[PATCH] drivers/perf: apple_m1: Force 63bit counters for M2 CPUs

Will Deacon will at kernel.org
Mon Jun 5 09:35:05 PDT 2023


On Sun, 28 May 2023 09:02:05 +0100, Marc Zyngier wrote:
> Sidharth reports that on M2, the PMU never generates any interrupt
> when using 'perf record', which is a annoying as you get no sample.
> I'm temped to say "no sample, no problem", but others may have
> a different opinion.
> 
> Upon investigation, it appears that the counters on M2 are
> significantly different from the ones on M1, as they count on
> 64 bits instead of 48. Which of course, in the fine M1 tradition,
> means that we can only use 63 bits, as the top bit is used to signal
> the interrupt...
> 
> [...]

Applied to will (for-next/perf), thanks!

[1/1] drivers/perf: apple_m1: Force 63bit counters for M2 CPUs
      https://git.kernel.org/will/c/8be3593b9efa

Cheers,
-- 
Will

https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev



More information about the linux-arm-kernel mailing list