This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/71615] wrong float point result with {-Ofast, -march=native, and valgrind}
- From: "wangmianzhi1 at linuxmail dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 22 Jun 2016 15:01:11 +0000
- Subject: [Bug target/71615] wrong float point result with {-Ofast, -march=native, and valgrind}
- Auto-submitted: auto-generated
- References: <bug-71615-4 at http dot gcc dot gnu dot org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71615
Mianzhi Wang <wangmianzhi1 at linuxmail dot org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution|--- |INVALID
--- Comment #4 from Mianzhi Wang <wangmianzhi1 at linuxmail dot org> ---
By executing:
$ valgrind --show-emwarns=yes ./test
It turned out:
==16833== Emulation warning: unsupported action:
==16833== Setting %mxcsr.fz (SSE flush-underflows-to-zero mode)
==16833== at 0x4006D2: set_fast_math (crtfastmath.c:93)
==16833== by 0x400A8C: __libc_csu_init (in /home/mianzhi/Desktop/test/test)
==16833== by 0x51857BE: (below main) (libc-start.c:247)
So it IS an known issue of valgrind.
See:
``Valgrind has the following limitations in its implementation of x86/AMD64
SSE2 FP arithmetic, relative to IEEE754''
of:
http://valgrind.org/docs/manual/manual-core.html#manual-core.limits
Here I'm closing the ticket.