[Bug target/114272] New: SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[Bug target/114272] New: SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[ricbal02 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

            Bug ID: 114272
           Summary: SCHEDULER_IDENT incorrect for Cortex-A520 and
                    Cortex-A510
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ricbal02 at gcc dot gnu.org
  Target Milestone: ---

The SCHEDULER_IDENT for these two CPUs was incorrectly
set to cortexa55, which is incorrect. This can cause
incorrect asm to be generated.

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[ricbal02 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

Richard Ball <ricbal02 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2024-03-07
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |ASSIGNED
             Target|                            |aarch64
           Assignee|unassigned at gcc dot gnu.org      |ricbal02 at gcc dot gnu.org

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Richard Ball from comment #0)
> The SCHEDULER_IDENT for these two CPUs was incorrectly
> set to cortexa55, which is incorrect. This can cause
> incorrect asm to be generated.

SCHEDULER_IDENT will/should NOT cause incorrect assembly to be generated. It
might produce worse code though as the scheduler model will be that rather than
a decent one.

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[ricbal02 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

--- Comment #2 from Richard Ball <ricbal02 at gcc dot gnu.org> ---
Bad wording on my part there, apologies. You're correct, not incorrect perhaps
sub-optimal is a better wording.

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

--- Comment #3 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Richard Ball <ricbal02@gcc.gnu.org>:

https://gcc.gnu.org/g:cab53aae43cf94171b01320c08302e47a5daa391

commit r14-9672-gcab53aae43cf94171b01320c08302e47a5daa391
Author: Richard Ball <richard.ball@arm.com>
Date:   Tue Mar 26 13:54:31 2024 +0000

    aarch64: Fix SCHEDULER_IDENT for Cortex-A510 and Cortex-A520

    The SCHEDULER_IDENT for these two CPUs
    was incorrectly set to cortexa55.
    This can cause sub-optimal asm
     to be generated.

    gcc/ChangeLog:
            PR target/114272
            * config/aarch64/aarch64-cores.def (AARCH64_CORE):
            Change SCHEDULER_IDENT from cortexa55 to cortexa53
            for Cortex-A510 and Cortex-A520.

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[ricbal02 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

--- Comment #4 from Richard Ball <ricbal02 at gcc dot gnu.org> ---
Fixed on Trunk so far

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

--- Comment #5 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-13 branch has been updated by Richard Ball
<ricbal02@gcc.gnu.org>:

https://gcc.gnu.org/g:28b3b8a2fe55521a43f3710cf6ced1ab63f1ea03

commit r13-8654-g28b3b8a2fe55521a43f3710cf6ced1ab63f1ea03
Author: Richard Ball <richard.ball@arm.com>
Date:   Fri Apr 26 18:15:23 2024 +0100

    aarch64: Fix SCHEDULER_IDENT for Cortex-A510

    The SCHEDULER_IDENT for this CPU was incorrectly
    set to cortexa55. This can cause
    sub-optimal asm to be generated.

    gcc/ChangeLog:
            PR target/114272
            * config/aarch64/aarch64-cores.def (AARCH64_CORE):
            Change SCHEDULER_IDENT from cortexa55 to cortexa53
            for Cortex-A510.

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

--- Comment #6 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-12 branch has been updated by Richard Ball
<ricbal02@gcc.gnu.org>:

https://gcc.gnu.org/g:751a0f54345b7e037db7f0389c19c1f87e0ae4de

commit r12-10397-g751a0f54345b7e037db7f0389c19c1f87e0ae4de
Author: Richard Ball <richard.ball@arm.com>
Date:   Fri Apr 26 18:21:07 2024 +0100

    aarch64: Fix SCHEDULER_IDENT for Cortex-A510

    The SCHEDULER_IDENT for this CPU was incorrectly
    set to cortexa55. This can cause
    sub-optimal asm to be generated.

    gcc/ChangeLog:
            PR target/114272
            * config/aarch64/aarch64-cores.def (AARCH64_CORE):
            Change SCHEDULER_IDENT from cortexa55 to cortexa53
            for Cortex-A510.

[Bug target/114272] SCHEDULER_IDENT incorrect for Cortex-A520 and Cortex-A510

[ricbal02 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114272

Richard Ball <ricbal02 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #7 from Richard Ball <ricbal02 at gcc dot gnu.org> ---
Fixed on Trunk for both and backported Cortex-A510 fix to gcc-12 and gcc-13