From bugzilla-daemon at cs.uiuc.edu  Wed Nov  1 09:06:59 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 1 Nov 2006 09:06:59 -0600
Subject: [LLVMbugs] [Bug 979] NEW: condprop pass asserts on test case
Message-ID: <200611011506.kA1F6x5V013277@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=979

           Summary: condprop pass asserts on test case
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Scalar Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: criswell at uiuc.edu


The condprop pass asserts out with the following error on the given testcase:

opt: CondPropagate.cpp:203: void
<unnamed>::CondProp::RevectorBlockTo(llvm::BasicBlock*, llvm::BasicBlock*):
Assertion `!isa<PHINode>(ToBB->begin()) && "Critical Edge Found!"' failed.
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::PrintStackTrace()+0x1a)[0x870c752]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::SignalHandler(int)+0x102)[0x870c9fa]
/lib/tls/libc.so.6[0x4caa58]
/lib/tls/libc.so.6(abort+0x1d2)[0x4cc342]
/lib/tls/libc.so.6(__assert_fail+0x108)[0x4c4338]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::CondProp::RevectorBlockTo(llvm::BasicBlock*,
llvm::BasicBlock*)+0xb7)[0x84f807d]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::CondProp::SimplifyPredecessors(llvm::BranchInst*)+0x156)[0x84f7e1e]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::CondProp::SimplifyBlock(llvm::BasicBlock*)+0x99)[0x84f7a8d]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::CondProp::runOnFunction(llvm::Function&)+0x6b)[0x84f79c9]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::FunctionPassManagerT::runPass(llvm::FunctionPass*,
llvm::Function*)+0x1f)[0x86ab359]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManagerT<llvm::FTraits>::runPasses(llvm::Function*,
std::map<llvm::Pass*, std::vector<llvm::Pass*, std::allocator<llvm::Pass*> >,
std::less<llvm::Pass*>, std::allocator<std::pair<llvm::Pass* const,
std::vector<llvm::Pass*, std::allocator<llvm::Pass*> > > > >&)+0x136)[0x86a9b96]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManagerT<llvm::FTraits>::runOnUnit(llvm::Function*)+0x177)[0x86a9a4b]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::FunctionPassManagerT::runOnFunction(llvm::Function&)+0x25)[0x86ab3f7]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::FunctionPass::runOnModule(llvm::Module&)+0xa5)[0x869858f]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::ModulePassManager::runPass(llvm::ModulePass*,
llvm::Module*)+0x1f)[0x86acd91]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManagerT<llvm::MTraits>::runPasses(llvm::Module*,
std::map<llvm::Pass*, std::vector<llvm::Pass*, std::allocator<llvm::Pass*> >,
std::less<llvm::Pass*>, std::allocator<std::pair<llvm::Pass* const,
std::vector<llvm::Pass*, std::allocator<llvm::Pass*> > > > >&)+0x136)[0x86abc10]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManagerT<llvm::MTraits>::runOnUnit(llvm::Module*)+0x177)[0x86abac5]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::ModulePassManager::runOnModule(llvm::Module&)+0x25)[0x869781d]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManager::run(llvm::Module&)+0x23)[0x8697b23]
/localhome/criswell/latestllvm/Debug/bin/opt(main+0xaaa)[0x841c636]
/lib/tls/libc.so.6(__libc_start_main+0xf0)[0x4b8770]
/localhome/criswell/latestllvm/Debug/bin/opt(std::__throw_logic_error(char
const*)+0x59)[0x840f8e1]
Abort (core dumped)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov  1 15:13:29 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 1 Nov 2006 15:13:29 -0600
Subject: [LLVMbugs] [Bug 980] NEW: llvm-gcc4 crash (yet another in out linux
	kernel series)
Message-ID: <200611012113.kA1LDT1x008563@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=980

           Summary: llvm-gcc4 crash (yet another in out linux kernel series)
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: alenhar2 at uiuc.edu


The following reduction (from the linux kernel) causes:

cc1: ../../source/gcc/llvm-types.cpp:724: unsigned int
StructTypeConversionInfo::getLLVMFieldFor(uint64_t, unsigned int&, bool):
Assertion `0 && "Could not find field!"' failed.


typedef struct {} rwlock_t;

struct inet_hashinfo {
 struct inet_ehash_bucket *ehash;
 rwlock_t lhash_lock __attribute__((__aligned__((1 << (5)))));
};

void __inet6_hash(struct inet_hashinfo *hashinfo)//, struct sock *sk)
{ }



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov  1 16:32:23 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 1 Nov 2006 16:32:23 -0600
Subject: [LLVMbugs] [Bug 979] condprop pass asserts on test case
Message-ID: <200611012232.kA1MWNI6010347@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=979

dpatel at apple.com changed:

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



------- Additional Comments From dpatel at apple.com  2006-11-01 16:32 -------
Fixed.

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061030/039327.html



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov  1 21:47:50 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 1 Nov 2006 21:47:50 -0600
Subject: [LLVMbugs] [Bug 975] crtstuff.c breaks llvm-gcc build
Message-ID: <200611020347.kA23lomK016650@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=975

rspencer at x10sys.com changed:

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



------- Additional Comments From rspencer at x10sys.com  2006-11-01 21:47 -------
Duplicate.

*** This bug has been marked as a duplicate of 971 ***



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov  1 22:06:49 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 1 Nov 2006 22:06:49 -0600
Subject: [LLVMbugs] [Bug 486] llvm-test Makefile rules not based on
	llvm/Makefile.rules
Message-ID: <200611020406.kA246nBe016985@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=486

rspencer at x10sys.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  BugsThisDependsOn|731                         |
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |WONTFIX
   Target Milestone|---                         |1.9



------- Additional Comments From rspencer at x10sys.com  2006-11-01 22:06 -------
llvm and llvm-test have diverged too far to converge them again. The
functionality they offer is significantly different and it seems to be fine with
everyone that they are different. Perhaps when the next upgrade to the LLVM
Makefile system is done (bug 731) we will revisit this. Or, maybe not.  

This bug has been outstanding for nearly two years. Time to assume that we're
not going to fix this.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  2 14:30:06 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 2 Nov 2006 14:30:06 -0600
Subject: [LLVMbugs] [Bug 786] Should build with -Wunused and -pedantic
Message-ID: <200611022030.kA2KU60W010134@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=786

rspencer at x10sys.com changed:

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



------- Additional Comments From rspencer at x10sys.com  2006-11-02 14:30 -------
The -Wunused and -Wno-unused-parameter flags are now turned on by default. Most
of the changes necessary to clean up LLVM have been made. A few remain, in these
categories:

1. Unused functions that I didn't want to delete without review.
2. Unused variables in generated code (not much we can do)

Developers and maintainers should strive to clean up the remaining few warnings
as they encounter them.

This bug is done.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  2 19:04:22 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 2 Nov 2006 19:04:22 -0600
Subject: [LLVMbugs] [Bug 955] setjmp prototype issue causes siod test to fail
Message-ID: <200611030104.kA314MD6015640@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=955

rspencer at x10sys.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|setjmp issue causes siod    |setjmp prototype issue
                   |test to fail                |causes siod test to fail
   Target Milestone|---                         |1.9



------- Additional Comments From rspencer at x10sys.com  2006-11-02 19:04 -------
This got fixed with this patch:

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061016/038923.html

siod and other tests no long fail on _setjmp.




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  2 19:16:14 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 2 Nov 2006 19:16:14 -0600
Subject: [LLVMbugs] [Bug 981] NEW: PPC backend build emits warning
Message-ID: <200611030116.kA31GEq8015979@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=981

           Summary: PPC backend build emits warning
           Product: tools
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: TableGen
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sabre at nondot.org
                CC: asl at math.spbu.ru,rspencer at x10sys.com


This is a regression due to PR786.

Tblgen should not emit the 'NoItineraries' global if it is not going to use it.

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  2 19:29:41 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 2 Nov 2006 19:29:41 -0600
Subject: [LLVMbugs] [Bug 981] PPC backend build emits warning
Message-ID: <200611030129.kA31TfCo016356@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=981

rspencer at x10sys.com changed:

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



------- Additional Comments From rspencer at x10sys.com  2006-11-02 19:29 -------
Fixed.

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061030/039420.html



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  2 20:00:30 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 2 Nov 2006 20:00:30 -0600
Subject: [LLVMbugs] [Bug 806] LTDL_SHLIB_EXT set to .so on darwin instead of
	.dylib
Message-ID: <200611030200.kA320U55017043@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=806

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-02 20:00 -------
bugpoint works and I don't remember what the original issue was.  If it reoccurs we can file a new bug.

Thanks Reid,

-Chris




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov  3 10:39:36 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 3 Nov 2006 10:39:36 -0600
Subject: [LLVMbugs] [Bug 982] NEW: bugpoint crash
Message-ID: <200611031639.kA3Gdatb019480@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=982

           Summary: bugpoint crash
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: bugpoint
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: alenhar2 at uiuc.edu


the attached program (probably not helpful) crashes when removing a function with:

andrewl at fenris:~/llvm/projects/llvm-test/MultiSource/Benchmarks/mediabench/gsm/toast$
make bugpoint-llc
/home/andrewl/llvm/Debug/bin/bugpoint Output/toast.llvm.bc -run-llc  
-input=/dev/null -output=Output/toast.out-nat -timeout=500 --tool-args  --args
-- -cpl
/home/andrewl/llvm/projects/llvm-test/MultiSource/Benchmarks/mediabench/gsm/toast/../data/clinton.pcm
Read input file      : 'Output/toast.llvm.bc'
*** All input ok
Found gcc: /usr/bin/gcc
Initializing execution environment: Running the code generator to test for a
crash: <llc>
*** Checking the code generator...
<llc><gcc><program>
*** Input program does not match reference diff!
Debugging code generator problem!
Checking to see if the program is misoptimized when these functions are run
through the passes: gsm_norm APCM_quantization_xmaxc_to_exp_mant
APCM_inverse_quantization RPE_grid_positioning
Decoding_of_the_coded_Log_Area_Ratios Coefficients_0_12 Coefficients_13_26
Coefficients_27_39 LARp_to_rp Short_term_analysis_filtering... <29 total>
While deleting: int () * %generic_init
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([7 x sbyte]* %str20, int 0, int 0),
    sbyte* getelementptr ([44 x sbyte]* %l21_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %l9_str4, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %linear_input,
    int (short*)* %linear_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([7 x sbyte]* %str20, int 0, int 0),
    sbyte* getelementptr ([44 x sbyte]* %l21_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %l9_str4, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %linear_input,
    int (short*)* %linear_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str2, int 0, int 0),
    sbyte* getelementptr ([28 x sbyte]* %str18, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str319, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %alaw_input,
    int (short*)* %alaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str2, int 0, int 0),
    sbyte* getelementptr ([28 x sbyte]* %str18, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str319, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %alaw_input,
    int (short*)* %alaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str1, int 0, int 0),
    sbyte* getelementptr ([34 x sbyte]* %l20_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str17, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %ulaw_input,
    int (short*)* %ulaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str1, int 0, int 0),
    sbyte* getelementptr ([34 x sbyte]* %l20_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str17, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %ulaw_input,
    int (short*)* %ulaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str1, int 0, int 0),
    sbyte* getelementptr ([34 x sbyte]* %l20_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str17, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %ulaw_input,
    int (short*)* %ulaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str2, int 0, int 0),
    sbyte* getelementptr ([28 x sbyte]* %str18, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str319, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %alaw_input,
    int (short*)* %alaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([7 x sbyte]* %str20, int 0, int 0),
    sbyte* getelementptr ([44 x sbyte]* %l21_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %l9_str4, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %linear_input,
    int (short*)* %linear_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([7 x sbyte]* %str20, int 0, int 0),
    sbyte* getelementptr ([44 x sbyte]* %l21_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %l9_str4, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %linear_input,
    int (short*)* %linear_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([7 x sbyte]* %str20, int 0, int 0),
    sbyte* getelementptr ([44 x sbyte]* %l21_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %l9_str4, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %linear_input,
    int (short*)* %linear_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str2, int 0, int 0),
    sbyte* getelementptr ([28 x sbyte]* %str18, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str319, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %alaw_input,
    int (short*)* %alaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str2, int 0, int 0),
    sbyte* getelementptr ([28 x sbyte]* %str18, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str319, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %alaw_input,
    int (short*)* %alaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str1, int 0, int 0),
    sbyte* getelementptr ([34 x sbyte]* %l20_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str17, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %ulaw_input,
    int (short*)* %ulaw_output }
Use still stuck around after Def is destroyed: { sbyte *, sbyte *, sbyte *, int
() *, int () *, int (short *) *, int (short *) * } {
    sbyte* getelementptr ([6 x sbyte]* %l11_str1, int 0, int 0),
    sbyte* getelementptr ([34 x sbyte]* %l20_str, int 0, int 0),
    sbyte* getelementptr ([3 x sbyte]* %str17, int 0, int 0),
    int ()* %generic_init,
    int ()* %generic_init,
    int (short*)* %ulaw_input,
    int (short*)* %ulaw_output }
bugpoint: Value.cpp:59: virtual llvm::Value::~Value(): Assertion `use_begin() ==
use_end() && "Uses remain when a value is destroyed!"' failed.
make: *** [Output/toast.bugpoint-llc] Aborted



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov  3 11:37:37 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 3 Nov 2006 11:37:37 -0600
Subject: [LLVMbugs] [Bug 974] -fno-unit-at-a-time should disable all ipo
Message-ID: <200611031737.kA3Hbbfh009972@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=974

dpatel at apple.com changed:

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



------- Additional Comments From dpatel at apple.com  2006-11-03 11:37 -------
I installed the patch. Will do separate patch for raiseallocs and SimplifyLibCallsPass.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov  3 11:37:51 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 3 Nov 2006 11:37:51 -0600
Subject: [LLVMbugs] [Bug 978] X86/ELF Dwarf Support Is Lacking
Message-ID: <200611031737.kA3HbpuN010725@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=978

dpatel at apple.com changed:

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





------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov  3 12:57:08 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 3 Nov 2006 12:57:08 -0600
Subject: [LLVMbugs] [Bug 978] X86/ELF Dwarf Support Is Lacking
Message-ID: <200611031857.kA3Iv8FI026554@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=978

dpatel at apple.com changed:

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



------- Additional Comments From dpatel at apple.com  2006-11-03 12:57 -------
I accidently resolved this bug. Reopening.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sun Nov  5 10:48:16 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 5 Nov 2006 10:48:16 -0600
Subject: [LLVMbugs] [Bug 983] NEW: Tool dependencies are broken
Message-ID: <200611051648.kA5GmGNi007089@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=983

           Summary: Tool dependencies are broken
           Product: Build scripts
           Version: cvs
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Makefiles
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


After the change was made to use llvm-config to generate the set of libraries
for linking a tool, the dependency on those libraries in the makefile was
broken. You can now rebuild a library that a tool needs and not get a new
executable. For example, make a change to the CBackend and llc won't re-link.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sun Nov  5 17:25:32 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 5 Nov 2006 17:25:32 -0600
Subject: [LLVMbugs] [Bug 984] NEW:
	test/Regression/C++Frontend/2003-11-02-WeakLinkage.cpp.tr xfailed
Message-ID: <200611052325.kA5NPWit014052@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=984

           Summary: test/Regression/C++Frontend/2003-11-02-
                    WeakLinkage.cpp.tr xfailed
           Product: tools
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sabre at nondot.org
                CC: dpatel at apple.com


llvm-gcc is compiling the template in this testcase to have weak linkage.  The testcase expects it to be 
linkonce.  We should investigate to find out if llvm-gcc can be enhanced to get this back, post 1.9.

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From isanbard at gmail.com  Mon Nov  6 00:55:00 2006
From: isanbard at gmail.com (Bill Wendling)
Date: Sun, 5 Nov 2006 22:55:00 -0800
Subject: [LLVMbugs] Trouble Building llvm-gcc
Message-ID: <ED95A16D-E5F4-4EF0-A0C0-FE40E7DA79F4@gmail.com>

Is anyone getting an ICE while building LLVM-GCC on a Mac? I reduced  
it down at least one case which fails. Here's the code:

typedef signed _Unwind_Sword __attribute__((__mode__(__word__)));
typedef unsigned _Unwind_Ptr __attribute__((__mode__(__pointer__)));
typedef unsigned _Unwind_Internal_Ptr __attribute__((__mode__ 
(__pointer__)));

int read_encoded_value_with_base (unsigned char encoding, _Unwind_Ptr  
base,
                                   const unsigned char *p,  
_Unwind_Ptr *val)
{
   union unaligned
   {
     unsigned u8 __attribute__ ((mode (DI)));
   } __attribute__((__packed__));
   const union unaligned *u = (const union unaligned *) p;
   _Unwind_Internal_Ptr result;

   switch (encoding & 0x0f)
     {
     case 0x00:
     case 0x01:
     case 0x09:
       {
         _Unwind_Sword tmp;
         p = read_sleb128 (p, &tmp);
       }
     case 0x02:
     case 0x03:
     case 0x04:
       result = u->u8;
     }

   if (result != 0)
     result += ((encoding & 0x70) == 0x10
                ? (_Unwind_Internal_Ptr) u : base);
}

This occurs when it recompiles the llvm-gcc source (so when it's  
compiled with xgcc), making it crazy to figure out what's happening.  
Has anyone seen this before or might have a clue what's going on?

Thanks!
-bw


From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 02:36:56 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 02:36:56 -0600
Subject: [LLVMbugs] [Bug 983] Tool dependencies are broken
Message-ID: <200611060836.kA68auiG031289@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=983

rspencer at x10sys.com changed:

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



------- Additional Comments From rspencer at x10sys.com  2006-11-06 02:36 -------
This is bogus. I must have been trying to build a tool that didn't actually
depend on the library that changed.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 07:05:13 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 07:05:13 -0600
Subject: [LLVMbugs] [Bug 985] NEW: -fno-inline-functions should disable
	function inlining
Message-ID: <200611061305.kA6D5DRB003334@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=985

           Summary: -fno-inline-functions should disable function inlining
           Product: tools
           Version: trunk
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rafael.espindola at gmail.com


Compiling crtstuff.c fails on ARM because the call to __do_global_dtors_aux () 
is inlined in the following function

static void __attribute__((__used__)) call___do_global_dtors_aux (void) { asm 
("\t.section\t.fini"); __do_global_dtors_aux (); asm ("\t.text"); }

The gcc build script uses the -fno-inline-functions option when compiling 
crtstuff.c, but it doesn't work with llvm-gcc



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 11:38:09 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 11:38:09 -0600
Subject: [LLVMbugs] [Bug 983] llvm-config build order can break build with
	parallel build
Message-ID: <200611061738.kA6Hc9Wc008402@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=983

rspencer at x10sys.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|INVALID                     |
            Summary|Tool dependencies are broken|llvm-config build order can
                   |                            |break build with parallel
                   |                            |build



------- Additional Comments From rspencer at x10sys.com  2006-11-06 11:38 -------
There actually is a problem, but its not exactly the one I thought. If you do this:

make -C lib/Target/CBackend
make -j 3 -C tools

Then you can end up with a broken build. The CBackend library gets updated which
means that llvm-config needs to rebuild. However, that rebuild is normally done
at the end of building lib. With a parallel build llvm-config gets rebuilt at
the same time as other things. However the tools depend on llvm-config to build.
If timing is just right, llvm-config will produce no output (error) and the link
command line for one of the tools will not contain any libraries (or worse)
consequently the tool link either fails outright or you end up with unresolved
symbols (lots of them).



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 12:05:29 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 12:05:29 -0600
Subject: [LLVMbugs] [Bug 983] llvm-config build order can break build with
	parallel build
Message-ID: <200611061805.kA6I5TCA009411@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=983

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-06 12:05 -------
The build in general only works if you do a top-level build.  This is what the logic in llvm/Makefile is all 
about.  Use of make -C or 'makellvm' is only safe if you "know what you're doing".  Otherwise, do a top-
level build always.

-Chris




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From sabre at nondot.org  Mon Nov  6 15:01:53 2006
From: sabre at nondot.org (Chris Lattner)
Date: Mon, 6 Nov 2006 13:01:53 -0800 (PST)
Subject: [LLVMbugs] Trouble Building llvm-gcc
In-Reply-To: <ED95A16D-E5F4-4EF0-A0C0-FE40E7DA79F4@gmail.com>
References: <ED95A16D-E5F4-4EF0-A0C0-FE40E7DA79F4@gmail.com>
Message-ID: <Pine.LNX.4.62.0611061301370.24964@nondot.org>

On Sun, 5 Nov 2006, Bill Wendling wrote:
> Is anyone getting an ICE while building LLVM-GCC on a Mac? I reduced
> it down at least one case which fails. Here's the code:

I'm not sure.  llvm-gcc doesn't build for me at all on darwin-x86.  Evan 
is investigating.

-Chris

> typedef signed _Unwind_Sword __attribute__((__mode__(__word__)));
> typedef unsigned _Unwind_Ptr __attribute__((__mode__(__pointer__)));
> typedef unsigned _Unwind_Internal_Ptr __attribute__((__mode__
> (__pointer__)));
>
> int read_encoded_value_with_base (unsigned char encoding, _Unwind_Ptr
> base,
>                                   const unsigned char *p,
> _Unwind_Ptr *val)
> {
>   union unaligned
>   {
>     unsigned u8 __attribute__ ((mode (DI)));
>   } __attribute__((__packed__));
>   const union unaligned *u = (const union unaligned *) p;
>   _Unwind_Internal_Ptr result;
>
>   switch (encoding & 0x0f)
>     {
>     case 0x00:
>     case 0x01:
>     case 0x09:
>       {
>         _Unwind_Sword tmp;
>         p = read_sleb128 (p, &tmp);
>       }
>     case 0x02:
>     case 0x03:
>     case 0x04:
>       result = u->u8;
>     }
>
>   if (result != 0)
>     result += ((encoding & 0x70) == 0x10
>                ? (_Unwind_Internal_Ptr) u : base);
> }
>
> This occurs when it recompiles the llvm-gcc source (so when it's
> compiled with xgcc), making it crazy to figure out what's happening.
> Has anyone seen this before or might have a clue what's going on?
>
> Thanks!
> -bw
>
> _______________________________________________
> LLVMbugs mailing list
> LLVMbugs at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
>

-Chris

-- 
http://nondot.org/sabre/
http://llvm.org/

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 16:27:29 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 16:27:29 -0600
Subject: [LLVMbugs] [Bug 986] NEW: Extremely ugly code is generated for
	integer division in some cases.
Message-ID: <200611062227.kA6MRTeH015201@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=986

           Summary: Extremely ugly code is generated for integer division in
                    some cases.
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru


Consider the folowing code:

uint %test_div_func7(uint %u) {
entry:
        %tmp1 = udiv uint %u, 7
        ret uint %tmp1
}

llvm currently generates the following code (well, actually it was generated via
llvm-gcc from C source with -O3):

test_div_func7:
        subl $4, %esp
        movl %esi, (%esp)
        movl $613566757, %ecx
        movl 8(%esp), %esi
        movl %esi, %eax
        mull %ecx
        subl %edx, %esi
        shrl %esi
        addl %edx, %esi
        shrl $2, %esi
        movl %esi, %eax
        movl (%esp), %esi
        addl $4, %esp
        ret

This is really ugly:

1. "General" algorithm for division-via-multiplication is ok, but it can be
tweaked more for x86 platform.
2. Codegenerator transforms LLVM code to something not so good.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 19:56:50 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 19:56:50 -0600
Subject: [LLVMbugs] [Bug 987] NEW: testcase failures on x86/linux
Message-ID: <200611070156.kA71uolV023576@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=987

           Summary: testcase failures on x86/linux
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca


I'm encountering a lot of nightly testcase failures and am bugpointing them.
I'll attach examples of .ll and matching .s for testcases that I think are all
the same bug.

I have no idea what the bug is yet.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From isanbard at gmail.com  Mon Nov  6 14:02:50 2006
From: isanbard at gmail.com (Bill Wendling)
Date: Mon, 6 Nov 2006 12:02:50 -0800
Subject: [LLVMbugs] Trouble Building llvm-gcc
In-Reply-To: <Pine.LNX.4.62.0611061301370.24964@nondot.org>
References: <ED95A16D-E5F4-4EF0-A0C0-FE40E7DA79F4@gmail.com>
	<Pine.LNX.4.62.0611061301370.24964@nondot.org>
Message-ID: <16e5fdf90611061202s77a6a000p5bcc4ebbba83d9fa@mail.gmail.com>

On 11/6/06, Chris Lattner <sabre at nondot.org> wrote:
> On Sun, 5 Nov 2006, Bill Wendling wrote:
> > Is anyone getting an ICE while building LLVM-GCC on a Mac? I reduced
> > it down at least one case which fails. Here's the code:
>
> I'm not sure.  llvm-gcc doesn't build for me at all on darwin-x86.  Evan
> is investigating.
>
Okay. Thanks!

-bw

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 21:50:04 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 21:50:04 -0600
Subject: [LLVMbugs] [Bug 988] NEW: bugpoint fails to point to bug
Message-ID: <200611070350.kA73o4i4025488@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=988

           Summary: bugpoint fails to point to bug
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: bugpoint
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


Bug point said I should report this:

*** The following function is being miscompiled:  MergeToLeft
Extracted a loop from the breaking portion of the program.
<cbe><gcc><program>  *** ERROR: Loop extraction broke the program. :( Please
report a bug!
      Continuing on with un-loop-extracted version.
Please submit the bugpoint-loop-extract-fail-*.bc files.

I'll attach that in a minute. The failure was generated by "make bugpoint-llc"
in the MultiSource/Benchmarks/McCat/18-imp test.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 21:58:28 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 21:58:28 -0600
Subject: [LLVMbugs] [Bug 988] bugpoint fails to point to bug
Message-ID: <200611070358.kA73wS41025681@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=988

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-06 21:58 -------
Despite what it claims, this is almost certainly not an issue with the loop extract pass.  I'll change bugpoint 
to not ask for bugs to be filed.

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov  6 22:13:31 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 6 Nov 2006 22:13:31 -0600
Subject: [LLVMbugs] [Bug 987] llvm.memcpy.i64 miscompiled on 32-bit target
Message-ID: <200611070413.kA74DViB026051@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=987

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
          Component|Backend: X86                |Common Code Generator Code
           Keywords|                            |miscompilation
         OS/Version|Linux                       |All
           Platform|PC                          |All
         Resolution|                            |FIXED
            Summary|testcase failures on        |llvm.memcpy.i64 miscompiled
                   |x86/linux                   |on 32-bit target
   Target Milestone|---                         |1.9



------- Additional Comments From sabre at nondot.org  2006-11-06 22:13 -------
Fixed.  Testcase here: CodeGen/Generic/2006-11-06-MemIntrinsicExpand.ll

Patch here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039555.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov  7 10:47:19 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 7 Nov 2006 10:47:19 -0600
Subject: [LLVMbugs] [Bug 989] NEW: gcc4 front-end, darwin-crt2.o
Message-ID: <200611071647.kA7GlJBu015658@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=989

           Summary: gcc4 front-end, darwin-crt2.o
           Product: tools
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: major
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: scottm at aero.org


# APPLE LOCAL use -mlongcall for large text support
/Users/scottm/play/llvm-cfrontend/obj/powerpc-apple-darwin7.9.0/gcc/xgcc -B/Users/scottm/play/
llvm-cfrontend/obj/powerpc-apple-darwin7.9.0/gcc/ -B/Users/scottm/play/llvm-cfrontend/powerpc-
apple-darwin7.9.0/powerpc-apple-darwin7.9.0/bin/ -B/Users/scottm/play/llvm-cfrontend/powerpc-
apple-darwin7.9.0/powerpc-apple-darwin7.9.0/lib/ -isystem /Users/scottm/play/llvm-cfrontend/
powerpc-apple-darwin7.9.0/powerpc-apple-darwin7.9.0/include -isystem /Users/scottm/play/llvm-
cfrontend/powerpc-apple-darwin7.9.0/powerpc-apple-darwin7.9.0/sys-include -DIN_GCC    -W -Wall 
-Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition  -isystem ./include  
-I. -I. -I../../../gcc -I../../../gcc/. -I../../../gcc/../include -I./../intl -I../../../gcc/../libcpp/include -I/
opt/local/include -I/opt/local/include -I/Users/scottm/play/llvm/include -I/Users/scottm/play/llvm/
obj/powerpc-apple-darwin7.9.0/include  -mlongcall \
  -c ../../../gcc/config/darwin-crt2.c -o crt2.o
/Users/scottm/play/llvm/lib/VMCore/Instructions.cpp:1250: failed assertion `0 && "Unknown setcc 
opcode!"'
../../../gcc/config/darwin-crt2.c: In function '__darwin_gcc3_preregister_frame_info':
../../../gcc/config/darwin-crt2.c:143: internal compiler error: Abort trap
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://llvm.org/bugs> for instructions.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov  7 12:32:33 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 7 Nov 2006 12:32:33 -0600
Subject: [LLVMbugs] [Bug 936] MultiSource/Applications/viterbi suffers from
	poor register allocation on X86
Message-ID: <200611071832.kA7IWXBQ018004@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=936

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|MultiSource/Applications/vit|MultiSource/Applications/vit
                   |erbi slow on X86/PPC with   |erbi suffers from poor
                   |LLC                         |register allocation on X86



------- Additional Comments From sabre at nondot.org  2006-11-07 12:32 -------
The remaining PPC issue is PPC specific (not generating bdnz).  I added it to the PPC README.

The fix for this bug sped up viterbi by 40% on X86, I'm declaring success :)

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov  7 18:29:37 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 7 Nov 2006 18:29:37 -0600
Subject: [LLVMbugs] [Bug 990] NEW: codegen problem with 'complex' type on
	x86/linux
Message-ID: <200611080029.kA80TbYP028801@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=990

           Summary: codegen problem with 'complex' type on x86/linux
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca


I've noticed a pattern in my testcase failures involving "csretcc". The bug
occurs with a simple testcase from the "cexp" man page:

// RUN: %llvmgcc -O2 -lm %s -o %s.exe
// RUN: ./%s.exe | grep "-1.000000+0.000000*i"
#include <stdio.h>

/* check that exp(i*pi) == -1 */
#include <math.h>   /* for atan */
#include <complex.h>
int main(void) {
  double pi = 4*atan(1);
  complex z = cexp(I*pi);
  printf("%f+%f*i\n", creal(z), cimag(z));
}

With LLVM, this produces:

$ llvm-gcc -O0 csretcc.c -lm -o csretcc
$ ./csretcc
0.000000+0.000000*i
Segmentation fault

while with GCC, it works:

$ gcc -O0 csretcc.c -lm -o csretcc
$ ./csretcc
-1.000000+0.000000*i

In the bytecode, the call to "cexp" is called with "csretcc" calling convention.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov  8 18:39:46 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 8 Nov 2006 18:39:46 -0600
Subject: [LLVMbugs] [Bug 991] NEW: NightlyTester should not submit report if
	error in script occured.
Message-ID: <200611090039.kA90dkBP029169@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=991

           Summary: NightlyTester should not submit report if error in
                    script occured.
           Product: Test Suite
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Nightly Tester
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org


The new nightly tester should submit report back to the server if the script
itself had an error.

We also should have a webform (password protected) so the core llvm team can
easily delete test results that are not useful or due to errors in the script or
script usage.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  9 00:27:33 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 9 Nov 2006 00:27:33 -0600
Subject: [LLVMbugs] [Bug 992] NEW: crash in opt -
	llvm::CallGraphNode::removeCallEdgeTo(llvm::CallGraphNode*)
Message-ID: <200611090627.kA96RXFx032549@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=992

           Summary: crash in opt -
                    llvm::CallGraphNode::removeCallEdgeTo(llvm::CallGraphNod
                    e*)
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Interprocedural Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org


opt: CallGraph.cpp:277: void llvm::CallGraphNode::removeCallEdgeTo(llvm::CallGraphNode*): Assertion `i 
&& "Cannot find callee to remove!"' failed.


to run testcase:
opt bugpoint-reduced-simplified.bc -inline -prune-eh



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  9 08:29:56 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 9 Nov 2006 08:29:56 -0600
Subject: [LLVMbugs] [Bug 993] NEW: Running -inline -prune-eh on testcase
	crashes with assert
Message-ID: <200611091429.kA9ETuNc008944@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=993

           Summary: Running -inline -prune-eh on testcase crashes with
                    assert
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Interprocedural Analyses
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: criswell at uiuc.edu


The error is this:

opt: CallGraph.cpp:277: void
llvm::CallGraphNode::removeCallEdgeTo(llvm::CallGraphNode*): Assertion `i &&
"Cannot find callee to remove!"' failed.
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::PrintStackTrace()+0x1a)[0x870fa1a]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::SignalHandler(int)+0x102)[0x870fcc2]
/lib/tls/libc.so.6[0x138a58]
/lib/tls/libc.so.6(abort+0x1d2)[0x13a342]
/lib/tls/libc.so.6(__assert_fail+0x108)[0x132338]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::CallGraphNode::removeCallEdgeTo(llvm::CallGraphNode*)+0x44)[0x85c8b7a]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::PruneEH::DeleteBasicBlock(llvm::BasicBlock*)+0x17b)[0x84d83b5]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::PruneEH::SimplifyFunction(llvm::Function*)+0x2c7)[0x84d8057]
/localhome/criswell/latestllvm/Debug/bin/opt((anonymous
namespace)::PruneEH::runOnSCC(std::vector<llvm::CallGraphNode*,
std::allocator<llvm::CallGraphNode*> > const&)+0x72)[0x84d798c]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::CallGraphSCCPass::runOnModule(llvm::Module&)+0x9c)[0x85cb0aa]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::ModulePassManager::runPass(llvm::ModulePass*,
llvm::Module*)+0x1f)[0x86af8b9]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManagerT<llvm::MTraits>::runPasses(llvm::Module*,
std::map<llvm::Pass*, std::vector<llvm::Pass*, std::allocator<llvm::Pass*> >,
std::less<llvm::Pass*>, std::allocator<std::pair<llvm::Pass* const,
std::vector<llvm::Pass*, std::allocator<llvm::Pass*> > > > >&)+0x136)[0x86ae738]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManagerT<llvm::MTraits>::runOnUnit(llvm::Module*)+0x177)[0x86ae5ed]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::ModulePassManager::runOnModule(llvm::Module&)+0x25)[0x869a345]
/localhome/criswell/latestllvm/Debug/bin/opt(llvm::PassManager::run(llvm::Module&)+0x23)[0x869a64b]
/localhome/criswell/latestllvm/Debug/bin/opt(main+0xa65)[0x841ef61]
/lib/tls/libc.so.6(__libc_start_main+0xf0)[0x126770]
/localhome/criswell/latestllvm/Debug/bin/opt(std::__throw_logic_error(char
const*)+0x59)[0x8412251]
Abort (core dumped)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  9 11:09:01 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 9 Nov 2006 11:09:01 -0600
Subject: [LLVMbugs] [Bug 993] Running -inline -prune-eh on testcase crashes
	with assert
Message-ID: <200611091709.kA9H91Am011712@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=993

tonic at nondot.org changed:

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



------- Additional Comments From tonic at nondot.org  2006-11-09 11:09 -------
Please search bugzilla before filing a bug.

*** This bug has been marked as a duplicate of 992 ***



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  9 12:17:31 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 9 Nov 2006 12:17:31 -0600
Subject: [LLVMbugs] [Bug 994] NEW: cannot disassemble CINT95
	126.gcc.linked.bc file
Message-ID: <200611091817.kA9IHV4I014508@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=994

           Summary: cannot disassemble CINT95 126.gcc.linked.bc file
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-dis
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org


[zion 126.gcc]$ llvm-dis Output/126.gcc.linked.bc
llvm-dis:
/mounts/zion/disks/0/localhome/tbrethou/llvm/include/llvm/Support/Casting.h:199:
typename llvm::cast_retty<To, From>::ret_type llvm::cast(const Y&) [with X =
llvm::ConstantInt, Y = llvm::Value*]: Assertion `isa<X>(Val) && "cast<Ty>()
argument of incompatible type!"' failed.
llvm-dis((anonymous namespace)::PrintStackTrace()+0x1f)[0x8166dff]
/lib/tls/libc.so.6(abort+0xe9)[0xb01199]
/lib/tls/libc.so.6(__assert_fail+0x101)[0xaf8dd1]
llvm-dis(llvm::BytecodeReader::ParseInstruction(std::vector<unsigned int,
std::allocator<unsigned int> >&, llvm::BasicBlock*)+0x3c32)[0x80b8b32]
Abort



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  9 17:37:13 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 9 Nov 2006 17:37:13 -0600
Subject: [LLVMbugs] [Bug 992] crash in opt -
	llvm::CallGraphNode::removeCallEdgeTo(llvm::CallGraphNode*)
Message-ID: <200611092337.kA9NbDpL002449@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=992

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-09 17:37 -------
Fixed.  Testcases here:
Transforms/Inline/2006-11-09-InlineCGUpdate.ll
Transforms/Inline/2006-11-09-InlineCGUpdate-2.ll

Patches here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039741.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039743.html

Tanya, I think it makes sense to merge these two patches into the 1.9 branch.  Despite what I thought at 
the time, the first patch isn't sufficient to fully fix the bug.

-Chris




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov  9 18:56:34 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 9 Nov 2006 18:56:34 -0600
Subject: [LLVMbugs] [Bug 990] codegen problem with 'complex' type on
	x86/linux
Message-ID: <200611100056.kAA0uYgb021929@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=990

asl at math.spbu.ru changed:

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



------- Additional Comments From asl at math.spbu.ru  2006-11-09 18:56 -------
Fixed with
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039746.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039747.html

Will check linux nightlytesters for newly passed tests (should be at least
fftbench & bigfib).



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 02:13:22 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 02:13:22 -0600
Subject: [LLVMbugs] [Bug 985] -fno-inline-functions should disable function
	inlining
Message-ID: <200611100813.kAA8DM3G003188@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=985

dpatel at apple.com changed:

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



------- Additional Comments From dpatel at apple.com  2006-11-10 02:13 -------
Applied.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 11:44:21 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 11:44:21 -0600
Subject: [LLVMbugs] [Bug 995] NEW: SPEC CFP2000 188.ammp llc crash
Message-ID: <200611101744.kAAHiLrJ019721@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=995

           Summary: SPEC CFP2000 188.ammp llc crash
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org


This is a regression from llvm 1.8. Testcase will be attached.

*** The following functions are being miscompiled:  v_nonbon_entry_2E_ce
fv_update_nonbon_bb190_2E_i_bb105_2E_i
fv_update_nonbon_bb333_2E_outer_2E_i_bb337_2E_i
<cbe><gcc>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
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 -shared
-fno-strict-aliasing
===> Leaving Output/bugpoint-train



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 12:12:38 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 12:12:38 -0600
Subject: [LLVMbugs] [Bug 996] NEW:
	MultiSource/Benchmarks/Prolangs-C/allroots broken on x86
Message-ID: <200611101812.kAAICcWC020407@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=996

           Summary: MultiSource/Benchmarks/Prolangs-C/allroots broken on x86
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru


This test is definitely broken. Please find attached LLVM bytecode & generated
assembler code.

Generated code segfaults at first "movaps" instruction in newton function. llc
-march=pentium3 generates working code.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 12:47:54 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 12:47:54 -0600
Subject: [LLVMbugs] [Bug 997] NEW: SingleSource/CustomChecked/oopack_v1p8
	llc fails
Message-ID: <200611101847.kAAIlsZ9021620@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=997

           Summary: SingleSource/CustomChecked/oopack_v1p8 llc fails
           Product: tools
           Version: 1.9
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: tonic at nondot.org


llc, jit both fail but only llc is a regression

Will try to provide testcase.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 13:42:19 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 13:42:19 -0600
Subject: [LLVMbugs] [Bug 997] SingleSource/CustomChecked/oopack_v1p8 llc
	fails
Message-ID: <200611101942.kAAJgJA6023897@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=997

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
          Component|llc                         |Programs Tests
           Keywords|                            |build-problem
         OS/Version|Linux                       |All
            Product|tools                       |Test Suite
           Platform|PC                          |All
         Resolution|                            |FIXED
   Target Milestone|---                         |1.9



------- Additional Comments From sabre at nondot.org  2006-11-10 13:42 -------
This also failed on darwin.  Fix here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039775.html

Tanya, please pull this into the 1.9 release.

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 14:45:12 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 14:45:12 -0600
Subject: [LLVMbugs] [Bug 998] NEW: llvm-ld infinite loop on bc archive
Message-ID: <200611102045.kAAKjCNY009506@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=998

           Summary: llvm-ld infinite loop on bc archive
           Product: tools
           Version: trunk
          Platform: All
        OS/Version: Linux
            Status: NEW
          Severity: major
          Priority: P2
         Component: llvm-ld
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: scottm at aero.org


llvm-ld can (and did) go into an infinite loop while processing a bytecode 
archive. Cleaned up the loop in LinkArchive.cpp -- it shouldn't be an infinite 
loop. The code should have implemented a loop that keeps processing this 
archive until the set of undefined symbols doesn't change.

Also, bytecode archives improperly get fed to the native linker. At least Linux 
bitches about malformed archives when this happens, YMMV.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 15:44:31 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 15:44:31 -0600
Subject: [LLVMbugs] [Bug 996] MultiSource/Benchmarks/Prolangs-C/allroots
	broken on x86
Message-ID: <200611102144.kAALiVUi000523@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=996

evan.cheng at apple.com changed:

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



------- Additional Comments From evan.cheng at apple.com  2006-11-10 15:44 -------
We should not have been generating spills of vector values in this test case.
This was a bug in the X86 max / min dag combine.
Fixed.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 17:30:39 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 17:30:39 -0600
Subject: [LLVMbugs] [Bug 995] SPEC CFP2000 188.ammp llc crash
Message-ID: <200611102330.kAANUd3m010721@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=995

tonic at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |1.9



------- Additional Comments From tonic at nondot.org  2006-11-10 17:30 -------
I have verified that the fix for 996 also fixed this one.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 18:38:24 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 18:38:24 -0600
Subject: [LLVMbugs] [Bug 999] NEW: X86 Assert: Cannot pop empty stack
Message-ID: <200611110038.kAB0cO7s007325@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=999

           Summary: X86 Assert: Cannot pop empty stack
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


While building llvm-gcc4, I get:

X86FloatingPoint.cpp:434: void
<unnamed>::FPS::popStackAfter(llvm::ilist_iterator<llvm::MachineInstr>&):
Assertion `StackTop > 0 && "Cannot pop empty stack!"' failed.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 10 18:40:09 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 10 Nov 2006 18:40:09 -0600
Subject: [LLVMbugs] [Bug 1000] NEW: Sabre Is Buggy
Message-ID: <200611110040.kAB0e9rO008009@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1000

           Summary: Sabre Is Buggy
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: C
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


There's no known fix for this bug.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 11 00:28:55 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 11 Nov 2006 00:28:55 -0600
Subject: [LLVMbugs] [Bug 1001] NEW: DAGISelEmitter not looking up
	instruction namespace accurately.
Message-ID: <200611110628.kAB6StpL020059@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1001

           Summary: DAGISelEmitter not looking up instruction namespace
                    accurately.
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: TableGen
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: patil.nikhil at gmail.com


I was trying to write my own simple LLVM backend for my own simple ISA when I 
ran into the following error message:

MCGenDAGISel.inc: In member function 
'llvm::SDNode*<unnamed>::MCDAGToDAGISel::SelectCode(llvm::SDOperand)':
MCGenDAGISel.inc:463: error: 'INSTRUCTION_LIST_END' is not a member of 
'llvm::TargetInstrInfo'

My target description had only one instruction NOP. The problem disappears when 
I change NOP to ADD. In fact anything *alphabetically* before INLINEASM is 
okay. (INLINEASM & PHINODE are in namespace TargetInstrInfo.) I have traced 
down the problem to utils/TableGen/DAGISelEmitter.cpp:3390, where we only look 
at the first instruction in the list of all instructions. So if there happen to 
be no instructions alphabetically before INLINEASM, we get screwed.

Please let me know if you need any more information :)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 11 01:19:51 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 11 Nov 2006 01:19:51 -0600
Subject: [LLVMbugs] [Bug 999] X86 Assert: Cannot pop empty stack
Message-ID: <200611110719.kAB7JpCQ020842@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=999

evan.cheng at apple.com changed:

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



------- Additional Comments From evan.cheng at apple.com  2006-11-11 01:19 -------
Fixed.




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 11 05:59:55 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 11 Nov 2006 05:59:55 -0600
Subject: [LLVMbugs] [Bug 998] llvm-ld infinite loop on bc archive
Message-ID: <200611111159.kABBxt2r002034@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=998

rspencer at x10sys.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|major                       |normal
             Status|NEW                         |RESOLVED
           Keywords|                            |quality-of-implementation
         OS/Version|Linux                       |All
         Resolution|                            |FIXED
            Version|trunk                       |1.4



------- Additional Comments From rspencer at x10sys.com  2006-11-11 05:59 -------
Scott's first patch passed muster and fixes the issue. Patches here:

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039823.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039824.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039821.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061106/039822.html

Thanks, Scott!




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 11 16:42:51 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 11 Nov 2006 16:42:51 -0600
Subject: [LLVMbugs] [Bug 971] llvm-gcc build fails due to incorrect patch
Message-ID: <200611112242.kABMgpNG012641@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=971

nicholas at mxc.ca changed:

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



------- Additional Comments From nicholas at mxc.ca  2006-11-11 16:42 -------
It works now.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sun Nov 12 20:16:43 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 12 Nov 2006 20:16:43 -0600
Subject: [LLVMbugs] [Bug 1002] NEW: LICM doesn't update alias sets correctly
Message-ID: <200611130216.kAD2GhmW023648@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1002

           Summary: LICM doesn't update alias sets correctly
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Loop Optimizer
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: alenhar2 at uiuc.edu


LICM doesn't update the alias sets with all the new instructions it inserts. 
See LICM::sink for examples.  Allocas and loads are created without being placed
in the Alias set tracker.  The results of the loads should alias the old values
they are replacing.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 14 08:58:20 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 14 Nov 2006 08:58:20 -0600
Subject: [LLVMbugs] [Bug 1003] NEW: ICE when compiling LLVM TOT
Message-ID: <200611141458.kAEEwKUn021335@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1003

           Summary: ICE when compiling LLVM TOT
           Product: libraries
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: major
          Priority: P2
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: isanbard at gmail.com


I get the following ICE during compilation of LLVM.

GEP is not of right type for indices!
        %i = getelementptr long* %i             ; <{  }*> [#uses=1]
 longInstruction does not dominate all uses!
        %i = getelementptr long* %i             ; <{  }*> [#uses=1]
        call void %llvm.dbg.declare( {  }* %i, {  }* cast (%llvm.dbg.variable.type* %llvm.dbg.variable to {  }*) )
Broken module found, compilation aborted!
/Users/wendling/llvm/llvm.src/projects/Stacker/lib/runtime/stacker_rt.c: In function 
'_stacker_dump_stack_':
/Users/wendling/llvm/llvm.src/projects/Stacker/lib/runtime/stacker_rt.c:31: internal compiler error: 
Abort trap
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://llvm.org/bugs> for instructions.

In the preprocessed file, it gives this ICE:

$ /Users/wendling/llvm/llvm-gcc4.install/bin/llvm-gcc -g bad.ii -o bad.ll -S -emit-llvmGEP is not of 
right type for indices!        %_c = getelementptr int* %_c_addr               ; <{  }*> [#uses=1]
 intInstruction does not dominate all uses!
        %_c = getelementptr int* %_c_addr               ; <{  }*> [#uses=1]
        call void %llvm.dbg.declare( {  }* %_c, {  }* cast (%llvm.dbg.variable.type* %llvm.dbg.variable to {  }*) 
)
GEP is not of right type for indices!
        %_f = getelementptr uint* %_f_addr              ; <{  }*> [#uses=1]
 uintInstruction does not dominate all uses!
        %_f = getelementptr uint* %_f_addr              ; <{  }*> [#uses=1]
        call void %llvm.dbg.declare( {  }* %_f, {  }* cast (%llvm.dbg.variable.type* %llvm.dbg.variable1 to {  }
*) )
GEP is not of right type for indices!
        %retval = getelementptr int* %retval            ; <{  }*> [#uses=1]
 intInstruction does not dominate all uses!
        %retval = getelementptr int* %retval            ; <{  }*> [#uses=1]
        call void %llvm.dbg.declare( {  }* %retval, {  }* cast (%llvm.dbg.variable.type* %llvm.dbg.variable4 to 
{  }*) )
GEP is not of right type for indices!
        %iftmp.2 = getelementptr int* %iftmp.2          ; <{  }*> [#uses=1]
 intInstruction does not dominate all uses!
        %iftmp.2 = getelementptr int* %iftmp.2          ; <{  }*> [#uses=1]
        call void %llvm.dbg.declare( {  }* %iftmp.2, {  }* cast (%llvm.dbg.variable.type* %llvm.dbg.variable5 
to {  }*) )
Broken module found, compilation aborted!
/usr/include/ctype.h: In function '__darwin_ct_rune_t __isctype(__darwin_ct_rune_t, long unsigned int)':
/usr/include/ctype.h:225: internal compiler error: Abort trap
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://llvm.org/bugs> for instructions.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 14 17:44:20 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 14 Nov 2006 17:44:20 -0600
Subject: [LLVMbugs] [Bug 1003] ICE when compiling LLVM TOT
Message-ID: <200611142344.kAENiKId031364@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1003

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-14 17:44 -------
This was due to llvm-gcc .o files being out of date w.r.t. llvm headers.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov 15 05:09:57 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 15 Nov 2006 05:09:57 -0600
Subject: [LLVMbugs] [Bug 1004] NEW: Problem with CodeExtractor.cpp
Message-ID: <200611151109.kAFB9vu7020405@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1004

           Summary: Problem with CodeExtractor.cpp
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Transformation Utilities
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: ashalimov at lvk.cs.msu.su


I have this code:

loopexit.6:   
... // something
no_exit.7:  
... // something
loopexit.7:             ; preds = %no_exit.7, %loopexit.6
        %n.3.1 = phi int [ 0, %loopexit.6 ], [ %n.2, %no_exit.7 ]   
...

I want extract code region, which contains loopexit.6 and no_exit.7. 
CodeExtractor creates new function and puts those basic blocks to this function.
CodeExtractor creates codeReplacer block containing call-instr to new function.
So CodeExtractor modifies phi-node instruction in loopexit.7 :
 %n.3.1 = phi int [ 0, %codeRepl1 ], [ %n.2.reload, %codeRepl1 ]
And verifier  say 'Error' :

PHINode should have one entry for each predecessor of its parent basic block!
        %n.3.1 = phi int [ 0, %codeRepl1 ], [ %n.2.reload, %codeRepl1 ]        
; <int> [#uses=1]
Broken module found, compilation aborted!



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 16 11:54:14 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 16 Nov 2006 11:54:14 -0600
Subject: [LLVMbugs] [Bug 994] cannot disassemble CINT95 126.gcc.linked.bc
	file
Message-ID: <200611161754.kAGHsECx017433@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=994

tonic at nondot.org changed:

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



------- Additional Comments From tonic at nondot.org  2006-11-16 11:54 -------
This bug has been fixed I guess. Not seeing it anymore.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 16 14:21:24 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 16 Nov 2006 14:21:24 -0600
Subject: [LLVMbugs] [Bug 1005] NEW: pod2html links unresolvable
Message-ID: <200611162021.kAGKLOcL031912@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1005

           Summary: pod2html links unresolvable
           Product: Documentation
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: minor
          Priority: P2
         Component: General docs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


/usr/bin/pod2html: llvm-nm.pod: cannot resolve L<ar(1)> in paragraph 51.
/usr/bin/pod2html: llvm-nm.pod: cannot resolve L<nm(1)> in paragraph 51.
/usr/bin/pod2html: llvm-ar.pod: cannot resolve L<ar(1)> in paragraph 114.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 16 17:23:20 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 16 Nov 2006 17:23:20 -0600
Subject: [LLVMbugs] [Bug 1006] NEW: llvm-gcc libstdc++.so fails to have Std
	C++ Symbol
Message-ID: <200611162323.kAGNNKI2010752@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1006

           Summary: llvm-gcc libstdc++.so fails to have Std C++ Symbol
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: minor
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


If you configure llvm-gcc4 without --disable-shared option and then attempt to
use it as a linker for a C++ program, you will encounter:

/proj/llvm/cfe/install-1/lib/gcc/i686-pc-linux-gnu/4.0.1/libstdc++.so: undefined
reference to `std::basic_istream<wchar_t, std::char_traits<wchar_t> >::ignore(int)'

In C, the symbol is: _ZNSt13basic_istreamIwSt11char_traitsIwEE6ignoreEi

This was a known issue with the 4.0.x release. The patch suggested on the gcc
patches list has been applied to llvm-gcc4. Its not clear why this symbol
doesn't show up when building shared libs. After digging into it for a while, I
believe the problem lies in the libstdc++-v3/config/linker-map.gnu file in
llvm-gcc4. You will find that there is no section for the 3.4.6 compiler. It
stops at 3.4.5.

There are several references to this problem on our llvm-dev mailing list:
http://lists.cs.uiuc.edu/pipermail/llvmdev/2006-August/006522.html
http://lists.cs.uiuc.edu/pipermail/llvmdev/2006-August/006587.html
http://lists.cs.uiuc.edu/pipermail/llvmdev/2006-August/006588.html 

The workaround is: use --disable-shared when you configure llvm-gcc4.

This issue is quite likely Linux specific.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 16 19:54:38 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 16 Nov 2006 19:54:38 -0600
Subject: [LLVMbugs] [Bug 1007] NEW: predsimplify rips out constant,
	replaces it with variable.
Message-ID: <200611170154.kAH1scmQ013328@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1007

           Summary: predsimplify rips out constant, replaces it with
                    variable.
           Product: libraries
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Scalar Optimizations
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca


predsimplify replaces "uint 0" with "uint %tmp75" in a GEP. This immediately
causes a failure.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 16 23:58:50 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 16 Nov 2006 23:58:50 -0600
Subject: [LLVMbugs] [Bug 1008] NEW: internal compiler error when building
	firefox: nsMemoryImpl.cpp
Message-ID: <200611170558.kAH5wofQ017428@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1008

           Summary: internal compiler error when building firefox:
                    nsMemoryImpl.cpp
           Product: tools
           Version: 1.9
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-g++
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: kremenek at cs.stanford.edu


llvm-g++ triggers an internal compiler error when trying to compile
nsMemoryImpl.cpp:

llvm-g++ -o nsMemoryImpl.o -c  -DMOZILLA_INTERNAL_API -DOSTYPE=\"Darwin\"
-DOSARCH=\"Darwin\" -DBUILD_ID=0000000000 -D_IMPL_NS_COM 
-I../../dist/include/string -I../../dist/include/xpcom -I../../dist/include
-I../../dist/include/nspr    -I/usr/X11R6/include   -fPIC  -I/usr/X11R6/include
-fno-rtti -fno-exceptions -Wall -Wconversion -Wpointer-arith -Wcast-align
-Woverloaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-non-virtual-dtor
-Wno-long-long -fpascal-strings -no-cpp-precomp -fno-common -fshort-wchar
-I/Developer/Headers/FlatCarbon  -DNDEBUG -DTRIMMED -O2  -I/usr/X11R6/include
-DMOZILLA_CLIENT -include ../../mozilla-config.h -Wp,-MD,.deps/nsMemoryImpl.pp
/Users/kremenek/build/firefox/mozilla/xpcom/base/nsMemoryImpl.cpp
/Users/kremenek/build/firefox/mozilla/xpcom/base/nsMemoryImpl.cpp: In function
'void* NS_Realloc_P(void*, PRSize)':
/Users/kremenek/build/firefox/mozilla/xpcom/base/nsMemoryImpl.cpp:346: internal
compiler error: Bus error
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://llvm.org/bugs> for instructions.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 17 13:36:20 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 17 Nov 2006 13:36:20 -0600
Subject: [LLVMbugs] [Bug 1008] internal compiler error when building
	firefox: nsMemoryImpl.cpp
Message-ID: <200611171936.kAHJaKWW011215@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1008

dpatel at apple.com changed:

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



------- Additional Comments From dpatel at apple.com  2006-11-17 13:36 -------
Fixed.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Fri Nov 17 14:20:56 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Fri, 17 Nov 2006 14:20:56 -0600
Subject: [LLVMbugs] [Bug 1009] NEW: bootstrap failure building libgcc2
Message-ID: <200611172020.kAHKKuiP011998@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1009

           Summary: bootstrap failure building libgcc2
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-gcc
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: alenhar2 at uiuc.edu


The attched is a reduction of a boot strap failure in llvm-gcc2 on x86/linux.

internal compiler error: in prune_unused_types_update_strings, at dwarf2out.c:14372
"/home/vadve/alenhar2/cfe/build/gcc/cc1" "-g" "-O2"     -o /dev/null min.i

removing either -g or -O2 makes the problem go away.  -O1 still exhibits the
problem.

where min.i is:

  __attribute__ ((__nothrow__)) strtol (__const char *__restrict __nptr, char
**__restrict __endptr, int __base) {  }
  __attribute__ ((__nothrow__)) strtof (__const char *__restrict __nptr, char
**__restrict __endptr) {  }


a back trace of cc1:
#0  fancy_abort (file=0x861d3f0 "../../source/gcc/dwarf2out.c", line=14372,
function=0x8621800 "prune_unused_types_update_strings")
    at ../../source/gcc/diagnostic.c:588
During symbol reading, Incomplete CFI data; unspecified register eax at 0x082272a1.
During symbol reading, Incomplete CFI data; unspecified register ecx at 0x082272a1.
During symbol reading, Incomplete CFI data; unspecified register edx at 0x082272a1.
During symbol reading, Incomplete CFI data; unspecified register ebx at 0x082272a1.
During symbol reading, Incomplete CFI data; unspecified register esi at 0x082272a1.
During symbol reading, Incomplete CFI data; unspecified register edi at 0x082272a1.
#1  0x0824668f in prune_unused_types_update_strings (die=0xb7c2a9f4) at
../../source/gcc/dwarf2out.c:14372
During symbol reading, Incomplete CFI data; unspecified register eax at 0x08246665.
During symbol reading, Incomplete CFI data; unspecified register ecx at 0x08246665.
During symbol reading, Incomplete CFI data; unspecified register edx at 0x08246665.
During symbol reading, Incomplete CFI data; unspecified register ebx at 0x08246665.
During symbol reading, Incomplete CFI data; unspecified register esi at 0x08246665.
During symbol reading, Incomplete CFI data; unspecified register edi at 0x08246665.
#2  0x08246758 in prune_unused_types_prune (die=0xb7c2a958) at
../../source/gcc/dwarf2out.c:14410
During symbol reading, Incomplete CFI data; unspecified register eax at 0x082466c4.
During symbol reading, Incomplete CFI data; unspecified register ecx at 0x082466c4.
During symbol reading, Incomplete CFI data; unspecified register edx at 0x082466c4.
During symbol reading, Incomplete CFI data; unspecified register ebx at 0x082466c4.
During symbol reading, Incomplete CFI data; unspecified register esi at 0x082466c4.
During symbol reading, Incomplete CFI data; unspecified register edi at 0x082466c4.
#3  0x08246766 in prune_unused_types_prune (die=0xb7c2a0d0) at
../../source/gcc/dwarf2out.c:14411
During symbol reading, Incomplete CFI data; unspecified register eax at 0x082466c4.
During symbol reading, Incomplete CFI data; unspecified register ecx at 0x082466c4.
#4  0x0824688d in prune_unused_types () at ../../source/gcc/dwarf2out.c:14446
#5  0x08246bb9 in dwarf2out_finish (filename=0xbff36a2c "min.i") at
../../source/gcc/dwarf2out.c:14557
#6  0x0849a81b in compile_file () at ../../source/gcc/toplev.c:1138
#7  0x0849c565 in do_compile () at ../../source/gcc/toplev.c:2327
#8  0x0849c5cc in toplev_main (argc=34, argv=0xbfec69e4) at
../../source/gcc/toplev.c:2359
#9  0x080d52f2 in main (argc=34, argv=0xbfec69e4) at ../../source/gcc/main.c:35

fr 1
print s->str
$1 = 0x9cc58f0 "__nptr"
print slot
$2 = (void **) 0xb7c265f8
print *(struct indirect_string_node*)*slot
$6 = {str = 0x9cc58f0 "__nptr", refcount = 2, form = 0, label = 0x0}



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 18 00:31:18 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 18 Nov 2006 00:31:18 -0600
Subject: [LLVMbugs] [Bug 1000] Chris Is Buggy
Message-ID: <200611180631.kAI6VIBA022261@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1000

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-18 00:31 -------
I don't think this is fixable.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 18 01:01:00 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 18 Nov 2006 01:01:00 -0600
Subject: [LLVMbugs] [Bug 1000] Chris Is Buggy
Message-ID: <200611180701.kAI710ok022701@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1000

rspencer at x10sys.com changed:

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



------- Additional Comments From rspencer at x10sys.com  2006-11-18 01:00 -------
It may not be fixable, but WORKSFORME? I don't think so, mister.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 18 01:01:27 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 18 Nov 2006 01:01:27 -0600
Subject: [LLVMbugs] [Bug 1000] Chris Is Buggy
Message-ID: <200611180701.kAI71Rcp022746@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1000

rspencer at x10sys.com changed:

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



------- Additional Comments From rspencer at x10sys.com  2006-11-18 01:01 -------
This is much more appropriate.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 18 01:14:35 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 18 Nov 2006 01:14:35 -0600
Subject: [LLVMbugs] [Bug 765] Properly describe CR architecture
Message-ID: <200611180714.kAI7EZE0022935@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=765

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-18 01:14 -------
This is in the PPC README.txt file.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 18 01:15:35 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 18 Nov 2006 01:15:35 -0600
Subject: [LLVMbugs] [Bug 752] How to build the C front-end should be updated
	to describe llvmgcc4
Message-ID: <200611180715.kAI7FZ93022972@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=752

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |doc-obsolete
         Resolution|                            |FIXED
   Target Milestone|---                         |1.9



------- Additional Comments From sabre at nondot.org  2006-11-18 01:15 -------
The README.LLVM in the src distro fully explains this now.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 18 01:16:45 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 18 Nov 2006 01:16:45 -0600
Subject: [LLVMbugs] [Bug 700] [llvm-g++] Templates instantiated on internal
	classes should be internal
Message-ID: <200611180716.kAI7GjhC023014@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=700

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-18 01:16 -------
This will be fixed when we merge in gcc 4.2.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sun Nov 19 12:50:01 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 19 Nov 2006 12:50:01 -0600
Subject: [LLVMbugs] [Bug 1010] NEW: x86-64 build failure lowering intrinsic
Message-ID: <200611191850.kAJIo1oO004905@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1010

           Summary: x86-64 build failure lowering intrinsic
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca


Regression/CodeGen/X86/xmm-r64.ll has regressed:

$ llvm/Debug/bin/llvm-as <
/home/nicholas/llvm-commit/test/Regression/CodeGen/X86/xmm-r64.ll |
llvm/Debug/bin/llc -march=x86-64
Warning: Generation of 64-bit code for a 32-bit processor requested.
llc: SelectionDAGISel.cpp:1775: void
llvm::SelectionDAGLowering::visitTargetIntrinsic(llvm::CallInst&, unsigned int):
Assertion `TLI.isTypeLegal(Op.getValueType()) && "Intrinsic uses a non-legal
type?"' failed.
llvm/Debug/bin/llc((anonymous namespace)::PrintStackTrace()+0x1a)[0x8814248]
llvm/Debug/bin/llc((anonymous namespace)::SignalHandler(int)+0x110)[0x881450c]
[0xffffe500]
Aborted

Not bugpointing because it's already a reduced testcase.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sun Nov 19 14:59:51 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 19 Nov 2006 14:59:51 -0600
Subject: [LLVMbugs] [Bug 1011] NEW: assertion failed while compiling libmng
Message-ID: <200611192059.kAJKxp4A007684@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1011

           Summary: assertion failed while compiling libmng
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru


While compiling libmng (via llvm-gcc4 -O2) I've got an assertion:

cc1: /home/asl/proj/llvm/src/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:1116:
llvm::SDOperand llvm::SelectionDAG::getNode(unsigned int, llvm::MVT::ValueType,
llvm::SDOperand): Assertion `Operand.getValueType() > VT && "Invalid truncate
node, src < dst!"' failed.
libmng_pixels.i:137: internal compiler error: Aborted
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://llvm.org/bugs> for instructions.

Please find:
1. Original source
2. Reduced sources (via delta).

Everything was ok without any optimizations (but no ok with just -O), so the
problem itself can be not in backend, but in some optimization passes.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sun Nov 19 15:54:04 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sun, 19 Nov 2006 15:54:04 -0600
Subject: [LLVMbugs] [Bug 1012] NEW: sgefa fails on linux
Message-ID: <200611192154.kAJLs4Y3009255@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1012

           Summary: sgefa fails on linux
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: nicholas at mxc.ca


Latest CVS (since about a week ago) broke a dozen tests in my nightly. Testcase
attached.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 02:17:53 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 02:17:53 -0600
Subject: [LLVMbugs] [Bug 1013] NEW: Invalid assembly generated with DWARF2
	and Obj-C GC
Message-ID: <200611200817.kAK8HrhQ025669@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1013

           Summary: Invalid assembly generated with DWARF2 and Obj-C GC
           Product: tools
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: major
          Priority: P2
         Component: llvm-g++
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: mark+llvm.org at bdash.net.nz


To reproduce:

-- objc_instance.m --
namespace {
  class JSValue {
    ~JSValue();
  };

  JSValue::~JSValue()
  {
  }
}
-----

$ llvm-gcc -save-temps -fobjc-gc -gdwarf-2 -c objc_instance.m
objc_instance.s:168:Rest of line ignored. 1st junk character valued 1 ().
$ head -n 168 objc_instance.s | tail -n 1 | xxd
0000000: 092e 6c6f 6e67 095f 014c 5f4f 424a 435f  ..long._.L_OBJC_
0000010: 494d 4147 455f 494e 464f 0a              IMAGE_INFO.
$

It appears that 0x01 has snuck into the symbol name between the _ and L in _L_OBJC_IMAGE_INFO.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 11:59:04 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 11:59:04 -0600
Subject: [LLVMbugs] [Bug 1012] fp stackifier broken
Message-ID: <200611201759.kAKHx4lY008484@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1012

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
            Summary|sgefa fails on linux        |fp stackifier broken
   Target Milestone|---                         |2.0



------- Additional Comments From sabre at nondot.org  2006-11-20 11:59 -------
Fixed.  Testcase here: CodeGen/X86/fp-stack-compare.ll
Patch here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061120/040133.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061120/040134.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 12:55:25 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 12:55:25 -0600
Subject: [LLVMbugs] [Bug 1001] DAGISelEmitter not looking up instruction
	namespace accurately.
Message-ID: <200611201855.kAKItPZJ009824@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1001

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |build-problem
         Resolution|                            |FIXED
   Target Milestone|---                         |2.0



------- Additional Comments From sabre at nondot.org  2006-11-20 12:55 -------
Option #1 sounds good to me!  Thanks, I applied your patch:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061120/040138.html

Thanks,

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 12:06:37 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 12:06:37 -0600
Subject: [LLVMbugs] [Bug 1011] assertion failed while compiling libmng
Message-ID: <200611201806.kAKI6b6J008728@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1011

sabre at nondot.org changed:

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



------- Additional Comments From sabre at nondot.org  2006-11-20 12:06 -------
Fixed, patch here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061120/040136.html

Testcase here: CodeGen/Generic/2006-11-20-DAGCombineCrash.ll

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 12:17:29 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 12:17:29 -0600
Subject: [LLVMbugs] [Bug 1010] CodeGen/X86/xmm-r64.ll failure on non-x86-64
	hosts
Message-ID: <200611201817.kAKIHTPI009113@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1010

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |compile-fail
         Resolution|                            |FIXED
            Summary|x86-64 build failure        |CodeGen/X86/xmm-r64.ll
                   |lowering intrinsic          |failure on non-x86-64 hosts
   Target Milestone|---                         |2.0



------- Additional Comments From sabre at nondot.org  2006-11-20 12:17 -------
Fixed, patch here: 
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061120/040137.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 12:58:28 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 12:58:28 -0600
Subject: [LLVMbugs] [Bug 1005] pod2html links unresolvable
Message-ID: <200611201858.kAKIwSTc009967@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1005

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |documentation
         Resolution|                            |FIXED
   Target Milestone|---                         |2.0



------- Additional Comments From sabre at nondot.org  2006-11-20 12:58 -------
Fixed, patch here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061120/040139.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 20 14:30:07 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 20 Nov 2006 14:30:07 -0600
Subject: [LLVMbugs] [Bug 1013] Invalid assembly generated with DWARF2 and
	Obj-C GC
Message-ID: <200611202030.kAKKU7qb011694@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1013

jlaskey at apple.com changed:

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



------- Additional Comments From jlaskey at apple.com  2006-11-20 14:30 -------
Index: lib/CodeGen/AsmPrinter.cpp
===============================================================
====
RCS file: /var/cvs/llvm/llvm/lib/CodeGen/AsmPrinter.cpp,v
retrieving revision 1.116
diff -u -1 -0 -d -r1.116 AsmPrinter.cpp
--- lib/CodeGen/AsmPrinter.cpp	1 Nov 2006 09:23:08 -0000
	1.116
+++ lib/CodeGen/AsmPrinter.cpp	20 Nov 2006 20:28:17 -0000
@@ -327,23 +327,31 @@
       // Emit the function pointer.
       EmitGlobalConstant(CS->getOperand(1));
     }
 }
 
 /// getGlobalLinkName - Returns the asm/link name of of the specified
 /// global variable.  Should be overridden by each target asm printer to
 /// generate the appropriate value.
 const std::string AsmPrinter::getGlobalLinkName(const GlobalVariable *GV) const{
   std::string LinkName;
-  // Default action is to use a global symbol.                              
-  LinkName = TAI->getGlobalPrefix();
-  LinkName += GV->getName();
+  
+  if (isa<Function>(GV)) {
+    LinkName += TAI->getFunctionAddrPrefix();
+    LinkName += Mang->getValueName(GV);
+    LinkName += TAI->getFunctionAddrSuffix();
+  } else {
+    LinkName += TAI->getGlobalVarAddrPrefix();
+    LinkName += Mang->getValueName(GV);
+    LinkName += TAI->getGlobalVarAddrSuffix();
+  }  
+  
   return LinkName;
 }
 
 // EmitAlignment - Emit an alignment directive to the specified power of two.
 void AsmPrinter::EmitAlignment(unsigned NumBits, const GlobalValue *GV) const {
   if (GV && GV->getAlignment())
     NumBits = Log2_32(GV->getAlignment());
   if (NumBits == 0) return;   // No need to emit alignment.
   if (TAI->getAlignmentIsInBytes()) NumBits = 1 << NumBits;
   O << TAI->getAlignDirective() << NumBits << "\n";




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov 22 19:36:52 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 22 Nov 2006 19:36:52 -0600
Subject: [LLVMbugs] [Bug 966] Turning on predicate simplifier pass breaks
	llvm-gcc bootstrapping
Message-ID: <200611230136.kAN1aqlg018189@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=966

nicholas at mxc.ca changed:

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



------- Additional Comments From nicholas at mxc.ca  2006-11-22 19:36 -------
New code has been checked in. I can bootstrap with predsimplify enabled. If you
can't, please reopen this bug.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Sat Nov 25 15:57:22 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Sat, 25 Nov 2006 15:57:22 -0600
Subject: [LLVMbugs] [Bug 821] llvm-gcc4 doesn't build: pthread_once not found
Message-ID: <200611252157.kAPLvMRL016205@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=821

asl at math.spbu.ru changed:

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





------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 10:31:40 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 10:31:40 -0600
Subject: [LLVMbugs] [Bug 1014] NEW: Bad codegen (possible due to bug on
	InstCombine)
Message-ID: <200611271631.kARGVeh9007022@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1014

           Summary: Bad codegen (possible due to bug on InstCombine)
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru


Consider the attached .ll file.

It will ask a number as input. Input "32".
Correct answer is "41". It can be obtained via cbackend. Both llc and lli
generate code, which produces "186".



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 12:10:22 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 12:10:22 -0600
Subject: [LLVMbugs] [Bug 1009] bootstrap failure building libgcc2
Message-ID: <200611271810.kARIAM9m008885@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1009

jlaskey at apple.com changed:

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





------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 13:56:12 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 13:56:12 -0600
Subject: [LLVMbugs] [Bug 1014] instcombine miscompilation
Message-ID: <200611271956.kARJuCgE011113@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1014

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |miscompilation
         OS/Version|Linux                       |All
           Platform|PC                          |All
         Resolution|                            |FIXED
            Summary|Bad codegen (possible due to|instcombine miscompilation
                   |bug on InstCombine)         |
   Target Milestone|---                         |2.0



------- Additional Comments From sabre at nondot.org  2006-11-27 13:56 -------
Fixed.  Testcase here: InstCombine/2006-11-27-XorBug.ll  Patch here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040292.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 14:47:19 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 14:47:19 -0600
Subject: [LLVMbugs] [Bug 1014] instcombine miscompilation
Message-ID: <200611272047.kARKlJUb012257@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1014

asl at math.spbu.ru changed:

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



------- Additional Comments From asl at math.spbu.ru  2006-11-27 14:47 -------
Attached testcase is still broken.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 15:50:51 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 15:50:51 -0600
Subject: [LLVMbugs] [Bug 1014] instcombine + dagcombine miscompilation
Message-ID: <200611272150.kARLopMo013930@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1014

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED
            Summary|instcombine miscompilation  |instcombine + dagcombine
                   |                            |miscompilation



------- Additional Comments From sabre at nondot.org  2006-11-27 15:50 -------
This same bug also exists in the dag combiner.  Fixed like this:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040294.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 16:41:14 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 16:41:14 -0600
Subject: [LLVMbugs] [Bug 1015] NEW: Possible bug in indvars simplify
Message-ID: <200611272241.kARMfEOc015536@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1015

           Summary: Possible bug in indvars simplify
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Loop Optimizer
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru


Consider the input file qappl.ll. Look into bb101 and next BBs. This code
corresponds to the following code in cpp (this is extract from Qt):

    for (i=0; (UINT)mouseTbl[i] != msg.message || !mouseTbl[i]; i += 3)
        ;
    if (!mouseTbl[i])
        return false;
    type = (QEvent::Type)mouseTbl[++i];
    button = mouseTbl[++i];

qappl.out.ll is output from opt -indvars. We have:

        %tmp107 = seteq uint %tmp103, %tmp106           ; <bool> [#uses=1]
        br bool %tmp107, label %cond_next109, label %bb98

cond_next109:           ; preds = %bb101
        %exitcond = setne uint %indvar, 0               ; <bool> [#uses=1]
        br bool %exitcond, label %bb98, label %bb114

%tmp103 corresponds to mouseTbl[i], %tmp106 - to msg.message. %indvar
corresponds to i/3.

Let us assume, that they equals, when i!=0. We have:

%tmp107 = true => branch to cond_next109 => %exitcond = true => branch to bb98

And we have infinite loop here :(



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 18:55:09 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 18:55:09 -0600
Subject: [LLVMbugs] [Bug 1016] NEW: Codegen wrong for: ptrtoint X* %y to bool
Message-ID: <200611280055.kAS0t9J3011234@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1016

           Summary: Codegen wrong for: ptrtoint X* %y to bool
           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: rspencer at x10sys.com


For this code:
bb122:
  %tmp124 = ptrtoint %struct.TypHeader* %hdR to bool
  %i.3.in.in = select bool %tmp124, %struct.TypHeader* %hdR, %struct.TypHeader* %hdL
  %hdR_addr.0 = select bool %tmp124, %struct.TypHeader* %hdL, %struct.TypHeader*
%hdR

llc is generating:
.BB1_3: #bb122
        movl %eax, %ebx
        # TRUNCATE movb %bl, %bl
        testb %bl, %bl
        movl %eax, %edi
        cmovne %esi, %edi
        testb %bl, %bl
        cmovne %eax, %esi

It looks like the recent DAGCombiner changes missed a case for the select
instruction.

Test cases will be attached.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Mon Nov 27 19:04:19 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Mon, 27 Nov 2006 19:04:19 -0600
Subject: [LLVMbugs] [Bug 1016] Codegen wrong for: ptrtoint X* %y to bool
Message-ID: <200611280104.kAS14J8q011531@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1016

sabre at nondot.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |miscompilation
         OS/Version|Linux                       |All
           Platform|PC                          |All
         Resolution|                            |FIXED
            Summary|Codegen wrong for: ptrtoint |Codegen wrong for: ptrtoint
                   |X* %y to bool               |X* %y to bool
   Target Milestone|---                         |2.0



------- Additional Comments From sabre at nondot.org  2006-11-27 19:04 -------
Fixed, patch here:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040313.html

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 01:55:38 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 01:55:38 -0600
Subject: [LLVMbugs] [Bug 1017] NEW: LLVM should support symbol aliasing
Message-ID: <200611280755.kAS7tcx2018684@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1017

           Summary: LLVM should support symbol aliasing
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Core LLVM classes
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: asl at math.spbu.ru


LLVM should support symbol aliasing (the one, produced via
__attribute((alias))__ in gcc).

It seems, this is the main reason of PR1006, since compatibility is achieved via
symbol aliasing.

Aliasing, seems, can be done in some painless way for functions: just clone it
or output the "stub", which will call original one.

Anyway, this is just cheap workaround. Aliasing should be introduced for GV's also.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 08:42:46 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 08:42:46 -0600
Subject: [LLVMbugs] [Bug 1018] NEW: wrong code on amd64
Message-ID: <200611281442.kASEgkPg002461@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1018

           Summary: wrong code on amd64
           Product: libraries
           Version: trunk
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rafael.espindola at gmail.com


The code:

------------------------
extern void * xcalloc (size_t, size_t) __attribute__ ((__malloc__));

typedef void * (*htab_alloc) (size_t, size_t);

int
main(void)
{

  htab_alloc alloc_f = xcalloc;
  (*alloc_f) (1, 5);
  return 0;
}
--------------------------------

compiles to
-------------------------------
main:
        subq $32, %rsp
        movq %rbp, 24(%rsp)
        leaq 24(%rsp), %rbp
        fnstcw -18(%rbp)
        movb $2, -17(%rbp)
        fldcw -18(%rbp)
        leaq xcalloc at GOTPCREL(%rip), %rax
        movq %rax, -16(%rbp)
        xorb %al, %al
        movl $5, %esi
        movl $1, %edi
        call xcalloc
        movl $0, -8(%rbp)
        movl $0, -4(%rbp)
.BB1_1: #return
        movl -4(%rbp), %eax
        movq %rbp, %rsp
        popq %rbp
        ret
-----------------------------
with gcc 4.1.2 it compiles to

------------------------------
main:
.LFB2:
        pushq   %rbp
.LCFI0:
        movq    %rsp, %rbp
.LCFI1:
        subq    $16, %rsp
.LCFI2:
        movq    $xcalloc, -8(%rbp)
        movq    -8(%rbp), %rdx
        movl    $5, %esi
        movl    $1, %edi
        movl    $0, %eax
        call    *%rdx
        movl    $0, %eax
        leave
        ret
-----------------------------
This is a reduction of a segmentation fault during a bootstrap (running 
genmodes).



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 13:57:04 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 13:57:04 -0600
Subject: [LLVMbugs] [Bug 1019] NEW: ParallelJIT requires pthread library,
	makefiles don't handle it right.
Message-ID: <200611281957.kASJv4J5024685@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1019

           Summary: ParallelJIT requires pthread library, makefiles don't
                    handle it right.
           Product: Build scripts
           Version: 1.6
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: minor
          Priority: P2
         Component: Makefiles
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


On windows, without a pthread.h or libpthread.so we get:

make[2]: Entering directory `/d/Apps/llvm/examples/ParallelJIT'
> llvm[2]: Compiling ParallelJIT.cpp for Release build 
> ParallelJIT.cpp:20:21: pthread.h: No such file or directory
> ParallelJIT.cpp:212: error: `pthread_cond_t' does not name a type
> ParallelJIT.cpp:212: error: extra semicolon
> ParallelJIT.cpp:213: error: `pthread_mutex_t' does not name a type
> ParallelJIT.cpp:213: error: extra semicolon
> .... lots more errors

The configuration checks are there, but they are not exposed to the Makefiles
and consequently not being used by the ParallelJIT Makefile.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From isanbard at gmail.com  Tue Nov 28 13:15:27 2006
From: isanbard at gmail.com (Bill Wendling)
Date: Tue, 28 Nov 2006 11:15:27 -0800
Subject: [LLVMbugs] Problem with new CastInst
Message-ID: <16e5fdf90611281115y5f07d8ech86783f355ffed3de@mail.gmail.com>

Hi all,

When doing the nightly build of LLVM, I got these errors which stopped
the build. This was on an iMac x86-64.

-bw

c++ -c -g -DENABLE_CHECKING -DENABLE_ASSERT_CHECKING -DIN_GCC -W -Wall
-Wwrite-strings -pedantic -Wno-long-long -Wno-variadic-macros
-mdynamic-no-pic -DHAVE_CONFIG_H -Wno-unuse\
d -DTARGET_NAME=\"i686-apple-darwin8\" -DNDEBUG -DENABLE_LLVM
-D__STDC_LIMIT_MACROS -DLLVM_VERSION_INFO='"1024"'   -I. -I.
-I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission\
/build/obj/src/gcc
-I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/.
-I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/../include
\
-I./../intl -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/../libcpp/include
 -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/\
include -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/obj-llvm/include
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp
\
-o llvm-convert.o
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In member function
'void<unnamed>::FunctionPrologArgumentConversion::HandleScalarArgumen\
t(const llvm::Type*)':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:216:
error: cannot allocate an object of abstract type 'llvm::CastInst'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
note:   because the following virtual functions are pure within
'llvm::Cas\
tInst':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/Instruction.h:67:
note:  virtual llvm::Instruction* llvm::Instruction::clone() const
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:221:
error: cannot allocate an object of abstract type 'llvm::CastInst'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
note:   since type 'llvm::CastInst' has pure virtual functions
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In member function
'void<unnamed>::FunctionPrologArgumentConversion::EnterField(unsigned\
 int, const llvm::Type*)':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:235:
error: cannot allocate an object of abstract type 'llvm::CastInst'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
note:   since type 'llvm::CastInst' has pure virtual functions
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In member function 'llvm::Value* TreeToLLVM::CastToType(llvm::Value*,
const llvm::Type*)\
':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:665:
error: cannot allocate an object of abstract type 'llvm::CastInst'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
note:   since type 'llvm::CastInst' has pure virtual functions
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In static member function 'static bool
TreeToLLVM::isNoopCast(llvm::Value*, const llvm::\
Type*)':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:673:
error: 'const class llvm::Type' has no member named
'isLosslesslyConvertibleTo'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In member function 'llvm::AllocaInst*
TreeToLLVM::CreateTemporary(const llvm::Type*)':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:699:
error: cannot allocate an object of abstract type 'llvm::CastInst'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
note:   since type 'llvm::CastInst' has pure virtual functions
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In member function
'void<unnamed>::FunctionCallArgumentConversion::HandleScalarArgument(\
const llvm::Type*)':
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:1867:
error: cannot allocate an object of abstract type 'llvm::CastInst'
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
note:   since type 'llvm::CastInst' has pure virtual functions
/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
In member function
'void<unnamed>::FunctionCallArgumentConversion::EnterField(unsigned i\
nt, const llvm::Type*)':

From rspencer at reidspencer.com  Tue Nov 28 15:17:53 2006
From: rspencer at reidspencer.com (Reid Spencer)
Date: Tue, 28 Nov 2006 13:17:53 -0800
Subject: [LLVMbugs] Problem with new CastInst
In-Reply-To: <16e5fdf90611281115y5f07d8ech86783f355ffed3de@mail.gmail.com>
References: <16e5fdf90611281115y5f07d8ech86783f355ffed3de@mail.gmail.com>
Message-ID: <1164748673.2837.101.camel@bashful.x10sys.com>

Hi Bill,

On Tue, 2006-11-28 at 11:15 -0800, Bill Wendling wrote:
> Hi all,
> 
> When doing the nightly build of LLVM, I got these errors which stopped
> the build. This was on an iMac x86-64.

Looks to me like you haven't applied the llvm-gcc patch or updates from
Subversion. The changes are now in subversion so you should just be able
to "svn update" and you'll get the changes you need to allow llvm-gcc to
compile.

Note that these changes didn't hit SVN until early this morning,
probably before you attempted to rebuild llvm-gcc for your nightly test.

Reid.

> 
> -bw
> 
> c++ -c -g -DENABLE_CHECKING -DENABLE_ASSERT_CHECKING -DIN_GCC -W -Wall
> -Wwrite-strings -pedantic -Wno-long-long -Wno-variadic-macros
> -mdynamic-no-pic -DHAVE_CONFIG_H -Wno-unuse\
> d -DTARGET_NAME=\"i686-apple-darwin8\" -DNDEBUG -DENABLE_LLVM
> -D__STDC_LIMIT_MACROS -DLLVM_VERSION_INFO='"1024"'   -I. -I.
> -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission\
> /build/obj/src/gcc
> -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/.
> -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/../include
> \
> -I./../intl -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/../libcpp/include
>  -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/\
> include -I/Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/obj-llvm/include
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp
> \
> -o llvm-convert.o
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In member function
> 'void<unnamed>::FunctionPrologArgumentConversion::HandleScalarArgumen\
> t(const llvm::Type*)':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:216:
> error: cannot allocate an object of abstract type 'llvm::CastInst'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
> note:   because the following virtual functions are pure within
> 'llvm::Cas\
> tInst':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/Instruction.h:67:
> note:  virtual llvm::Instruction* llvm::Instruction::clone() const
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:221:
> error: cannot allocate an object of abstract type 'llvm::CastInst'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
> note:   since type 'llvm::CastInst' has pure virtual functions
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In member function
> 'void<unnamed>::FunctionPrologArgumentConversion::EnterField(unsigned\
>  int, const llvm::Type*)':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:235:
> error: cannot allocate an object of abstract type 'llvm::CastInst'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
> note:   since type 'llvm::CastInst' has pure virtual functions
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In member function 'llvm::Value* TreeToLLVM::CastToType(llvm::Value*,
> const llvm::Type*)\
> ':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:665:
> error: cannot allocate an object of abstract type 'llvm::CastInst'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
> note:   since type 'llvm::CastInst' has pure virtual functions
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In static member function 'static bool
> TreeToLLVM::isNoopCast(llvm::Value*, const llvm::\
> Type*)':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:673:
> error: 'const class llvm::Type' has no member named
> 'isLosslesslyConvertibleTo'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In member function 'llvm::AllocaInst*
> TreeToLLVM::CreateTemporary(const llvm::Type*)':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:699:
> error: cannot allocate an object of abstract type 'llvm::CastInst'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
> note:   since type 'llvm::CastInst' has pure virtual functions
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In member function
> 'void<unnamed>::FunctionCallArgumentConversion::HandleScalarArgument(\
> const llvm::Type*)':
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:1867:
> error: cannot allocate an object of abstract type 'llvm::CastInst'
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/llvm/include/llvm/InstrTypes.h:256:
> note:   since type 'llvm::CastInst' has pure virtual functions
> /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-submission/build/obj/src/gcc/llvm-convert.cpp:
> In member function
> 'void<unnamed>::FunctionCallArgumentConversion::EnterField(unsigned i\
> nt, const llvm::Type*)':
> _______________________________________________
> LLVMbugs mailing list
> LLVMbugs at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs


From isanbard at gmail.com  Tue Nov 28 15:21:49 2006
From: isanbard at gmail.com (Bill Wendling)
Date: Tue, 28 Nov 2006 13:21:49 -0800
Subject: [LLVMbugs] Problem with new CastInst
In-Reply-To: <1164748673.2837.101.camel@bashful.x10sys.com>
References: <16e5fdf90611281115y5f07d8ech86783f355ffed3de@mail.gmail.com>
	<1164748673.2837.101.camel@bashful.x10sys.com>
Message-ID: <16e5fdf90611281321r78036d41l64048bec3ebfaf8d@mail.gmail.com>

On 11/28/06, Reid Spencer <rspencer at reidspencer.com> wrote:
> Hi Bill,
>
> On Tue, 2006-11-28 at 11:15 -0800, Bill Wendling wrote:
> > Hi all,
> >
> > When doing the nightly build of LLVM, I got these errors which stopped
> > the build. This was on an iMac x86-64.
>
> Looks to me like you haven't applied the llvm-gcc patch or updates from
> Subversion. The changes are now in subversion so you should just be able
> to "svn update" and you'll get the changes you need to allow llvm-gcc to
> compile.
>
> Note that these changes didn't hit SVN until early this morning,
> probably before you attempted to rebuild llvm-gcc for your nightly test.
>
Hi Reid,

That would explain the failure. I was using the OpenSource SVN
repository from 3AM. I'm doing a rebuild now, so it should succeed.

-bw

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 15:38:35 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 15:38:35 -0600
Subject: [LLVMbugs] [Bug 1020] NEW: CBE throw exception compiling inline asm
Message-ID: <200611282138.kASLcZiK026609@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1020

           Summary: CBE throw exception compiling inline asm
           Product: libraries
           Version: 1.8
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: C
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sabre at nondot.org


Andrew, with your recent patches, the following .bc throws an exception due to some broken string 
manipulation.

llc -march=c foo.bc

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 15:47:16 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 15:47:16 -0600
Subject: [LLVMbugs] [Bug 1021] NEW: bugpoint reduced test case doesn't
	reproduce bug
Message-ID: <200611282147.kASLlGvT026784@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1021

           Summary: bugpoint reduced test case doesn't reproduce bug
           Product: tools
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: bugpoint
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: djg at cray.com


While reducing a bad-generated-code bug (which I'll submit a separate report
for), bugpoint produced a reduced test case which is supposed to reproduce the 
problem, but it didn't. The bug ended up being with the folding of static
initializer data into instructions that use it. The reduced test case bugpoint
produced had the static initializers removed, so it didn't exhibit the problem.

Here's the reduced test case that reproduces the orignial bug, and which
bugpoint reduces into a test case that doesn't reproduce it.

target datalayout = "e-p:32:32"
target endian = little
target pointersize = 32
target triple = "i686-pc-linux-gnu"
%fmt = constant [4 x sbyte] c"%x\0A\00"
%bytes = constant [4 x sbyte] c"\AA\BB\CC\DD"

implementation

int %main() {
        %y = alloca uint
        %c = cast uint* %y to sbyte*
        %z = getelementptr [4 x sbyte]* %bytes, int 0, int 0
        call void %llvm.memcpy.i32( sbyte* %c, sbyte* %z, uint 4, uint 1 )
        %r = load uint* %y
        %t = cast [4 x sbyte]* %fmt to sbyte*
        %tmp = call int (sbyte*, ...)* %printf( sbyte* %t, uint %r )
        ret int 0
}

declare void %llvm.memcpy.i32(sbyte*, sbyte*, uint, uint)

declare int %printf(sbyte*, ...)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 15:52:55 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 15:52:55 -0600
Subject: [LLVMbugs] [Bug 1022] NEW: bug in folding load of static
	initializer data
Message-ID: <200611282152.kASLqt3j026918@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1022

           Summary: bug in folding load of static initializer data
           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: djg at cray.com


There's a bug in folding loads of static initializers. If there are bytes with
the high bit set, adjacent bytes are getting set to all ones when the compiler
is running on a host with signed char. Here's a patch that fixes it:

Index: SelectionDAGISel.cpp
===================================================================
RCS file: /var/cvs/llvm/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp,v
retrieving revision 1.319
diff -u -r1.319 SelectionDAGISel.cpp
--- SelectionDAGISel.cpp
+++ SelectionDAGISel.cpp
@@ -3179,7 +3179,7 @@
   if (TLI.isLittleEndian())
     Offset = Offset + MSB - 1;
   for (unsigned i = 0; i != MSB; ++i) {
-    Val = (Val << 8) | Str[Offset];
+    Val = (Val << 8) | (unsigned char) Str[Offset];
     Offset += TLI.isLittleEndian() ? -1 : 1;
   }
   return DAG.getConstant(Val, VT);



And here's a testcase that reproduces the problem.

target datalayout = "e-p:32:32"
target endian = little
target pointersize = 32
target triple = "i686-pc-linux-gnu"
%fmt = constant [4 x sbyte] c"%x\0A\00"
%bytes = constant [4 x sbyte] c"\AA\BB\CC\DD"

implementation

int %main() {
        %y = alloca uint
        %c = cast uint* %y to sbyte*
        %z = getelementptr [4 x sbyte]* %bytes, int 0, int 0
        call void %llvm.memcpy.i32( sbyte* %c, sbyte* %z, uint 4, uint 1 )
        %r = load uint* %y
        %t = cast [4 x sbyte]* %fmt to sbyte*
        %tmp = call int (sbyte*, ...)* %printf( sbyte* %t, uint %r )
        ret int 0
}

declare void %llvm.memcpy.i32(sbyte*, sbyte*, uint, uint)

declare int %printf(sbyte*, ...)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 15:57:23 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 15:57:23 -0600
Subject: [LLVMbugs] [Bug 1023] NEW: folding load of non-constant initializer
Message-ID: <200611282157.kASLvN1d027054@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1023

           Summary: folding load of non-constant initializer
           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: djg at cray.com


The logic that folds loads of static initializers doesn't take into account
whether the variable being initialized may be modified at runtime.

Here's a patch that fixes it:


Index: SelectionDAGISel.cpp
===================================================================
RCS file: /var/cvs/llvm/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp,v
retrieving revision 1.319
diff -u -r1.319 SelectionDAGISel.cpp
--- SelectionDAGISel.cpp
+++ SelectionDAGISel.cpp
@@ -3298,7 +3298,7 @@
         }
         if (G) {
           GlobalVariable *GV = dyn_cast<GlobalVariable>(G->getGlobal());
-          if (GV) {
+          if (GV && GV->isConstant()) {
             Str = GV->getStringValue(false);
             if (!Str.empty()) {
               CopyFromStr = true;



And here's a testcase that reproduces it:

target datalayout = "e-p:32:32"
target endian = little
target pointersize = 32
target triple = "i686-pc-linux-gnu"
%fmt = global [4 x sbyte] c"%x\0A\00"
%bytes = global [4 x sbyte] c"\AA\BB\CC\DD"

implementation

void %foo() {
        %y = alloca uint
        %c = cast uint* %y to sbyte*
        %z = getelementptr [4 x sbyte]* %bytes, int 0, int 0
        call void %llvm.memcpy.i32( sbyte* %c, sbyte* %z, uint 4, uint 1 )
        %r = load uint* %y
        %t = cast [4 x sbyte]* %fmt to sbyte*
        %tmp = call int (sbyte*, ...)* %printf( sbyte* %t, uint %r )
        ret void
}

declare void %llvm.memcpy.i32(sbyte*, sbyte*, uint, uint)

declare int %printf(sbyte*, ...)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 16:19:07 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 16:19:07 -0600
Subject: [LLVMbugs] [Bug 1025] NEW: compiler abort with
	llvm.readcyclecounter on x86-64
Message-ID: <200611282219.kASMJ7ND027585@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1025

           Summary: compiler abort with llvm.readcyclecounter on x86-64
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: djg at cray.com


The llvm.readcyclecounter intrinsic gets an abort on x86-64. This message is
printed:

Cannot yet select: 0x894a7d8: i64,ch = ReadCycleCounter 0x894a208

prior to a stack trace and abort.

Here's a testcase:

target datalayout = "e-p:64:64"
target endian = little
target pointersize = 64
target triple = "x86-64-unknown-linux-gnu"

implementation

int %main() {
        %y = call ulong %llvm.readcyclecounter()
        ret int 0
}

declare ulong %llvm.readcyclecounter()



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 16:27:39 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 16:27:39 -0600
Subject: [LLVMbugs] [Bug 1020] CBE throw exception compiling inline asm
Message-ID: <200611282227.kASMRd6D027851@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1020

alenhar2 at uiuc.edu changed:

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



------- Additional Comments From alenhar2 at uiuc.edu  2006-11-28 16:27 -------
fixed.

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040351.html



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 16:06:31 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 16:06:31 -0600
Subject: [LLVMbugs] [Bug 1024] NEW: use 16-byte stack alignment on x86-64 ELF
Message-ID: <200611282206.kASM6VUc027301@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1024

           Summary: use 16-byte stack alignment on x86-64 ELF
           Product: libraries
           Version: trunk
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: X86
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: djg at cray.com


The ELF gABI for x86-64 requires the stack-pointer to be 16-byte aligned.
Right now LLVM is not doing this and I'm seeing lots of segfaults on
x86_64-unknown-linux-gnu without this.

Here's the reference:
http://www.x86-64.org/documentation/abi.pdf
Section 3.2.2, second paragraph.

Here's a patch:

Index: X86Subtarget.cpp
===================================================================
RCS file: /var/cvs/llvm/llvm/lib/Target/X86/X86Subtarget.cpp,v
retrieving revision 1.40
diff -u -r1.40 X86Subtarget.cpp
--- X86Subtarget.cpp    20 Nov 2006 18:16:05 -0000      1.40
+++ X86Subtarget.cpp    28 Nov 2006 21:59:54 -0000
@@ -255,6 +255,8 @@
     }
   }
 
-  if (TargetType == isDarwin || TargetType == isCygwin)
+  if (TargetType == isDarwin ||
+      TargetType == isCygwin ||
+      (TargetType == isELF && Is64Bit))
     stackAlignment = 16;
 }



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 19:38:18 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 19:38:18 -0600
Subject: [LLVMbugs] [Bug 1022] bug in folding load of static initializer data
Message-ID: <200611290138.kAT1cIvp031870@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1022

evan.cheng at apple.com changed:

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



------- Additional Comments From evan.cheng at apple.com  2006-11-28 19:38 -------
Patch applied. Thanks!




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 19:58:24 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 19:58:24 -0600
Subject: [LLVMbugs] [Bug 1023] folding load of non-constant initializer
Message-ID: <200611290158.kAT1wOxh032548@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1023

evan.cheng at apple.com changed:

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



------- Additional Comments From evan.cheng at apple.com  2006-11-28 19:58 -------
Patch applied. Thanks!




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Tue Nov 28 20:00:51 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Tue, 28 Nov 2006 20:00:51 -0600
Subject: [LLVMbugs] [Bug 1024] use 16-byte stack alignment on x86-64 ELF
Message-ID: <200611290200.kAT20pMp032630@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1024

evan.cheng at apple.com changed:

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



------- Additional Comments From evan.cheng at apple.com  2006-11-28 20:00 -------
Applied. Thanks!




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov 29 02:28:22 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 29 Nov 2006 02:28:22 -0600
Subject: [LLVMbugs] [Bug 1025] compiler abort with llvm.readcyclecounter on
	x86-64
Message-ID: <200611290828.kAT8SMI5015541@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1025

evan.cheng at apple.com changed:

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



------- Additional Comments From evan.cheng at apple.com  2006-11-29 02:28 -------
Fixed.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Wed Nov 29 15:29:52 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Wed, 29 Nov 2006 15:29:52 -0600
Subject: [LLVMbugs] [Bug 1026] NEW: InstVisit duplicates instruction visitors
Message-ID: <200611292129.kATLTqso015114@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1026

           Summary: InstVisit duplicates instruction visitors
           Product: libraries
           Version: 1.0
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: minor
          Priority: P2
         Component: Support Libraries
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: rspencer at x10sys.com


The InstVisitor class currently defines the individual instruction visitors
twice, with one defaulting to a delegation to the other. This is a bit silly and
should be corrected. However, it might mean cleanup up the users of InstVisitor
as well.

The change is fairly simple. In Instruction.def, change the various
HANDLE_*_INST macros to take three arguments. From:

HANDLE_*_INST(num, opcode, CLASS)

to

HANDLE_*_INT(num, opcode, CLASS, BASE_CLASS)

for example:

HANDLE_TERM_INST(1, Ret, ReturnInst, TerminatorInst)

The HANDLE_INST class would also need to change.

Then the macro in InstVisitor.h could then be defined as:

#define HANDLE_INST(NUM, OPCODE, CLASS, BASECLASS) \
    RetTy visit##CLASS(CLASS &I) { DELEGATE(BASECLASS); }

and the duplicates deleted.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 30 18:38:12 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 30 Nov 2006 18:38:12 -0600
Subject: [LLVMbugs] [Bug 1019] ParallelJIT requires pthread library,
	makefiles don't handle it right.
Message-ID: <200612010038.kB10cCLW028518@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1019

rspencer at x10sys.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Keywords|                            |build-problem
         Resolution|                            |FIXED
   Target Milestone|---                         |2.0



------- Additional Comments From rspencer at x10sys.com  2006-11-30 18:38 -------
Patches committed. Thanks, Anton.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 30 21:54:41 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 30 Nov 2006 21:54:41 -0600
Subject: [LLVMbugs] [Bug 1027] NEW: llvm+g++ compilation error when
	compiling MySQL 5.2.27
Message-ID: <200612010354.kB13sfVf031706@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1027

           Summary: llvm+g++ compilation error when compiling MySQL 5.2.27
           Product: tools
           Version: trunk
          Platform: Macintosh
        OS/Version: MacOS X
            Status: NEW
          Severity: normal
          Priority: P2
         Component: llvm-g++
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: kremenek at cs.stanford.edu


compiling set_var.cc in MySQL 5.2.27 triggers an internal compilation error:

make[4]: Nothing to be done for `all'.
if llvm-g++ -DMYSQL_SERVER -DDEFAULT_MYSQL_HOME="\"/usr/local\""
-DDATADIR="\"/usr/local/var\"" -DSHAREDIR="\"/usr/local/share/mysql\""
-DHAVE_CONFIG_H -I. -I. -I.. -I../innobase/include -I../include -I../include
-I../regex -I.     -O3 -DDBUG_OFF    -fno-implicit-templates -fno-exceptions
-fno-rtti -D_P1003_1B_VISIBLE -DSIGNAL_WITH_VIO_CLOSE -DSIGNALS_DONT_BREAK_READ
-DIGNORE_SIGHUP_SIGQUIT -MT set_var.o -MD -MP -MF ".deps/set_var.Tpo" -c -o
set_var.o set_var.cc; \
then mv -f ".deps/set_var.Tpo" ".deps/set_var.Po"; else rm -f
".deps/set_var.Tpo"; exit 1; fi
set_var.cc:3690: internal compiler error: in EmitLV_COMPONENT_REF, at
llvm-convert.cpp:4841
Please submit a full bug report,

Attached is the preprocessed .i file.  compile with:

llvm-g++  -fno-implicit-templates -fno-exceptions -fno-rtti  -c -o set_var.o
set_var.i



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

From bugzilla-daemon at cs.uiuc.edu  Thu Nov 30 22:56:03 2006
From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu)
Date: Thu, 30 Nov 2006 22:56:03 -0600
Subject: [LLVMbugs] [Bug 1028] NEW: CBE should support
	llvm.stacksave/stackrestore
Message-ID: <200612010456.kB14u3VH032736@zion.cs.uiuc.edu>

http://llvm.org/bugs/show_bug.cgi?id=1028

           Summary: CBE should support llvm.stacksave/stackrestore
           Product: libraries
           Version: trunk
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Backend: C
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: sabre at nondot.org


The CBE should translate llvm.stacksave/stackrestore into calls to GCC's __builtin_stack_save/
__builtin_stack_restore functions.  The only tricky part of this is figuring out which versions of GCC have 
these builtins (if we care).

This would get 176.gcc working with the CBE, which would be very useful when bugpointing.

-Chris



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.