* icache_is_aliasing and big.LITTLE
@ 2019-05-08 18:45 Salman Qazi
2019-05-09 8:50 ` Catalin Marinas
0 siblings, 1 reply; 4+ messages in thread
From: Salman Qazi @ 2019-05-08 18:45 UTC (permalink / raw)
To: Catalin Marinas, Linux Kernel Mailing List
Hi,
What is the intention behind icache_is_aliasing on big.LITTLE systems
where some icaches are VIPT and others are PIPT? Is it meant to be
conservative in some sense or should it be made per-CPU?
Thanks for your help,
Salman
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: icache_is_aliasing and big.LITTLE
2019-05-08 18:45 icache_is_aliasing and big.LITTLE Salman Qazi
@ 2019-05-09 8:50 ` Catalin Marinas
2019-05-09 18:14 ` Salman Qazi
2019-05-10 11:04 ` Mark Rutland
0 siblings, 2 replies; 4+ messages in thread
From: Catalin Marinas @ 2019-05-09 8:50 UTC (permalink / raw)
To: Salman Qazi; +Cc: Linux Kernel Mailing List
Hi,
On Wed, May 08, 2019 at 11:45:03AM -0700, Salman Qazi wrote:
> What is the intention behind icache_is_aliasing on big.LITTLE systems
> where some icaches are VIPT and others are PIPT? Is it meant to be
> conservative in some sense or should it be made per-CPU?
It needs to cover the worst case scenario across all CPUs, i.e. aliasing
VIPT if one of the CPUs has this. We can't make it per-CPU because a
thread performing cache maintenance might be migrated to another CPU
with different cache policy (e.g. sync_icache_aliases()).
--
Catalin
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: icache_is_aliasing and big.LITTLE
2019-05-09 8:50 ` Catalin Marinas
@ 2019-05-09 18:14 ` Salman Qazi
2019-05-10 11:04 ` Mark Rutland
1 sibling, 0 replies; 4+ messages in thread
From: Salman Qazi @ 2019-05-09 18:14 UTC (permalink / raw)
To: Catalin Marinas; +Cc: Linux Kernel Mailing List
Thank you.
On Thu, May 9, 2019 at 1:50 AM Catalin Marinas <catalin.marinas@arm.com> wrote:
>
> Hi,
>
> On Wed, May 08, 2019 at 11:45:03AM -0700, Salman Qazi wrote:
> > What is the intention behind icache_is_aliasing on big.LITTLE systems
> > where some icaches are VIPT and others are PIPT? Is it meant to be
> > conservative in some sense or should it be made per-CPU?
>
> It needs to cover the worst case scenario across all CPUs, i.e. aliasing
> VIPT if one of the CPUs has this. We can't make it per-CPU because a
> thread performing cache maintenance might be migrated to another CPU
> with different cache policy (e.g. sync_icache_aliases()).
>
> --
> Catalin
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: icache_is_aliasing and big.LITTLE
2019-05-09 8:50 ` Catalin Marinas
2019-05-09 18:14 ` Salman Qazi
@ 2019-05-10 11:04 ` Mark Rutland
1 sibling, 0 replies; 4+ messages in thread
From: Mark Rutland @ 2019-05-10 11:04 UTC (permalink / raw)
To: Catalin Marinas; +Cc: Salman Qazi, Linux Kernel Mailing List
On Thu, May 09, 2019 at 09:50:04AM +0100, Catalin Marinas wrote:
> Hi,
>
> On Wed, May 08, 2019 at 11:45:03AM -0700, Salman Qazi wrote:
> > What is the intention behind icache_is_aliasing on big.LITTLE systems
> > where some icaches are VIPT and others are PIPT? Is it meant to be
> > conservative in some sense or should it be made per-CPU?
>
> It needs to cover the worst case scenario across all CPUs, i.e. aliasing
> VIPT if one of the CPUs has this. We can't make it per-CPU because a
> thread performing cache maintenance might be migrated to another CPU
> with different cache policy (e.g. sync_icache_aliases()).
It's slightly more subtle than that -- for broadcast maintenance the
policy of the CPU receiving the broadcast matters.
So even if all i-cache maintenance were performed on a thread pinned to
a CPU with PIPT caches, to correctly affect any VIPT i-caches in the
system it would be necessary to perform maintenance as-if the CPU
performing the maintenance had VIPT i-caches.
Thanks,
Mark.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2019-05-10 11:04 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-05-08 18:45 icache_is_aliasing and big.LITTLE Salman Qazi
2019-05-09 8:50 ` Catalin Marinas
2019-05-09 18:14 ` Salman Qazi
2019-05-10 11:04 ` Mark Rutland
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).