[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-09 Thread erozen at microsoft dot com via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

Eugene Rozenfeld  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED

--- Comment #5 from Eugene Rozenfeld  ---
Fixed by https://gcc.gnu.org/g:7410032a772a9e77b620b091c2b551b68113a179 .

[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-08 Thread cvs-commit at gcc dot gnu.org via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

--- Comment #4 from CVS Commits  ---
The master branch has been updated by Eugene Rozenfeld :

https://gcc.gnu.org/g:7410032a772a9e77b620b091c2b551b68113a179

commit r13-4567-g7410032a772a9e77b620b091c2b551b68113a179
Author: Eugene Rozenfeld 
Date:   Tue Dec 6 17:05:18 2022 -0800

Fix count comparison in ipa-cp

The existing comparison was incorrect for non-PRECISE counts
(e.g., AFDO): we could end up with a 0 base_count, which could
lead to asserts, e.g., in good_cloning_opportunity_p.

Tested on x86_64-pc-linux-gnu.

gcc/ChangeLog:
PR ipa/108000
* ipa-cp.cc (ipcp_propagate_stage): Fix profile count comparison

gcc/testsuite
* gcc.dg/tree-prof/pr108000.c: Regression test

[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-07 Thread erozen at microsoft dot com via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

--- Comment #3 from Eugene Rozenfeld  ---
Repro steps:

gcc -O2 -g -o ./pr108000 ./pr108000.c

perf record -e cpu/event=0xC4,umask=0x20/puk -b -m8 -o perf.data ./pr108000

create_gcov --binary ./pr108000 --profile=./perf.data -gcov_version=2
--gcov=./pr108000.gcda

gcc ./pr108000.c -O2 -fauto-profile=./pr108000.gcda -O2 -o ./pr108000.afdo

[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-06 Thread rguenth at gcc dot gnu.org via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

Richard Biener  changed:

   What|Removed |Added

   Assignee|unassigned at gcc dot gnu.org  |erozen at microsoft dot 
com
 Status|WAITING |ASSIGNED

[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-06 Thread erozen at microsoft dot com via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

--- Comment #2 from Eugene Rozenfeld  ---
I actually have a fix for this, just wanted to file a simple test case. I
initially ran into this failure while reviving autoprofiledbootstrap build. The
failure happens while using the profile and the bug is an incorrect count
comparison. 

Here is the patch:
https://gcc.gnu.org/pipermail/gcc-patches/2022-November/606935.html

I will add this test to the patch before submitting it.

[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-06 Thread pinskia at gcc dot gnu.org via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed||2022-12-07
 Status|UNCONFIRMED |WAITING
 Ever confirmed|0   |1

[Bug ipa/108000] Assert during ipa-cp with AutoFDO

2022-12-06 Thread pinskia at gcc dot gnu.org via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108000

--- Comment #1 from Andrew Pinski  ---
Can you provide the exact command lines being used and does it fail during
while using the profile or before?