SPEC 2017 profiling question (502.gcc_r and 505.mcf_r fail)

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

SPEC 2017 profiling question (502.gcc_r and 505.mcf_r fail)

Steve Ellcey-10
I am curious if anyone has tried running 'peak' SPEC 2017 numbers using
profiling.  Now that the cactus lto bug has been fixed I can run all
the SPEC intrate and fprate benchmarks with '-Ofast -flto -march=native'
on my aarch64 box and get accurate results but when I try to use these
options along with -fprofile-generate/-fprofile-use I get two
verification errors: 502.gcc_r and 505.mcf_r. The gcc benchmark is
generating different assembly language for some of its tests and mcf is
generating different numbers that look too large to just be due to
unsafe math optimizations.

Has anyone else seen these failures?

Steve Ellcey
[hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: SPEC 2017 profiling question (502.gcc_r and 505.mcf_r fail)

Bill Schmidt-3
On 10/4/19 10:13 AM, Steve Ellcey wrote:

> I am curious if anyone has tried running 'peak' SPEC 2017 numbers using
> profiling.  Now that the cactus lto bug has been fixed I can run all
> the SPEC intrate and fprate benchmarks with '-Ofast -flto -march=native'
> on my aarch64 box and get accurate results but when I try to use these
> options along with -fprofile-generate/-fprofile-use I get two
> verification errors: 502.gcc_r and 505.mcf_r. The gcc benchmark is
> generating different assembly language for some of its tests and mcf is
> generating different numbers that look too large to just be due to
> unsafe math optimizations.
>
> Has anyone else seen these failures?


Have you tried -fno-strict-aliasing?  There is a known issue with
spec_qsort() that affects both of these benchmarks.  See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83201.

Hope this helps,

Bill

>
> Steve Ellcey
> [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: SPEC 2017 profiling question (502.gcc_r and 505.mcf_r fail)

Thomas Koenig-6
Am 04.10.19 um 22:58 schrieb Bill Schmidt:

>
> Have you tried -fno-strict-aliasing?  There is a known issue with
> spec_qsort() that affects both of these benchmarks.  See
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83201.

Reading that bug report (another refusal to correct a glaring
standards violation) convinces me just a tiny little bit more that
SPEC is a cancer.
Reply | Threaded
Open this post in threaded view
|

Re: SPEC 2017 profiling question (502.gcc_r and 505.mcf_r fail)

Steve Ellcey-10
In reply to this post by Bill Schmidt-3
On Fri, 2019-10-04 at 15:58 -0500, Bill Schmidt wrote:
>
> > Has anyone else seen these failures?
>
>
> Have you tried -fno-strict-aliasing?  There is a known issue with
> spec_qsort() that affects both of these benchmarks.  See
>
https://urldefense.proofpoint.com/v2/url?u=https-3A__gcc.gnu.org_bugzilla_show-5Fbug.cgi-3Fid-3D83201&d=DwIDaQ&c=nKjWec2b6R0mOyPaz7xtfQ&r=Kj0CuWu6MgrNHos80CzrFt4fiXgwrFhMWDTO9Ue_lRU&m=M5tfnhGt9QWxrZvk7eKa9J_EonLqJs6YezVWveUtFhM&s=gesldYv1Oq8frkNSrX4O912SsKENeUKBZZruZ5UZ-NM&e=
>  .
>
> Hope this helps,
>
> Bill

Ah, of course, thank you.  I verified that this fixes my mcf failure,
gcc is still running.  I already had -fno-strict-aliasing for
perlbench, I should have figured out that it could be affecting other
tests too.

Steve Ellcey
[hidden email]