From bugzilla-daemon at cs.uiuc.edu Fri Dec 1 00:38:14 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 1 Dec 2006 00:38:14 -0600 Subject: [LLVMbugs] [Bug 1027] llvm-g++ crash compiling MySQL 5.2.27 Message-ID: <200612010638.kB16cED1002170@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1027 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED OS/Version|MacOS X |All Platform|Macintosh |All Resolution| |FIXED Summary|llvm+g++ compilation error |llvm-g++ crash compiling |when compiling MySQL 5.2.27 |MySQL 5.2.27 Target Milestone|--- |2.0 ------- Additional Comments From sabre at nondot.org 2006-12-01 00:38 ------- Fixed, testcase here: Regression/C++Frontend/2006-11-30-ConstantExprCrash.cpp Patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040515.html Thanks Ted! -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 Fri Dec 1 01:25:10 2006 From: isanbard at gmail.com (Bill Wendling) Date: Thu, 30 Nov 2006 23:25:10 -0800 Subject: [LLVMbugs] llvm-gcc Build Failure Message-ID: <7C6FE855-F4AD-4042-8D6D-FAA9A475C79E@gmail.com> There's a failure when I build llvm-gcc4. Here's the error: c++ -c -g -O2 -DIN_GCC -W -Wall -Wwrite-strings -pedantic -Wno-long- long -Wno-variadic-macros -fno-common -DHAVE_CONFIG_H -Wno-unused - DTARGET_NAME=\"powerpc-apple-darwin8.8.0\" -DENABLE_LLVM - D__STDC_LIMIT_MACROS -I. -I. -I../../llvm-gcc4.src/gcc -I../../llvm- gcc4.src/gcc/. -I../../llvm-gcc4.src/gcc/../include -I./../intl - I../../llvm-gcc4.src/gcc/../libcpp/include -I/Users/wendling/llvm/ llvm.src/include -I/Users/wendling/llvm/llvm.obj/include ../../llvm- gcc4.src/gcc/llvm-debug.cpp -o llvm-debug.o ../../llvm-gcc4.src/gcc/llvm-debug.cpp: In member function 'void llvm::DebugInfo::EmitFunctionStart(tree_node*, llvm::Function*, llvm::BasicBlock*)': ../../llvm-gcc4.src/gcc/llvm-debug.cpp:272: error: 'class llvm::SubprogramDesc' has no member named 'setDisplayName' ../../llvm-gcc4.src/gcc/llvm-debug.cpp: In member function 'void llvm::DebugInfo::EmitGlobalVariable(llvm::GlobalVariable*, tree_node*)': ../../llvm-gcc4.src/gcc/llvm-debug.cpp:429: error: 'class llvm::GlobalVariableDesc' has no member named 'setDisplayName' ../../llvm-gcc4.src/gcc/llvm-debug.cpp: In member function 'llvm::TypeDesc* llvm::DebugInfo::getOrCreateType(tree_node*, llvm::CompileUnitDesc*)': ../../llvm-gcc4.src/gcc/llvm-debug.cpp:739: error: 'class llvm::GlobalVariableDesc' has no member named 'setDisplayName' ../../llvm-gcc4.src/gcc/llvm-debug.cpp:779: error: 'class llvm::SubprogramDesc' has no member named 'setDisplayName' make[1]: *** [llvm-debug.o] Error 1 make: *** [all-gcc] Error 2 Any ideas? -bw From rafael.espindola at gmail.com Fri Dec 1 07:05:43 2006 From: rafael.espindola at gmail.com (=?UTF-8?Q?Rafael_Esp=C3=ADndola?=) Date: Fri, 1 Dec 2006 10:05:43 -0300 Subject: [LLVMbugs] llvm-gcc Build Failure In-Reply-To: <7C6FE855-F4AD-4042-8D6D-FAA9A475C79E@gmail.com> References: <7C6FE855-F4AD-4042-8D6D-FAA9A475C79E@gmail.com> Message-ID: <564d96fb0612010505u1fc51ec1xa6b1c3a7a3e1bec9@mail.gmail.com> This patch I think: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040473.html Best Regards, Rafael From bugzilla-daemon at cs.uiuc.edu Fri Dec 1 11:50:01 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 1 Dec 2006 11:50:01 -0600 Subject: [LLVMbugs] [Bug 1029] NEW: Couldn't allocate output reg for contraint 'r' (AMD64) Message-ID: <200612011750.kB1Ho1H0023143@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1029 Summary: Couldn't allocate output reg for contraint 'r' (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 ------------------------------------ void frame_dummy (void) { void (*register_classes) (void *) = 0; __asm ("" : "+r" (register_classes)); } ------------------------------------ produces the following error: --------------------------------------------- Couldn't allocate output reg for contraint 'r'! --------------------------------------------- ------- 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 Dec 1 13:31:45 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 1 Dec 2006 13:31:45 -0600 Subject: [LLVMbugs] [Bug 1030] NEW: __APPLE_CC__ is incorrect defined on Linux Message-ID: <200612011931.kB1JVj7U025435@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1030 Summary: __APPLE_CC__ is incorrect defined on Linux 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 It seems, that this define is incorrectly set (al least) on Linux platform Running llvm-gcc -E -dM -xc /dev/null -o - | grep APPLE produces: #define __APPLE_CC__ 5421 ------- 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 Fri Dec 1 13:17:32 2006 From: isanbard at gmail.com (Bill Wendling) Date: Fri, 1 Dec 2006 11:17:32 -0800 Subject: [LLVMbugs] llvm-gcc Build Failure In-Reply-To: <564d96fb0612010505u1fc51ec1xa6b1c3a7a3e1bec9@mail.gmail.com> References: <7C6FE855-F4AD-4042-8D6D-FAA9A475C79E@gmail.com> <564d96fb0612010505u1fc51ec1xa6b1c3a7a3e1bec9@mail.gmail.com> Message-ID: <16e5fdf90612011117o25c23bc9g52447e9bf4ffc924@mail.gmail.com> On 12/1/06, Rafael Esp?ndola wrote: > This patch I think: > http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061127/040473.html > Thanks. :-) -bw From bugzilla-daemon at cs.uiuc.edu Sat Dec 2 22:44:31 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 2 Dec 2006 22:44:31 -0600 Subject: [LLVMbugs] [Bug 1031] NEW: make install relinks tools Message-ID: <200612030444.kB34iVGA013551@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1031 Summary: make install relinks tools Product: Build scripts Version: cvs Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Makefiles AssignedTo: unassignedbugs at nondot.org ReportedBy: nicholas at mxc.ca Running "make" compiles and links the tools, but then "make install" will relink them even if nothing changed. Could this please be avoided? ------- 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 Dec 2 22:52:40 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 2 Dec 2006 22:52:40 -0600 Subject: [LLVMbugs] [Bug 1031] make install relinks tools Message-ID: <200612030452.kB34qepi013730@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1031 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |INVALID Target Milestone|--- |2.0 ------- Additional Comments From rspencer at x10sys.com 2006-12-02 22:52 ------- There are no unnecessary steps. The re-link at install time is done by libtool's "finalize" step and is required to ensure that runtime library locations are correct for the tools in their installed locations. If its something other than libtool (mklib), please re-open and provide more details. ------- 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 Dec 3 10:42:16 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 3 Dec 2006 10:42:16 -0600 Subject: [LLVMbugs] [Bug 1030] __APPLE_CC__ is incorrect defined on Linux Message-ID: <200612031642.kB3GgGuk023689@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1030 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |INVALID ------- Additional Comments From rspencer at x10sys.com 2006-12-03 10:42 ------- This isn't correct. The __APPLE_CC__ symbol is defined by the compiler (header or not) to identify the compiler as the apple version of gcc. It doesn't tell you anything about the platform. I have fixed the test/Regression/CFrontend/extern-weak.ll test to check for the symbol __linux__ instead. ------- 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 Dec 3 14:43:40 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 3 Dec 2006 14:43:40 -0600 Subject: [LLVMbugs] [Bug 1030] __APPLE_CC__ is incorrect defined on Linux Message-ID: <200612032043.kB3Khe6e012882@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1030 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|INVALID | ------- Additional Comments From sabre at nondot.org 2006-12-03 14:43 ------- This isn't quite true. Mainline GCC also defines __APPLE_CC__ when targetting darwin. If llvm-gcc defines __APPLE_CC__ on non-darwin targets, that is a 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 Mon Dec 4 08:33:30 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 4 Dec 2006 08:33:30 -0600 Subject: [LLVMbugs] [Bug 1032] NEW: junk `(%rip)' after expression Message-ID: <200612041433.kB4EXUCs021485@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1032 Summary: junk `(%rip)' after expression 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 Compiling the attached test with -relocation-model=static produces an assembly file with the line: movq __CTOR_END__(%rip)(,%rax,8), %rax This fails to assembly with the error: test.s:18: Error: junk `(%rip)' after expression ------- 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 Dec 4 11:58:42 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 4 Dec 2006 11:58:42 -0600 Subject: [LLVMbugs] [Bug 1033] NEW: bitcast between long and double on x86-64 Message-ID: <200612041758.kB4Hwg6c025515@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1033 Summary: bitcast between long and double 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 On x86-64 a bitcast from long to double or double to long results triggers an llc abort. Among other things, this comes up in the expansion of copysign. long %p(double %t) { %u = bitcast double %t to long ret long %u } double %q(long %t) { %u = bitcast long %t to double ret double %u } With llc -march=x86-64, this gets Cannot yet select: 0x8813318: f64 = bit_convert 0x88135c8 ------- 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 Dec 4 16:04:26 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 4 Dec 2006 16:04:26 -0600 Subject: [LLVMbugs] [Bug 1034] NEW: SCCP and Packed Tyep Message-ID: <200612042204.kB4M4Qc4030834@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1034 Summary: SCCP and Packed Tyep Product: libraries Version: trunk Platform: Macintosh OS/Version: MacOS X Status: NEW Severity: normal Priority: P2 Component: Scalar Optimizations AssignedTo: unassignedbugs at nondot.org ReportedBy: dpatel at apple.com SCCP does not work will with Vectors. Fix it. ------- 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 Dec 4 16:39:23 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 4 Dec 2006 16:39:23 -0600 Subject: [LLVMbugs] [Bug 1029] Couldn't allocate output reg for contraint 'r' (AMD64) Message-ID: <200612042239.kB4MdNjU031704@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1029 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Keywords| |compile-fail Resolution| |FIXED Target Milestone|--- |2.0 Version|trunk |1.9 ------- Additional Comments From sabre at nondot.org 2006-12-04 16:39 ------- Patch applied, thanks Rafael! Patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/040773.html Testcase here: Regression/CodeGen/X86/x86-64-asm.ll -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 Dec 4 18:59:22 2006 From: isanbard at gmail.com (Bill Wendling) Date: Mon, 4 Dec 2006 16:59:22 -0800 Subject: [LLVMbugs] Problems with Stacker Message-ID: <16e5fdf90612041659v43be216enf6853ce0a6de370f@mail.gmail.com> Iwhen building the top of tree, I keep getting this error: llvm[4]: Compiling stacker_rt.c for Debug build (bytecode) llvm[4]: Compiling stacker_rt.ll to stacker_rt.bc for Debug build (bytecode) /Volumes/Gir/devel/llvm/llvm.clean/Debug/bin/gccas: /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.ll:18,0: syntax error, unexpected $undefined, expecting STRINGCONSTANT /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.ll:18: while reading token: 'a' make[4]: *** [/Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.bc] Error 1 make[3]: *** [all] Error 1 make[2]: *** [all] Error 1 make[1]: *** [all] Error 1 make: *** [all] Error 1 Anyone else receiving these? -bw From rspencer at reidspencer.com Mon Dec 4 19:33:39 2006 From: rspencer at reidspencer.com (Reid Spencer) Date: Mon, 04 Dec 2006 17:33:39 -0800 Subject: [LLVMbugs] Problems with Stacker In-Reply-To: <16e5fdf90612041659v43be216enf6853ce0a6de370f@mail.gmail.com> References: <16e5fdf90612041659v43be216enf6853ce0a6de370f@mail.gmail.com> Message-ID: <1165282419.26545.102.camel@bashful.x10sys.com> On Mon, 2006-12-04 at 16:59 -0800, Bill Wendling wrote: > Iwhen building the top of tree, I keep getting this error: > > llvm[4]: Compiling stacker_rt.c for Debug build (bytecode) > llvm[4]: Compiling stacker_rt.ll to stacker_rt.bc for Debug build (bytecode) > /Volumes/Gir/devel/llvm/llvm.clean/Debug/bin/gccas: > /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.ll:18,0: > syntax error, unexpected $undefined, expecting STRINGCONSTANT > /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.ll:18: > while reading token: 'a' > make[4]: *** [/Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.bc] Please see my email about llvm-gcc3 not being supported any more. llvm-gcc3 cannot produce assembly code that gccas can read any more. You'll need to upgrade to llvm-gcc4 if you wish to work on the LLVM head. > Error 1 > make[3]: *** [all] Error 1 > make[2]: *** [all] Error 1 > make[1]: *** [all] Error 1 > make: *** [all] Error 1 > > > Anyone else receiving these? > > -bw > _______________________________________________ > LLVMbugs mailing list > LLVMbugs at cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs From isanbard at gmail.com Mon Dec 4 19:41:19 2006 From: isanbard at gmail.com (Bill Wendling) Date: Mon, 4 Dec 2006 17:41:19 -0800 Subject: [LLVMbugs] Problems with Stacker In-Reply-To: <1165282419.26545.102.camel@bashful.x10sys.com> References: <16e5fdf90612041659v43be216enf6853ce0a6de370f@mail.gmail.com> <1165282419.26545.102.camel@bashful.x10sys.com> Message-ID: <16e5fdf90612041741j304f56a0k619cec6e7c146403@mail.gmail.com> On 12/4/06, Reid Spencer wrote: > On Mon, 2006-12-04 at 16:59 -0800, Bill Wendling wrote: > > Iwhen building the top of tree, I keep getting this error: > > > > llvm[4]: Compiling stacker_rt.c for Debug build (bytecode) > > llvm[4]: Compiling stacker_rt.ll to stacker_rt.bc for Debug build (bytecode) > > /Volumes/Gir/devel/llvm/llvm.clean/Debug/bin/gccas: > > /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.ll:18,0: > > syntax error, unexpected $undefined, expecting STRINGCONSTANT > > /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.ll:18: > > while reading token: 'a' > > make[4]: *** [/Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/Debug/stacker_rt.bc] > > Please see my email about llvm-gcc3 not being supported any more. > llvm-gcc3 cannot produce assembly code that gccas can read any more. > You'll need to upgrade to llvm-gcc4 if you wish to work on the LLVM > head. > I've only been using llvm-gcc4. I reconfigured/built my tree and it works now. -bw From dpatel at apple.com Mon Dec 4 19:25:15 2006 From: dpatel at apple.com (Devang Patel) Date: Mon, 4 Dec 2006 17:25:15 -0800 Subject: [LLVMbugs] Problems with Stacker In-Reply-To: <16e5fdf90612041659v43be216enf6853ce0a6de370f@mail.gmail.com> References: <16e5fdf90612041659v43be216enf6853ce0a6de370f@mail.gmail.com> Message-ID: On Dec 4, 2006, at 4:59 PM, Bill Wendling wrote: > Iwhen building the top of tree, I keep getting this error: > > llvm[4]: Compiling stacker_rt.c for Debug build (bytecode) > llvm[4]: Compiling stacker_rt.ll to stacker_rt.bc for Debug build > (bytecode) > /Volumes/Gir/devel/llvm/llvm.clean/Debug/bin/gccas: > /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/ > Debug/stacker_rt.ll:18,0: > syntax error, unexpected $undefined, expecting STRINGCONSTANT > /Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/lib/runtime/ > Debug/stacker_rt.ll:18: > while reading token: 'a' > make[4]: *** [/Volumes/Gir/devel/llvm/llvm.clean/projects/Stacker/ > lib/runtime/Debug/stacker_rt.bc] > Error 1 I got this error earlier today. It disappeared when I reconfigured my llvm tree using newer llvm-gcc. - Devang From bugzilla-daemon at cs.uiuc.edu Tue Dec 5 08:38:20 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 5 Dec 2006 08:38:20 -0600 Subject: [LLVMbugs] [Bug 1035] NEW: stage1 compiler miscompiles the stage2 on amd64 Message-ID: <200612051438.kB5EcKLR025215@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1035 Summary: stage1 compiler miscompiles the stage2 on amd64 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 The stage2 compiler generated during a bootstrap on a linux amd64 segfaults while trying to compile the source: ------------------ typedef __PTRDIFF_TYPE__ ptrdiff_t; ------------------ The stage1 compiler works on this input, so i suppose that stage2 was miscompiled. Any ideas on how to reduce this test case? ------- 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 Dec 5 12:23:53 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 5 Dec 2006 12:23:53 -0600 Subject: [LLVMbugs] [Bug 1033] bitcast between long and double on x86-64 Message-ID: <200612051823.kB5INrER012693@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1033 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From sabre at nondot.org 2006-12-05 12:23 ------- The preferred fix for this is to expand it into movd/movq. However, Evan is busy with other things, so this won't happen in the near term. In the short-term, this patch gets us correct, but suboptimal (goes through memory) code: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/040803.html Testcase here: CodeGen/X86/bitcast.ll -Chris _test1: subq $8, %rsp movsd %xmm0, (%rsp) movq (%rsp), %rax addq $8, %rsp ret _test2: subq $8, %rsp movq %rdi, (%rsp) movsd (%rsp), %xmm0 addq $8, %rsp ret ------- 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 Dec 5 13:52:37 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 5 Dec 2006 13:52:37 -0600 Subject: [LLVMbugs] [Bug 1032] x86-64: junk `(%rip)' after expression Message-ID: <200612051952.kB5JqbQ8014636@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1032 evan.cheng at apple.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED ------- Additional Comments From evan.cheng at apple.com 2006-12-05 13:52 ------- 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 Tue Dec 5 13:53:42 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 5 Dec 2006 13:53:42 -0600 Subject: [LLVMbugs] [Bug 1018] wrong code on amd64 Message-ID: <200612051953.kB5JrgfR014696@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1018 evan.cheng at apple.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED ------- Additional Comments From evan.cheng at apple.com 2006-12-05 13:53 ------- Rafael, this should be fixed. Please verify. ------- 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 Wed Dec 6 13:58:01 2006 From: isanbard at gmail.com (Bill Wendling) Date: Wed, 6 Dec 2006 11:58:01 -0800 Subject: [LLVMbugs] Problems in Build Message-ID: <16e5fdf90612061158w429d5860uc8763a89814658ed@mail.gmail.com> There was a failure in last night's build. Here's the error message: /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-nightly/build/obj/src/gcc/llvm-backend.cpp: In function 'void print_llvm(FILE*\ , void*)': /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-nightly/build/obj/src/gcc/llvm-backend.cpp:821: error: cannot convert 'bool' t\ o 'const llvm::Module*' for argument '4' to 'std::ostream& llvm::WriteAsOperand(std::ostream&, const llvm::Value*, bool\ , const llvm::Module*)' make[3]: *** [llvm-backend.o] Error 1 make[3]: *** Waiting for unfinished jobs.... rm cpp.pod gpl.pod fsf-funding.pod gcc.pod gfdl.pod gcov.pod make[2]: *** [stage1_build] Error 2 make[1]: *** [bootstrap] Error 2 Any comments? Thanks! -bw From dpatel at apple.com Wed Dec 6 14:09:59 2006 From: dpatel at apple.com (Devang Patel) Date: Wed, 6 Dec 2006 12:09:59 -0800 Subject: [LLVMbugs] Problems in Build In-Reply-To: <16e5fdf90612061158w429d5860uc8763a89814658ed@mail.gmail.com> References: <16e5fdf90612061158w429d5860uc8763a89814658ed@mail.gmail.com> Message-ID: <24A79FD9-0378-4175-AA82-C209C10EAFA2@apple.com> On Dec 6, 2006, at 11:58 AM, Bill Wendling wrote: > There was a failure in last night's build. Here's the error message: > > /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-nightly/build/obj/src/gcc/ > llvm-backend.cpp: > In function 'void print_llvm(FILE*\ > , void*)': > /Volumes/Gir/devel/llvm/SUBMISSIONS/llvm-nightly/build/obj/src/gcc/ > llvm-backend.cpp:821: > error: cannot convert 'bool' t\ > o 'const llvm::Module*' for argument '4' to 'std::ostream& > llvm::WriteAsOperand(std::ostream&, const llvm::Value*, bool\ > , const llvm::Module*)' > make[3]: *** [llvm-backend.o] Error 1 > make[3]: *** Waiting for unfinished jobs.... > rm cpp.pod gpl.pod fsf-funding.pod gcc.pod gfdl.pod gcov.pod > make[2]: *** [stage1_build] Error 2 > make[1]: *** [bootstrap] Error 2 > > Any comments? Jim fixed this... http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/040889.html - Devang From bugzilla-daemon at cs.uiuc.edu Wed Dec 6 22:02:23 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Wed, 6 Dec 2006 22:02:23 -0600 Subject: [LLVMbugs] [Bug 1036] NEW: Weird Failure: test/Regression/CodeGen/Generic/print-arith-int.ll Message-ID: <200612070402.kB742N3c008993@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1036 Summary: Weird Failure: test/Regression/CodeGen/Generic/print- arith-int.ll 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 This test started failing on the cvs head today: llc: /proj/llvm/llvm-1/include/llvm/Target/MRegisterInfo.h:236: static bool llvm::MRegisterInfo::isVirtualRegister(unsigned int): Assertion `Reg && "this is not a register!"' failed. llc((anonymous namespace)::PrintStackTrace()+0x1a)[0x886f136] llc((anonymous namespace)::SignalHandler(int)+0x110)[0x886f45e] [0x346420] /lib/libc.so.6(abort+0x101)[0x15e4f1] /lib/libc.so.6(__assert_fail+0xfd)[0x156859] llc(llvm::MRegisterInfo::isVirtualRegister(unsigned int)+0x30)[0x84e838a] llc(llvm::VirtRegMap::getStackSlot(unsigned int) const+0x11)[0x86e40eb] llc(llvm::VirtRegMap::hasStackSlot(unsigned int) const+0x18)[0x86e4146] llc[0x86e2083] llc[0x86e9396] llc[0x86cee7d] llc(llvm::MachineFunctionPass::runOnFunction(llvm::Function&)+0x28)[0x83edb7c] llc(llvm::FunctionPassManagerT::runPass(llvm::FunctionPass*, llvm::Function*)+0x1f)[0x87f25e3] llc(llvm::PassManagerT::runPasses(llvm::Function*, std::map >, std::less, std::allocator > > > >&)+0x12a)[0x8803960] llc(llvm::PassManagerT::runOnUnit(llvm::Function*)+0x184)[0x8806114] llc(llvm::FunctionPassManagerT::runOnFunction(llvm::Function&)+0x23)[0x87f26c5] llc(llvm::FunctionPassManager::run(llvm::Function&)+0x91)[0x87f0abb] llc(main+0x925)[0x83dcfbd] /lib/libc.so.6(__libc_start_main+0xdc)[0x14a4e4] llc(__gxx_personality_v0+0x16d)[0x83dbd61] 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 Wed Dec 6 22:03:11 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Wed, 6 Dec 2006 22:03:11 -0600 Subject: [LLVMbugs] [Bug 1037] NEW: Weird Failure: test/Regression/CodeGen/Generic/print-arith-int.ll Message-ID: <200612070403.kB743Bpw009016@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1037 Summary: Weird Failure: test/Regression/CodeGen/Generic/print- arith-int.ll 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 This test started failing on the cvs head today: llc: /proj/llvm/llvm-1/include/llvm/Target/MRegisterInfo.h:236: static bool llvm::MRegisterInfo::isVirtualRegister(unsigned int): Assertion `Reg && "this is not a register!"' failed. llc((anonymous namespace)::PrintStackTrace()+0x1a)[0x886f136] llc((anonymous namespace)::SignalHandler(int)+0x110)[0x886f45e] [0x346420] /lib/libc.so.6(abort+0x101)[0x15e4f1] /lib/libc.so.6(__assert_fail+0xfd)[0x156859] llc(llvm::MRegisterInfo::isVirtualRegister(unsigned int)+0x30)[0x84e838a] llc(llvm::VirtRegMap::getStackSlot(unsigned int) const+0x11)[0x86e40eb] llc(llvm::VirtRegMap::hasStackSlot(unsigned int) const+0x18)[0x86e4146] llc[0x86e2083] llc[0x86e9396] llc[0x86cee7d] llc(llvm::MachineFunctionPass::runOnFunction(llvm::Function&)+0x28)[0x83edb7c] llc(llvm::FunctionPassManagerT::runPass(llvm::FunctionPass*, llvm::Function*)+0x1f)[0x87f25e3] llc(llvm::PassManagerT::runPasses(llvm::Function*, std::map >, std::less, std::allocator > > > >&)+0x12a)[0x8803960] llc(llvm::PassManagerT::runOnUnit(llvm::Function*)+0x184)[0x8806114] llc(llvm::FunctionPassManagerT::runOnFunction(llvm::Function&)+0x23)[0x87f26c5] llc(llvm::FunctionPassManager::run(llvm::Function&)+0x91)[0x87f0abb] llc(main+0x925)[0x83dcfbd] /lib/libc.so.6(__libc_start_main+0xdc)[0x14a4e4] llc(__gxx_personality_v0+0x16d)[0x83dbd61] 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 Wed Dec 6 22:04:22 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Wed, 6 Dec 2006 22:04:22 -0600 Subject: [LLVMbugs] [Bug 1036] Weird Failure: test/Regression/CodeGen/Generic/print-arith-int.ll Message-ID: <200612070404.kB744Mc3009053@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1036 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |DUPLICATE ------- Additional Comments From rspencer at x10sys.com 2006-12-06 22:04 ------- *** This bug has been marked as a duplicate of 1037 *** ------- 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 Dec 7 10:22:14 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Thu, 7 Dec 2006 10:22:14 -0600 Subject: [LLVMbugs] [Bug 1037] Weird Failure: test/Regression/CodeGen/Generic/print-arith-int.ll Message-ID: <200612071622.kB7GMEXd031452@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1037 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From rspencer at x10sys.com 2006-12-07 10:22 ------- It turns out that this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/040942.html was the cause of the problem. I've reverted it. Evan: please test your patches with at least llvm/test before committing. ------- 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 Dec 7 23:18:34 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Thu, 7 Dec 2006 23:18:34 -0600 Subject: [LLVMbugs] [Bug 1038] NEW: Linker cannot link external weak symbols Message-ID: <200612080518.kB85IYPM013635@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1038 Summary: Linker cannot link external weak symbols Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Linker AssignedTo: unassignedbugs at nondot.org ReportedBy: alenhar2 at uiuc.edu The following file cannot link with itself though gccld: %kallsyms_names = extern_weak global [0 x ubyte] implementation ------- 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 Dec 9 13:31:50 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 9 Dec 2006 13:31:50 -0600 Subject: [LLVMbugs] [Bug 1039] NEW: timing info printed from static dtor Message-ID: <200612091931.kB9JVoJR010508@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1039 Summary: timing info printed from static dtor Product: libraries Version: trunk Platform: Macintosh OS/Version: MacOS X Status: NEW Severity: normal Priority: P2 Component: Support Libraries AssignedTo: unassignedbugs at nondot.org ReportedBy: sabre at nondot.org -time-passes and related output is printed from the TimingInfo static dtor. This breaks --info-output- file, which is (correctly) shut down from llvm_shutdown. I'll move timinginfo to work with llvm_shutdown. -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 Sat Dec 9 22:06:01 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 9 Dec 2006 22:06:01 -0600 Subject: [LLVMbugs] [Bug 799] Should use llvm-config to build the llvm tools Message-ID: <200612100406.kBA461qk019136@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=799 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED Target Milestone|--- |1.9 ------- Additional Comments From rspencer at x10sys.com 2006-12-09 22:06 ------- This was fixed before 1.9 was released. ------- 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 Dec 9 22:10:24 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 9 Dec 2006 22:10:24 -0600 Subject: [LLVMbugs] [Bug 739] Can't `make install prefix=/tmp/llvm'. DESTDIR not supported either. Message-ID: <200612100410.kBA4AOdP019229@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=739 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED Target Milestone|--- |2.0 ------- Additional Comments From rspencer at x10sys.com 2006-12-09 22:10 ------- Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/041149.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 Sat Dec 9 22:22:13 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 9 Dec 2006 22:22:13 -0600 Subject: [LLVMbugs] [Bug 1040] NEW: Using llvm-config creates useless rebuilds Message-ID: <200612100422.kBA4MDAp019462@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1040 Summary: Using llvm-config creates useless rebuilds Product: Build scripts Version: 1.6 Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: Makefiles AssignedTo: unassignedbugs at nondot.org ReportedBy: rspencer at x10sys.com Build times have increased after the tools were changed to use llvm-config to generate the list of libraries to link with. Here's the scenario: 1. llvm-config depends on all libraries so a change to any one of them results in llvm-config being rebuilt. 2. All tools depend on llvm-config so any change to it results in all tools being rebuilt. For example, consider a change to lib/Transforms/Scalar. It should only affect opt, gccas, gccld, and maybe llc and lli. However, when you rebuild, all tools are relinked. There's no reason that llvm-dis, llvm-as, etc. need to be relinked. Previously the makefiles did not re-link these tools for which a dependency was not changed. Its unclear what the correct solution is. We need an "existence" dependency for the tools on llvm-config. ------- 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 Dec 10 01:42:28 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 10 Dec 2006 01:42:28 -0600 Subject: [LLVMbugs] [Bug 1039] timing info printed from static dtor Message-ID: <200612100742.kBA7gSb2022698@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1039 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From sabre at nondot.org 2006-12-10 01:42 ------- Fixed: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/041151.html This should fix the gccas failures on the nightly tester. -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 Sun Dec 10 08:00:34 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 10 Dec 2006 08:00:34 -0600 Subject: [LLVMbugs] [Bug 739] Can't `make install prefix=/tmp/llvm'. DESTDIR not supported either. Message-ID: <200612101400.kBAE0YH5030469@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=739 ralph at inputplus.co.uk changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED | ------- Additional Comments From ralph at inputplus.co.uk 2006-12-10 08:00 ------- The patch doesn't really fix the bug. It defines DESTDIR in the makefile, something the GNU standards I linked to on opening the bug explicitly rule out: "You should not set the value of DESTDIR in your Makefile at all". And there still isn't a variable called `prefix' that can be overridden on the command line. So the patch has implemented its own non-standard scheme which will just cause confusion in the long run for packagers. May as well get it right now, if at all. ------- 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 Dec 10 12:13:58 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 10 Dec 2006 12:13:58 -0600 Subject: [LLVMbugs] [Bug 1039] timing info printed from static dtor Message-ID: <200612101813.kBAIDwkK001896@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1039 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED | ------- Additional Comments From rspencer at x10sys.com 2006-12-10 12:13 ------- The patch resolves gccas and other compile timings. But, now "JIT Codegen" is not getting picked up. All tests reported that JIT codegen failed. JIT codegen succeeded before your change. ------- 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 Dec 10 13:02:30 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 10 Dec 2006 13:02:30 -0600 Subject: [LLVMbugs] [Bug 1039] timing info printed from static dtor Message-ID: <200612101902.kBAJ2UdV002685@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1039 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |RESOLVED Resolution| |FIXED ------- Additional Comments From sabre at nondot.org 2006-12-10 13:02 ------- Fixed: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061204/041154.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 Sun Dec 10 18:57:14 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 10 Dec 2006 18:57:14 -0600 Subject: [LLVMbugs] [Bug 1041] NEW: Attached bytecode causes bugpoint to crash Message-ID: <200612110057.kBB0vElM009014@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1041 Summary: Attached bytecode causes bugpoint to crash Product: tools Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: bugpoint AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru Running ./bugpoint -run-llc test.bc results to: Checking instruction 'cv2237': Checking instruction 'cv2133': Checking instruction 'cv2030': Checking instruction 'mphi1567': bugpoint: /home/asl/proj/llvm/src/lib/VMCore/Value.cpp:59: virtual llvm::Value::~Value(): Assertion `use_begin() == use_end() && "Uses remain when a value is destroyed!"' failed. Checking instruction 'mcmp1568': ./bugpoint((anonymous namespace)::PrintStackTrace()+0x1f)[0x844a7df] /lib/libc.so.6(abort+0xeb)[0xb7ce1133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7cd94f3] ./bugpoint(llvm::Value::~Value()+0x294)[0x8406f24] [0x8d9ea18] ------- 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 Dec 10 19:03:04 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 10 Dec 2006 19:03:04 -0600 Subject: [LLVMbugs] [Bug 1042] NEW: Assertion failed in LiveVariables.cpp:157 Message-ID: <200612110103.kBB1345o009169@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1042 Summary: Assertion failed in LiveVariables.cpp:157 Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Common Code Generator Code AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru Running ./llc test.bc results to (bytecode passes verifier): llc: /home/asl/proj/llvm/src/lib/CodeGen/LiveVariables.cpp:157: void llvm::LiveVariables::HandleVirtRegUse(llvm::LiveVariables::VarInfo&, llvm::MachineBasicBlock*, llvm::MachineInstr*): Assertion `VRInfo.DefInst && "Register use before def!"' failed. ./llc((anonymous namespace)::PrintStackTrace()+0x1f)[0x859b96f] /lib/libc.so.6(abort+0xeb)[0xb7d83133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7d7b4f3] ./llc(llvm::LiveVariables::HandleVirtRegUse(llvm::LiveVariables::VarInfo&, llvm::MachineBasicBlock*, llvm::MachineInstr*)+0x13c)[0x842bbec] ------- 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 Dec 11 02:43:07 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 11 Dec 2006 02:43:07 -0600 Subject: [LLVMbugs] [Bug 1043] NEW: Core LLVM Extensions Message-ID: <200612110843.kBB8h7g4025533@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1043 Summary: Core LLVM Extensions Product: libraries Version: trunk Platform: All OS/Version: All Status: NEW Severity: enhancement Priority: P2 Component: Core LLVM classes AssignedTo: zhousheng00 at gmail.com ReportedBy: zhousheng00 at gmail.com CC: rspencer at x10sys.com,sabre at nondot.org This PR is a follow-up of pr950. It extends the LLVM core libraries to accept Bit Accurate Types (BATs) into the IR. The BAT is a extension to integral types. It supports not only traditional integral types like BoolTy, ByteTy, IntTy, and ShortTy, LongTy (these common cases will be renamed into Int1Ty, Int8Ty like), but also non-byte-width BATs like Int13. The LLVM core libraries will include BATs without modifying the existing types except rename. To do this, a subclass of Type, saying IntType will be added into LLVM to represent BATs. Bytecode Reader and LLVM Assembly parser will also be updated to handle BATs. Since this is a fairly extensive change to LLVM, this bug will be used to trackthe progress of the work as it proceeds. ------- 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 Dec 11 14:27:39 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 11 Dec 2006 14:27:39 -0600 Subject: [LLVMbugs] [Bug 1044] NEW: LLVM Streams do not support stream manipulators Message-ID: <200612112027.kBBKRdU3017612@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1044 Summary: LLVM Streams do not support stream manipulators Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Support Libraries AssignedTo: unassignedbugs at nondot.org ReportedBy: criswell at uiuc.edu The LLVM stream classes do not support stream manipulators like std::hex and std::dec. This causes compilation failures in at least one source file and may lead to other compiliation problems down the road. A current work around is to get the stream from the LLVM stream using the stream() method and then to use this std::[o|i] stream directly, but this seems like a hack. I did this in llvm/lib/CodeGen/SelectionDAG/ScheduleDAGSimple.cpp. ------- 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 Dec 11 18:44:47 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 11 Dec 2006 18:44:47 -0600 Subject: [LLVMbugs] [Bug 1040] Using llvm-config creates useless rebuilds Message-ID: <200612120044.kBC0ilQK024530@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1040 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED ------- Additional Comments From rspencer at x10sys.com 2006-12-11 18:44 ------- Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041220.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 Mon Dec 11 21:07:45 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 11 Dec 2006 21:07:45 -0600 Subject: [LLVMbugs] [Bug 1045] NEW: ScalarReplAggregates.cpp:474 assert tripped Message-ID: <200612120307.kBC37jlZ027163@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1045 Summary: ScalarReplAggregates.cpp:474 assert tripped Product: tools Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: gccld AssignedTo: unassignedbugs at nondot.org ReportedBy: scottm at aero.org ScalarReplAggregates.cpp:474 assert tripped while compiling the gcc4 frontend. Also occurs when running gccld directly on the resulting bytecode (with -O2 -g removed and -emit-llvm inserted on the compiler's command line): [compiler ugliness, started in gdb] Starting program: /work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/gcc/xgcc -O2 -g -shared-libgcc -B/work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/gcc/ -nostdinc++ -L/work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/i686-pc-linux-gnu/libstdc++-v3/src -L/work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/i686-pc-linux-gnu/libstdc++-v3/src/.libs -B/work/scottm/llvm-cfrontend/i686-unknown-linux-gnu/i686-pc-linux-gnu/bin/ -B/work/scottm/llvm-cfrontend/i686-unknown-linux-gnu/i686-pc-linux-gnu/lib/ -isystem /work/scottm/llvm-cfrontend/i686-unknown-linux-gnu/i686-pc-linux-gnu/include -isystem /work/scottm/llvm-cfrontend/i686-unknown-linux-gnu/i686-pc-linux-gnu/sys-include -I/work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/i686-pc-linux-gnu/libstdc++-v3/include/i686-pc-linux-gnu -I/work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/i686-pc-linux-gnu/libstdc++-v3/include -I/work/scottm/llvm-cfrontend/libstdc++-v3/libsupc++ -D_GNU_SOURCE -fno-implicit-templates -Wall -Wextra -Wwrite-strings -Wcast-qual -fdiagnostics-show-location=once -ffunction-sections -fdata-sections -c ../../../../../libstdc++-v3/src/bitmap_allocator.cc -o bitmap_allocator.o [Thread debugging using libthread_db enabled] [New Thread -1479231808 (LWP 20699)] WARNING: EH not supported yet! cc1plus: /work/scottm/llvm/lib/Transforms/Scalar/ScalarReplAggregates.cpp:474: bool MergeInType(const llvm::Type*, const llvm::Type*&, const llvm::TargetData&): Assertion `In->isIntegral() && "Unknown FP type!"' failed. /work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/i686-pc-linux-gnu/libstdc++-v3/include/ext/bitmap_allocator.h: In member function '_Tp* __gnu_cxx::bitmap_allocator<_Tp>::_M_allocate_single_object() [with _Tp = wchar_t]': /work/scottm/llvm-cfrontend/obj/i686-unknown-linux-gnu/i686-pc-linux-gnu/libstdc++-v3/include/ext/bitmap_allocator.h:998: internal compiler error: Aborted Please submit a full bug report, with preprocessed source if appropriate. See for instructions. [gccld ugliness] (gemsbok) src gccld bitmap_allocator.bc gccld: /work/scottm/llvm/lib/Transforms/Scalar/ScalarReplAggregates.cpp:474: bool MergeInType(const llvm::Type*, const llvm::Type*&, const llvm::TargetData&): Assertion `In->isIntegral() && "Unknown FP type!"' failed. gccld((anonymous namespace)::PrintStackTrace()+0x1a)[0x827732a] /lib/tls/i686/cmov/libc.so.6(abort+0x109)[0xa7cbefb9] /lib/tls/i686/cmov/libc.so.6(__assert_fail+0x10f)[0xa7cb6fbf] gccld[0x8116987] zsh: abort gccld bitmap_allocator.bc ------- 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 Dec 11 22:26:05 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 11 Dec 2006 22:26:05 -0600 Subject: [LLVMbugs] [Bug 1045] ScalarReplAggregates.cpp:474 assert tripped Message-ID: <200612120426.kBC4Q53C028810@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1045 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Component|gccld |Scalar Optimizations Product|tools |libraries Resolution| |FIXED ------- Additional Comments From sabre at nondot.org 2006-12-11 22:26 ------- Fixed, testcase here: Transforms/ScalarRepl/2006-12-11-SROA-Crash.ll Patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041239.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 Tue Dec 12 13:40:09 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 12 Dec 2006 13:40:09 -0600 Subject: [LLVMbugs] [Bug 1046] NEW: Cannot promote operator Message-ID: <200612121940.kBCJe9Us003723@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1046 Summary: Cannot promote operator Product: tools Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: lli AssignedTo: unassignedbugs at nondot.org ReportedBy: scottm at aero.org Fairly simple matrix multiplication code; lli chokes with: lli: /work/scottm/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp:2987: llvm::SDOperand::SelectionDAGLegalize::PromoteOp(llvm::SDOperand): Assertion `0 && "Do not know how to promote this operator!"' failed. bytecode file 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 Tue Dec 12 17:07:34 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 12 Dec 2006 17:07:34 -0600 Subject: [LLVMbugs] [Bug 1047] NEW: String litteral being optimized away to nothing Message-ID: <200612122307.kBCN7YkX007534@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1047 Summary: String litteral being optimized away to nothing Product: libraries Version: 1.9 Platform: Macintosh URL: http://gendou.com/crap/temp/llvm-bug.zip OS/Version: MacOS X Status: NEW Severity: normal Priority: P2 Component: Backend: X86 AssignedTo: unassignedbugs at nondot.org ReportedBy: jww132 at hotmail.com To reproduce the bug: 1. unzip llvm-bug.zip 2. cd into the directory 3. run make you will see the diff of hello1.ll and hello2.ll hello1.c and hello2.c are similar hello world examples using our custom libc the makefile should compile them without using the default system libraries notice that the string literal "Hello World!\n" is not an argument to printf in hello1. instead, int 0 is the only argument. it looks like it has been optimized to nothing. rather than being a bug in my library, i believe this is an optimizer problem, because you can edit the Makefile, removing the -O3 optimization argument, and the problem seems to disappear. this is my first bugzilla report, pardon me if my n00bness is apparent! :P 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 Dec 12 22:33:32 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 12 Dec 2006 22:33:32 -0600 Subject: [LLVMbugs] [Bug 1047] attempt to define llvm intrinsic not rejected by verifier Message-ID: <200612130433.kBD4XWlx014680@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1047 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From sabre at nondot.org 2006-12-12 22:33 ------- Fixed, patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041334.html Testcase here: Verifier/2006-12-12-IntrinsicDefine.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 Thu Dec 14 17:16:32 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Thu, 14 Dec 2006 17:16:32 -0600 Subject: [LLVMbugs] [Bug 1048] NEW: Invalid operand # of TargetInstrInfo Message-ID: <200612142316.kBENGWIG018339@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1048 Summary: Invalid operand # of TargetInstrInfo Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Common Code Generator Code AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru This is first "bug" output from my Qt-compiling series. Running llc produce: llc: /home/asl/proj/llvm/src/include/llvm/Target/TargetInstrInfo.h:135: int llvm::TargetInstrDescriptor::getOperandConstraint(unsigned int, llvm::TOI::OperandConstraint) const: Assertion `OpNum < numOperands && "Invalid operand # of TargetInstrInfo"' failed. ./llc((anonymous namespace)::PrintStackTrace()+0x1f)[0x85a7f6f] /lib/libc.so.6(abort+0xeb)[0xb7cd8133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7cd04f3] ./llc((anonymous namespace)::LocalSpiller::RewriteMBB(llvm::MachineBasicBlock&, llvm::VirtRegMap&)+0x1d5c)[0x846abcc] ------- 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 Dec 15 02:09:00 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 02:09:00 -0600 Subject: [LLVMbugs] [Bug 1048] Invalid operand # of TargetInstrInfo Message-ID: <200612150809.kBF890fa028524@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1048 evan.cheng at apple.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From evan.cheng at apple.com 2006-12-15 02:08 ------- Fixed: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041440.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 Fri Dec 15 10:25:46 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 10:25:46 -0600 Subject: [LLVMbugs] [Bug 1049] NEW: Assertion `isa(Val) && "cast() argument of incompatible type!"' failed Message-ID: <200612151625.kBFGPk8g011417@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1049 Summary: Assertion `isa(Val) && "cast() argument of incompatible type!"' failed Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Common Code Generator Code AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru Qt-compiling series is continuing... Attached bytecode forces llc to crash with assertion: ./llc bugpoint-reduced-simplified.bc llc: /home/asl/proj/llvm/src/include/llvm/Support/Casting.h:199: typename llvm::cast_retty::ret_type llvm::cast(const Y&) [with X = llvm::ConstantSDNode, Y = llvm::SDOperand]: Assertion `isa(Val) && "cast() argument of incompatible type!"' failed. ./llc((anonymous namespace)::PrintStackTrace()+0x1f)[0x85a8cef] /lib/libc.so.6(abort+0xeb)[0xb7d6d133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7d654f3] ./llc(llvm::ScheduleDAG::EmitNode(llvm::SDNode*, std::map, std::allocator > >&)+0x1079)[0x840dcd9] ------- 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 Dec 15 11:36:53 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 11:36:53 -0600 Subject: [LLVMbugs] [Bug 1038] Linker cannot link external weak symbols Message-ID: <200612151736.kBFHarF9008139@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1038 alenhar2 at uiuc.edu changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED ------- Additional Comments From alenhar2 at uiuc.edu 2006-12-15 11:36 ------- fixed. patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041454.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 Fri Dec 15 13:15:34 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 13:15:34 -0600 Subject: [LLVMbugs] [Bug 1050] NEW: Empty archive crashes linker Message-ID: <200612151915.kBFJFY72017520@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1050 Summary: Empty archive crashes linker Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Linker AssignedTo: unassignedbugs at nondot.org ReportedBy: alenhar2 at uiuc.edu if an archive is empty, the linker (gccld in this case) crashes. It should manage to procede. ------- 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 Dec 15 13:20:23 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 13:20:23 -0600 Subject: [LLVMbugs] [Bug 1051] NEW: Assertion "Register class of operand and regclass of use don't agree!"' failed. Message-ID: <200612151920.kBFJKNKj017633@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1051 Summary: Assertion "Register class of operand and regclass of use don't agree!"' failed. Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Common Code Generator Code AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru And another assertion in Qt series :) Attached bytecode: 1. Produce assertion: llc: /home/asl/proj/llvm/src/lib/CodeGen/SelectionDAG/ScheduleDAG.cpp:367: void llvm::ScheduleDAG::AddOperand(llvm::MachineInstr*, llvm::SDOperand, unsigned int, const llvm::TargetInstrDescriptor*, std::map, std::allocator > >&): Assertion `RegMap->getRegClass(VReg) == RC && "Register class of operand and regclass of use don't agree!"' failed. ./llc((anonymous namespace)::PrintStackTrace()+0x1f)[0x85a8cef] /lib/libc.so.6(abort+0xeb)[0xb7d03133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7cfb4f3] ./llc(llvm::ScheduleDAG::AddOperand(llvm::MachineInstr*, llvm::SDOperand, unsigned int, llvm::TargetInstrDescriptor const*, std::map, std::allocator > >&)+0x6b9)[0x840c479] 2. Bugpoint just hangs at some step reducing it. -timeout don't work... ------- 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 Dec 15 13:45:53 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 13:45:53 -0600 Subject: [LLVMbugs] [Bug 1050] Empty archive crashes linker Message-ID: <200612151945.kBFJjrWL018315@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1050 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From rspencer at x10sys.com 2006-12-15 13:45 ------- Several asserts in the Archive Library should have been producing error messages but weren't. This is now fixed with these patches: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041458.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 Fri Dec 15 16:15:28 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 16:15:28 -0600 Subject: [LLVMbugs] [Bug 1051] Assertion "Register class of operand and regclass of use don't agree!"' failed. Message-ID: <200612152215.kBFMFSWq011801@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1051 asl at math.spbu.ru changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |DUPLICATE ------- Additional Comments From asl at math.spbu.ru 2006-12-15 16:15 ------- Lack of inline FP asm support can lead to really funny results. *** This bug has been marked as a duplicate of 879 *** ------- 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 Dec 15 18:26:06 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 18:26:06 -0600 Subject: [LLVMbugs] [Bug 1047] attempt to define llvm intrinsic not rejected by verifier Message-ID: <200612160026.kBG0Q6NJ014213@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1047 jww132 at hotmail.com changed: What |Removed |Added ---------------------------------------------------------------------------- URL|http://gendou.com/crap/temp/|http://gendou.com/crap/temp/ |llvm-bug.zip |llvm-bug2.zip Status|RESOLVED |REOPENED Resolution|FIXED | ------- Additional Comments From jww132 at hotmail.com 2006-12-15 18:26 ------- I have made the two changes you mentioned. I got rid of llvm.o which re-defined intrinsics. Shame on me for doing that in the first place! Also, __stdio_puts() now has a thrid parameter. However, the problem still exists. >> tail call void (int, ...)* %printf( int 0 ) This problem is only apparent when using optimizer passes. Am I still doing something stupid? Should I open a new bug report and let you close this one? I uploaded llvm-bug2.zip (see bug URL) so you can see it for yourself. ------- 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 Dec 15 18:39:55 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 18:39:55 -0600 Subject: [LLVMbugs] [Bug 1047] attempt to define llvm intrinsic not rejected by verifier Message-ID: <200612160039.kBG0dtjL014419@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1047 jww132 at hotmail.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |RESOLVED Resolution| |FIXED ------- Additional Comments From jww132 at hotmail.com 2006-12-15 18:39 ------- Ok, I just talked to nicholas, and he realized that the string literal is being accessed directly in printf() instead of being passed as an argument. So, I was chasing a red haring. Sorry!! I set the bug back to 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 Dec 15 18:44:46 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 18:44:46 -0600 Subject: [LLVMbugs] [Bug 1052] NEW: missed optimization opportunity: var-args Message-ID: <200612160044.kBG0ik9O014516@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1052 Summary: missed optimization opportunity: var-args Product: libraries Version: 1.9 Platform: Macintosh URL: http://llvm.org/bugs/show_bug.cgi?id=1047 OS/Version: MacOS X Status: NEW Severity: enhancement Priority: P2 Component: Loop Optimizer AssignedTo: unassignedbugs at nondot.org ReportedBy: jww132 at hotmail.com Nicholas advised I file this bug I noticed while working on bug1047. Apparently, when the optimizer in-lined vprintf() into printf(), it left the parameter in there. However, this parameter is not being used. Rather, the [14 x sbyte] is loaded directly inside the function. This argument can be dropped. Cheers! ------- 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 Dec 15 20:28:07 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 15 Dec 2006 20:28:07 -0600 Subject: [LLVMbugs] [Bug 1042] Verifier doesn't catch invalid invoke use Message-ID: <200612160228.kBG2S73n016401@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1042 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Component|Common Code Generator Code |Core LLVM classes Resolution| |FIXED Summary|Assertion failed in |Verifier doesn't catch |LiveVariables.cpp:157 |invalid invoke use Target Milestone|--- |2.0 Version|trunk |1.0 ------- Additional Comments From sabre at nondot.org 2006-12-15 20:28 ------- This is actually a verifier bug not rejecting broken code. Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041486.html Testcases here: test/Regression/Verifier/invoke-[12].ll Note that the broken LLVM code was made by some transformation that is buggy. That means that if you try to compile the source that produced this again, you'll get an assert. Please file another bug so we can fix that. 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 Sat Dec 16 11:57:57 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 11:57:57 -0600 Subject: [LLVMbugs] [Bug 1053] NEW: Processor autodetection incorrect Message-ID: <200612161757.kBGHvvrG011517@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1053 Summary: Processor autodetection incorrect Product: tools Version: trunk Platform: Macintosh OS/Version: MacOS X Status: NEW Severity: normal Priority: P2 Component: TableGen AssignedTo: unassignedbugs at nondot.org ReportedBy: scottm at aero.org This bug initially manifests itself while building the c frontend, but it isn't really a cfe bug (sorta). The CFE sets the rs6000_cpu variable to PROCESSOR_DEFAULT, which turns out to be PROCESSOR_POWER4. llvm::PPCSubtarget::ParseSubtargetFeatures eventually gets a string that represents the default processor architecture, which starts with "-970", corresponding to PROCESSOR_POWER4. This effectively nails the processor architecture down to always being a G5/970. tblgen emits code that will only set the specific CPU iff the CPU isn't set yet (call setCPUIfNone()) Meanwhile, llvm has correctly identified the host CPU, which is a G4 in my case. Because setCPUIfNone() is called and the default features string specified a 970, the processor is de-facto set to a G5 irrespective of the actual, detected processor. This patch is an attempt to fix/remedy the bug (but it might not be entirely correct): Index: utils/TableGen/SubtargetEmitter.cpp =================================================================== RCS file: /var/cvs/llvm/llvm/utils/TableGen/SubtargetEmitter.cpp,v retrieving revision 1.24 diff -u -r1.24 SubtargetEmitter.cpp --- utils/TableGen/SubtargetEmitter.cpp 12 Dec 2006 20:55:58 -0000 1.24 +++ utils/TableGen/SubtargetEmitter.cpp 16 Dec 2006 17:50:10 -0000 @@ -461,7 +461,8 @@ OS << "Subtarget::ParseSubtargetFeatures(const std::string &FS,\n" " const std::string &CPU) {\n" " SubtargetFeatures Features(FS);\n" - " Features.setCPUIfNone(CPU);\n" + " if (CPU != \"\");\n" + " Features.setCPU(CPU);\n" " uint32_t Bits = Features.getBits(SubTypeKV, SubTypeKVSize,\n" " FeatureKV, FeatureKVSize);\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 Sat Dec 16 12:03:42 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 12:03:42 -0600 Subject: [LLVMbugs] [Bug 1054] NEW: Enums attempt to redefine manifest constants Message-ID: <200612161803.kBGI3gwp011660@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1054 Summary: Enums attempt to redefine manifest constants Product: libraries Version: trunk Platform: Macintosh OS/Version: MacOS X Status: NEW Severity: normal Priority: P2 Component: Backend: PowerPC AssignedTo: unassignedbugs at nondot.org ReportedBy: scottm at aero.org Shouldn't try to redefine manifest constants/#defines that system header files might be providing. See patch. ------- 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 Dec 16 14:25:08 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 14:25:08 -0600 Subject: [LLVMbugs] [Bug 1054] Enums attempt to redefine manifest constants in macho writer Message-ID: <200612162025.kBGKP87e014204@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1054 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Component|Backend: PowerPC |Common Code Generator Code Keywords| |build-problem Resolution| |FIXED Summary|Enums attempt to redefine |Enums attempt to redefine |manifest constants |manifest constants in macho | |writer Target Milestone|--- |2.0 Version|trunk |1.9 ------- Additional Comments From sabre at nondot.org 2006-12-16 14:25 ------- Side note: please don't put tags like "(scottm)" into the patches. Patches applied: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041494.html http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041493.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 Sat Dec 16 14:38:41 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 14:38:41 -0600 Subject: [LLVMbugs] [Bug 1055] NEW: LLVMlto library does not build on Linux (amd64) Message-ID: <200612162038.kBGKcfka014477@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1055 Summary: LLVMlto library does not build on Linux (amd64) Product: Build scripts Version: cvs Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Makefiles AssignedTo: unassignedbugs at nondot.org ReportedBy: chandlerc at gmail.com The library LLVMlto is only built on Darwin. Commenting out lines 26 and 28 of tools/Makefile, to add lto to the build, results in the following error: ... make[2]: Entering directory `/home/chandlerc/code/compilers/build/llvm/tools/lto' llvm[2]: Linking Debug Loadable Module LLVMlto.so *** Warning: Linking the shared library /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMlto.la against the non-libtool *** objects /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMARM.o /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMIA64.o /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMAlpha.o /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMPowerPC.o /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMSparc.o /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMX86.o /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMCBackend.o is not portable! /usr/lib/gcc/x86_64-pc-linux-gnu/4.1.1/../../../../x86_64-pc-linux-gnu/bin/ld: /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMARM.o: relocation R_X86_64_32 against `llvm::SupportDwarfLinkVar' can not be used when making a shared object; recompile with -fPIC /home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMARM.o: could not read symbols: Bad value collect2: ld returned 1 exit status make[2]: *** [/home/chandlerc/code/compilers/build/llvm/Debug/lib/LLVMlto.la] Error 1 make[2]: Leaving directory `/home/chandlerc/code/compilers/build/llvm/tools/lto' ... The same error is generated using './configure --with-pic' and without. I am running Gentoo Linux, amd64 architecture, gcc (GCC) 4.1.1 (Gentoo 4.1.1-r3) . -Chandler ------- 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 Dec 16 15:17:25 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 15:17:25 -0600 Subject: [LLVMbugs] [Bug 1049] crash compiling inline asm Message-ID: <200612162117.kBGLHPj9015214@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1049 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Keywords| |compile-fail Resolution| |FIXED Summary|Assertion `isa(Val) && |crash compiling inline asm |"cast() argument of | |incompatible type!"' failed | Target Milestone|--- |2.0 Version|trunk |1.8 ------- Additional Comments From sabre at nondot.org 2006-12-16 15:17 ------- Fixed. This was a subtle nasty bug with a simple fix. Patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041497.html Testcase here: CodeGen/Generic/2006-12-16-InlineAsmCrash.ll 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 Sat Dec 16 15:20:05 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 15:20:05 -0600 Subject: [LLVMbugs] [Bug 1052] missed optimization opportunity: var-args Message-ID: <200612162120.kBGLK57F015294@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1052 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |WONTFIX ------- Additional Comments From sabre at nondot.org 2006-12-16 15:20 ------- You're right, but we can't change this. The compiler actually explicitly has code to support this behavior, because the CBE cannot support varargs functions with no fixed parameters. I will fix the code to pass undef instead of 'int 0', which will reduce the cost of this. -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 Sat Dec 16 15:38:47 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 15:38:47 -0600 Subject: [LLVMbugs] [Bug 1056] NEW: Crash building llvm-gcc on FreeBSD x86_64 Message-ID: <200612162138.kBGLclZM015692@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1056 Summary: Crash building llvm-gcc on FreeBSD x86_64 Product: libraries Version: trunk Platform: PC OS/Version: FreeBSD Status: NEW Severity: normal Priority: P2 Component: Interprocedural Optimizations AssignedTo: unassignedbugs at nondot.org ReportedBy: jeffc at jolt-lang.org The last file compiled during the build of llvm-gcc when it crashes: /usr/home/jeffc/llvm-gcc/obj/gcc/xgcc -B/usr/home/jeffc/llvm-gcc/obj/gcc/ -B/home/jeffc/llvm-gcc/install/amd64-unknown-freebsd6.1/bin/ -B/home/jeffc/llvm-gcc/install/amd64-unknown-freebsd6.1/lib/ -isystem /home/jeffc/llvm-gcc/install/amd64-unknown-freebsd6.1/include -isystem /home/jeffc/llvm-gcc/install/amd64-unknown-freebsd6.1/sys-include -O2 -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition -isystem ./include -fPIC -pthread -g -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2 -D__GCC_FLOAT_NOT_NEEDED -I. -I. -I../../gcc -I../../gcc/. -I../../gcc/../include -I./../intl -I../../gcc/../libcpp/include -I/usr/home/jeffc/llvm/include -I/home/jeffc/llvm/obj/include -DL_lshrdi3 -c ../../gcc/libgcc2.c -o libgcc/./_lshrdi3.o WARNING: 128-bit integers not supported! ../../gcc/libgcc2.c: In function '__lshrti3': ../../gcc/libgcc2.c:412: internal compiler error: Segmentation fault: 11 This is apparently due to an optimization bug in LLVM. The crash goes away if the -O2 option is removed. The preprocessed version of the file, as well as the bytecode file produced with -emit-llvm -O0 is 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 Sat Dec 16 19:53:15 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 16 Dec 2006 19:53:15 -0600 Subject: [LLVMbugs] [Bug 1057] NEW: patch for Mac OS X 10.3 Message-ID: <200612170153.kBH1rFqL019930@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1057 Summary: patch for Mac OS X 10.3 Product: libraries Version: trunk Platform: Macintosh OS/Version: MacOS X Status: NEW Severity: normal Priority: P2 Component: Target-Independent JIT AssignedTo: unassignedbugs at nondot.org ReportedBy: scottm at aero.org Following patch required to successfully compile (and link) lli - problem traced to JIT.cpp: Index: lib/ExecutionEngine/JIT/JIT.cpp =================================================================== RCS file: /var/cvs/llvm/llvm/lib/ExecutionEngine/JIT/JIT.cpp,v retrieving revision 1.83 diff -u -r1.83 JIT.cpp --- lib/ExecutionEngine/JIT/JIT.cpp 12 Dec 2006 01:17:41 -0000 1.83 +++ lib/ExecutionEngine/JIT/JIT.cpp 17 Dec 2006 01:51:08 -0000 @@ -31,9 +31,10 @@ #ifdef __APPLE__ #include -#if (MAC_OS_X_VERSION_MIN_REQUIRED > MAC_OS_X_VERSION_10_4) || \ - (MAC_OS_X_VERSION_MIN_REQUIRED == MAC_OS_X_VERSION_10_4 && \ - __APPLE_CC__ >= 5330) +#if defined(MAC_OS_X_VERSION_10_4) && \ + ((MAC_OS_X_VERSION_MIN_REQUIRED > MAC_OS_X_VERSION_10_4) || \ + (MAC_OS_X_VERSION_MIN_REQUIRED == MAC_OS_X_VERSION_10_4 && \ + __APPLE_CC__ >= 5330)) // __dso_handle is resolved by Mac OS X dynamic linker. extern void *__dso_handle __attribute__ ((__visibility__ ("hidden"))); #endif @@ -309,9 +310,10 @@ // If the global is external, just remember the address. if (GV->isExternal()) { #ifdef __APPLE__ -#if (MAC_OS_X_VERSION_MIN_REQUIRED > MAC_OS_X_VERSION_10_4) || \ - (MAC_OS_X_VERSION_MIN_REQUIRED == MAC_OS_X_VERSION_10_4 && \ - __APPLE_CC__ >= 5330) +#if defined(MAC_OS_X_VERSION_10_4) && \ + ((MAC_OS_X_VERSION_MIN_REQUIRED > MAC_OS_X_VERSION_10_4) || \ + (MAC_OS_X_VERSION_MIN_REQUIRED == MAC_OS_X_VERSION_10_4 && \ + __APPLE_CC__ >= 5330)) // Apple gcc defaults to -fuse-cxa-atexit (i.e. calls __cxa_atexit instead // of atexit). It passes the address of linker generated symbol __dso_handle // to the function. ------- 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 Dec 17 00:55:27 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 17 Dec 2006 00:55:27 -0600 Subject: [LLVMbugs] [Bug 1058] NEW: CBE Not Generating BITCAST_TEMPORARY Declarations Message-ID: <200612170655.kBH6tRom024870@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1058 Summary: CBE Not Generating BITCAST_TEMPORARY Declarations Product: libraries Version: trunk Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: Backend: C AssignedTo: rspencer at x10sys.com ReportedBy: rspencer at x10sys.com In some cases, the CBE is not generating the BITCAST_TEMPORARY variable needed to perform a fp<->int bitcast. This was originally reported by Evan Cheng. The symptom is compile errors like this: Output/462.libquantum.cbe.c: In function 'quantum_objcode_put': Output/462.libquantum.cbe.c:674: error: 'ltmp_118_4__BITCAST_TEMPORARY' undeclared (first use in this function) ------- 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 Dec 17 12:51:41 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 17 Dec 2006 12:51:41 -0600 Subject: [LLVMbugs] [Bug 1058] CBE Not Generating BITCAST_TEMPORARY Declarations Message-ID: <200612171851.kBHIpfHJ005788@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1058 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED ------- Additional Comments From rspencer at x10sys.com 2006-12-17 12:51 ------- Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041524.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 Sun Dec 17 13:55:11 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 17 Dec 2006 13:55:11 -0600 Subject: [LLVMbugs] [Bug 1041] Attached bytecode causes bugpoint to crash Message-ID: <200612171955.kBHJtBQR006793@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1041 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- BugsThisDependsOn| |1042 Status|NEW |RESOLVED Resolution| |INVALID ------- Additional Comments From sabre at nondot.org 2006-12-17 13:55 ------- No, you're exactly right, the bc file isn't valid. Nice catch! -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 Sun Dec 17 15:04:58 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 17 Dec 2006 15:04:58 -0600 Subject: [LLVMbugs] [Bug 1057] patch for Mac OS X 10.3 Message-ID: <200612172104.kBHL4wSG007954@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1057 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Keywords| |portability Resolution| |FIXED ------- Additional Comments From sabre at nondot.org 2006-12-17 15:04 ------- Patch looks great, thanks! http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061211/041526.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 Dec 18 11:31:58 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 18 Dec 2006 11:31:58 -0600 Subject: [LLVMbugs] [Bug 1046] Cannot promote operator Message-ID: <200612181731.kBIHVwVv008105@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1046 scottm at aero.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED ------- Additional Comments From scottm at aero.org 2006-12-18 11:31 ------- Cannot reproduce with current tree. Ugh! It might have been fixed. Will close out this bug for the time being. ------- 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 Dec 18 15:38:32 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 18 Dec 2006 15:38:32 -0600 Subject: [LLVMbugs] [Bug 1042] Verifier doesn't catch invalid invoke use Message-ID: <200612182138.kBILcWHr012073@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1042 asl at math.spbu.ru changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED | ------- Additional Comments From asl at math.spbu.ru 2006-12-18 15:38 ------- invoke-1.ll is not rejected by verifier. Everything is ok for invoke-2.ll ------- 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 Dec 18 15:57:40 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 18 Dec 2006 15:57:40 -0600 Subject: [LLVMbugs] [Bug 1042] Verifier doesn't catch invalid invoke use Message-ID: <200612182157.kBILveM3012423@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1042 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |RESOLVED Resolution| |FIXED ------- Additional Comments From rspencer at x10sys.com 2006-12-18 15:57 ------- Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061218/041558.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 Mon Dec 18 23:55:44 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Mon, 18 Dec 2006 23:55:44 -0600 Subject: [LLVMbugs] [Bug 1059] NEW: intel-style asm printer for x86 doesn't support jump tables Message-ID: <200612190555.kBJ5ti9q020300@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1059 Summary: intel-style asm printer for x86 doesn't support jump tables Product: libraries Version: trunk Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: Backend: X86 AssignedTo: unassignedbugs at nondot.org ReportedBy: sabre at nondot.org -x86-asm-syntax=intel yields: $BB1_2451: ;bb740 mov EAX, mov ECX, DWORD PTR [ESP + 1372] jmp DWORD PTR [EAX + 4*ECX] AT&T yields: LBB1_2451: #bb740 movl $LJTI1_3, %eax movl 1372(%esp), %ecx jmpl *(%eax,%ecx,4) ------- 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 Dec 19 00:54:50 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 00:54:50 -0600 Subject: [LLVMbugs] [Bug 1004] Problem with CodeExtractor.cpp Message-ID: <200612190654.kBJ6soWn021224@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1004 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |INVALID ------- Additional Comments From sabre at nondot.org 2006-12-19 00:54 ------- The problem is that the code extractor expects critical edges to be broken and they are not. -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 Dec 19 01:46:12 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 01:46:12 -0600 Subject: [LLVMbugs] [Bug 1060] NEW: ParseAssemblyString does not allow allow already existing names Message-ID: <200612190746.kBJ7kCup022100@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1060 Summary: ParseAssemblyString does not allow allow already existing names Product: libraries Version: trunk Platform: All OS/Version: All Status: NEW Severity: enhancement Priority: P2 Component: Target-Independent JIT AssignedTo: unassignedbugs at nondot.org ReportedBy: eric at vanrietpaap.nl When working with the JIT it is sometimes necessary to recompile a new implementation of a function. I use ParseAssemblyString to feed new code into the JIT. However, Encountering an already defined function gives a parse error. Could we have another ParseAssemblyString function (or with an additional parameter) that "does the right thing" (tm) so things like getNamedFunction, freeMachineCodeForFunction and recompileAndRelinkFunction can be used with expected results. ------- 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 Dec 19 03:20:09 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 03:20:09 -0600 Subject: [LLVMbugs] [Bug 1061] NEW: X86IntelAsmPrinter is broken/incomplete Message-ID: <200612190920.kBJ9K96b032530@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1061 Summary: X86IntelAsmPrinter is broken/incomplete 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 $ ./llc -f test-j.bc -x86-asm-syntax=intel llc: /home/asl/proj/llvm/src/include/llvm/CodeGen/MachineInstr.h:138: int64_t llvm::MachineOperand::getImmedValue() const: Assertion `isImm() && "Wrong MachineOperand accessor"' failed. ./llc((anonymous namespace)::PrintStackTrace()+0x1f)[0x85a77cf] /lib/libc.so.6(abort+0xeb)[0xb7cf6133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7cee4f3] ./llc(llvm::X86IntelAsmPrinter::printMemReference(llvm::MachineInstr const*, unsigned int, char const*)+0x49c)[0x827468c] Don't know MF machinery yet to investigate/fix :( ------- 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 Dec 19 12:22:53 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 12:22:53 -0600 Subject: [LLVMbugs] [Bug 1062] NEW: address constants forced to registers in small-model static-reloc x86-64 Message-ID: <200612191822.kBJIMrOV009656@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1062 Summary: address constants forced to registers in small-model static-reloc 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 Weak, extern and linkonce symbols are currently excluded from address operands in x86-64 with the small memory model and static linking. With this test case: %x = external global int implementation int %foo(long %i) { %a = getelementptr int* %x, long %i %t = load int* %a ret int %t } Currently llc -march=x86-64 -code-model=small -relocation-model=static emits this: leaq x(%rip), %rcx movl (%rcx,%rax,4), %eax I believe should emit this: movl x(,%rax,4), %eax Unless I'm missing something, it should be safe to let llc do this, with the following patch. Index: X86ISelDAGToDAG.cpp =================================================================== RCS file: /var/cvs/llvm/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp,v retrieving revision 1.138 diff -u -r1.138 X86ISelDAGToDAG.cpp --- X86ISelDAGToDAG.cpp +++ X86ISelDAGToDAG.cpp @@ -604,9 +604,7 @@ SDOperand N0 = N.getOperand(0); if (GlobalAddressSDNode *G = dyn_cast(N0)) { GlobalValue *GV = G->getGlobal(); - bool isAbs32 = !is64Bit || - (isStatic && !(GV->isExternal() || GV->hasWeakLinkage() || - GV->hasLinkOnceLinkage())); + bool isAbs32 = !is64Bit || isStatic; if (isAbs32 || isRoot) { AM.GV = G->getGlobal(); AM.Disp += G->getOffset(); ------- 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 Dec 19 13:30:55 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 13:30:55 -0600 Subject: [LLVMbugs] [Bug 1061] X86IntelAsmPrinter is broken/incomplete Message-ID: <200612191930.kBJJUtAP010798@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1061 sabre at nondot.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Keywords| |compile-fail Resolution| |FIXED Target Milestone|--- |2.0 Version|trunk |1.9 ------- Additional Comments From sabre at nondot.org 2006-12-19 13:30 ------- Fixed. Testcase here: CodeGen/X86/2006-12-19-IntelSyntax.ll Patch here: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061218/041591.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 Dec 19 15:08:49 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 15:08:49 -0600 Subject: [LLVMbugs] [Bug 1059] intel-style asm printer for x86 doesn't support jump tables Message-ID: <200612192108.kBJL8nU1012905@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1059 asl at math.spbu.ru changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |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 Tue Dec 19 15:31:57 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 19 Dec 2006 15:31:57 -0600 Subject: [LLVMbugs] [Bug 1062] address constants forced to registers in small-model static-reloc x86-64 Message-ID: <200612192131.kBJLVvBj013417@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1062 evan.cheng at apple.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From evan.cheng at apple.com 2006-12-19 15:31 ------- 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 Wed Dec 20 16:40:36 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Wed, 20 Dec 2006 16:40:36 -0600 Subject: [LLVMbugs] [Bug 1053] Processor autodetection incorrect Message-ID: <200612202240.kBKMeaSS016883@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1053 jlaskey at apple.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Additional Comments From jlaskey at apple.com 2006-12-20 16:40 ------- The CFE has been modified to provide a more appropriate choice of cpu. Depending on -m64, "ppc" or "ppc64" is chosen by default. "ppc" is equivalent to G3, and "ppc64" is equivalent to power4 (or ppc970 in 64-bit mode.) If "-mcpu=xyz" is selected then "xyz" will be used. if "-maltivec" is specified then the cpu is upgraded to minimally "ppc7400" (G4.) Changes should be on the mirror Dec 21/2006 at 6am PST. -- Jim ------- 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 Dec 20 20:20:59 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Wed, 20 Dec 2006 20:20:59 -0600 Subject: [LLVMbugs] [Bug 1063] NEW: test/Regression/Transforms/InstCombine/2006-04-01-InfLoop fails on amd64 Message-ID: <200612210220.kBL2Kxun020376@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1063 Summary: test/Regression/Transforms/InstCombine/2006-04-01- InfLoop fails on amd64 Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Transformation Utilities AssignedTo: unassignedbugs at nondot.org ReportedBy: chandlerc at gmail.com test/Regression/Transforms/InstCombine/2006-04-01-InfLoop.ll failure on amd64, on LLVM being built with gcc 3.4.6 A snippet of the run output is below, with more detailed output in attachments. -Chandler ------- 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 Dec 22 10:27:07 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Fri, 22 Dec 2006 10:27:07 -0600 Subject: [LLVMbugs] [Bug 1064] NEW: ppc linux doesn't handle float parameter to varargs function Message-ID: <200612221627.kBMGR7Cl008889@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1064 Summary: ppc linux doesn't handle float parameter to varargs function Product: libraries Version: trunk Platform: Macintosh OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Backend: PowerPC AssignedTo: unassignedbugs at nondot.org ReportedBy: nicholas at mxc.ca The Linux ABI requires that you set control register 6 before calling a variadic function with a float parameter. Similarly, it recommends that you clear it before calling a variadic function without one. A simple example is: #include int main(void) { printf("%f\n", 1.2); return 0; } which will print "0.0" on PPC Linux. In the assembly, right before "bl printf", add "creqv 6,6,6" and it will work properly. This fix will require changes to the PPCISelLowering.cpp:LowerCALL function. ------- 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 Dec 23 12:18:07 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 23 Dec 2006 12:18:07 -0600 Subject: [LLVMbugs] [Bug 1065] NEW: Crash in InstCombine Message-ID: <200612231818.kBNII7ie013305@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1065 Summary: Crash in InstCombine Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Scalar Optimizations AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru This is the reduction from libmng & Qt. Everythign was ok some weeks ago. I've tried to rebuild Qt after SETCC-removal patch landed. Steps to reproduce: ~/proj/llvm/build/Release/bin $ ./opt bugpoint-reduced-simplified.bc -instcombine -f opt: /home/asl/proj/llvm/src/include/llvm/Support/Casting.h:199: typename llvm::cast_retty::ret_type llvm::cast(const Y&) [with X = llvm::ShiftInst, Y = llvm::Instruction*]: Assertion `isa(Val) && "cast() argument of incompatible type!"' failed. ./opt((anonymous namespace)::PrintStackTrace()+0x1f)[0x836a96f] /lib/libc.so.6(abort+0xeb)[0xb7d96133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7d8e4f3] ./opt((anonymous namespace)::InstCombiner::FoldSelectOpOp(llvm::SelectInst&, llvm::Instruction*, llvm::Instruction*)+0x963)[0x81843a3] Bytecode 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 Sat Dec 23 12:59:41 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 23 Dec 2006 12:59:41 -0600 Subject: [LLVMbugs] [Bug 1065] Crash in InstCombine Message-ID: <200612231859.kBNIxf55014001@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1065 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Keywords| |crash-on-invalid OS/Version|Linux |All Platform|PC |All Resolution| |FIXED Target Milestone|--- |2.0 ------- Additional Comments From rspencer at x10sys.com 2006-12-23 12:59 ------- Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061218/041759.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 Sat Dec 23 18:07:40 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 23 Dec 2006 18:07:40 -0600 Subject: [LLVMbugs] [Bug 1066] NEW: Assertion failed in InstCombine Message-ID: <200612240007.kBO07eA5018979@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1066 Summary: Assertion failed in InstCombine Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: Scalar Optimizations AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru $ ./opt -instcombine bugpoint-reduced-simplified.bc -o test_bc.bc -f opt: /home/asl/proj/llvm/src/lib/VMCore/Value.cpp:157: void llvm::Value::replaceAllUsesWith(llvm::Value*): Assertion `New->getType() == getType() && "replaceAllUses of value with new value of different type!"' failed. ./opt((anonymous namespace)::PrintStackTrace()+0x1f)[0x836a7ff] /lib/libc.so.6(abort+0xeb)[0xb7d81133] /lib/libc.so.6(__assert_fail+0xeb)[0xb7d794f3] ./opt[0x832f26c] Bytecode 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 Sat Dec 23 18:42:12 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sat, 23 Dec 2006 18:42:12 -0600 Subject: [LLVMbugs] [Bug 1066] Assertion failed in InstCombine Message-ID: <200612240042.kBO0gCbp019518@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1066 rspencer at x10sys.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Keywords| |compile-fail OS/Version|Linux |All Platform|PC |All Resolution| |FIXED Target Milestone|--- |2.0 ------- Additional Comments From rspencer at x10sys.com 2006-12-23 18:42 ------- Fixed with this patch: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20061218/041763.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 Sun Dec 24 06:02:38 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 24 Dec 2006 06:02:38 -0600 Subject: [LLVMbugs] [Bug 1067] NEW: PIC codegen for Linux/X86 Message-ID: <200612241202.kBOC2cJF031573@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1067 Summary: PIC codegen for Linux/X86 Product: libraries Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: enhancement Priority: P2 Component: Backend: X86 AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru We really should support PIC codegen for Linux. Probably, we need to add additional entry to TargetData to distinguish pic & non-pic codegen (propagating -fPIC flag from llvm-gcc). It seems, that X86/Darwin codegen is already PIC (everything there is PIC), so, this code should be inspected. Also, ELF-specific codegen (GOT, PLT) should be introduced. This will allow us to build rather complex applications, e.g. mozilla/seamonkey. ------- 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 Dec 24 14:01:06 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 24 Dec 2006 14:01:06 -0600 Subject: [LLVMbugs] [Bug 1068] NEW: ICE with "-g" option specified Message-ID: <200612242001.kBOK16R5007017@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1068 Summary: ICE with "-g" option specified Product: tools Version: trunk Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: llvm-g++ AssignedTo: unassignedbugs at nondot.org ReportedBy: asl at math.spbu.ru $ llvm-g++ -c -g test.ii test.ii:13: warning: inline function 'QVarLengthArray::QVarLengthArray(int) [with T = char, int Prealloc = 4096]' used but never defined test.ii: In member function 'void QInotifyFileSystemWatcherEngine::readFromInotify()': test.ii:20: internal compiler error: in tree_low_cst, at tree.c:3890 Please submit a full bug report, with preprocessed source if appropriate. See for instructions. Everything is ok without "-g". ------- 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 Dec 26 15:42:33 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Tue, 26 Dec 2006 15:42:33 -0600 Subject: [LLVMbugs] [Bug 1069] NEW: Incorrect Post-Dominance Frontier Calculation Message-ID: <200612262142.kBQLgXwb007261@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1069 Summary: Incorrect Post-Dominance Frontier Calculation Product: libraries Version: trunk Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: Core LLVM classes AssignedTo: unassignedbugs at nondot.org ReportedBy: lefever at crhc.uiuc.edu To the best of my knowledge, I have come across a bug in the post-dominance frontier computation in LLVM. In particular, LLVM computes the following post-dominance frontier for the function at the end of this message: Basic Block Post-Dominance-Frontier ----------- ----------------------- entry - bb - cond_next8 cond_next cond_next12 cond_next cond_next - bb21 cond_next8, cond_next12 cond_true27 cond_next8, cond_next12 pop.exit cond_next8, cond_next12 cond_true.i cond_true27 return ??? return was not even listed in LLVM's PDF. However, I believe the post-dominance frontier should be computed as such: Basic Block Post-Dominance-Frontier ----------- ----------------------- entry - cond_next entry cond_next8 cond_next cond_next12 cond_next8 bb21 cond_next8, cond_next12 cond_true27 bb21 cond_true.i cond_true27 pop.exit bb21 bb cond_next, cond_next12 return entry, bb21 Below is the function. ----------------------------------------- void %load_array(int %var_name) { entry: %tmp = call sbyte %check_stack( int 1 ) ; [#uses=1] %tmp = seteq sbyte %tmp, 0 ; [#uses=1] br bool %tmp, label %return, label %cond_next cond_next: ; preds = %entry %tmp = load %struct.bc_var** %ex_stack ; <%struct.bc_var*> [#uses=1] %tmp = getelementptr %struct.bc_var* %tmp, int 0, uint 0 ; <%struct.bc_struct**> [#uses=1] %tmp = load %struct.bc_struct** %tmp ; <%struct.bc_struct*> [#uses=2] %tmp = call int %bc_num2long( %struct.bc_struct* %tmp ) ; [#uses=4] %tmp2 = setlt int %tmp, 0 ; [#uses=1] %tmp6 = setgt int %tmp, 65535 ; [#uses=1] %bothcond = or bool %tmp2, %tmp6 ; [#uses=1] br bool %bothcond, label %bb, label %cond_next8 cond_next8: ; preds = %cond_next %tmp10 = seteq int %tmp, 0 ; [#uses=1] br bool %tmp10, label %cond_next12, label %bb21 cond_next12: ; preds = %cond_next8 %tmp16 = call sbyte %bc_is_zero( %struct.bc_struct* %tmp ) ; [#uses=1] %tmp17 = seteq sbyte %tmp16, 0 ; [#uses=1] br bool %tmp17, label %bb, label %bb21 bb: ; preds = %cond_next12, %cond_next %tmp = load sbyte*** %a_names ; [#uses=1] %tmp19 = getelementptr sbyte** %tmp, int %var_name ; [#uses=1] %tmp = load sbyte** %tmp19 ; [#uses=1] call void (sbyte*, ...)* %rt_error( sbyte* getelementptr ([34 x sbyte]* %str, int 0, uint 0), sbyte* %tmp ) ret void bb21: ; preds = %cond_next12, %cond_next8 %tmp24 = call %struct.bc_struct** %get_array_num( int %var_name, int %tmp ) ; <%struct.bc_struct**> [#uses=2] %tmp26 = seteq %struct.bc_struct** %tmp24, null ; [#uses=1] br bool %tmp26, label %return, label %cond_true27 cond_true27: ; preds = %bb21 %tmp.i = load %struct.bc_var** %ex_stack ; <%struct.bc_var*> [#uses=4] %tmp.i = seteq %struct.bc_var* %tmp.i, null ; [#uses=1] br bool %tmp.i, label %pop.exit, label %cond_true.i cond_true.i: ; preds = %cond_true27 %tmp.i = getelementptr %struct.bc_var* %tmp.i, int 0, uint 1 ; <%struct.bc_var**> [#uses=1] %tmp3.i = load %struct.bc_var** %tmp.i ; <%struct.bc_var*> [#uses=1] store %struct.bc_var* %tmp3.i, %struct.bc_var** %ex_stack %tmp.i = getelementptr %struct.bc_var* %tmp.i, int 0, uint 0 ; <%struct.bc_struct**> [#uses=1] call void %bc_free_num( %struct.bc_struct** %tmp.i ) %tmp5.i = bitcast %struct.bc_var* %tmp.i to sbyte* ; [#uses=1] call void %free( sbyte* %tmp5.i ) br label %pop.exit pop.exit: ; preds = %cond_true.i, %cond_true27 %tmp29 = load %struct.bc_struct** %tmp24 ; <%struct.bc_struct*> [#uses=1] call void %push_copy( %struct.bc_struct* %tmp29 ) ret void return: ; preds = %bb21, %entry ret void } ----------------------------------------- Regards, Ryan ------- 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 Dec 31 16:44:16 2006 From: bugzilla-daemon at cs.uiuc.edu (bugzilla-daemon at cs.uiuc.edu) Date: Sun, 31 Dec 2006 16:44:16 -0600 Subject: [LLVMbugs] [Bug 1070] NEW: llvm-upgrade can cause redefinition errors Message-ID: <200612312244.kBVMiG7X001856@zion.cs.uiuc.edu> http://llvm.org/bugs/show_bug.cgi?id=1070 Summary: llvm-upgrade can cause redefinition errors Product: tools Version: trunk Platform: All OS/Version: All Status: NEW Severity: minor Priority: P2 Component: llvm-upgrade AssignedTo: unassignedbugs at nondot.org ReportedBy: rspencer at x10sys.com Current, llvm-upgrade is not handling redefinition of named values properly. This invariably is the result of the collapsed type planes. For example, code that previously defined a value in both the UInt and Int type planes will now not work because those type planes have merged to just i32. The two values will appear as redefinitions in the i32 type plane for the same symbol. One quick thing to help with this is to drop useless bitcasts, like: %tmp = bitcast i32 %tmp to i32 These result from old signedness conversions where the bitcast used to be: %tmp = bitcast uint %tmp to int Such bitcasts can now just be dropped by llvm-upgrade. However, this isn't sufficient, its still quite possible that code like this: %tmp = load uint* %uptr %tmp = load int* %sptr won't work because it now looks like: %tmp = load i32* %uptr %tmp = load i32* %sptr and a redefinition error occurs. The solution involves tracking the symbol table in llvm-upgrade and making sure that no re-definitions occur. If they do, a simple name uniqueness algorithm should be sufficient. ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.