public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [r12-5236 Regression] FAIL: gcc.dg/tree-prof/merge_block.c scan-tree-dump-not optimized "Invalid sum" on Linux/x86_64
@ 2021-11-14  2:58 sunil.k.pandey
  2021-11-14 10:49 ` Jan Hubicka
  0 siblings, 1 reply; 2+ messages in thread
From: sunil.k.pandey @ 2021-11-14  2:58 UTC (permalink / raw)
  To: gcc-patches, gcc-regression, jh

On Linux/x86_64,

5aa91072e24c1e16a5ec641b48b64c9c9f199f13 is the first bad commit
commit 5aa91072e24c1e16a5ec641b48b64c9c9f199f13
Author: Jan Hubicka <jh@suse.cz>
Date:   Sat Nov 13 22:25:23 2021 +0100

    Implement DSE of dead functions calls storing memory.

caused

FAIL: c-c++-common/tsan/free_race2.c   -O2  execution test
FAIL: c-c++-common/tsan/free_race.c   -O2  execution test
FAIL: gcc.dg/ipa/ipa-sra-4.c scan-ipa-dump-times sra "Will split parameter" 2
FAIL: gcc.dg/tree-prof/merge_block.c scan-tree-dump-not optimized "Invalid sum"

with GCC configured with

../../gcc/configure --prefix=/local/skpandey/gccwork/toolwork/gcc-bisect-master/master/r12-5236/usr --enable-clocale=gnu --with-system-zlib --with-demangler-in-ld --with-fpmath=sse --enable-languages=c,c++,fortran --enable-cet --without-isl --enable-libmpx x86_64-linux --disable-bootstrap

To reproduce:

$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race2.c --target_board='unix{-m64}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race2.c --target_board='unix{-m64\ -march=cascadelake}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race.c --target_board='unix{-m64}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race.c --target_board='unix{-m64\ -march=cascadelake}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m32}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m32\ -march=cascadelake}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m64}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m64\ -march=cascadelake}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m32}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m32\ -march=cascadelake}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m64}'"
$ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m64\ -march=cascadelake}'"

(Please do not reply to this email, for question about this report, contact me at skpgkp2 at gmail dot com)

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [r12-5236 Regression] FAIL: gcc.dg/tree-prof/merge_block.c scan-tree-dump-not optimized "Invalid sum" on Linux/x86_64
  2021-11-14  2:58 [r12-5236 Regression] FAIL: gcc.dg/tree-prof/merge_block.c scan-tree-dump-not optimized "Invalid sum" on Linux/x86_64 sunil.k.pandey
@ 2021-11-14 10:49 ` Jan Hubicka
  0 siblings, 0 replies; 2+ messages in thread
From: Jan Hubicka @ 2021-11-14 10:49 UTC (permalink / raw)
  To: sunil.k.pandey; +Cc: gcc-patches, gcc-regression

> On Linux/x86_64,
> 
> 5aa91072e24c1e16a5ec641b48b64c9c9f199f13 is the first bad commit
> commit 5aa91072e24c1e16a5ec641b48b64c9c9f199f13
> Author: Jan Hubicka <jh@suse.cz>
> Date:   Sat Nov 13 22:25:23 2021 +0100
> 
>     Implement DSE of dead functions calls storing memory.
> 
> caused
> 
> FAIL: c-c++-common/tsan/free_race2.c   -O2  execution test
> FAIL: c-c++-common/tsan/free_race.c   -O2  execution test
> FAIL: gcc.dg/ipa/ipa-sra-4.c scan-ipa-dump-times sra "Will split parameter" 2
> FAIL: gcc.dg/tree-prof/merge_block.c scan-tree-dump-not optimized "Invalid sum"
> 
> with GCC configured with
> 
> ../../gcc/configure --prefix=/local/skpandey/gccwork/toolwork/gcc-bisect-master/master/r12-5236/usr --enable-clocale=gnu --with-system-zlib --with-demangler-in-ld --with-fpmath=sse --enable-languages=c,c++,fortran --enable-cet --without-isl --enable-libmpx x86_64-linux --disable-bootstrap
> 
> To reproduce:
> 
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race2.c --target_board='unix{-m64}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race2.c --target_board='unix{-m64\ -march=cascadelake}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race.c --target_board='unix{-m64}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tsan.exp=c-c++-common/tsan/free_race.c --target_board='unix{-m64\ -march=cascadelake}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m32}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m32\ -march=cascadelake}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m64}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="ipa.exp=gcc.dg/ipa/ipa-sra-4.c --target_board='unix{-m64\ -march=cascadelake}'"

In these two cases we do DSE and the testcase is no longer getting what
it wants (invalid store or SRA transform).  I had patch adding
-fno-ipa-modref and will commit it.

> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m32}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m32\ -march=cascadelake}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m64}'"
> $ cd {build_dir}/gcc && make check RUNTESTFLAGS="tree-prof.exp=gcc.dg/tree-prof/merge_block.c --target_board='unix{-m64\ -march=cascadelake}'"

This seems real bug in complete loop unrolling.  Not sure what broke
here, but cunrolli now does not get frequencies right.

Honza

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-11-14 10:49 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-11-14  2:58 [r12-5236 Regression] FAIL: gcc.dg/tree-prof/merge_block.c scan-tree-dump-not optimized "Invalid sum" on Linux/x86_64 sunil.k.pandey
2021-11-14 10:49 ` Jan Hubicka

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).