[Bug target/115024] New: 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[Bug target/115024] New: 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[colin.king at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

            Bug ID: 115024
           Summary: 128 bit division performance regression, x86, between
                    gcc-14 and gcc-13 using target clones on skylake
                    platform
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: colin.king at intel dot com
  Target Milestone: ---

Created attachment 58158
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58158&action=edit
reproducer source code for __int128_t division regression

I'm seeing a 5% performance regression in gcc-14 compared to gcc-13, using gcc
on Ubuntu 24.04:

Versions:
gcc version 13.2.0 (Ubuntu 13.2.0-23ubuntu4) 
gcc version 14.0.1 20240412 (experimental) [master r14-9935-g67e1433a94f]
(Ubuntu 14-20240412-0ubuntu1) 

cking@skylake:~$ CFLAGS="" gcc-13 -O2 reproducer-div128.c 
cking@skylake:~$ ./a.out 
1650.83 div128 ops per sec

cking@skylake:~$ CFLAGS="" gcc-14 -O2 reproducer-div128.c 
cking@skylake:~$ ./a.out 
1567.48 div128 ops per sec

The original issue appeared when regression testing stress-ng cpu div128
stressor [1]. I've managed to extract the attached reproducer from the original
code (see attached).

Salient point to focus on:

1. The issue is also dependant on the TARGET_CLONES macro being defined as
__attribute__((target_clones("avx,default")))  - the avx target clones seems to
be an issue in reproducing this problem.

Attached are the reproducer C source and disassembled object code. 

References: [1]
https://github.com/ColinIanKing/stress-ng/blob/master/stress-cpu.c

[Bug target/115024] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[colin.king at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

--- Comment #1 from Colin Ian King <colin.king at intel dot com> ---
Created attachment 58159
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58159&action=edit
gcc-13 disassembly

gcc-13 disassembly

[Bug target/115024] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[colin.king at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

--- Comment #2 from Colin Ian King <colin.king at intel dot com> ---
Created attachment 58160
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58160&action=edit
gcc-14 disassembly

gcc-14 disassembly

[Bug target/115024] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[colin.king at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

--- Comment #3 from Colin Ian King <colin.king at intel dot com> ---
Created attachment 58161
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58161&action=edit
perf output for gcc-13 compiled code

[Bug target/115024] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[colin.king at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

--- Comment #4 from Colin Ian King <colin.king at intel dot com> ---
Created attachment 58162
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58162&action=edit
perf output for gcc-14 compiled code

[Bug target/115024] [14/15 regression] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[sjames at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

Sam James <sjames at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |14.2

[Bug target/115024] [14/15 regression] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[haochen.jiang at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

Haochen Jiang <haochen.jiang at intel dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |haochen.jiang at intel dot com

--- Comment #5 from Haochen Jiang <haochen.jiang at intel dot com> ---
From my test, trunk only has <1% regression if I calculated right.

[haochenj@shgcc101 ~]$ ./13.exe
1240.97 div128 ops per sec
[haochenj@shgcc101 ~]$ ./13.exe
1235.78 div128 ops per sec
[haochenj@shgcc101 ~]$ ./13.exe
1236.95 div128 ops per sec

[haochenj@shgcc101 ~]$ ./trunk.exe
1228.43 div128 ops per sec
[haochenj@shgcc101 ~]$ ./trunk.exe
1227.11 div128 ops per sec
[haochenj@shgcc101 ~]$ ./trunk.exe
1225.42 div128 ops per sec

[Bug target/115024] [14/15 regression] 128 bit division performance regression, x86, between gcc-14 and gcc-13 using target clones on skylake platform

[haochen.jiang at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115024

--- Comment #6 from Haochen Jiang <haochen.jiang at intel dot com> ---
I have got a machine to reproduce the regression.

Seem like a DSB miss from my data, but don't know why. Need more investigation.