Hi, http://llvm.org/bugs/show_bug.cgi?id=3367 http://llvm.org/bugs/show_bug.cgi?id=3831 These bugs are both cases where the optimizers are generating incorrect code. They have simple test cases and reasonably straightforward fixes. I've just done a successful test-suite run on i686-pc-linux-gnu with both fixes included. Can you please consider applying the fixes? Thanks, Jay.
Jay, On Mar 30, 2009, at 10:29 AM, Jay Foad wrote:> Hi, > > http://llvm.org/bugs/show_bug.cgi?id=3367 > http://llvm.org/bugs/show_bug.cgi?id=3831 > > These bugs are both cases where the optimizers are generating > incorrect code. They have simple test cases and reasonably > straightforward fixes. I've just done a successful test-suite run on > i686-pc-linux-gnu with both fixes included. > > Can you please consider applying the fixes?I'll take a look at these fixes. - Devang
Jay,> On Mar 30, 2009, at 10:29 AM, Jay Foad wrote: > >> Hi, >> >> http://llvm.org/bugs/show_bug.cgi?id=3367Do you have updated test case that fails with mainline? The current test case in bugzilla does not fail for me. Thanks, - Devang>> >> http://llvm.org/bugs/show_bug.cgi?id=3831 >> >> These bugs are both cases where the optimizers are generating >> incorrect code. They have simple test cases and reasonably >> straightforward fixes. I've just done a successful test-suite run on >> i686-pc-linux-gnu with both fixes included. >> >> Can you please consider applying the fixes?