On Oct 12, 2007, at 4:50 AM, Török Edvin wrote: > On 10/12/07, Chris Lattner <[EMAIL PROTECTED]> wrote: >> >> It is somewhat convoluted, but len is set when the identifier is >> lexed. The codepath is through the >> StringMap<IdentifierInfo>::GetOrCreateValue method, which calls >> StringMapEntry<IdentifierInfo>::Create, which sets the size on line >> 155 of llvm/include/llvm/ADT/StringMap.h. >> >> I don't see how there could be a problem here, everything looks >> reasonable. >> > > Ok, it is something wrong with StringMap.h, but the testcases don't > catch it. > See patch at end of this mail. > Ahhh, I see.
> When I apply this patch clang can successfully compile t.c!! Nice! I applied this patch, please verify it also fixes it for you: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of- Mon-20071008/054469.html > The CodeGen failure (both w/, and w/o my patch) is this one: > FAIL: /home/edwin/llvm-svn/llvm/test/CodeGen/X86/2007-09-17- > ObjcFrameEH.ll > Failed with exit(1) at line 1 > while running: llvm-as < > /home/edwin/llvm-svn/llvm/test/CodeGen/X86/2007-09-17-ObjcFrameEH.ll | > llc -march=x86 -enable-eh | grep {isNullOrNil].eh"} | wc -l | grep 2 > child process exited abnormally This was failing for other people as well, I think the testcase should be fixed on mainline now. Thanks for the help! -Chris _______________________________________________ cfe-dev mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
