Hi, It looks like the default for -fomit-frame-pointer has recently changed from "no" to "yes at O1 and higher", but I did not see the commit. Was that intentional? (AddressSanitizer uses frame pointers to unwind stack, that's how we noticed). Thanks, --kcc -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20111227/2ce3c2d9/attachment.html>
On Tue, Dec 27, 2011 at 11:48 AM, Kostya Serebryany <kcc at google.com> wrote:> Hi, > > It looks like the default forĀ -fomit-frame-pointer has recently changed from > "no" to "yes at O1 and higher", but I did not see the commit. > Was that intentional?Yes; r146586. -Eli
On Tue, Dec 27, 2011 at 12:06 PM, Eli Friedman <eli.friedman at gmail.com>wrote:> On Tue, Dec 27, 2011 at 11:48 AM, Kostya Serebryany <kcc at google.com> > wrote: > > Hi, > > > > It looks like the default for -fomit-frame-pointer has recently changed > from > > "no" to "yes at O1 and higher", but I did not see the commit. > > Was that intentional? > > Yes; r146586. >What would be the best fix for asan? Force -mdisable-fp-elim when asan is enabled or document that asan requires -fomit-frame-pointer for better warning messages? Thanks! --kcc> > -Eli >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20111227/553e7470/attachment.html>
> It looks like the default forĀ -fomit-frame-pointer has recently changed from > "no" to "yes at O1 and higher", but I did not see the commit. > Was that intentional?Yes. See r146586 and PR8186 -- With best regards, Anton Korobeynikov Faculty of Mathematics and Mechanics, Saint Petersburg State University