Hi Jordan, Yes, I'm still alive! :) Sorry, I was a bit lazy and didn't include a comment in this commit. This test case was first optimizing to strcpy, and then crashing. The check there is ok. I've added a comment in r161172.
Nuno ----- Original Message ----- > Good to see you still around for a bit. Is this supposed to only match > __strcpy_chk and not __builtin___strcpy_chk? Because then you need a > CHECK-NOT. (And to ensure that the test isn't "fixed" otherwise, please > add a comment! > > On Aug 1, 2012, at 10:02 , Nuno Lopes <[email protected]> wrote: > >> Author: nlopes >> Date: Wed Aug 1 12:02:30 2012 >> New Revision: 161119 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=161119&view=rev >> Log: >> add test for PR13497 >> >> Added: >> cfe/trunk/test/CodeGen/nobuiltin.c >> >> Added: cfe/trunk/test/CodeGen/nobuiltin.c >> URL: >> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/CodeGen/nobuiltin.c?rev=161119&view=auto >> ============================================================================== >> --- cfe/trunk/test/CodeGen/nobuiltin.c (added) >> +++ cfe/trunk/test/CodeGen/nobuiltin.c Wed Aug 1 12:02:30 2012 >> @@ -0,0 +1,7 @@ >> +// RUN: %clang_cc1 -fno-builtin -O1 -S -o - %s | FileCheck %s >> + >> +void fn() { >> + char content[2]; >> + // CHECK: __strcpy_chk >> + __builtin___strcpy_chk(content, "", 1); >> +} >> _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
