[Bug target/114412] New: [14 Regression] 7% slowdown of 436.cactusADM on aarch64

[Bug target/114412] New: [14 Regression] 7% slowdown of 436.cactusADM on aarch64

[pheeck at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

            Bug ID: 114412
           Summary: [14 Regression] 7% slowdown of 436.cactusADM on
                    aarch64
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Keywords: missed-optimization, needs-bisection
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pheeck at gcc dot gnu.org
            Blocks: 26163
  Target Milestone: ---
              Host: aarch64-gnu-linux
            Target: aarch64-gnu-linux

As seen here

https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=586.100.0

there is a 7% execution time slowdown of the 436.cactusADM SPEC 2006 benchmark
between commits

g:2d8cd712b17e88f2
g:46666b9bde5903d8

The compilation options are: -Ofast -march=native -mtune=native -flto PGO.

The CPU is Ampere Altra - Neoverse N1.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26163
[Bug 26163] [meta-bug] missed optimization in SPEC (2k17, 2k and 2k6 and 95)

[Bug target/114412] [14 Regression] 7% slowdown of 436.cactusADM on aarch64

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |14.0

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Compared to GCC 13 (~april in the graph) it's even 16% combined.

[Bug target/114412] [14 Regression] 7% slowdown of 436.cactusADM on aarch64

[law at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

Jeffrey A. Law <law at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2
                 CC|                            |law at gcc dot gnu.org

[Bug target/114412] [14/15 Regression] 7% slowdown of 436.cactusADM on aarch64

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|14.0                        |14.2

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 14.1 is being released, retargeting bugs to GCC 14.2.

[Bug target/114412] [14/15 Regression] 7% slowdown of 436.cactusADM on aarch64

[tnfchris at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

Tamar Christina <tnfchris at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tnfchris at gcc dot gnu.org

--- Comment #3 from Tamar Christina <tnfchris at gcc dot gnu.org> ---
Hi Filip,

Do you generate these runs with counters based PGO or compiler instrumentation?

Just so I know before I start trying to reproduce them.

[Bug target/114412] [14/15 Regression] 7% slowdown of 436.cactusADM on aarch64

[pheeck at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

--- Comment #4 from Filip Kastl <pheeck at gcc dot gnu.org> ---
(In reply to Tamar Christina from comment #3)
> Hi Filip,
> 
> Do you generate these runs with counters based PGO or compiler
> instrumentation?
> 
> Just so I know before I start trying to reproduce them.

Hi Tamar,

By counters you mean some sort of hardware counters? I didn't know there were
multiple ways to do PGO with GCC.

I think that the answer to your question is "compiler instrumentation". I just
do -fprofile-generate, run the instrumented binary and then -fprofile-use.

[Bug target/114412] [14/15 Regression] 7% slowdown of 436.cactusADM on aarch64

[tnfchris at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114412

--- Comment #5 from Tamar Christina <tnfchris at gcc dot gnu.org> ---
(In reply to Filip Kastl from comment #4)
> (In reply to Tamar Christina from comment #3)
> > Hi Filip,
> > 
> > Do you generate these runs with counters based PGO or compiler
> > instrumentation?
> > 
> > Just so I know before I start trying to reproduce them.
> 
> Hi Tamar,
> 
> By counters you mean some sort of hardware counters? I didn't know there
> were multiple ways to do PGO with GCC.
> 
> I think that the answer to your question is "compiler instrumentation". I
> just do -fprofile-generate, run the instrumented binary and then
> -fprofile-use.

Yeah, with some elbow grease the perf record method works too, but it's not
very accurate on Armv8.

I'll try to reproduce and bisect these over the weekend!