, clang::SourceLocation) + 589
7   clang     0x000000000087d28b clang::Sema::LookupName(clang::Scope*,
clang::DeclarationName, clang::Sema::LookupNameKind, bool, bool,
clang::SourceLocation) + 539
8   clang     0x000000000083432b
clang::Sema::ActOnParamDeclarator(clang::Scope*, clang::Declarator&) + 187
9   clang     0x0000000000924099
clang::Parser::ParseFunctionDeclarator(clang::SourceLocation,
clang::Declarator&, clang::AttributeList*, bool) + 985
10  clang     0x0000000000926180
clang::Parser::ParseDirectDeclarator(clang::Declarator&) + 624
11  clang     0x0000000000920b6a
clang::Parser::ParseDeclaratorInternal(clang::Declarator&, void
(clang::Parser::*)(clang::Declarator&)) + 186
12  clang     0x000000000091c206
clang::Parser::ParseDeclarationOrFunctionDefinition(llvm::SmallVector*) + 710
13  clang     0x000000000091c860 clang::Parser::ParseExternalDeclaration() +
112
14  clang     0x000000000091cbe6 clang::Parser::ParseTopLevelDecl(void*&) + 22
15  clang     0x0000000000822ee3 clang::ParseAST(clang::Preprocessor&,
clang::ASTConsumer*, clang::TranslationUnit*, bool) + 163
16  clang     0x00000000005b1831
17  clang     0x00000000005b6bcc main + 2060
18  libc.so.6 0x00007f5bd7a23546 __libc_start_main + 230
19  clang     0x0000000000571159
Aborted
$ 

I enclosed the preprocessed file.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 18:14:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 18:14:29 -0600
Subject: [LLVMbugs] [Bug 3399] lsr crash
In-Reply-To: 
Message-ID: <200902170014.n1H0ETIA026148@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3399

Evan Cheng  changed:

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

--- Comment #4 from Evan Cheng   2009-02-16 18:14:28 ---
Fixed.
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090216/073825.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 cs.uiuc.edu  Mon Feb 16 20:03:13 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 20:03:13 -0600
Subject: [LLVMbugs] [Bug 3598] New: [clang codegen perf] Don' t pass
 _Complex with byval if you can avoid it
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3598

           Summary: [clang codegen perf] Don't pass _Complex with byval if
                    you can avoid it
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Keywords: code-quality
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu

clang codegens a call to cacos with byval, which pins an alloca to the stack:

_Complex double foo2(_Complex double x) {
  return cacosl(x);
}

This causes us to compile this to:

_foo2:
        subl    $60, %esp
        movsd   76(%esp), %xmm0
        movsd   68(%esp), %xmm1
        movsd   %xmm1, 24(%esp)
        movsd   %xmm0, 32(%esp)
        movl    36(%esp), %eax
        movl    %eax, 16(%esp)
        movl    32(%esp), %eax
        movl    %eax, 12(%esp)
        movl    28(%esp), %eax
        movl    %eax, 8(%esp)
        movl    24(%esp), %eax
        movl    %eax, 4(%esp)
        leal    40(%esp), %eax
        movl    %eax, (%esp)
        call    _cacos
        subl    $4, %esp
        movsd   48(%esp), %xmm0
        movsd   40(%esp), %xmm1
        movl    64(%esp), %eax
        movsd   %xmm1, (%eax)
        movsd   %xmm0, 8(%eax)
        addl    $60, %esp
        ret     $4

llvm-gcc doesn't do 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 cs.uiuc.edu  Mon Feb 16 20:56:20 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 20:56:20 -0600
Subject: [LLVMbugs] [Bug 3598] [clang codegen perf] Don' t pass _Complex
 with byval if you can avoid it
In-Reply-To: 
Message-ID: <200902170256.n1H2uKjR000318@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3598

Daniel Dunbar  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |daniel at zuster.org
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Daniel Dunbar   2009-02-16 20:56:20 ---

*** This bug has been marked as a duplicate of bug 3474 ***

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 21:04:54 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 21:04:54 -0600
Subject: [LLVMbugs] [Bug 3501] clang x86-64 ABI bug with bit-fields
In-Reply-To: 
Message-ID: <200902170304.n1H34sWm000621@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3501

Daniel Dunbar  changed:

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

--- Comment #2 from Daniel Dunbar   2009-02-16 21:04:53 ---
Implemented here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012512.html

--
ddunbar at ozzy:CodeGen$ cat x.c
struct s0 {int a : 5, b : 5;};
unsigned a(struct s0 x) { return x.a; }
ddunbar at ozzy:CodeGen$ gcc -O3 -fomit-frame-pointer -m64 -S -o - x.c | head -10
        .text
        .align 4,0x90
.globl _a
_a:
LFB2:
        sall    $3, %edi
        sarb    $3, %dil
        movsbl  %dil,%eax
        ret
LFE2:
ddunbar at ozzy:CodeGen$ xcc -O3 -fomit-frame-pointer -m64 -S -o - x.c | head -10

        .text
        .align  4,0x90
        .globl  _a
_a:
        shll    $27, %edi
        movl    %edi, %eax
        sarl    $27, %eax
        ret
--

Not sure why you regard this as a tricky case, things like 
--
typedef struct {
  float a;
  int b : 3;
  int c : 3;
  int d : 31;
} T;
--
are much nastier from a classification point of view.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 16 22:36:02 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 22:36:02 -0600
Subject: [LLVMbugs] [Bug 3599] New: Bad-quality code passing union as
	argument on x86_64
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3599

           Summary: Bad-quality code passing union as argument on x86_64
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu

union UGeckoInstruction
{
unsigned hex;
};
unsigned a(union UGeckoInstruction a) {return a.hex;}

Run through clang -emit-llvm -O2 -arch=x86_64, gives the following:
; ModuleID = '-'
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-f80:128:128"
target triple = "x86_64-pc-linux-gnu"

define i32 @a(i64) nounwind readnone {
entry:
        %tmp5 = trunc i64 %0 to i8              ;  [#uses=1]
        %tmp7 = lshr i64 %0, 8          ;  [#uses=1]
        %tmp8 = trunc i64 %tmp7 to i8           ;  [#uses=1]
        %tmp10 = lshr i64 %0, 16                ;  [#uses=1]
        %tmp11 = trunc i64 %tmp10 to i8         ;  [#uses=1]
        %tmp13 = lshr i64 %0, 24                ;  [#uses=1]
        %tmp14 = trunc i64 %tmp13 to i8         ;  [#uses=1]
        %1 = zext i8 %tmp5 to i32               ;  [#uses=1]
        %2 = zext i8 %tmp8 to i32               ;  [#uses=1]
        %3 = shl i32 %2, 8              ;  [#uses=1]
        %4 = zext i8 %tmp11 to i32              ;  [#uses=1]
        %5 = shl i32 %4, 16             ;  [#uses=1]
        %6 = zext i8 %tmp14 to i32              ;  [#uses=1]
        %7 = shl i32 %6, 24             ;  [#uses=1]
        %8 = or i32 %7, %1              ;  [#uses=1]
        %9 = or i32 %8, %5              ;  [#uses=1]
        %10 = or i32 %9, %3             ;  [#uses=1]
        ret i32 %10
}

This is rather silly.  I'm not sure whether to blame clang or the LLVM
optimizers for this result, 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 cs.uiuc.edu  Mon Feb 16 23:12:20 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 16 Feb 2009 23:12:20 -0600
Subject: [LLVMbugs] [Bug 3593] LLVM optimizes infinite loop to
 return^W^W^W^W^Wruns infinite loops in less than 6 seconds ?
In-Reply-To: 
Message-ID: <200902170512.n1H5CKjt005136@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3593

Nick Lewycky  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |nicholas at mxc.ca
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #2 from Nick Lewycky   2009-02-16 23:12:19 ---

*** This bug has been marked as a duplicate of bug 965 ***

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 00:52:56 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 00:52:56 -0600
Subject: [LLVMbugs] [Bug 3215] need a new "diagnostic continuation" kind,
 to silence secondary info when the primary info is silenced
In-Reply-To: 
Message-ID: <200902170652.n1H6quN2009216@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3215

Chris Lattner  changed:

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

--- Comment #5 from Chris Lattner   2009-02-17 00:52:56 ---
Fixed:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012520.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 cs.uiuc.edu  Tue Feb 17 01:17:26 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 01:17:26 -0600
Subject: [LLVMbugs] [Bug 2788] Track #line and # [0-9]+ directives in
	SourceLocation
In-Reply-To: 
Message-ID: <200902170717.n1H7HQii010461@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2788

Chris Lattner  changed:

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

--- Comment #2 from Chris Lattner   2009-02-17 01:17:26 ---
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 cs.uiuc.edu  Tue Feb 17 04:58:25 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 04:58:25 -0600
Subject: [LLVMbugs] [Bug 3601] New: loop passes vs call graph
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3601

           Summary: loop passes vs call graph
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Loop Optimizer
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jay.foad at antixlabs.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2572)
 --> (http://llvm.org/bugs/attachment.cgi?id=2572)
test case

This came up on bug 3367. If you apply my suggested fix for that bug
(http://llvm.org/bugs/attachment.cgi?id=2424) and then run this on the test
case attached here:

$ llvm-as -f -o - mine.ll | opt -f -o /dev/null -inline -loop-rotate

I get:

opt: /home/foad/svn/llvm-project/llvm/trunk/lib/Analysis/IPA/CallGraph.cpp:306:
void llvm::CallGraphNode::replaceCallSite(llvm::CallSite, llvm::CallSite):
Assertion `I != CalledFunctions.end() && "Cannot find callsite to replace!"'
failed.
0   opt       0x0863d6ae
1   opt       0x0863d9f0
[...]
Aborted

I think the problem is that -loop-rotate doesn't preserve the call graph (which
is OK, because it doesn't claim to), but the CallGraphSCCPassManager runs it
interleaved with other CallGraphSCCPasses, which assume that the call graph is
always up to date.

I raised the issue on LLVMdev. Duncan summarised:

> given the callgraph F -> G, the pass manager currently does the following:
> run inliner on G, run loop passes on G, run inliner on F, run loop
> passes on F.  Presumably what is happening is this: the loop passes change
> the functions that G calls (but don't update the callgraph).  Now the
> inliner visits F and decides to inline G into F.  When it does this, it
> presumably merges the callgraph info for G (i.e. what G calls) into that of
> F.  But this info is wrong, so F ends up having invalid callgraph info which
> at some point causes trouble.

and suggested a way to fix it:

> I think what should happen is: if a SCC pass (eg: inline) is followed
> by function passes that preserve the callgraph, then it should schedule
> them together like above.  However if the SCC pass is followed by a
> function pass that does not preserve the callgraph then it should be
> scheduled entirely after the SCC pass.

Devang said:

> This will defeat the goal of applying loop transformations before
> inlining leaf functions. Note, Loop transformations are not aware of
> call graph. They do not claim to preserve call graph. However, loop
> passes are run by a loop pass manager (LPPassManager) which is itself
> a function pass. The pass manager is not doing the right thing  here
> because LPPassManager is incorrectly claiming to preserve call graph.
> The right approach is to teach LPPassManager to really preserve call
> graph.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 09:46:49 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 09:46:49 -0600
Subject: [LLVMbugs] [Bug 3597] linux 2.4.37 compilation fails in linux.
In-Reply-To: 
Message-ID: <200902171546.n1HFknuw006674@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3597

Douglas Gregor  changed:

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

--- Comment #2 from Douglas Gregor   2009-02-17 09:46:48 ---
The code that triggered this issue is gone, so the same assertion can't crop
up. Closing this unless we see some related problem pop 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 cs.uiuc.edu  Tue Feb 17 10:51:16 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 10:51:16 -0600
Subject: [LLVMbugs] [Bug 3602] New: ccc: doesn't pass -Wp,-MD, to clang
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3602

           Summary: ccc: doesn't pass -Wp,-MD, to clang
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

I tried to build the linux kernel with clang, it gets this far:

ccc -Wp,-MD,kernel/.bounds.s.d  -nostdinc -isystem include -Iinclude 
-I/home/edwin/builds/linux-2.6/arch/x86/include -include
include/linux/autoconf.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes
-Wno-trigraphs -fno-strict-aliasing -fno-common
-Werror-implicit-function-declaration -O2 -m64 -march=core2 -mno-red-zone
-mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -DCONFIG_AS_CFI=1
-DCONFIG_AS_CFI_SIGNAL_FRAME=1 -pipe -Wno-sign-compare
-fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow
-Iarch/x86/include/asm/mach-default -Wframe-larger-than=2048
-fno-stack-protector -fno-omit-frame-pointer -fno-optimize-sibling-calls
-Wdeclaration-after-statement -Wno-pointer-sign   -D"KBUILD_STR(s)=#s"
-D"KBUILD_BASENAME=KBUILD_STR(bounds)"  -D"KBUILD_MODNAME=KBUILD_STR(bounds)" 
-fverbose-asm -S -o kernel/bounds.s kernel/bounds.c

ccc: include: linker input file unused when -S is present
fixdep: kernel/.bounds.s.d: No such file or directory
make[1]: *** [kernel/bounds.s] Error 2

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 13:00:41 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 13:00:41 -0600
Subject: [LLVMbugs] [Bug 3603] New: Finish dependency file support
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3603

           Summary: Finish dependency file support
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang is still missing various things needed for full dependency file
generation support.

1. Change -M* options to be sane and force driver to convert -M* to the sane
set.

2. Support -M, -MG, -MM, -MQ.

3. Some corner cases differ vs gcc. For example:
--
ddunbar at lordcrumb:tmp$ rm -f foo && xcc -S -MD -MF foo -m32 -o y -x c
/dev/null; cat foo
y.o: /dev/null
ddunbar at lordcrumb:tmp$ rm -f foo && gcc -S -MD -MF foo -m32 -o y -x c
/dev/null; cat foo
y: /dev/null

ddunbar at lordcrumb:tmp$ rm -f foo && xcc -E -MD -MF foo -m32 -o y -x c
/dev/null; cat foo
cat: foo: No such file or directory
ddunbar at lordcrumb:tmp$ rm -f foo && gcc -E -MD -MF foo -m32 -o y -x c
/dev/null; cat foo
null.o: /dev/null
--

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 14:48:57 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 14:48:57 -0600
Subject: [LLVMbugs] [Bug 3599] Bad-quality code passing union as argument on
	x86_64
In-Reply-To: 
Message-ID: <200902172048.n1HKmv6k018397@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3599

Chris Lattner  changed:

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

--- Comment #7 from Chris Lattner   2009-02-17 14:48:56 ---
Ah, with -std-compile-opts and that patch, this is completely simplified. 
Instcombine needs CSE to see the equivalence of the truncates.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 15:29:10 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 15:29:10 -0600
Subject: [LLVMbugs] [Bug 3606] New: undefined symbol with -O2
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3606

           Summary: undefined symbol with -O2
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

hi.. I am getting this:

witten /tmp# ccc -c -O0 acpi_wakeup.c
ccc: Unknown host 'freebsd', using generic host information.
witten /tmp# nm acpi_wakeup.o | grep r_ds
00000030 b r_ds
witten /tmp# ccc -c -O1 acpi_wakeup.c
ccc: Unknown host 'freebsd', using generic host information.
witten /tmp# nm acpi_wakeup.o | grep r_ds
         U r_ds
witten /tmp# 

when the optimization is turned on the symbol is optimized out, this is most
likely llvm bug.

test case attached

this prevents i386 freebsd kernel from linking

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 15:34:27 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 15:34:27 -0600
Subject: [LLVMbugs] [Bug 3607] New: crash on bitfield with incomplete type
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3607

           Summary: crash on bitfield with incomplete type
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang crashes on this
--
enum e0;
struct s0 {
  enum e0 f : 1;
};
--

It would be nice to get gcc's extra warning here about width:
--
ddunbar at lordcrumb:Sema$ gcc -fsyntax-only boom.c
boom.c:3: warning: 'f' is narrower than values of its type
boom.c:3: error: field 'f' has incomplete type
--

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 17 16:09:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 16:09:14 -0600
Subject: [LLVMbugs] [Bug 3606] undefined symbol with -O2
In-Reply-To: 
Message-ID: <200902172209.n1HM9E5C022113@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3606

Chris Lattner  changed:

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

--- Comment #4 from Chris Lattner   2009-02-17 16:09:14 ---
Yes, this is definitely a bug in the source: you have to mark things used by an
inline asm as attribute(used).  You're just lucky it works with 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 cs.uiuc.edu  Tue Feb 17 19:37:16 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 17 Feb 2009 19:37:16 -0600
Subject: [LLVMbugs] [Bug 3499] llvmc dejagnu tests require llvm-g++ in the
	path
In-Reply-To: 
Message-ID: <200902180137.n1I1bGBX021987@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3499

Tanya Lattner  changed:

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

--- Comment #5 from Tanya Lattner   2009-02-17 19:37:15 ---
Verified that this is 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 cs.uiuc.edu  Wed Feb 18 01:19:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 01:19:29 -0600
Subject: [LLVMbugs] [Bug 3609] New: diagnostics: [missing terminating "
 character] should say > when needed
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3609

           Summary: diagnostics: [missing terminating " character] should
                    say > when needed
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: alexei.svitkine at gmail.com
                CC: llvmbugs at cs.uiuc.edu

When the #include directive uses the  form, but is not properly
terminated, e.g.:

#include  character

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 07:51:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 07:51:14 -0600
Subject: [LLVMbugs] [Bug 3610] New: VAARG not implemented for soft floats.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3610

           Summary: VAARG not implemented for soft floats.
           Product: new-bugs
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rich at pennware.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2576)
 --> (http://llvm.org/bugs/attachment.cgi?id=2576)
.ll file using vaarg

[~/ellcc/ellcc] main% llvm-as -f t0056.ll         
[~/ellcc/ellcc] main% llc t0056.bc 
SoftenFloatResult #0: 0x9f55c44: f32,ch = vaarg 0x9f55b3c, 0x9f5571c, 0x9f55bc0
llc:
/home/rich/llvm-trunk-new/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp:54:
void llvm::DAGTypeLegalizer::SoftenFloatResult(llvm::SDNode*, unsigned int):
Assertion `0 && "Do not know how to soften the result of this operator!"'
failed.                                                                         
0   llc       0x08c970ae                                                        
1   llc       0x08c973f0                                                        
2             0x00110400 __kernel_sigreturn + 0                                 
3   libc.so.6 0x00bcce28 abort + 392                                            
4   libc.so.6 0x00bc440e __assert_fail + 238                                    
5   llc       0x089dd961
6   llc       0x089c64ea
7   llc       0x089c6f69 llvm::SelectionDAG::LegalizeTypes() + 45
8   llc       0x0895ea4a llvm::SelectionDAGISel::CodeGenAndEmitDAG() + 1180
9   llc       0x089612d4
llvm::SelectionDAGISel::SelectBasicBlock(llvm::BasicBlock*,
llvm::ilist_iterator,
llvm::ilist_iterator) + 598
10  llc       0x08961c87
llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function&,
llvm::MachineFunction&, llvm::MachineModuleInfo*, llvm::DwarfWriter*,
llvm::TargetInstrInfo const&) + 2219
11  llc       0x08962963 llvm::SelectionDAGISel::runOnFunction(llvm::Function&)
+ 841
12  llc       0x08c259f5 llvm::FPPassManager::runOnFunction(llvm::Function&) +
249
13  llc       0x08c2646a llvm::FunctionPassManagerImpl::run(llvm::Function&) +
116
14  llc       0x08c265cb llvm::FunctionPassManager::run(llvm::Function&) + 137
15  llc       0x083d0d22 main + 2465
16  libc.so.6 0x00bb66e5 __libc_start_main + 229
17  llc       0x083cf441
Abort
[~/ellcc/ellcc] 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 cs.uiuc.edu  Wed Feb 18 11:44:38 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 11:44:38 -0600
Subject: [LLVMbugs] [Bug 3611] New: AsmWriter produces wrong output for
 functions with no name
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3611

           Summary: AsmWriter produces wrong output for functions with no
                    name
           Product: new-bugs
           Version: unspecified
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu

Same goes for global variables, it seems.

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

From baldrick at free.fr  Wed Feb 18 11:50:16 2009
From: baldrick at free.fr (Duncan Sands)
Date: Wed, 18 Feb 2009 18:50:16 +0100
Subject: [LLVMbugs] [Bug 3611] New: AsmWriter produces wrong output for
	functions with no name
In-Reply-To: 
References: 
Message-ID: <200902181850.17665.baldrick@free.fr>

Hit the "commit" button too soon!  Testcase and comments
in the PR.

Ciao,

Duncan.

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 12:16:41 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 12:16:41 -0600
Subject: [LLVMbugs] [Bug 3554] [2.5] SPEC/CFP2006/447.dealII LLC Regression
In-Reply-To: 
Message-ID: <200902181816.n1IIGf8v022366@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3554

Tanya Lattner  changed:

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

--- Comment #16 from Tanya Lattner   2009-02-18 12:16:40 ---
Verified that this is fixed. It is not related to 3553.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 12:43:03 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 12:43:03 -0600
Subject: [LLVMbugs] [Bug 3612] New: printf(printf);
 leads to invalid codegen and no diagnosis
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3612

           Summary: printf(printf); leads to invalid codegen and no
                    diagnosis
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: alexei.svitkine at gmail.com
                CC: llvmbugs at cs.uiuc.edu

This is a new bug. It used to generated expected results a week ago
(diagnostics and expected code gen).

Compiling the following code:

#include 

int foo(void) {
  return printf(printf);
}

Results in the following codegen'd function (and no diagnostics from clang!):

define i32 @__ccons_anon1() nounwind {
entry:
        %retval = alloca i32            ;  [#uses=2]
        %call = call i32 (i8*, ...)* @printf(i8* bitcast (i32 (i8*, ...)*
@printf1 to i8*))             ;  [#uses=1]
        store i32 %call, i32* %retval
        %0 = load i32* %retval          ;  [#uses=1]
        ret i32 %0
}

When this function is then attempted to be executed with llvm, it results in
the following assert:

ERROR: Program used external function 'printf1' which could not be resolved!

The expected behaviour (which used to be the case 1 week ago or so) is:

First of all, diagnostics produced due to incorrect pointer type being passed.
Second of all, the calling of foo() succeeding and printf treating a pointer to
itself as a string and printing some garbage (until it encounters a zero).

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 15:01:26 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 15:01:26 -0600
Subject: [LLVMbugs] [Bug 3613] New: struct members optimized out ?
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3613

           Summary: struct members optimized out ?
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Basic
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2578)
 --> (http://llvm.org/bugs/attachment.cgi?id=2578)
Pre-processed devfs_vnops.c

In the attached file devfs_ops_f.fo_ioctl is null when compiled with clang,
with llvm-gcc it correctly points to devfs_ioctl_f().

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 15:49:05 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 15:49:05 -0600
Subject: [LLVMbugs] [Bug 3611] AsmWriter produces wrong output for functions
 with no name
In-Reply-To: 
Message-ID: <200902182149.n1ILn5np012451@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3611

Chris Lattner  changed:

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

--- Comment #3 from Chris Lattner   2009-02-18 15:49:04 ---
Fixed:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090216/073952.html

Please verify the global case works as well.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 19:18:02 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 19:18:02 -0600
Subject: [LLVMbugs] [Bug 3616] New: Bug in BranchFolding.cpp:OptimizeBlock
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3616

           Summary: Bug in BranchFolding.cpp:OptimizeBlock
           Product: libraries
           Version: 2.4
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         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=2579)
 --> (http://llvm.org/bugs/attachment.cgi?id=2579)
Rar file including cfg graphs, bitcode and possible patch

I???ve ran across an issue in BranchFolding.cpp where it is incorrectly folding
a branch to the wrong fallthrough location. This is in LLVM 2.4 and seems to be
in 2.5 also.
The code in question is:
void BranchFolder::OptimizeBlock(MachineBasicBlock *MBB) {
  MachineFunction::iterator FallThrough = MBB;
  ++FallThrough;

  // If this block is empty, make everyone use its fall-through, not the block
  // explicitly.  Landing pads should not do this since the landing-pad table
  // points to this block.
  if (MBB->empty() && !MBB->isLandingPad()) {
    // Dead block?  Leave for cleanup later.
    if (MBB->pred_empty()) return;

    if (FallThrough == MBB->getParent()->end()) {
      // TODO: Simplify preds to not branch here if possible!
    } else {
      // Rewrite all predecessors of the old block to go to the fallthrough
      // instead.
      while (!MBB->pred_empty()) {
        MachineBasicBlock *Pred = *(MBB->pred_end()-1);
        Pred->ReplaceUsesOfBlockWith(MBB, FallThrough);
      }

      // If MBB was the target of a jump table, update jump tables to go to the
      // fallthrough instead.
      MBB->getParent()->getJumpTableInfo()->
        ReplaceMBBInJumpTables(MBB, FallThrough);
      MadeChange = true;
    }
    return;
  }

The problem with this section of code is that FallThrough is not guaranteed to
be a successor of MBB or even a descendent of MBB.
The bitcode I???ve attached is a case where there are 5 basic blocks, where the
first four end with conditional branches to an early return, as specified with
initial.dot. 
TailMergeBlocks in BranchFolding::runOnMachineFunction merges the 4 early
return blocks to a single basic block and numbers renumbers them, as specified
with tailmerge.dot.
When it runs optimize block on the if.end14 block, it enters the above segment
of code, removing it and replacing it with FallThrough, which is NOT its
successor block and links two blocks changing the structure of the program as
shown in Optimizeblock.dot.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 22:15:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 22:15:08 -0600
Subject: [LLVMbugs] [Bug 3617] New: llvm-gcc and clang can' t compile
 negative shift in O3 correctly
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3617

           Summary: llvm-gcc and clang can't compile negative shift in O3
                    correctly
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Keywords: miscompilation
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: dodohack at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2580)
 --> (http://llvm.org/bugs/attachment.cgi?id=2580)
c source file

Source Code: shift.c
=============================
int func(int x, int y) {
  x = x << y;
  return x;
}

int main() {
  int a = 120;
  int b = -2;
  printf("%d\n", func(a,b));
}
=============================
compiled with:
[hwalin at earth UnitTests]$ /opt/llvm-svn/obj/Debug/bin/clang -O=3 --emit-llvm
shift.c
=============================
[hwalin at earth UnitTests]$ cat shift.ll
; ModuleID = 'shift.c'
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"
target triple = "i386-pc-linux-gnu"
@.str = internal constant [4 x i8] c"%d\0A\00"          ; <[4 x i8]*> [#uses=1]

define i32 @func(i32 %x, i32 %y) nounwind readnone {
entry:
        %shl = shl i32 %x, %y           ;  [#uses=1]
        ret i32 %shl
}

define i32 @main(...) nounwind {
entry:
        %call2 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]*
@.str, i32 0, i32 0), i32 undef)             ;  [#uses=0]
        ret i32 undef
}

declare i32 @printf(i8* nocapture, ...) nounwind

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 22:17:47 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 22:17:47 -0600
Subject: [LLVMbugs] [Bug 3617] llvm-gcc and clang can't compile negative
 shift in O3 correctly
In-Reply-To: 
Message-ID: <200902190417.n1J4HliO026266@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3617

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-18 22:17:45 ---
Negative shifts are undefined.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 18 23:37:07 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 18 Feb 2009 23:37:07 -0600
Subject: [LLVMbugs] [Bug 3613] struct members optimized out ?
In-Reply-To: 
Message-ID: <200902190537.n1J5b73o029070@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3613

Daniel Dunbar  changed:

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

--- Comment #2 from Daniel Dunbar   2009-02-18 23:37:07 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012753.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 cs.uiuc.edu  Thu Feb 19 00:31:30 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 00:31:30 -0600
Subject: [LLVMbugs] [Bug 3553] [2.5] FPGrowth LLC Regression
In-Reply-To: 
Message-ID: <200902190631.n1J6VUZk020432@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3553

Chris Lattner  changed:

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

--- Comment #7 from Chris Lattner   2009-02-19 00:31:30 ---
This turned out to be a bug in the program.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 03:12:09 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 03:12:09 -0600
Subject: [LLVMbugs] [Bug 3618] New: misdiagnosis of scalar initializer for
 flexible array
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3618

           Summary: misdiagnosis of scalar initializer for flexible array
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

--
ddunbar at milton:min$ cat /tmp/t.c
struct { int x; int y[]; } a = { 1, 2 };
ddunbar at milton:min$ ccc -fsyntax-only /tmp/t.c
/tmp/t.c:1:37: warning: excess elements in struct initializer
struct { int x; int y[]; } a = { 1, 2 };
                                    ^
1 diagnostic generated.
--

gcc treats "2" as a single initializer for the flexible array, I don't think we
need that level of compatibility but we should at least error on this (vs
miscompiling).

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 05:58:18 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 05:58:18 -0600
Subject: [LLVMbugs] [Bug 3619] New: ConstantArrayType's profile does not
 contain storage class.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3619

           Summary: ConstantArrayType's profile does not contain storage
                    class.
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bolzoni at cs.unipr.it
                CC: llvmbugs at cs.uiuc.edu, bagnara at cs.unipr.it

Test case:
---->
void f(int b[5][static 5]);
int a[5][5];
----<
the latter array type contains the static storage class.

I know this code is invalid for C99 6.7.5.2 (refused by gcc, accepted by
clang,) but once clang will manage the static keyword in the array declaration
correctly (without replacing the array with a pointer) the problem will occur
in this similar code:
---->
void f(int b[static 5][5]);
int a[5][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 cs.uiuc.edu  Thu Feb 19 06:03:10 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 06:03:10 -0600
Subject: [LLVMbugs] [Bug 3534] Add paq8p to llvm testsuite
In-Reply-To: 
Message-ID: <200902191203.n1JC3Aju013225@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3534

T??r??k Edwin  changed:

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

--- Comment #3 from T??r??k Edwin   2009-02-19 06:03:09 ---
Added in SVN r65042.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 06:38:59 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 06:38:59 -0600
Subject: [LLVMbugs] [Bug 3533] Add mafft to testsuite
In-Reply-To: 
Message-ID: <200902191238.n1JCcxEF014476@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3533

T??r??k Edwin  changed:

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

--- Comment #1 from T??r??k Edwin   2009-02-19 06:38:58 ---
Added in SVN r65043.
mafft is BSD licensed, and the datafiles are from the phoronix testsuite which
is GPL-ed.

Testreport:

Program        | GCCAS  Bytecode LLC compile LLC-BETA compile JIT codegen | GCC
    CBE     LLC     LLC-BETA JIT     | GCC/CBE GCC/LLC GCC/LLC-BETA
LLC/LLC-BETA
pairlocalalign | 2.0721 318656   0.3200      *                0.1640      |  
22.92   58.96   26.69 *          28.10 | 0.39    0.86    n/a          n/a

LLVM is about 15% slower.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 06:57:53 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 06:57:53 -0600
Subject: [LLVMbugs] [Bug 3611] AsmWriter produces wrong output for functions
 with no name
In-Reply-To: 
Message-ID: <200902191257.n1JCvrf2015234@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3611

Duncan Sands  changed:

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

--- Comment #4 from Duncan Sands   2009-02-19 06:57:51 ---
Thanks a lot.  Unfortunately globals still don't parse.
Consider

  @0 = global i32 0

Then llvm-as says:

  llvm-as: noname.ll:1:1: expected top-level entity
  @0 = global i32 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 cs.uiuc.edu  Thu Feb 19 11:02:34 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 11:02:34 -0600
Subject: [LLVMbugs] [Bug 3502] Cross-compile llvm-gcc Linux x86 to arm-linux
In-Reply-To: 
Message-ID: <200902191702.n1JH2YmT023726@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3502

Evan Cheng  changed:

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

--- Comment #10 from Evan Cheng   2009-02-19 11:02:33 ---
Both test cases should have been fixed by r64745. Please check.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 11:31:32 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 11:31:32 -0600
Subject: [LLVMbugs] [Bug 3619] ConstantArrayType's profile does not contain
 storage class.
In-Reply-To: 
Message-ID: <200902191731.n1JHVW2q025086@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3619

Chris Lattner  changed:

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

--- Comment #3 from Chris Lattner   2009-02-19 11:31:31 ---
Fixed:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012781.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 cs.uiuc.edu  Thu Feb 19 12:18:39 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 12:18:39 -0600
Subject: [LLVMbugs] [Bug 3611] AsmWriter produces wrong output for functions
 with no name
In-Reply-To: 
Message-ID: <200902191818.n1JIIdvr027054@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3611

Chris Lattner  changed:

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

--- Comment #5 from Chris Lattner   2009-02-19 12:18:38 ---
Anonymous globals don't accept a name like that right now, this works fine
though:

global i32 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 cs.uiuc.edu  Thu Feb 19 12:30:15 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 12:30:15 -0600
Subject: [LLVMbugs] [Bug 3609] diagnostics: [missing terminating "
 character] should say > when needed
In-Reply-To: 
Message-ID: <200902191830.n1JIUFmE027506@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3609

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-19 12:30:15 ---
Fixed, thanks:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012786.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 cs.uiuc.edu  Thu Feb 19 13:05:59 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 13:05:59 -0600
Subject: [LLVMbugs] [Bug 3622] New: Assertion failed: (success && " Illegal
 argument to getIntegerConstantExpr"), function getIntegerConstantExprValue,
 file /Users/ddunbar/llvm/tools/clang /lib/AST/../../include/clang/AST/Expr.h,
 line 172.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3622

           Summary: Assertion failed: (success && "Illegal argument to
                    getIntegerConstantExpr"), function
                    getIntegerConstantExprValue, file
                    /Users/ddunbar/llvm/tools/clang/lib/AST/../../include/cl
                    ang/AST/Expr.h, line 172.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang asserts on the following code. offsetof is not always evaluatable, we
need to change evaluateOffsetOf (and preferably suck into Evaluate?).
--
struct s {
  int arr[10];
};

int b;
int a[__builtin_offsetof(struct s, arr[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 cs.uiuc.edu  Thu Feb 19 15:42:49 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 15:42:49 -0600
Subject: [LLVMbugs] [Bug 3623] New: duplicate "extractvalue" not eliminated
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3623

           Summary: duplicate "extractvalue" not eliminated
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Scalar Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nlewycky at google.com
                CC: llvmbugs at cs.uiuc.edu

In this .ll file:

  define double @zsqr({ double, double } %a) {
    %X = extractvalue { double, double } %a, 0
    %Y = extractvalue { double, double } %a, 0
    %Z = mul double %X, %Y
    ret double %Z
  } 

no optimization pass in -std-compile-opts folds %X and %Y.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 16:12:23 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 16:12:23 -0600
Subject: [LLVMbugs] [Bug 3622] Assertion failed: (success && " Illegal
 argument to getIntegerConstantExpr"), function getIntegerConstantExprValue,
 file /Users/ddunbar/llvm/tools/clang /lib/AST/../../include/clang/AST/Expr.h,
 line 172.
In-Reply-To: 
Message-ID: <200902192212.n1JMCNhK003016@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3622

Daniel Dunbar  changed:

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

--- Comment #1 from Daniel Dunbar   2009-02-19 16:12:22 ---

*** This bug has been marked as a duplicate of bug 3396 ***

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 16:24:42 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 16:24:42 -0600
Subject: [LLVMbugs] [Bug 3398] IRGen crash with pointer cast to int in
	static var
In-Reply-To: 
Message-ID: <200902192224.n1JMOgBO003436@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3398

Daniel Dunbar  changed:

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

--- Comment #4 from Daniel Dunbar   2009-02-19 16:24:42 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012802.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 cs.uiuc.edu  Thu Feb 19 16:24:51 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 16:24:51 -0600
Subject: [LLVMbugs] [Bug 3463] error: cannot codegen this constant
	expression yet
In-Reply-To: 
Message-ID: <200902192224.n1JMOp1J003459@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3463

Daniel Dunbar  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |daniel at zuster.org
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from Daniel Dunbar   2009-02-19 16:24:50 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012802.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 cs.uiuc.edu  Thu Feb 19 17:30:48 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 17:30:48 -0600
Subject: [LLVMbugs] [Bug 3624] New: Tablegen incorrectly transoforming a
	pattern
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3624

           Summary: Tablegen incorrectly transoforming a pattern
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         Component: TableGen
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: micah.villmow at amd.com
                CC: llvmbugs at cs.uiuc.edu

The following pattern:
def DIV_i32 : Pat<(sdiv GPRI32:$src0, GPRI32:$src1),
    (CMOVLOG_i32 
     (BINARY_AND_i32 
      (BINARY_XOR_i32 GPRI32:$src0, GPRI32:$src1), 
      (LOADCONST_i32 0x80000000)), 
     (NEGATE_i32 
      (UDIV_i32 
       (MAX_i32 GPRI32:$src0, 
        (NEGATE_i32 GPRI32:$src0)), 
       (MAX_i32 GPRI32:$src1, 
        (NEGATE_i32 GPRI32:$src1)))), 
     (UDIV_i32 
      (MAX_i32 GPRI32:$src0, 
       (NEGATE_i32 GPRI32:$src0)), 
      (MAX_i32 GPRI32:$src1, 
       (NEGATE_i32 GPRI32:$src1))))>; 

Is incorrectly transformed to:
SDNode *Emit_29(const SDValue &N, unsigned Opc0, unsigned Opc1, unsigned Opc2,
unsigned Opc3, unsigned Opc4, unsigned Opc5, unsigned Opc6, unsigned Opc7,
unsigned Opc8, unsigned Opc9, unsigned Opc10, unsigned Opc11, unsigned Opc12,
unsigned Opc13, unsigned Opc14, MVT VT0, MVT VT1, MVT VT2, MVT VT3, MVT VT4,
MVT VT5, MVT VT6, MVT VT7, MVT VT8, MVT VT9, MVT VT10, MVT VT11, MVT VT12, MVT
VT13, MVT VT14) DISABLE_INLINE {
  SDValue N0 = N.getOperand(0);
  SDValue N1 = N.getOperand(1);
  AddToISelQueue(N0);
  AddToISelQueue(N1);
  SDValue Tmp2(CurDAG->getTargetNode(Opc0, VT0, N0, N1), 0);
  SDValue Tmp3 = CurDAG->getTargetConstant(0x7FFFFFFFULL, MVT::i32);
  SDValue Tmp4(CurDAG->getTargetNode(Opc1, VT1, Tmp3), 0);
  SDValue Tmp5(CurDAG->getTargetNode(Opc2, VT2, Tmp2, Tmp4), 0);
  AddToISelQueue(N0);
  AddToISelQueue(N0);
  SDValue Tmp8(CurDAG->getTargetNode(Opc3, VT3, N0), 0);
  SDValue Tmp9(CurDAG->getTargetNode(Opc4, VT4, N0, Tmp8), 0);
  AddToISelQueue(N1);
  AddToISelQueue(N1);
  SDValue Tmp12(CurDAG->getTargetNode(Opc5, VT5, N1), 0);
  SDValue Tmp13(CurDAG->getTargetNode(Opc6, VT6, N1, Tmp12), 0);
  SDValue Tmp14(CurDAG->getTargetNode(Opc7, VT7, Tmp9, Tmp13), 0);
  SDValue Tmp15(CurDAG->getTargetNode(Opc8, VT8, Tmp14), 0);
  AddToISelQueue(N0);
  AddToISelQueue(N0);
  SDValue Tmp18(CurDAG->getTargetNode(Opc9, VT9, N0), 0);
  SDValue Tmp19(CurDAG->getTargetNode(Opc10, VT10, N0, Tmp18), 0);
  AddToISelQueue(N1);
  AddToISelQueue(N1);
  SDValue Tmp22(CurDAG->getTargetNode(Opc11, VT11, N1), 0);
  SDValue Tmp23(CurDAG->getTargetNode(Opc12, VT12, N1, Tmp22), 0);
  SDValue Tmp24(CurDAG->getTargetNode(Opc13, VT13, Tmp19, Tmp23), 0);
  return CurDAG->SelectNodeTo(N.getNode(), Opc14, VT14, Tmp5, Tmp15, Tmp24);
}
SDNode *Select_ISD_SDIV_i32(const SDValue &N) {
  return Emit_29(N, AMDIL::BINARY_XOR_i32, AMDIL::LOADCONST_i32,
AMDIL::BINARY_AND_i32, AMDIL::NEGATE_i32, AMDIL::MAX_i32, AMDIL::NEGATE_i32,
AMDIL::MAX_i32, AMDIL::UDIV_i32, AMDIL::NEGATE_i32, AMDIL::NEGATE_i32,
AMDIL::MAX_i32, AMDIL::NEGATE_i32, AMDIL::MAX_i32, AMDIL::UDIV_i32,
AMDIL::CMOVLOG_i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32,
MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32,
MVT::i32);
}

When it should be transformed into:
SDNode *Emit_29(const SDValue &N, unsigned Opc0, unsigned Opc1, unsigned Opc2,
unsigned Opc3, unsigned Opc4, unsigned Opc5, unsigned Opc6, unsigned Opc7,
unsigned Opc8, unsigned Opc9, unsigned Opc10, unsigned Opc11, unsigned Opc12,
unsigned Opc13, unsigned Opc14, MVT VT0, MVT VT1, MVT VT2, MVT VT3, MVT VT4,
MVT VT5, MVT VT6, MVT VT7, MVT VT8, MVT VT9, MVT VT10, MVT VT11, MVT VT12, MVT
VT13, MVT VT14) DISABLE_INLINE {
  SDValue N0 = N.getOperand(0);
  SDValue N1 = N.getOperand(1);
  AddToISelQueue(N0);
  AddToISelQueue(N1);
  SDValue Tmp2(CurDAG->getTargetNode(Opc0, VT0, N0, N1), 0);
  SDValue Tmp3 = CurDAG->getTargetConstant(0x80000000ULL, MVT::i32);
  SDValue Tmp4(CurDAG->getTargetNode(Opc1, VT1, Tmp3), 0);
  SDValue Tmp5(CurDAG->getTargetNode(Opc2, VT2, Tmp2, Tmp4), 0);
  AddToISelQueue(N0);
  AddToISelQueue(N0);
  SDValue Tmp8(CurDAG->getTargetNode(Opc3, VT3, N0), 0);
  SDValue Tmp9(CurDAG->getTargetNode(Opc4, VT4, N0, Tmp8), 0);
  AddToISelQueue(N1);
  AddToISelQueue(N1);
  SDValue Tmp12(CurDAG->getTargetNode(Opc5, VT5, N1), 0);
  SDValue Tmp13(CurDAG->getTargetNode(Opc6, VT6, N1, Tmp12), 0);
  SDValue Tmp14(CurDAG->getTargetNode(Opc7, VT7, Tmp9, Tmp13), 0);
  SDValue Tmp15(CurDAG->getTargetNode(Opc8, VT8, Tmp14), 0);
  AddToISelQueue(N0);
  AddToISelQueue(N0);
  SDValue Tmp18(CurDAG->getTargetNode(Opc9, VT9, N0), 0);
  SDValue Tmp19(CurDAG->getTargetNode(Opc10, VT10, N0, Tmp18), 0);
  AddToISelQueue(N1);
  AddToISelQueue(N1);
  SDValue Tmp22(CurDAG->getTargetNode(Opc11, VT11, N1), 0);
  SDValue Tmp23(CurDAG->getTargetNode(Opc12, VT12, N1, Tmp22), 0);
  SDValue Tmp24(CurDAG->getTargetNode(Opc13, VT13, Tmp19, Tmp23), 0);
  return CurDAG->SelectNodeTo(N.getNode(), Opc14, VT14, Tmp5, Tmp15, Tmp24);
}
SDNode *Select_ISD_SDIV_i32(const SDValue &N) {
  return Emit_29(N, AMDIL::BINARY_XOR_i32, AMDIL::LOADCONST_i32,
AMDIL::BINARY_AND_i32, AMDIL::NEGATE_i32, AMDIL::MAX_i32, AMDIL::NEGATE_i32,
AMDIL::MAX_i32, AMDIL::UDIV_i32, AMDIL::NEGATE_i32, AMDIL::NEGATE_i32,
AMDIL::MAX_i32, AMDIL::NEGATE_i32, AMDIL::MAX_i32, AMDIL::UDIV_i32,
AMDIL::CMOVLOG_i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32,
MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32, MVT::i32,
MVT::i32);
}

The problem is the TargetConstant is generated incorrectly.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 17:57:53 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 17:57:53 -0600
Subject: [LLVMbugs] [Bug 3616] Bug in BranchFolding.cpp:OptimizeBlock
In-Reply-To: 
Message-ID: <200902192357.n1JNvreo006912@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3616

Dale Johannesen  changed:

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

--- Comment #15 from Dale Johannesen   2009-02-19 17:57:52 ---
About all we could do is test whether the function works and give up on tail
merging completely if it doesn't.  I'm against that, now that the documentation
describes what's needed correctly; it's not as if InsertBranch is hard to
implement.

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 19 19:19:45 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 19 Feb 2009 19:19:45 -0600
Subject: [LLVMbugs] [Bug 3625] New: _IntSaturate() macro from FixMath. h
 fails to compile when a PCH importing Foundation.h is used
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3625

           Summary: _IntSaturate() macro from FixMath.h fails to compile
                    when a PCH importing Foundation.h is used
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: catfish.man at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2588)
 --> (http://llvm.org/bugs/attachment.cgi?id=2588)
Xcode project that demonstrates the issue

The attached project compiles successfully with gcc, fails with clang. 

Built via xcodebuild CC=$CCC LD=$CCC, with CCC=/llvm/tools/clang/tools/ccc/ccc
and PATH=/Volumes/Other/llvm/Release-Asserts/bin/:$PATH

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 00:47:36 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 00:47:36 -0600
Subject: [LLVMbugs] [Bug 3626] New: clang rejects K&R call which doesn't
 match # of arguments
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3626

           Summary: clang rejects K&R call which doesn't match # of
                    arguments
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang rejects the following code gcc accepts; this occurs in emacs.
--
void f0(a, b, c, d) int a,b,c,d; {}
void f1(void) { f0(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 cs.uiuc.edu  Fri Feb 20 00:51:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 00:51:29 -0600
Subject: [LLVMbugs] [Bug 3627] New: vector widening crash
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3627

           Summary: vector widening crash
           Product: libraries
           Version: 1.0
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Common Code Generator Code
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu

test/CodeGen/Generic/vector-casts.ll  crashes due to unsupported cases in
vector widening in legalizetypes.  Reproduce with:

llvm-as < vector-casts.ll | llc -march=ppc32 -mcpu=g5

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 01:09:20 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 01:09:20 -0600
Subject: [LLVMbugs] [Bug 3628] New: PPC Backend crashes with cannot select
 v4i32 SRA instruction
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3628

           Summary: PPC Backend crashes with cannot select v4i32 SRA
                    instruction
           Product: libraries
           Version: 1.0
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: PowerPC
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: clattner at apple.com
                CC: llvmbugs at cs.uiuc.edu

cd test/CodeGen/X86; llvm-as < vshift_split2.ll | llc -march=ppc32 -mcpu=g5
Cannot yet select: 0x18074d0: v4i32 = sra 0x180702c, 0x1806e1c

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 01:25:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 01:25:08 -0600
Subject: [LLVMbugs] [Bug 2657] Error when accessing vector component from
 member variable.
In-Reply-To: 
Message-ID: <200902200725.n1K7P8eR021807@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2657

Syoyo Fujita  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|WORKSFORME                  |FIXED

--- Comment #5 from Syoyo Fujita   2009-02-20 01:25:07 ---
I've also confirmed that recent version of clang  compiles correctly the code. 

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 03:33:30 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 03:33:30 -0600
Subject: [LLVMbugs] [Bug 3463] error: cannot codegen this constant
	expression yet
In-Reply-To: 
Message-ID: <200902200933.n1K9XU9T004271@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3463

Roman Divacky  changed:

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

--- Comment #4 from Roman Divacky   2009-02-20 03:33:30 ---
the attached still cannot be compiled:

lev /tmp$ ccc -c devolatile.c 
ccc: Unknown host 'freebsd', using generic host information.
devolatile.c:42:1: error: cannot compile this constant expression yet

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 03:44:39 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 03:44:39 -0600
Subject: [LLVMbugs] [Bug 3629] New: __attribute__((weak)) not supported
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3629

           Summary: __attribute__((weak)) not supported
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

lev /tmp$ cat weak.c
int foo __attribute__((weak));
int bar;
lev /tmp$ gcc -c weak.c && nm weak.o
0000000000000004 C bar
0000000000000000 V foo
lev /tmp$ ccc -c weak.c && nm weak.o
ccc: Unknown host 'freebsd', using generic host information.
0000000000000004 C bar
0000000000000004 C foo

the attribute is not support, we need this in FreeBSD quite a lot

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 10:19:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 10:19:12 -0600
Subject: [LLVMbugs] [Bug 3631] New:
 test/FrontendFortran/2008-11-03-OptionOverride. f90 not portable
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3631

           Summary: test/FrontendFortran/2008-11-03-OptionOverride.f90 not
                    portable
           Product: new-bugs
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org
                CC: llvmbugs at cs.uiuc.edu

This fails on ppc. The test should be fixed to be portable.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 10:20:41 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 10:20:41 -0600
Subject: [LLVMbugs] [Bug 3632] New: test/FrontendC/2009-01-20-k8.c is not
	portable
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3632

           Summary: test/FrontendC/2009-01-20-k8.c is not portable
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org
                CC: llvmbugs at cs.uiuc.edu

Fails on ppc.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 10:24:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 10:24:14 -0600
Subject: [LLVMbugs] [Bug 3633] New:
 test/FrontendC/2008-03-24-BitField-And-Alloca. c not portable
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3633

           Summary: test/FrontendC/2008-03-24-BitField-And-Alloca.c not
                    portable
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org
                CC: llvmbugs at cs.uiuc.edu

Fails on ppc. Test should 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 cs.uiuc.edu  Fri Feb 20 12:55:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 12:55:14 -0600
Subject: [LLVMbugs] [Bug 3625] _IntSaturate() macro from FixMath. h fails to
 compile when a PCH importing Foundation.h is used
In-Reply-To: 
Message-ID: <200902201855.n1KItEJb025975@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3625

Ted Kremenek  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 cs.uiuc.edu  Fri Feb 20 13:10:16 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 13:10:16 -0600
Subject: [LLVMbugs] [Bug 3463] error: cannot codegen this constant
	expression yet
In-Reply-To: 
Message-ID: <200902201910.n1KJAGuQ026541@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3463

Daniel Dunbar  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |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 cs.uiuc.edu  Fri Feb 20 15:17:40 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 15:17:40 -0600
Subject: [LLVMbugs] [Bug 3636] New: build_gcc needs to be modified to build
	fortran
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3636

           Summary: build_gcc needs to be modified to build fortran
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org
                CC: llvmbugs at cs.uiuc.edu

build_gcc should be modified to build fortran. The tricky part is that packages
from macports (or equiv) need to be downloaded in order to build. I don't think
we would want that done automatically..

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 20 16:59:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 16:59:12 -0600
Subject: [LLVMbugs] [Bug 3502] Cross-compile llvm-gcc Linux x86 to arm-linux
In-Reply-To: 
Message-ID: <200902202259.n1KMxCq3003031@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3502

S.Bharadwaj Yadavalli  changed:

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

--- Comment #13 from S.Bharadwaj Yadavalli   2009-02-20 16:59:12 ---

I encountered the assertion failure while compiling my project using llvm based
x86_64->ARM cross compiler. I attached is a simplified versions of .bc and .ll.

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 cs.uiuc.edu  Fri Feb 20 17:57:28 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 20 Feb 2009 17:57:28 -0600
Subject: [LLVMbugs] [Bug 2305] CHECKER: track ranges for symbolic values
In-Reply-To: 
Message-ID: <200902202357.n1KNvSuI005621@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2305

Ted Kremenek  changed:

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

--- Comment #2 from Ted Kremenek   2009-02-20 17:57:28 ---
2305 and 2306 are really the same issue.

*** This bug has been marked as a duplicate of bug 2306 ***

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 00:58:20 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 00:58:20 -0600
Subject: [LLVMbugs] [Bug 3638] New: legalizedag_vec.ll: divdi3 occurs 4
	times, not 2
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3638

           Summary: legalizedag_vec.ll: divdi3 occurs 4 times, not 2
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: scooter.phd at gmail.com
                CC: llvmbugs at cs.uiuc.edu

The test/CodeGen/X85/legalizedag_vec.ll searches for divdi3 to occur twice (2x)
in llc's x86 assembler output. With shared library stubs, however, divdi3
occurs four times.

Before I change "2" to "4", what's the correct answer? Should divdi3 occur
twice or four times?

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 01:59:40 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 01:59:40 -0600
Subject: [LLVMbugs] [Bug 3639] New: @encode fails on structures ending in
	unsized array
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3639

           Summary: @encode fails on structures ending in unsized array
           Product: clang
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: AST
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: joe at alacatialabs.com
                CC: llvmbugs at cs.uiuc.edu

Clang currently asserts when handling @encode on a structure that ends in an
unsized array:
Assertion failed: (0 && "Unhandled array type!"), function
getObjCEncodingForTypeImpl, file ASTContext.cpp, line 2198.

With a stack trace of:
clang::ASTContext::getObjCEncodingForTypeImpl at ASTContext.cpp:2198
clang::ASTContext::getObjCEncodingForTypeImpl at ASTContext.cpp:2233
clang::ASTContext::getObjCEncodingForType at ASTContext.cpp:2029

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 04:51:25 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 04:51:25 -0600
Subject: [LLVMbugs] [Bug 3640] New: Assertion failed: (SI.isSigned() ==
 E->getType()-> isSignedIntegerType() && "Invalid evaluation result."),
 function Success, file ExprConstant.cpp, line 494.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3640

           Summary: Assertion failed: (SI.isSigned() == E->getType()-
                    >isSignedIntegerType() && "Invalid evaluation result."),
                    function Success, file ExprConstant.cpp, line 494.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: AST
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

I am getting this when compiling the attached:

witten /tmp# ccc -c -ffreestanding vga.c

.... warnings ...

Assertion failed: (SI.isSigned() == E->getType()->isSignedIntegerType() &&
"Invalid evaluation result."), function Success, file ExprConstant.cpp, line
494.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 11:57:26 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 11:57:26 -0600
Subject: [LLVMbugs] [Bug 3629] __attribute__((weak)) not supported
In-Reply-To: 
Message-ID: <200902211757.n1LHvQ1f024467@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3629

Anton Korobeynikov  changed:

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

--- Comment #9 from Anton Korobeynikov   2009-02-21 11:57:25 ---
(In reply to comment #8)
> Is this fixed Anton?
We still have differences w.r.t. gcc in 'normal weak' cases. I don't see, how
this can make stuff not working, so, probably we should close this PR now and
reopen, if necessary.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 12:14:43 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 12:14:43 -0600
Subject: [LLVMbugs] [Bug 3640] Assertion failed: (SI.isSigned() ==
 E->getType()-> isSignedIntegerType() && "Invalid evaluation result."),
 function Success, file ExprConstant.cpp, line 494.
In-Reply-To: 
Message-ID: <200902211814.n1LIEhbf025422@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3640

Daniel Dunbar  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |daniel at zuster.org
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Daniel Dunbar   2009-02-21 12:14:42 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012910.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 cs.uiuc.edu  Sat Feb 21 13:12:13 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 13:12:13 -0600
Subject: [LLVMbugs] [Bug 3610] VAARG not implemented for soft floats.
In-Reply-To: 
Message-ID: <200902211912.n1LJCDkj027597@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3610

Richard Pennington  changed:

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

--- Comment #1 from Richard Pennington   2009-02-21 13:12:13 ---
(In reply to comment #0)
> Created an attachment (id=2576)
 --> (http://llvm.org/bugs/attachment.cgi?id=2576) [details]
> .ll file using vaarg
> 
> [~/ellcc/ellcc] main% llvm-as -f t0056.ll         
> [~/ellcc/ellcc] main% llc t0056.bc 
> SoftenFloatResult #0: 0x9f55c44: f32,ch = vaarg 0x9f55b3c, 0x9f5571c, 0x9f55bc0
> llc:
> /home/rich/llvm-trunk-new/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp:54:
> void llvm::DAGTypeLegalizer::SoftenFloatResult(llvm::SDNode*, unsigned int):
> Assertion `0 && "Do not know how to soften the result of this operator!"'
> failed.                                                                         
> 0   llc       0x08c970ae                                                        
> 1   llc       0x08c973f0                                                        
> 2             0x00110400 __kernel_sigreturn + 0                                 
> 3   libc.so.6 0x00bcce28 abort + 392                                            
> 4   libc.so.6 0x00bc440e __assert_fail + 238                                    
> 5   llc       0x089dd961
> 6   llc       0x089c64ea
> 7   llc       0x089c6f69 llvm::SelectionDAG::LegalizeTypes() + 45
> 8   llc       0x0895ea4a llvm::SelectionDAGISel::CodeGenAndEmitDAG() + 1180
> 9   llc       0x089612d4
> llvm::SelectionDAGISel::SelectBasicBlock(llvm::BasicBlock*,
> llvm::ilist_iterator,
> llvm::ilist_iterator) + 598
> 10  llc       0x08961c87
> llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function&,
> llvm::MachineFunction&, llvm::MachineModuleInfo*, llvm::DwarfWriter*,
> llvm::TargetInstrInfo const&) + 2219
> 11  llc       0x08962963 llvm::SelectionDAGISel::runOnFunction(llvm::Function&)
> + 841
> 12  llc       0x08c259f5 llvm::FPPassManager::runOnFunction(llvm::Function&) +
> 249
> 13  llc       0x08c2646a llvm::FunctionPassManagerImpl::run(llvm::Function&) +
> 116
> 14  llc       0x08c265cb llvm::FunctionPassManager::run(llvm::Function&) + 137
> 15  llc       0x083d0d22 main + 2465
> 16  libc.so.6 0x00bb66e5 __libc_start_main + 229
> 17  llc       0x083cf441
> Abort
> [~/ellcc/ellcc] main%
> 

Fixed in revision 65239.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 14:51:45 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 14:51:45 -0600
Subject: [LLVMbugs] [Bug 3595] instcombine miscompilation
In-Reply-To: 
Message-ID: <200902212051.n1LKpiLG031105@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3595

Nick Lewycky  changed:

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

--- Comment #10 from Nick Lewycky   2009-02-21 14:51:43 ---
Fixed in r65246.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 16:13:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 16:13:50 -0600
Subject: [LLVMbugs] [Bug 3641] New: Assertion failed: (Ty->isSized() &&
 "Cannot getTypeInfo() on a type that is unsized!"),
 function getTypeSizeInBits, file TargetData.  cpp, line 421.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3641

           Summary: Assertion failed: (Ty->isSized() && "Cannot
                    getTypeInfo() on a type that is unsized!"), function
                    getTypeSizeInBits, file TargetData.cpp, line 421.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

I hit this assertion when compiling the attached, this must be fairly new as I
didnt hit this in the morning. I belive 24 hours old clang does not have this
bug

witten /tmp# ccc -c -ffreestanding subr_prof.c
ccc: Unknown host 'freebsd', using generic host information.
Assertion failed: (Ty->isSized() && "Cannot getTypeInfo() on a type that is
unsized!"), function getTypeSizeInBits, file TargetData.cpp, line 421.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 16:21:58 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 16:21:58 -0600
Subject: [LLVMbugs] [Bug 3642] New: Assertion failed: (CGT.getTargetData().
 getTypePaddedSizeInBits(STy) == RL.getSize()), function layoutStructFields,
 file CodeGenTypes.cpp, line 514.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3642

           Summary: Assertion failed:
                    (CGT.getTargetData().getTypePaddedSizeInBits(STy) ==
                    RL.getSize()), function layoutStructFields, file
                    CodeGenTypes.cpp, line 514.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

hitting this when compiling the attached:

ccc -c -ffreestanding ng_pppoe.c

... warnings ...

Assertion failed: (CGT.getTargetData().getTypePaddedSizeInBits(STy) ==
RL.getSize()), function layoutStructFields, file CodeGenTypes.cpp, line 514.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 19:39:16 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 19:39:16 -0600
Subject: [LLVMbugs] [Bug 3639] @encode asserts on structures ending in
	unsized array
In-Reply-To: 
Message-ID: <200902220139.n1M1dFnP008044@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3639

Anders Carlsson  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andersca at mac.com
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Anders Carlsson   2009-02-21 19:39:15 ---
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012937.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 cs.uiuc.edu  Sat Feb 21 20:15:00 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 20:15:00 -0600
Subject: [LLVMbugs] [Bug 3642] Assertion failed:
 (CGT.getTargetData().getTypePaddedSizeInBits( STy) == RL.getSize()),
 function layoutStructFields, file CodeGenTypes.cpp, line 514.
In-Reply-To: 
Message-ID: <200902220215.n1M2F0Zh009103@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3642

Anders Carlsson  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andersca at mac.com
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Anders Carlsson   2009-02-21 20:14:59 ---
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012938.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 cs.uiuc.edu  Sat Feb 21 21:24:38 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 21:24:38 -0600
Subject: [LLVMbugs] [Bug 3254] __attribute__ ((aligned)) is ignored
In-Reply-To: 
Message-ID: <200902220324.n1M3OcXw011644@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3254

Eli Friedman  changed:

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

--- Comment #2 from Eli Friedman   2009-02-21 21:24:37 ---
Fixed with r65258/r65259.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 21:35:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 21:35:06 -0600
Subject: [LLVMbugs] [Bug 3433] __alignof__ for basic types doesn't match gcc
In-Reply-To: 
Message-ID: <200902220335.n1M3Z6oi012030@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3433

Eli Friedman  changed:

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

--- Comment #2 from Eli Friedman   2009-02-21 21:35:05 ---
I'm going to call this fixed with r65260, since we now get the right result for
double on x86.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 21:36:48 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 21:36:48 -0600
Subject: [LLVMbugs] [Bug 2815] escaped chars in string literal cause
 incorrect caret position in diagnostic messages
In-Reply-To: 
Message-ID: <200902220336.n1M3am01012102@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=2815

Eli Friedman  changed:

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

--- Comment #2 from Eli Friedman   2009-02-21 21:36:46 ---
Fixed with sabre's recent work on string diagnostics.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 21 23:47:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 21 Feb 2009 23:47:08 -0600
Subject: [LLVMbugs] [Bug 3644] New: clang rejects redefinition of extern
	inline function
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3644

           Summary: clang rejects redefinition of extern inline function
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

clang rejects the following code gcc accepts:
--
extern inline void f0(void) { }
void f0(void) { }
--

This notably occurs in gcc-4.2. gcc itself rejects this in c99 mode. We may
well not want to support this, but filing for tracking purposes.

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 22 00:06:37 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 00:06:37 -0600
Subject: [LLVMbugs] [Bug 3645] New: clang incorrectly rejects "static int a;
 extern int a; "
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3645

           Summary: clang incorrectly rejects "static int a; extern int a;"
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu

Per summary, the following testcase is incorrectly rejected by clang:
static int a;
extern int a;

Per C99 6.2.2p4, the identifier "a" has internal linkage.

Related: bug 3425, a similar sort of error with function redeclarations.

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 22 02:36:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 02:36:33 -0600
Subject: [LLVMbugs] [Bug 3502] Cross-compile llvm-gcc Linux x86 to arm-linux
In-Reply-To: 
Message-ID: <200902220836.n1M8aXV4022224@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3502

Evan Cheng  changed:

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

--- Comment #14 from Evan Cheng   2009-02-22 02:36:33 ---
Fixed.
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090216/074104.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 cs.uiuc.edu  Sun Feb 22 02:56:17 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 02:56:17 -0600
Subject: [LLVMbugs] [Bug 3646] New: gcc 3.3.5 - broken?
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3646

           Summary: gcc 3.3.5 - broken?
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: OpenBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

On OpenBSD 4.4 using the gcc 3.3.5 (propolice) compiler I get this error:

In file included from X86TargetAsmInfo.cpp:14:
X86TargetAsmInfo.h: In constructor '
   llvm::X86TargetAsmInfo::X86TargetAsmInfo(const
   llvm::X86TargetMachine&)':
X86TargetAsmInfo.h:31:error: syntax error before ';' token

Compiling same file with gcc 4.2.0 20070307 (prerelease), it compiles
successfully.

Should this 3.3.5 compiler be listed in the broken gcc versions section?

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 22 03:16:22 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 03:16:22 -0600
Subject: [LLVMbugs] [Bug 3647] New: [llvm2.5prerelease2] utils/ unittests
 fails to build on OpenBSD 4.4 with gcc 4.2
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3647

           Summary: [llvm2.5prerelease2] utils/unittests fails to build on
                    OpenBSD 4.4 with gcc 4.2
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Using gcc 4.2.0 20070307 (prerelease) from the OpenBSD 4.4 package, and
configuring llvm like this (egcc is the 4.2.0 version, gcc is 3.3.5):
./configure CC=egcc CXX=eg++

It fails during make in utils/unittest/:
gmake[3]: Entering directory `/home/edwin/llvm-2.5/utils/unittest/googletest'
llvm[3]: Compiling gtest.cc for Release build
In file included from
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:65,
                 from gtest.cc:34:
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: 'testing::internal::FormatForComparisonFailureMessage' declared as an
'inline' variable
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: 'testing::internal::String
testing::internal::FormatForComparisonFailureMessage' redeclared as different
kind of symbol
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:271:
error: previous declaration of 'testing::internal::String
testing::internal::FormatForComparisonFailureMessage(const char*, const
std::string&)'
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: 'std::wstring' has not been declared
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: 'str' was not declared in this scope
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: expected primary-expression before 'const'
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: expected unqualified-id before '::' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: expected ',' or '...' before '::' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:
In function 'testing::internal::String
testing::internal::FormatForComparisonFailureMessage()':
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/internal/gtest-internal.h:274:
error: 'str' was not declared in this scope
In file included from
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:68,
                 from gtest.cc:34:
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest-message.h:
At global scope:
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest-message.h:172:
error: expected unqualified-id before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest-message.h:172:
error: expected ',' or '...' before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest-message.h:172:
error: 'testing::Message& testing::Message::operator<<()' must take exactly one
argument
In file included from gtest.cc:34:
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:575:
error: expected unqualified-id before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:575:
error: expected ',' or '...' before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h: In
function 'testing::internal::String
testing::internal::FormatForFailureMessage()':
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:576:
error: 'wstr' was not declared in this scope
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h: At global
scope:
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:824:
error: expected unqualified-id before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:824:
error: expected ',' or '...' before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:827:
error: expected unqualified-id before '&' token
/home/edwin/llvm-2.5/utils/unittest/googletest/include/gtest/gtest.h:827:
error: expected ',' or '...' before '&' token
gtest.cc:872: error: expected unqualified-id before '&' token
gtest.cc:872: error: expected ',' or '...' before '&' token
gtest.cc:872: error: 'testing::Message& testing::Message::operator<<()' must
take exactly one argument
gtest.cc:1280: error: expected unqualified-id before '&' token
gtest.cc:1280: error: expected ',' or '...' before '&' token
gtest.cc: In function 'testing::AssertionResult testing::IsSubstring(const
char*, const char*)':
gtest.cc:1281: error: 'needle' was not declared in this scope
gtest.cc:1281: error: 'haystack' was not declared in this scope
gtest.cc: At global scope:
gtest.cc:1286: error: expected unqualified-id before '&' token
gtest.cc:1286: error: expected ',' or '...' before '&' token
gtest.cc: In function 'testing::AssertionResult testing::IsNotSubstring(const
char*, const char*)':
gtest.cc:1287: error: 'needle' was not declared in this scope
gtest.cc:1287: error: 'haystack' was not declared in this scope
gtest.cc: At global scope:
gtest.cc:847: warning: 'void testing::internal::StreamWideCharsToMessage(const
wchar_t*, size_t, testing::Message*)' defined but not used
gmake[3]: *** [/home/edwin/llvm-2.5/utils/unittest/googletest/Release/gtest.o]
Error 1
gmake[3]: Leaving directory `/home/edwin/llvm-2.5/utils/unittest/googletest'
gmake[2]: *** [googletest/.makeall] Error 2

I have removed the unittest/unittests from the Makefile, and see if I can build
without it on OpenBSD.

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 22 06:24:43 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 06:24:43 -0600
Subject: [LLVMbugs] [Bug 3648] New: clang gives @encode a different type
 from gcc ( and related issues)
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3648

           Summary: clang gives @encode a different type from gcc (and
                    related issues)
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: sharparrow1 at yahoo.com
                CC: llvmbugs at cs.uiuc.edu

Per summary; testcase:

int a[sizeof(@encode(int)) == 2 ? 1 : -1]; // Type is char[2]
char b[] = @encode(int); // Initialize like a string
char (*c)[2] = &@encode(int); // @encode is an lvalue

I'm not sure whether the differences here are intentional or not... there isn't
any standards document to refer to here :)  I don't see any good reason to be
incompatible with gcc here, though.

That said, I'm guessing these sort of constructs aren't common, considering
that nobody has complained about it yet.  (I haven't seen this in real-world
code; I was just trying it out of curiosity.)

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 22 11:30:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 11:30:24 -0600
Subject: [LLVMbugs] [Bug 3649] New: stack is not saved/restored correctly
 with multiple VLAs
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3649

           Summary: stack is not saved/restored correctly with multiple VLAs
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu, andersca at mac.com

The printed addresses below should be constant:
--
ddunbar at frank:tmp$ cat t.c
#include 

int f0(int n) {
  int arr[n];
  for (unsigned i=0; i
Message-ID: <200902221847.n1MIlVeD010919@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3649

Anders Carlsson  changed:

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

--- Comment #1 from Anders Carlsson   2009-02-22 12:47:30 ---
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090216/012967.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 cs.uiuc.edu  Sun Feb 22 13:23:56 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 13:23:56 -0600
Subject: [LLVMbugs] [Bug 3641] Assertion failed: (Ty->isSized() && "Cannot
 getTypeInfo() on a type that is unsized!"), function getTypeSizeInBits,
 file TargetData.  cpp, line 421.
In-Reply-To: 
Message-ID: <200902221923.n1MJNu7n012104@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3641

Roman Divacky  changed:

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

--- Comment #3 from Roman Divacky   2009-02-22 13:23:55 ---
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 cs.uiuc.edu  Sun Feb 22 17:10:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 17:10:14 -0600
Subject: [LLVMbugs] [Bug 3638] legalizedag_vec.ll: divdi3 occurs 4 times,
	not 2
In-Reply-To: 
Message-ID: <200902222310.n1MNAEdw019490@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3638

Scott Michel  changed:

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

--- Comment #2 from Scott Michel   2009-02-22 17:10:13 ---
Works for me. Just wanted to ensure that I wasn't violating the intended spirit
of the test.

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

From bugzilla-daemon at cs.uiuc.edu  Sun Feb 22 18:09:28 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 22 Feb 2009 18:09:28 -0600
Subject: [LLVMbugs] [Bug 3642] Assertion failed:
 (CGT.getTargetData().getTypePaddedSizeInBits( STy) == RL.getSize()),
 function layoutStructFields, file CodeGenTypes.cpp, line 514.
In-Reply-To: 
Message-ID: <200902230009.n1N09SAN021521@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3642

Eli Friedman  changed:

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

--- Comment #8 from Eli Friedman   2009-02-22 18:09:26 ---
Wrong 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 cs.uiuc.edu  Mon Feb 23 01:20:09 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 01:20:09 -0600
Subject: [LLVMbugs] [Bug 3650] New: bugpoint mode to debug fast codegen
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3650

           Summary: bugpoint mode to debug fast codegen
           Product: tools
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: bugpoint
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca
                CC: llvmbugs at cs.uiuc.edu

It'd be nice if bugpoint could test 'llc' versus 'llc -fast' when reducing.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 23 06:28:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 06:28:50 -0600
Subject: [LLVMbugs] [Bug 3651] New: cast_or_null for op_iterator *does*
	dereference NULL
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3651

           Summary: cast_or_null for op_iterator *does* dereference NULL
           Product: libraries
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ggreif at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Using the type simplification mechanism with
cast_or_null(op_begin()) fails to check whether the dereferenced
op_begin() returns a non-null result before going on and doing the cast. This
is not the case when type simplification is not needed. I.e. this works:
cast_or_null(op_begin()->get()).

I see two possible places where the bug hides:
1) in the type simplifier that deals with operand_iterator and friends
specifically, or
2) the generic mechanism of type simplification.

I'll try to look into both in the next days, but maybe someone more
knowledgeable comes in first.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 23 13:40:21 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 13:40:21 -0600
Subject: [LLVMbugs] [Bug 3652] Missing sign-extend patterns for i128
In-Reply-To: 
Message-ID: <200902231940.n1NJeLFU007443@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3652

Tilmann Scheller  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 cs.uiuc.edu  Mon Feb 23 15:28:57 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 15:28:57 -0600
Subject: [LLVMbugs] [Bug 3654] New: tls is broken with -fast
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3654

           Summary: tls is broken with -fast
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llc
        AssignedTo: rafael.espindola at gmail.com
        ReportedBy: rafael.espindola at gmail.com
                CC: llvmbugs at cs.uiuc.edu, nlewycky at google.com

If the program

@v = thread_local global i32 0
define i32* @f() {
entry:
        ret i32* @v
}

Is compiled with llc -relocation-model=pic, a valid instruction sequence is
produced. If it is compiled with llc -relocation-model=pic -fast, the call to
__tls_get_addr is missing.

Running in gdb shows that X86TargetLowering::LowerGlobalTLSAddress is not
called when using -fast.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 23 16:05:42 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 16:05:42 -0600
Subject: [LLVMbugs] [Bug 3654] tls is broken with -fast
In-Reply-To: 
Message-ID: <200902232205.n1NM5g8j013660@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3654

Dan Gohman  changed:

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

--- Comment #1 from Dan Gohman   2009-02-23 16:05:41 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090223/074143.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 cs.uiuc.edu  Mon Feb 23 19:02:26 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 19:02:26 -0600
Subject: [LLVMbugs] [Bug 3650] bugpoint mode to debug fast codegen
In-Reply-To: 
Message-ID: <200902240102.n1O12Qru020476@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3650

Nick Lewycky  changed:

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

--- Comment #4 from Nick Lewycky   2009-02-23 19:02:25 ---
Aha! That even makes sense. Thanks Dan.

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

From bugzilla-daemon at cs.uiuc.edu  Mon Feb 23 19:28:28 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 23 Feb 2009 19:28:28 -0600
Subject: [LLVMbugs] [Bug 3425] Inaccurate storage class for redeclaration of
	function
In-Reply-To: 
Message-ID: <200902240128.n1O1SSYT021455@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3425

Douglas Gregor  changed:

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

--- Comment #1 from Douglas Gregor   2009-02-23 19:28:27 ---
Your patch was correct, but there are a few more places we needed to check or
merge storage specifiers on function declarations. This should do the trick:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013011.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 cs.uiuc.edu  Tue Feb 24 03:39:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 03:39:24 -0600
Subject: [LLVMbugs] [Bug 3656] New: Error with objc type conversion in
 __attribute__((cleanup)) function argument
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3656

           Summary: Error with objc type conversion in
                    __attribute__((cleanup)) function argument
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: devlists at shadowlab.org
                CC: llvmbugs at cs.uiuc.edu

gcc compile this file without complain, but clang (ccc / rev 65375) report an
error that prevent compilation.

============ objc-cleanup.m ============

@class NSString;

static __inline__
void $wb_scopeReleaseObject(id *scopeReleasedObject) {
  // commented to avoid 'may not responds  to release' warning.
  //[*scopeReleasedObject release];
}

int main() {
        __attribute__((cleanup($wb_scopeReleaseObject))) NSString *test = nil;
        return 0;
}

===================================
ccc objc-cleanup.m
error: 'cleanup' function '$wb_scopeReleaseObject' parameter has type 'id *',
expected type 'NSString **'
       __attribute__((cleanup($wb_scopeReleaseObject))) NSString *test;
                       ^

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 08:16:39 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 08:16:39 -0600
Subject: [LLVMbugs] [Bug 3657] New: clang -S coredumping
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3657

           Summary: clang -S coredumping
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: rdivacky at freebsd.org
                CC: llvmbugs at cs.uiuc.edu

I am getting this:

lev /tmp$ clang profile.c

lev /tmp$ clang -S profile.c
Segmentation fault (core dumped)

the assembler output makes the clang coredump. might be bug in llvm or clang
codegen for llvm. the test case attached

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 10:06:48 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 10:06:48 -0600
Subject: [LLVMbugs] [Bug 3658] New: Assertion failed:
 (C->getType()->getPrimitiveSizeInBits() < Ty->getPrimitiveSizeInBits()&& "
 SrcTy must be smaller than DestTy for ZExt!"), function getZExt,
 file Constants.cpp, line 1909.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3658

           Summary: Assertion failed: (C->getType()-
                    >getPrimitiveSizeInBits() < Ty-
                    >getPrimitiveSizeInBits()&& "SrcTy must be smaller than
                    DestTy for ZExt!"), function getZExt, file
                    Constants.cpp, line 1909.
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: parser
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2604)
 --> (http://llvm.org/bugs/attachment.cgi?id=2604)
Pre-processed sys_term.c

ccc -c -O2 -pipe -march=nocona -DLINEMODE -DUSE_TERMIO -DDIAGNOSTICS
-DOLD_ENVIRON -DENV_HACK -DSTREAMSPTY -DINET6
-I/usr/src/libexec/telnetd/../../contrib/telnet -DAUTHENTICATION -DENCRYPTION
-DKRB5 -DFORWARD -Dnet_write=telnet_net_write -fstack-protector
-Wsystem-headers -Werror -Wall -Wno-format-y2k -Wno-uninitialized
-Wno-pointer-sign -c sys_term.c
In file included from
/usr/src/libexec/telnetd/../../contrib/telnet/telnetd/sys_term.c:1:
/usr/src/libexec/telnetd/../../contrib/telnet/telnetd/sys_term.c:1212:17:
warning: invalid conversion '%m'
 syslog(3, "%s: %m", altlogin);
           ~~~~~^~~
Assertion failed: (C->getType()->getPrimitiveSizeInBits() <
Ty->getPrimitiveSizeInBits()&& "SrcTy must be smaller than DestTy for ZExt!"),
function getZExt, file Constants.cpp, line 1909.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 10:13:44 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 10:13:44 -0600
Subject: [LLVMbugs] [Bug 3659] New: Expensive checks failure in
	ListScheduleBottomUp
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3659

           Summary: Expensive checks failure in ListScheduleBottomUp
           Product: new-bugs
           Version: unspecified
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2605)
 --> (http://llvm.org/bugs/attachment.cgi?id=2605)
testcase .ll

If built with --enable-expensive-checks, the attached testcase
causes llc to abort with:

/usr/include/c++/4.3/bits/stl_heap.h:356:error: elements in iterator range [
    __first, __last) do not form a heap with respect to the predicate __comp    
    .                                                                           

Objects involved in the operation:
iterator "__first" @ 0x0x7fffb856b830 {
type =
N11__gnu_debug14_Safe_iteratorIN9__gnu_cxx17__normal_iteratorIPPN4llvm5SUnitENSt6__norm6vectorIS5_SaIS5_EEEEENSt7__debug6vectorIS5_S9_EEEE
(mutable iterator);                                                             
  state = dereferenceable (start-of-sequence);                                  
  references sequence with type `NSt7__debug6vectorIPN4llvm5SUnitESaIS3_EEE' @
0x0x7fffb856b830                                     
}                                                                               
iterator "__last" @ 0x0x7fffb856b800 {
type =
N11__gnu_debug14_Safe_iteratorIN9__gnu_cxx17__normal_iteratorIPPN4llvm5SUnitENSt6__norm6vectorIS5_SaIS5_EEEEENSt7__debug6vectorIS5_S9_EEEE
(mutable iterator);
  state = past-the-end;
  references sequence with type `NSt7__debug6vectorIPN4llvm5SUnitESaIS3_EEE' @
0x0x7fffb856b800
}
...
Aborted (core dumped)

This happens on the call to Queue.pop() in line 1024 below.
It's not clear to me what the problem is exactly.

1021        SUnit *pop() {
1022          if (empty()) return NULL;
1023          SUnit *V = Queue.top();
1024          Queue.pop();
1025          V->NodeQueueId = 0;
1026          return V;
1027        }

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 10:20:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 10:20:12 -0600
Subject: [LLVMbugs] [Bug 3660] New: Clang does not enforce the restrictions
	on main
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3660

           Summary: Clang does not enforce the restrictions on main
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: dgregor at apple.com
                CC: llvmbugs at cs.uiuc.edu

Clang's semantic analysis does not enforce any of the restrictions on the
function main, in C or in C++. For example, it happily accepts this bogus
definition:

  inline static float main(float) { }

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 11:05:38 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 11:05:38 -0600
Subject: [LLVMbugs] [Bug 3661] New: ccc -O1 miscompiles git
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3661

           Summary: ccc -O1 miscompiles git
           Product: libraries
           Version: trunk
          Platform: PC
               URL: http://pastie.textmate.org/private/dzzobqfokiltcqh5uoiem
                    a
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Interprocedural Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: frimmirf at gmail.com
                CC: llvmbugs at cs.uiuc.edu

When compiling git and running the test suite with `make CC=ccc CFLAGS="-O1"
test`, it fails on test 2050.

Apparently, the return value of a function gets messed up -- I pasted a summary
on http://pastie.textmate.org/private/dzzobqfokiltcqh5uoiema .

When compiling with -O0 everything works as expected. -O1 fails too.

This happens on OS X Leopard with trunk clang and llvm, but also with llvm-gcc
in /Developer/usr/bin

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 13:11:10 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 13:11:10 -0600
Subject: [LLVMbugs] [Bug 3658] Assertion failed:
 (C->getType()->getPrimitiveSizeInBits() < Ty-> getPrimitiveSizeInBits()&&
 "SrcTy must be smaller than DestTy for ZExt!"), function getZExt,
 file Constants.cpp, line 1909.
In-Reply-To: 
Message-ID: <200902241911.n1OJBAxk003489@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3658

Daniel Dunbar  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |daniel at zuster.org
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Daniel Dunbar   2009-02-24 13:11:10 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090223/074185.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 cs.uiuc.edu  Tue Feb 24 13:11:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 13:11:29 -0600
Subject: [LLVMbugs] [Bug 3657] clang -S coredumping
In-Reply-To: 
Message-ID: <200902241911.n1OJBTii003523@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3657

Daniel Dunbar  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 cs.uiuc.edu  Tue Feb 24 14:29:10 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 14:29:10 -0600
Subject: [LLVMbugs] [Bug 3662] New: crash in
 clang::CodeGen::CodeGenFunction:: GetAddrOfStaticLocalVar
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3662

           Summary: crash in
                    clang::CodeGen::CodeGenFunction::GetAddrOfStaticLocalVar
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: LLVM Codegen
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2606)
 --> (http://llvm.org/bugs/attachment.cgi?id=2606)
Pre-processed deps.c

ccc -v -march=nocona -fstack-protector -Wsystem-headers -Werror -Wall
-Wno-format-y2k -W -Wno-unused-parameter -Wstrict-prototypes
-Wmissing-prototypes -Wpointer-arith -Wno-uninitialized -Wno-pointer-sign
-Wformat=2 -Wno-format-extra-args -Werror -c
/usr/src/usr.sbin/pkg_install/lib/deps.c
ccc version 1.0 (http://llvm.org/svn/llvm-project/cfe/trunk)
clang -S -disable-free --relocation-model=static --disable-fp-elim
--nozero-initialized-in-bss --unwind-tables=0 --fmath-errno=1 -v -o
/tmp/tmpTxVRSU.s -x c /usr/src/usr.sbin/pkg_install/lib/deps.c
ignoring nonexistent directory
"/usr/obj/usr/src/usr.sbin/pkg_install/lib/clang/1.0/include"
ignoring nonexistent directory "/System/Library/Frameworks"
ignoring nonexistent directory "/Library/Frameworks"
#include "..." search starts here:
#include <...> search starts here:
 /usr/local/include
 /usr/include
End of search list.
*** Error code 245

(gdb) bt
#0  0x000000000084bfb9 in
clang::CodeGen::CodeGenFunction::GetAddrOfStaticLocalVar ()
#1  0x0000000000800bef in llvm::IRBuilder::CreateSub ()
#2  0x0000000000800ffc in clang::CodeGen::CodeGenModule::EmitConstantExpr ()
#3  0x0000000000802a93 in clang::CodeGen::CodeGenModule::EmitConstantExpr ()
#4  0x0000000000803084 in clang::CodeGen::CodeGenModule::EmitConstantExpr ()
#5  0x00000000007ffb04 in llvm::IRBuilder::CreateSub ()
#6  0x0000000000800e1f in clang::CodeGen::CodeGenModule::EmitConstantExpr ()
#7  0x00000000007e5250 in
clang::CodeGen::CodeGenFunction::GenerateStaticBlockVarDecl ()
#8  0x00000000007e6e01 in
clang::CodeGen::CodeGenFunction::EmitStaticBlockVarDecl ()
#9  0x0000000000840f63 in clang::CodeGen::CodeGenFunction::EmitDeclStmt ()
#10 0x0000000000845c75 in clang::CodeGen::CodeGenFunction::EmitStmt ()
#11 0x0000000000848970 in clang::CodeGen::CodeGenFunction::EmitCompoundStmt ()
#12 0x0000000000848f00 in clang::CodeGen::CodeGenFunction::EmitSimpleStmt ()
#13 0x0000000000845b57 in clang::CodeGen::CodeGenFunction::EmitStmt ()
#14 0x000000000084f2b3 in clang::CodeGen::CodeGenFunction::GenerateCode ()
#15 0x00000000007c6e5e in
clang::CodeGen::CodeGenModule::EmitGlobalFunctionDefinition ()
#16 0x00000000007c763b in clang::CodeGen::CodeGenModule::EmitGlobal ()
---Type  to continue, or q  to quit---
#17 0x00000000007c7a7c in clang::CodeGen::CodeGenModule::EmitTopLevelDecl ()
#18 0x00000000007c06dd in clang::CreateLLVMCodeGen ()
#19 0x000000000055c94a in llvm::cl::opt >::handleOccurrence ()
#20 0x00000000009162e9 in clang::ParseAST ()
#21 0x00000000005a2e8e in InitializeGCMode ()
#22 0x00000000005a5eb2 in 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 cs.uiuc.edu  Tue Feb 24 15:40:35 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 15:40:35 -0600
Subject: [LLVMbugs] [Bug 3663] New: clang rejects integer size expression
 that gcc accepts
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3663

           Summary: clang rejects integer size expression that gcc accepts
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: kremenek at apple.com
                CC: llvmbugs at cs.uiuc.edu

clang rejects the following top-level declaration that gcc accepts:

  static const unsigned array[((2 * (int)((((4) / 2) + 1.0/3.0) * (4) - 1e-8))
+ 1)];

the error:

$ clang t.c
t.c:1:23: error: variable length array declaration not allowed at file scope
static const unsigned array[((2 * (int)((((4) / 2) + 1.0/3.0) * (4) - 1e-8)) +
1)];
                      ^    
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 diagnostic generated.

This is related to .  The line above is the complete
test case.

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

From bugzilla-daemon at cs.uiuc.edu  Tue Feb 24 17:12:12 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 17:12:12 -0600
Subject: [LLVMbugs] [Bug 3648] clang gives @encode a different type from gcc
 ( and related issues)
In-Reply-To: 
Message-ID: <200902242312.n1ONCCgQ011916@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3648

Chris Lattner  changed:

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

--- Comment #4 from Chris Lattner   2009-02-24 17:12:11 ---
I did some refactoring and checked in this patch to wrap up array init with
@encode:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013055.html

Thanks for noticing this 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 cs.uiuc.edu  Tue Feb 24 20:52:20 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 24 Feb 2009 20:52:20 -0600
Subject: [LLVMbugs] [Bug 3664] New: Struct return values inhibit tail call
	elimination
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3664

           Summary: Struct return values inhibit tail call elimination
           Product: libraries
           Version: 2.4
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: jon at ffconsultancy.com
                CC: llvmbugs at cs.uiuc.edu

A call in tail position is not tail call eliminated if the return type of the
function is a struct:

define fastcc { { i8*, i8* }*, i8*} @init({ { i8*, i8* }*, i8*}, i32) {
entry:
      %2 = tail call fastcc { { i8*, i8* }*, i8* } @init({ { i8*, i8*}*, i8*}
%0, i32 %1)
      ret { { i8*, i8* }*, i8*} %2
}

Both tail calls and multiple return values are desirable so this interaction is
unwanted. However, it is easily worked around by resorting to sret form.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 01:39:47 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 01:39:47 -0600
Subject: [LLVMbugs] [Bug 3659] Expensive checks failure in
	ListScheduleBottomUp
In-Reply-To: 
Message-ID: <200902250739.n1P7dlHp029320@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3659

Duncan Sands  changed:

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

--- Comment #2 from Duncan Sands   2009-02-25 01:39:46 ---
Yes, it looks like it's the same as PR3387.

*** This bug has been marked as a duplicate of bug 3387 ***

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 04:44:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 04:44:06 -0600
Subject: [LLVMbugs] [Bug 3665] New: debuginfo: static variables declared
 inside functions are missing original name
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3665

           Summary: debuginfo: static variables declared inside functions
                    are missing original name
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Example, this works with gcc and not with llvm-gcc:

int main()
{
 static int a=4;
 return a;
}

$ llvm-gcc -O0 -g foo.c

$ gdb ./a.out

(gdb) b main
(gdb) run
(gdb) print a
No symbol "a" in current context.

The reason is that 'a' is actually called a.1441:
        %1 = load i32* @a.1441, align 4 

and the debug info descriptor doesn't contain the original name:

@.str5 = internal constant [7 x i8] c"a.1441\00", section "llvm.metadata"      
        ; <[7 x i8]*> [#uses=1]
@llvm.dbg.global_variable = internal constant %llvm.dbg.global_variable.type {
i32 458804, { }* bitcast (%llvm.dbg.anchor.type* @llvm.dbg.global_variables to
{ }*), { }* bitcast (%llvm.dbg.compile_unit.type* @llvm.dbg.compile_unit to {
}*), i8* getelementptr ([7 x i8]* @.str5, i32 0, i32 0), i8* getelementptr ([7
x i8]* @.str5, i32 0, i32 0), i8* null, { }* bitcast
(%llvm.dbg.compile_unit.type* @llvm.dbg.compile_unit to { }*), i32 2, { }*
bitcast (%llvm.dbg.basictype.type* @llvm.dbg.basictype to { }*), i1 true, i1
true, { }* bitcast (i32* @a.1441 to { }*) }, section "llvm.metadata"           
; <%llvm.dbg.global_variable.type*> [#uses=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 cs.uiuc.edu  Wed Feb 25 06:53:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 06:53:06 -0600
Subject: [LLVMbugs] [Bug 3666] New: X86 backend converts calls to constant
 addresses to indirect calls.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3666

           Summary: X86 backend converts calls to constant addresses to
                    indirect calls.
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: marks at dcs.gla.ac.uk
                CC: llvmbugs at cs.uiuc.edu

The following llvm code:

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

define i32 @main() nounwind {
entry:
        %0 = call i32 inttoptr (i32 12345678 to i32 (i32)*)(i32 0) nounwind    
        ;  [#uses=1]
        ret i32 %0
}

Produces the following assembler (with directives removed)

main:
        subl    $4, %esp
        movl    $0, (%esp)
        movl    $12345678, %eax
        call    *%eax
        addl    $4, %esp
        ret

Whereas it should produce this code:

main:
        subl    $4, %esp
        movl    $0, (%esp)
        call    *12345678
        addl    $4, %esp
        ret

This is important since all calls in JIT compiled code, will be in the above
form.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 07:40:28 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 07:40:28 -0600
Subject: [LLVMbugs] [Bug 3667] New: instcombine divide by zero on array with
 zero-length element type
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3667

           Summary: instcombine divide by zero on array with zero-length
                    element type
           Product: new-bugs
           Version: unspecified
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: baldrick at free.fr
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2608)
 --> (http://llvm.org/bugs/attachment.cgi?id=2608)
testcase .ll

$ opt -instcombine -disable-output divide_by_zero.bc
0   opt       0x0000000000d956b2
1   opt       0x0000000000d95a0c
2   libc.so.6 0x00007f2605387040
3   opt       0x0000000000a21c98
4   opt       0x0000000000a223bf
5   opt       0x0000000000a2a8af
6   opt       0x0000000000a2b27a
7   opt       0x0000000000a2b582
8   opt       0x0000000000cefb81
llvm::FPPassManager::runOnFunction(llvm::Function&) + 239
9   opt       0x0000000000cefce4
llvm::FPPassManager::runOnModule(llvm::Module&) + 82
10  opt       0x0000000000ceefb7
llvm::MPPassManager::runOnModule(llvm::Module&) + 189
11  opt       0x0000000000cf05c2 llvm::PassManagerImpl::run(llvm::Module&) +
116
12  opt       0x0000000000cf0625 llvm::PassManager::run(llvm::Module&) + 33
13  opt       0x00000000008f738c main + 4192
14  libc.so.6 0x00007f26053725a6 __libc_start_main + 230
15  opt       0x00000000008d1359
Floating point exception (core dumped)

The problem is that ArrayEltSize is zero here:

10836           // If the index will be to exactly the right offset with the
scale taken
10837           // out, perform the transformation. Note, we don't know whether
Scale is
10838           // signed or not. We'll use unsigned version of division/modulo
10839           // operation after making sure Scale doesn't have the sign bit
set.
10840           if (Scale && Scale->getSExtValue() >= 0LL &&
10841               Scale->getZExtValue() % ArrayEltSize == 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 cs.uiuc.edu  Wed Feb 25 10:33:45 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 10:33:45 -0600
Subject: [LLVMbugs] [Bug 3626] clang rejects K&R call which doesn't match #
	of arguments
In-Reply-To: 
Message-ID: <200902251633.n1PGXjvr026307@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3626

Douglas Gregor  changed:

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

--- Comment #1 from Douglas Gregor   2009-02-25 10:33:44 ---
C99 DR #316 implies that this code is well-formed. However, the C committee
declined to clarify the point, so we go with the behavior this is implied.
Fixed here:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013080.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 cs.uiuc.edu  Wed Feb 25 11:19:39 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 11:19:39 -0600
Subject: [LLVMbugs] [Bug 3656] Error with objc type conversion in
 __attribute__((cleanup)) function argument
In-Reply-To: 
Message-ID: <200902251719.n1PHJdQo028184@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3656

Anders Carlsson  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andersca at mac.com
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Anders Carlsson   2009-02-25 11:19:38 ---
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013082.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 cs.uiuc.edu  Wed Feb 25 12:20:26 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 12:20:26 -0600
Subject: [LLVMbugs] [Bug 3667] instcombine divide by zero on array with
 zero-length element type
In-Reply-To: 
Message-ID: <200902251820.n1PIKQEx030186@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3667

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-25 12:20:25 ---
Fixed:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090223/074225.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 cs.uiuc.edu  Wed Feb 25 12:37:13 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 12:37:13 -0600
Subject: [LLVMbugs] [Bug 3669] New: [clang on Wine]: constant integer size
 rejected as not being constant
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3669

           Summary: [clang on Wine]: constant integer size rejected as not
                    being constant
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: kremenek at apple.com
                CC: llvmbugs at cs.uiuc.edu

The following self-contained code example (reduced from a case in Wine) is
rejected by Clang because the size of the array is deemed not a constant:

typedef unsigned long UINT_PTR, *PUINT_PTR;
typedef struct HHOOK__ {} *HKEY;
static HKEY special_root_keys[((UINT_PTR)((HKEY) 0x80000006) -
(UINT_PTR)((HKEY) 0x80000000) + 1)];

The error emitted by clang:

t.c:3:13: error: variable length array declaration not allowed at file scope
static HKEY special_root_keys[((UINT_PTR)((HKEY) 0x80000006) -
(UINT_PTR)((HKEY) 0x80000000) + 1)];
            ^                
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 diagnostic 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 cs.uiuc.edu  Wed Feb 25 13:42:03 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 13:42:03 -0600
Subject: [LLVMbugs] [Bug 3665] debuginfo: static variables declared inside
 functions are missing original name
In-Reply-To: 
Message-ID: <200902251942.n1PJg3l8000844@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3665

Devang Patel  changed:

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

--- Comment #2 from Devang Patel   2009-02-25 13:42:02 ---
Thanks Dale. Applied your patch. rev 65465.
Updated DwarfWriter to handle second part. rev. 65468

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 14:09:42 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 14:09:42 -0600
Subject: [LLVMbugs] [Bug 3662] crash in clang::CodeGen::CodeGenFunction::
 GetAddrOfStaticLocalVar
In-Reply-To: 
Message-ID: <200902252009.n1PK9gPm001813@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3662

Daniel Dunbar  changed:

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

--- Comment #2 from Daniel Dunbar   2009-02-25 14:09:41 ---
Fixed here:
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013091.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 cs.uiuc.edu  Wed Feb 25 18:27:24 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 18:27:24 -0600
Subject: [LLVMbugs] [Bug 3671] New: clang fails parse stdio.h from (gentoo)
	glibc-2.6.1
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3671

           Summary: clang fails parse stdio.h from (gentoo) glibc-2.6.1
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: piotr.rak at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2611)
 --> (http://llvm.org/bugs/attachment.cgi?id=2611)
Reduced testcase.

As Eli Friedman requested in
http://lists.cs.uiuc.edu/pipermail/cfe-dev/2009-February/004474.html reporting
it here.

Here is a code causing problem:

struct __gconv_step_data {};

typedef struct __gconv_info
{
  void * dumb;
  __extension__ struct __gconv_step_data __data [];
} *__gconv_t;

typedef union
{
  struct __gconv_info __cd;
  struct
  {
    struct __gconv_info __cd;
    struct __gconv_step_data __data;
  } __combined;
} _G_iconv_t;

reduced.c:15:25: error: variable sized type '__cd' must be at end of struct or
class
    struct __gconv_info __cd;

union _G_iconv_t definition is in <_G_config.h>, struct _gconv_info in

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 19:46:54 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 19:46:54 -0600
Subject: [LLVMbugs] [Bug 3672] New: Syntax error not rejected with
	-pedantic-errors
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3672

           Summary: Syntax error not rejected with -pedantic-errors
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: NetBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: neil at daikokuya.co.uk
                CC: llvmbugs at cs.uiuc.edu

void g18(void) {
  extern int g19;
}
int *p=&g19;

g19 is undefined, therefore not an expression, so this is a syntax error and
must be rejected with -pedantic-errors.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 20:43:16 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 20:43:16 -0600
Subject: [LLVMbugs] [Bug 3669] [clang on Wine]: constant integer size
 rejected as not being constant
In-Reply-To: 
Message-ID: <200902260243.n1Q2hGJh014523@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3669

Eli Friedman  changed:

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

--- Comment #1 from Eli Friedman   2009-02-25 20:43:15 ---

*** This bug has been marked as a duplicate of bug 3663 ***

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 21:59:21 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 21:59:21 -0600
Subject: [LLVMbugs] [Bug 3663] clang rejects integer size expression that
	gcc accepts
In-Reply-To: 
Message-ID: <200902260359.n1Q3xLcL016720@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3663

Eli Friedman  changed:

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

--- Comment #5 from Eli Friedman   2009-02-25 21:59:20 ---
Fixed in r65500.

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

From bugzilla-daemon at cs.uiuc.edu  Wed Feb 25 22:16:50 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 25 Feb 2009 22:16:50 -0600
Subject: [LLVMbugs] [Bug 1890] clang allows illegal redeclaration of
	function?
In-Reply-To: 
Message-ID: <200902260416.n1Q4GopF017340@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1890

Eli Friedman  changed:

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

--- Comment #7 from Eli Friedman   2009-02-25 22:16:48 ---
This was fixed recently, at least for 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 cs.uiuc.edu  Thu Feb 26 01:16:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 01:16:06 -0600
Subject: [LLVMbugs] [Bug 3521] Handling of constant initializers needs
	cleaning
In-Reply-To: 
Message-ID: <200902260716.n1Q7G6HK026861@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3521

Daniel Dunbar  changed:

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

--- Comment #9 from Daniel Dunbar   2009-02-26 01:16:05 ---
I don't think so; at least we don't need the bug anymore. 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 cs.uiuc.edu  Thu Feb 26 08:51:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 08:51:33 -0600
Subject: [LLVMbugs] [Bug 3673] New: clang miscompilation
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3673

           Summary: clang miscompilation
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Basic
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2612)
 --> (http://llvm.org/bugs/attachment.cgi?id=2612)
diff of .s files

The following program SEGV's when built with clang, llvm-gcc and gcc work fine.

#include 
#include 
#include 

static DB *symtable;

int
main(void)
{
        printf("dbopen()\n");
        symtable = dbopen(/*filename*/NULL,
                          O_CREAT | O_NONBLOCK | O_RDWR, /*mode*/0, DB_HASH,
                          /*openinfo*/NULL);
        printf("after dbopen()\n");
}

(gdb) break main
Breakpoint 1 at 0x4005c8: file t.c, line 10.
(gdb) run
Starting program: /tmp/t

Breakpoint 1, main () at t.c:10
10              printf("dbopen()\n");
(gdb) step
dbopen()
11              symtable = dbopen(/*filename*/NULL,
(gdb)

Program received signal SIGSEGV, Segmentation fault.
0x00000008007152fc in __bt_open () from /lib/libc.so.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 cs.uiuc.edu  Thu Feb 26 09:56:45 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 09:56:45 -0600
Subject: [LLVMbugs] [Bug 3544] Incorrect warning when compiling AIUtilities
In-Reply-To: 
Message-ID: <200902261556.n1QFujl8022030@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3544

snaroff at apple.com  changed:

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

--- Comment #5 from snaroff at apple.com  2009-02-26 09:56:44 ---
Fixed in http://llvm.org/viewvc/llvm-project?rev=65532&view=rev.

Thanks for the report (and sorry for the inconvenience).

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 26 11:44:56 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 11:44:56 -0600
Subject: [LLVMbugs] [Bug 3673] clang miscompilation
In-Reply-To: 
Message-ID: <200902261744.n1QHiu34026881@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3673

Anders Carlsson  changed:

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

--- Comment #8 from Anders Carlsson   2009-02-26 11:44:56 ---
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013123.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 cs.uiuc.edu  Thu Feb 26 15:01:25 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 15:01:25 -0600
Subject: [LLVMbugs] [Bug 3410] Better 'expected ';
	' at end of statement diagnostic
In-Reply-To: 
Message-ID: <200902262101.n1QL1P44001808@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3410

Douglas Gregor  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 cs.uiuc.edu  Thu Feb 26 15:20:08 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 15:20:08 -0600
Subject: [LLVMbugs] [Bug 3674] New: scan-build ignores -ffreestanding
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3674

           Summary: scan-build ignores -ffreestanding
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ed at 80386.nl
                CC: llvmbugs at cs.uiuc.edu

When using scan-build to perform some analysis on the FreeBSD kernel, it seems
to ignore -ffreestanding, so we get a whole bunch of files that won't be
processed, because they call FreeBSD's kernelspace malloc(), which uses a
different set of arguments.

../../../dev/acpica/acpi_ec.c:327:52: error: too many arguments to function
call
    params = malloc(sizeof(struct acpi_ec_params), M_TEMP, M_WAITOK | M_ZERO);
             ~~~~~~                                ^~~~~~~~~~~~~~~~~~~~~~~~~

I guess it would eventually be nice if there was some kind of feature to test
if malloc() return values are tested for NULL. In the FreeBSD kernel, this is
only needed when M_NOWAIT is used, instead of M_WAITOK. This is probably a lot
harder...

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

From bugzilla-daemon at cs.uiuc.edu  Thu Feb 26 17:21:13 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 17:21:13 -0600
Subject: [LLVMbugs] [Bug 3674] scan-build ignores -ffreestanding
In-Reply-To: 
Message-ID: <200902262321.n1QNLD6i007122@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3674

Ted Kremenek  changed:

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

--- Comment #2 from Ted Kremenek   2009-02-26 17:21:12 ---
Fixed:

http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20090223/013160.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 cs.uiuc.edu  Thu Feb 26 17:44:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 17:44:19 -0600
Subject: [LLVMbugs] [Bug 3097] clang refuses wchar_t foo[]
In-Reply-To: 
Message-ID: <200902262344.n1QNiJdL008071@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3097

Chris Lattner  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|clang does not support wide |clang refuses wchar_t foo[]
                   |character strings           |

--- Comment #7 from Chris Lattner   2009-02-26 17:44:19 ---
With a series of patches, this works 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 cs.uiuc.edu  Thu Feb 26 18:01:59 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 18:01:59 -0600
Subject: [LLVMbugs] [Bug 3218] ternary confuses literal format string
	detection
In-Reply-To: 
Message-ID: <200902270001.n1R01xtM008791@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3218

Ted Kremenek  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|DUPLICATE                   |FIXED

--- Comment #4 from Ted Kremenek   2009-02-26 18:01:58 ---
This has been fixed for a few months now.  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 cs.uiuc.edu  Thu Feb 26 18:02:39 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 26 Feb 2009 18:02:39 -0600
Subject: [LLVMbugs] [Bug 1870] format string checker does not employ simple
 constant propagation
In-Reply-To: 
Message-ID: <200902270002.n1R02d4o008849@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1870

Ted Kremenek  changed:

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

--- Comment #3 from Ted Kremenek   2009-02-26 18:02:38 ---
This bug has been fixed for a while now.  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 cs.uiuc.edu  Fri Feb 27 00:29:29 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 00:29:29 -0600
Subject: [LLVMbugs] [Bug 3642] Assertion failed:
 (CGT.getTargetData().getTypePaddedSizeInBits( STy) == RL.getSize()),
 function layoutStructFields, file CodeGenTypes.cpp, line 514.
In-Reply-To: 
Message-ID: <200902270629.n1R6TTte021290@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3642

Daniel Dunbar  changed:

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

--- Comment #9 from Daniel Dunbar   2009-02-27 00:29:28 ---

*** This bug has been marked as a duplicate of bug 3671 ***

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 27 01:28:40 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 01:28:40 -0600
Subject: [LLVMbugs] [Bug 3675] New: potentially bogus error: " member of
 anonymous struct redeclares 'u'"
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3675

           Summary: potentially bogus error: "member of anonymous struct
                    redeclares 'u'"
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Semantic Analyzer
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: kremenek at apple.com
                CC: llvmbugs at cs.uiuc.edu

gcc-4.2 -fsyntax-only eats the following self-contained test case but clang
rejects it:

typedef struct _IRP {
  union {
    struct {
      union {} u;
      struct {
        union {} u;
      };
    } Overlay;
  } Tail;
} IRP;

This seemingly contrived test-case was reduced from actual code in Wine.

The error I see:

$ clang t.c
t.c:6:18: error: member of anonymous struct redeclares 'u'
        union {} u;
                 ^
t.c:4:16: note: previous declaration is here
      union {} u;
               ^
2 diagnostics 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 cs.uiuc.edu  Fri Feb 27 02:14:21 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 02:14:21 -0600
Subject: [LLVMbugs] [Bug 3676] New: ADT/hash_map raises warning on g++ 4.3
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3676

           Summary: ADT/hash_map raises warning on g++ 4.3
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: minor
          Priority: P2
         Component: Support Libraries
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca
                CC: llvmbugs at cs.uiuc.edu

ADT/hash_map.h maps to #include  on gcc 3.1+. However, by gcc
4.3, that's obsolete and GCC generates a warning.

The replacement is #include . We may need to typedef the
name or somesuch, but otherwise I expect it to work out of the box.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 27 08:02:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 08:02:06 -0600
Subject: [LLVMbugs] [Bug 3678] New: Invalid input constraint when compiling
 FreeBSD' s libm on AMD64
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3678

           Summary: Invalid input constraint when compiling FreeBSD's libm
                    on AMD64
           Product: new-bugs
           Version: unspecified
          Platform: Macintosh
               URL: http://svn.freebsd.org/viewvc/base/head/lib/msun/src/mat
                    h_private.h?view=markup
        OS/Version: Mac System 9.x
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ed at 80386.nl
                CC: llvmbugs at cs.uiuc.edu, rdivacky at freebsd.org

When compiling the attached test case, one gets the following compiler error:

test.c:10:36: error: invalid input constraint 'Y' in asm
        asm("cvtsd2si %1,%0" : "=r" (n) : "Y" (x));

This error does not occur when compiling with 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 cs.uiuc.edu  Fri Feb 27 09:47:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 09:47:19 -0600
Subject: [LLVMbugs] [Bug 3679] New: #pragma weak doesn't work
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3679

           Summary: #pragma weak doesn't work
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2619)
 --> (http://llvm.org/bugs/attachment.cgi?id=2619)
Test case

0>root at one /tmp# gcc -o t t.c
0>root at one /tmp# ./t
spasiba
0>root at one /tmp# ccc -o t t.c
/tmp/tmpTGEXC1.o(.text+0x2b): In function `main':
: undefined reference to `t'

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 27 10:44:22 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 10:44:22 -0600
Subject: [LLVMbugs] [Bug 3675] potentially bogus error: "member of anonymous
 struct redeclares 'u'"
In-Reply-To: 
Message-ID: <200902271644.n1RGiMjH019883@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3675

Ted Kremenek  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         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 cs.uiuc.edu  Fri Feb 27 11:01:49 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 11:01:49 -0600
Subject: [LLVMbugs] [Bug 3680] New: anonymous union: invalid error & poor
	diagnostic
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3680

           Summary: anonymous union: invalid error & poor diagnostic
           Product: clang
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: parser
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: daniel at zuster.org
                CC: llvmbugs at cs.uiuc.edu

This is wrong:
--
ddunbar at 67-218-110-109:EH$ echo "struct {};" | xcc -fsyntax-only -x c -
echo "struct {};" | xcc -fsyntax-only -x c -
:1:1: error: anonymous structs must be struct or union members
struct {};
^
1 diagnostic generated.
ddunbar at 67-218-110-109:EH$ echo "struct {};" | gcc -fsyntax-only -x c -
echo "struct {};" | gcc -fsyntax-only -x c -
:1: warning: unnamed struct/union that defines no instances
--

And the diagnostic here is wrong (';' works too):
--
ddunbar at 67-218-110-109:Sema$ cat t.c
struct s0 {
  union {
    int a;
  }
};
ddunbar at 67-218-110-109:Sema$ clang t.c
t.c:5:1: error: expected identifier or '('
};
^
1 diagnostic 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 cs.uiuc.edu  Fri Feb 27 11:35:52 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 11:35:52 -0600
Subject: [LLVMbugs] [Bug 3682] New: Assertion failed: (0 && "Operand number
 out of range!"), function ConvertAsmString, file CGStmt.cpp, line 720.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3682

           Summary: Assertion failed: (0 && "Operand number out of range!"),
                    function ConvertAsmString, file CGStmt.cpp, line 720.
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2621)
 --> (http://llvm.org/bugs/attachment.cgi?id=2621)
Pre-processed cmll_cbc.c

ccc -O2 -pipe -march=nocona -DTERMIOS -DANSI_SOURCE
-I/usr/src/secure/lib/libcrypto/../../../crypto/openssl
-I/usr/src/secure/lib/libcrypto/../../../crypto/openssl/crypto
-I/usr/obj/usr/src/secure/lib/libcrypto -DOPENSSL_THREADS -DDSO_DLFCN
-DHAVE_DLFCN_H -DOPENSSL_NO_IDEA -DL_ENDIAN -DNO_IDEA -std=gnu89
-fstack-protector  -c
/usr/src/secure/lib/libcrypto/../../../crypto/openssl/crypto/camellia/cmll_cbc.c
Assertion failed: (0 && "Operand number out of range!"), function
ConvertAsmString, file CGStmt.cpp, line 720.

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 27 12:12:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 12:12:06 -0600
Subject: [LLVMbugs] [Bug 3678] Invalid input constraint when compiling
 FreeBSD's libm/ OpenSSL on AMD64
In-Reply-To: 
Message-ID: <200902271812.n1RIC6Kr023169@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3678

Dale Johannesen  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dalej at apple.com
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #5 from Dale Johannesen   2009-02-27 12:12:05 ---
What gcc is that?   This may have changed in gcc mainline at some point; as of
4.2, "Y" is a prefix, and you need to say "Yt" to match any SSE2 register. 
gcc-4.2 behaves the same 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 cs.uiuc.edu  Fri Feb 27 15:42:46 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 15:42:46 -0600
Subject: [LLVMbugs] [Bug 3678] Invalid input constraint when compiling
 FreeBSD's libm/ OpenSSL on AMD64
In-Reply-To: 
Message-ID: <200902272142.n1RLgk27030382@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3678

Ed Schouten  changed:

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

--- Comment #7 from Ed Schouten   2009-02-27 15:42:45 ---
Even with the proposed Yt, I still get get the same compiler error. I'm
reopening this ticket, because in my opinion the current resolution is 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 cs.uiuc.edu  Fri Feb 27 16:45:06 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 16:45:06 -0600
Subject: [LLVMbugs] [Bug 3612] printf(printf);
	leads to invalid codegen and no diagnosis
In-Reply-To: 
Message-ID: <200902272245.n1RMj68h032699@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3612

Mike Stump  changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mrs at apple.com
             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 cs.uiuc.edu  Fri Feb 27 16:47:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 16:47:33 -0600
Subject: [LLVMbugs] [Bug 3683] New: Testcase failure:
	CodeGen/Generic/vector-casts.ll
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3683

           Summary: Testcase failure: CodeGen/Generic/vector-casts.ll
           Product: libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: PowerPC
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ggreif at gmail.com
                CC: llvmbugs at cs.uiuc.edu

This is a PPC mac mini. Xcode 2.5.

FAIL: /gdm/llvm/test/CodeGen/Generic/vector-casts.ll for PR2671
Failed with signal(SIGABRT) at line 1
while running:  llvm-as < /gdm/llvm/test/CodeGen/Generic/vector-casts.ll | llc
LegalizeVectorTypes.cpp:686: failed assertion `0 && "Unexpected type action!"'

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

From bugzilla-daemon at cs.uiuc.edu  Fri Feb 27 23:42:19 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 27 Feb 2009 23:42:19 -0600
Subject: [LLVMbugs] [Bug 3259] sema should diagnose illegal jump into scope
 of variably modified type
In-Reply-To: 
Message-ID: <200902280542.n1S5gJlg013041@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3259

Eli Friedman  changed:

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

--- Comment #1 from Eli Friedman   2009-02-27 23:42:14 ---
This is more "in progress" than actually fixed, but I committed the patch in
r65678 is in, so there isn't really anything to track 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 cs.uiuc.edu  Sat Feb 28 07:35:34 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 07:35:34 -0600
Subject: [LLVMbugs] [Bug 3684] New: Newline token has start of line flag.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3684

           Summary: Newline token has start of line flag.
           Product: clang
           Version: unspecified
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Basic
        AssignedTo: unassignedclangbugs at nondot.org
        ReportedBy: bolzoni at cs.unipr.it
                CC: llvmbugs at cs.uiuc.edu, bagnara at cs.unipr.it

Using the raw lexer newlines become a token. But the newline token itself has
start of line flag, not the one after.

e.g.,
---->
/* multi purpose static space. */
unsigned char buffer[0xff] __attribute__((aligned));

int f();

----<

becomes (in a nicely formatted output):
----->
comment('/* multi purpose static space. */', [start_of_line])
unknown('\n', [start_of_line])
identifier('unsigned', [])
unknown(' ', [])
identifier('char', [])
unknown(' ', [])
identifier('buffer', [])
l_square([, [])
numeric_constant('0xff', [])
r_square(']', [])
unknown(' ', [])
identifier('__attribute__', [])
l_paren('(', [])
l_paren('(', [])
identifier('aligned', [])
r_paren(')', [])
r_paren(')', [])
semi(';', [])
unknown('\n', [start_of_line])
identifier('int', [])
unknown(' ', [])
identifier('f', [])
l_paren('(', [])
r_paren(')', [])
semi(';', [])
unknown('\n', [start_of_line])
-----<

as you can see all the newlines have the start of line flags.
I think it was intended that `the comment,' `unsigned,' and `int' need the
start of line flag.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 28 10:52:09 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 10:52:09 -0600
Subject: [LLVMbugs] [Bug 3228] Bad behavior with labels+blocks extension
In-Reply-To: 
Message-ID: <200902281652.n1SGq9Fe011991@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3228

snaroff at apple.com  changed:

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

--- Comment #2 from snaroff at apple.com  2009-02-28 10:52:08 ---
Fixed by http://llvm.org/viewvc/llvm-project?rev=65694&view=rev.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 28 11:37:56 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 11:37:56 -0600
Subject: [LLVMbugs] [Bug 3685] New: llvm-as should catch zero width vector
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3685

           Summary: llvm-as should catch zero width vector
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-as
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca
                CC: llvmbugs at cs.uiuc.edu

This .ll file:

  @a = internal constant void bitcast(<0 x i8> <> to void)

produces an assertion:

llvm-as: Type.cpp:504: llvm::VectorType::VectorType(const llvm::Type*, unsigned
int): Assertion `NumEl > 0 && "NumEl of a VectorType must be greater than 0"'
failed.
0   llvm-as 0x081a40fd
1   llvm-as 0x081a447b
2           0xffffe400 __kernel_sigreturn + 0
Aborted

Well, yeah. :-) llvm-as should emit a polite error message 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 cs.uiuc.edu  Sat Feb 28 12:13:14 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 12:13:14 -0600
Subject: [LLVMbugs] [Bug 3685] llvm-as should catch zero width vector
In-Reply-To: 
Message-ID: <200902281813.n1SIDE9T014892@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=3685

Chris Lattner  changed:

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

--- Comment #1 from Chris Lattner   2009-02-28 12:13:13 ---
Fixed:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090223/074377.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 cs.uiuc.edu  Sat Feb 28 13:22:33 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 13:22:33 -0600
Subject: [LLVMbugs] [Bug 3686] New: Assertion failed: (0 && " Do not know
 how to promote this operator's operand!"), function PromoteIntegerOperand,
 file LegalizeIntegerTypes.cpp, line 666.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3686

           Summary: Assertion failed: (0 && "Do not know how to promote this
                    operator's operand!"), function PromoteIntegerOperand,
                    file LegalizeIntegerTypes.cpp, line 666.
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2623)
 --> (http://llvm.org/bugs/attachment.cgi?id=2623)
Pre-processed _hldtoa.c

ccc -O2 -pipe -march=nocona -I/usr/src/lib/libc/include
-I/usr/src/lib/libc/../../include -I/usr/src/lib/libc/amd64
-D__DBINTERFACE_PRIVATE -I/usr/src/lib/libc/../../contrib/gdtoa -DINET6
-I/usr/obj/usr/src/lib/libc -I/usr/src/lib/libc/resolv -DPOSIX_MISTAKE
-I/usr/src/lib/libc/locale -DBROKEN_DES -DPORTMAP -DDES_BUILTIN
-I/usr/src/lib/libc/rpc -DYP -DNS_CACHING -DSYMBOL_VERSIONING -fstack-protector
-Wsystem-headers -Werror -Wall -Wno-format-y2k -Wno-uninitialized
-Wno-pointer-sign -c /usr/src/lib/libc/gdtoa/_hldtoa.c
PromoteIntegerOperand Op #0: 0x801ef4378: f80 = bit_convert 0x801ef2850
Assertion failed: (0 && "Do not know how to promote this operator's operand!"),
function PromoteIntegerOperand, file LegalizeIntegerTypes.cpp, line 666.

This only happens at -O2. I tried to reduce the testcase but my delta-foo is to
weak.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 28 13:29:45 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 13:29:45 -0600
Subject: [LLVMbugs] [Bug 3687] New: Assertion failed: ((i >=
 FTy->getNumParams() || FTy-> getParamType(i) == Params[i]->getType()) && "
 Calling a function with a bad signature!"), function init,
 file Instructions.cpp, line 294.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3687

           Summary: Assertion failed: ((i >= FTy->getNumParams() || FTy-
                    >getParamType(i) == Params[i]->getType()) && "Calling a
                    function with a bad signature!"), function init, file
                    Instructions.cpp, line 294.
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2624)
 --> (http://llvm.org/bugs/attachment.cgi?id=2624)
Pre-processed env.c

ccc -O2 -pipe -march=nocona -DHAVE_CONFIG_H -I/usr/src/gnu/lib/libgomp -I.
-I/usr/src/gnu/lib/libgomp/../../../contrib/gcclibs/libgomp
-I/usr/src/gnu/lib/libgomp/../../../contrib/gcclibs/libgomp/config/posix
-fstack-protector  -c
/usr/src/gnu/lib/libgomp/../../../contrib/gcclibs/libgomp/env.c
Assertion failed: ((i >= FTy->getNumParams() || FTy->getParamType(i) ==
Params[i]->getType()) && "Calling a function with a bad signature!"), function
init, file Instructions.cpp, line 294.

Fails to build libgomp on FreeBSD.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 28 13:36:26 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 13:36:26 -0600
Subject: [LLVMbugs] [Bug 3688] New: crash in
	clang::NamedDecl::declarationReplaces
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3688

           Summary: crash in clang::NamedDecl::declarationReplaces
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2625)
 --> (http://llvm.org/bugs/attachment.cgi?id=2625)
Pre-processed annotate.c

clang crashes when building libgdb on FreeBSD. (gdb 6.1.1)

ccc  -march=nocona -Dxregcomp=regcomp -Dxre_exec=re_exec -Dxregexec=regexec
-Dxre_search=re_search -Dxre_compile_fastmap=re_compile_fastmap
-Dxregerror=regerror -Dxre_comp=re_comp -Dxre_set_syntax=re_set_syntax
-DHAVE_CONFIG_H -DRL_NO_COMPAT -DMI_OUT=1 -DTUI=1 -I.
-I/usr/src/gnu/usr.bin/gdb/libgdb/../arch/amd64
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../binutils/libbfd
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../binutils/libbfd/amd64
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../../../contrib/gdb/gdb
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../../../contrib/gdb/gdb/config
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../../../contrib/gdb/include
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../../../contrib/binutils/include
-I/usr/src/gnu/usr.bin/gdb/libgdb/../../../../contrib/binutils/bfd
-fstack-protector  -c
/usr/src/gnu/usr.bin/gdb/libgdb/../../../../contrib/gdb/gdb/annotate.c
*** Error code 245

(gdb) thread apply all bt

Thread 1 (Thread 801c021c0 (LWP 100136)):
#0  0x000000000070a5b1 in clang::NamedDecl::declarationReplaces ()
#1  0x00000000004957ba in 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 cs.uiuc.edu  Sat Feb 28 14:07:55 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 14:07:55 -0600
Subject: [LLVMbugs] [Bug 3689] New: clang: support -dD -E
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3689

           Summary: clang: support -dD -E
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: enhancement
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: edwintorok at gmail.com
                CC: llvmbugs at cs.uiuc.edu

clang currently supports -dM -E, however it doesn't support -dD -E.

I find -dD -E to be useful when tracing what has been expanded from headers,
and why with a single command.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 28 16:39:25 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 16:39:25 -0600
Subject: [LLVMbugs] [Bug 3690] New: Assertion failed: (StackTop == 0 && "
 Stack should be empty after a call!"), function handleSpecialFP,
 file X86FloatingPoint.cpp, line 952.
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3690

           Summary: Assertion failed: (StackTop == 0 && "Stack should be
                    empty after a call!"), function handleSpecialFP, file
                    X86FloatingPoint.cpp, line 952.
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

Created an attachment (id=2626)
 --> (http://llvm.org/bugs/attachment.cgi?id=2626)
Pre-processed ldexp.c

ccc -pipe -march=nocona -I/usr/src/lib/libc/include
-I/usr/src/lib/libc/../../include -I/usr/src/lib/libc/amd64
-D__DBINTERFACE_PRIVATE -I/usr/src/lib/libc/../../contrib/gdtoa -DINET6
-I/usr/obj/usr/src/lib/libc -I/usr/src/lib/libc/resolv -DPOSIX_MISTAKE
-I/usr/src/lib/libc/locale -DBROKEN_DES -DPORTMAP -DDES_BUILTIN
-I/usr/src/lib/libc/rpc -DYP -DNS_CACHING -DSYMBOL_VERSIONING -fstack-protector
-Wsystem-headers -Werror -Wall -Wno-format-y2k -Wno-uninitialized
-Wno-pointer-sign -c /usr/src/lib/libc/amd64/gen/ldexp.c
Assertion failed: (StackTop == 0 && "Stack should be empty after a call!"),
function handleSpecialFP, file X86FloatingPoint.cpp, line 952.

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

From bugzilla-daemon at cs.uiuc.edu  Sat Feb 28 16:42:52 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 16:42:52 -0600
Subject: [LLVMbugs] [Bug 3691] New: Support for complex modes missing
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3691

           Summary: Support for complex modes missing
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: FreeBSD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: pawel.worach at gmail.com
                CC: llvmbugs at cs.uiuc.edu

ccc -c -pipe -march=nocona -DIN_GCC -DIN_LIBGCC2 -D__GCC_FLOAT_NOT_NEEDED 
-DHAVE_GTHR_DEFAULT  -I/usr/src/gnu/lib/libgcc/../../../contrib/gcclibs/include
 -I/usr/src/gnu/lib/libgcc/../../../contrib/gcc/config
-I/usr/src/gnu/lib/libgcc/../../../contrib/gcc -I. 
-I/usr/src/gnu/lib/libgcc/../../usr.bin/cc/cc_tools -fstack-protector 
-fvisibility=hidden -DHIDE_EXPORTS -fPIC  -DL_muldi3 -o _muldi3.o
/usr/src/gnu/lib/libgcc/../../../contrib/gcc/libgcc2.c
In file included from
/usr/src/gnu/lib/libgcc/../../../contrib/gcc/libgcc2.c:65:
/usr/src/gnu/lib/libgcc/../../../contrib/gcc/libgcc2.h:159:47: error: unknown
machine mode 'SC'
typedef _Complex float SCtype   __attribute__ ((mode (SC)));
                                                ^
/usr/src/gnu/lib/libgcc/../../../contrib/gcc/libgcc2.h:163:47: error: unknown
machine mode 'DC'
typedef _Complex float DCtype   __attribute__ ((mode (DC)));
                                                ^
/usr/src/gnu/lib/libgcc/../../../contrib/gcc/libgcc2.h:166:39: error:
unsupported machine mode 'XF'
typedef         float XFtype    __attribute__ ((mode (XF)));
                                                ^
/usr/src/gnu/lib/libgcc/../../../contrib/gcc/libgcc2.h:167:47: error: unknown
machine mode 'XC'
typedef _Complex float XCtype   __attribute__ ((mode (XC)));
                                                ^
4 diagnostics 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 cs.uiuc.edu  Sat Feb 28 17:08:03 2009
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 28 Feb 2009 17:08:03 -0600
Subject: [LLVMbugs] [Bug 3692] New: llvm configure fails if current path
	includes spaces
Message-ID: 

http://llvm.org/bugs/show_bug.cgi?id=3692

           Summary: llvm configure fails if current path includes spaces
           Product: new-bugs
           Version: unspecified
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: luketheh at apple.com
                CC: llvmbugs at cs.uiuc.edu

The shell commands at the end of the config.status file fail on OS X when the
srcdir has a space in its path. We should handle this case and build
successfully.

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