Re: [PATCH 6.18.y] mm/damon/core: disallow time-quota setting zero esz

[SeongJae Park <sj@kernel.org>]

TL; DR: This patch is broken, please drop.  I will send v2 soon, as a reply to
this.  Please read below for details.

On Mon, 04 May 2026 17:06:06 +0000 sashiko-bot@kernel.org wrote:

> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] The newly added `ctx` parameter in `damos_set_effective_quota()` is unused, and the function incorrectly uses the `DAMON_MIN_REGION` macro instead of `ctx->min_sz_region`.
> --
> 
> commit 8afd735e4c663dcbcca090e71f37ad10c37c705a
> Author: SeongJae Park <sj@kernel.org>
> 
> mm/damon/core: disallow time-quota setting zero esz
> 
> When DAMOS scheme throughput is very slow, the time quota can make the
> effective size quota smaller than the context's minimum region size, causing
> actions to be skipped. This commit attempts to fix the problem by ensuring
> the effective size quota is at least the minimum region size.
> 
> Link: https://lore.kernel.org/20260407003153.79589-1-sj@kernel.org
> Link: https://lore.kernel.org/20260405192504.110014-1-sj@kernel.org [1]
> 
> > diff --git a/mm/damon/core.c b/mm/damon/core.c
> > --- a/mm/damon/core.c
> > +++ b/mm/damon/core.c
> > @@ -2124,7 +2124,8 @@ static unsigned long damos_quota_score(struct damos_quota *quota)
> >  /*
> >   * Called only if quota->ms, or quota->sz are set, or quota->goals is not empty
> >   */
> > -static void damos_set_effective_quota(struct damos_quota *quota)
> > +static void damos_set_effective_quota(struct damos_quota *quota,
> > +		struct damon_ctx *ctx)
> >  {
> >  	unsigned long throughput;
> >  	unsigned long esz = ULONG_MAX;
> 
> [ ... ]
> 
> > @@ -2150,6 +2151,7 @@ static void damos_set_effective_quota(struct damos_quota *quota)
> >  		else
> >  			throughput = PAGE_SIZE * 1024;
> >  		esz = min(throughput * quota->ms, esz);
> > +		esz = max(DAMON_MIN_REGION, esz);
> 
> Should this be using ctx->min_sz_region instead of DAMON_MIN_REGION?
> 
> The commit message mentions bounding the effective size quota to
> damon_ctx->min_region_sz, and the ctx parameter was successfully added
> to the function signature. 
> 
> However, the ctx parameter is left unused here and the macro is used as a
> hardcoded bound instead. Since ctx->min_sz_region can be dynamically
> configured, can the effective size quota still fall below it with this
> implementation?

Correct.  Cherry-picking the original commit was causing build issue since the
upstream commit is renamed field (min_region_sz).  When fixing the build issue
I mistakenly assumed the field itslef is not yet introduced to 6.18 and using
DAMON_MIN_REGION.  It should use the field that not yet renamed
(min_sz_region), as Sashiko pointed out.

Greg, please drop this.  I will send a v2 soon, as a reply to this mail.

> 
> >  	}
> >  
> >  	if (quota->sz && quota->sz < esz)
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260504130127.25364-1-sj@kernel.org?part=1
> 


Thanks,
SJ