Re: [PATCH 0/4] memcg: cleanup the memcg stats interfaces

[Shakeel Butt]

On Tue, Nov 11, 2025 at 11:01:47AM -0800, Roman Gushchin wrote:
> Shakeel Butt <shakeel.butt@xxxxxxxxx> writes:
> 
> > The memcg stats are safe against irq (and nmi) context and thus does not
> > require disabling irqs. However for some stats which are also maintained
> > at node level, it is using irq unsafe interface and thus requiring the
> > users to still disables irqs or use interfaces which explicitly disables
> > irqs. Let's move memcg code to use irq safe node level stats function
> > which is already optimized for architectures with HAVE_CMPXCHG_LOCAL
> > (all major ones), so there will not be any performance penalty for its
> > usage.
> 
> Do you have any production data for this or it's theory-based?

At the moment it is theory-based or more specifically based on the
comments on HAVE_CMPXCHG_LOCAL variants of stats update functions.

> 
> In general I feel we need a benchmark focused on memcg stats:
> there was a number of performance improvements and regressions in this
> code over last years, so a dedicated benchmark can help with measuring
> them.

Yeah it makes sense to have a benchmark. Let me see which benchmark
trigger this code paths a lot. At the high level, these interfaces are
used in reclaim and migration which are not really that performance
critical. I will try benchmarks with a lot of allocs/frees.

> 
> Nice cleanup btw, thanks!

Thanks.