The first buildbot failure I can readily find was Monday, 26oct2009 around 7PM PDT. The assertion is Assertion failed: ((i >= FTy->getNumParams() || FTy->getParamType(i) == Params[i]->getType()) && "Calling a function with a bad signature!"), function init, file /Volumes/Sandbox/Buildbot/llvm/ build.llvm-gcc-x86_64-darwin10-selfhost/llvmCore.roots/llvmCore~obj/ src/lib/VMCore/Instructions.cpp, line 312. It seems to be compiling function uw_frame_state_for() within unwind- dw2.c; apparently the parameter types weren't a good match for a call to __builtin_object_size(). Does this look familiar to anyone? stuart -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20091027/33e4d442/attachment.html>
On Oct 27, 2009, at 11:23 AMPDT, Stuart Hastings wrote:> The first buildbot failure I can readily find was Monday, 26oct2009 > around 7PM PDT. The assertion is > > Assertion failed: ((i >= FTy->getNumParams() || FTy->getParamType(i) > == Params[i]->getType()) && "Calling a function with a bad > signature!"), function init, file /Volumes/Sandbox/Buildbot/llvm/ > build.llvm-gcc-x86_64-darwin10-selfhost/llvmCore.roots/llvmCore~obj/ > src/lib/VMCore/Instructions.cpp, line 312.Yes, I've been fighting through it this morning. Reverting Eric's patch fixed it (you might have to remove stmp* files to get it to rebuild enough stuff).> It seems to be compiling function uw_frame_state_for() within unwind- > dw2.c; apparently the parameter types weren't a good match for a > call to __builtin_object_size(). > > Does this look familiar to anyone? > > stuart > _______________________________________________ > LLVM Developers mailing list > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20091027/665ee5d4/attachment.html>
On Oct 27, 2009, at 11:23 AM, Stuart Hastings wrote:> The first buildbot failure I can readily find was Monday, 26oct2009 around 7PM PDT. The assertion is > > Assertion failed: ((i >= FTy->getNumParams() || FTy->getParamType(i) == Params[i]->getType()) && "Calling a function with a bad signature!"), function init, file /Volumes/Sandbox/Buildbot/llvm/build.llvm-gcc-x86_64-darwin10-selfhost/llvmCore.roots/llvmCore~obj/src/lib/VMCore/Instructions.cpp, line 312. > > > It seems to be compiling function uw_frame_state_for() within unwind-dw2.c; apparently the parameter types weren't a good match for a call to __builtin_object_size(). > > Does this look familiar to anyone?Yeah, it's me. Unexpected behavior, I think Bob was backing it out. If not I will and fix. -eric
On Oct 27, 2009, at 11:26 AM, Dale Johannesen wrote:> > On Oct 27, 2009, at 11:23 AMPDT, Stuart Hastings wrote: > >> The first buildbot failure I can readily find was Monday, 26oct2009 around 7PM PDT. The assertion is >> >> Assertion failed: ((i >= FTy->getNumParams() || FTy->getParamType(i) == Params[i]->getType()) && "Calling a function with a bad signature!"), function init, file /Volumes/Sandbox/Buildbot/llvm/build.llvm-gcc-x86_64-darwin10-selfhost/llvmCore.roots/llvmCore~obj/src/lib/VMCore/Instructions.cpp, line 312. > > Yes, I've been fighting through it this morning. Reverting Eric's patch fixed it (you might have to remove stmp* files to get it to rebuild enough stuff).Bob already reverted on ToT. Please update. Thanks. -eric
Apparently Analagous Threads
- [LLVMdev] llvmgcc ToT broken
- Function attributes for LibFunc and its impact on GlobalsAA
- [LLVMdev] Calling a function with bad signature, possible bug.
- [LLVMdev] Queries regarding function's arguments data type
- [LLVMdev] Why asserts don't provide much information?