Hello,
On Mon, Jul 22, 2024 at 11:03:00AM -0700, Boqun Feng wrote:
> On Mon, Jul 22, 2024 at 07:52:22AM -1000, Tejun Heo wrote:
> > On Mon, Jul 22, 2024 at 10:47:30AM -0700, Boqun Feng wrote:
> > > This looks like a data race because we read pcpu_nr_empty_pop_pages out
> > > of the lock for a best effort checking, @Tejun, maybe you could confirm
> > > on this?
> >
> > That does sound plausible.
> >
> > > - if (pcpu_nr_empty_pop_pages < PCPU_EMPTY_POP_PAGES_LOW)
> > > + /*
> > > + * Checks pcpu_nr_empty_pop_pages out of the pcpu_lock, data races may
> > > + * occur but this is just a best-effort checking, everything is synced
> > > + * in pcpu_balance_work.
> > > + */
> > > + if (data_race(pcpu_nr_empty_pop_pages) < PCPU_EMPTY_POP_PAGES_LOW)
> > > pcpu_schedule_balance_work();
> >
> > Would it be better to use READ/WRITE_ONCE() for the variable?
> >
>
> For READ/WRITE_ONCE(), we will need to replace all write accesses and
> all out-of-lock read accesses to pcpu_nr_empty_pop_pages, like below.
> It's better in the sense that it doesn't rely on compiler behaviors on
> data races, not sure about the performance impact though.
>
I think a better alternative is we can move it up into the lock under
area_found. The value gets updated as part of pcpu_alloc_area() as the
code above populates percpu memory that is already allocated.
We should probably annotate pcpu_update_empty_pages() with:
lockdep_assert_held(&pcpu_lock);
Thanks,
Dennis
> Regards,
> Boqun
>
> ----->8
> diff --git a/mm/percpu.c b/mm/percpu.c
> index 20d91af8c033..729e8188238b 100644
> --- a/mm/percpu.c
> +++ b/mm/percpu.c
> @@ -570,7 +570,8 @@ static void pcpu_isolate_chunk(struct pcpu_chunk *chunk)
>
> if (!chunk->isolated) {
> chunk->isolated = true;
> - pcpu_nr_empty_pop_pages -= chunk->nr_empty_pop_pages;
> + WRITE_ONCE(pcpu_nr_empty_pop_pages,
> + pcpu_nr_empty_pop_pages - chunk->nr_empty_pop_pages);
> }
> list_move(&chunk->list, &pcpu_chunk_lists[pcpu_to_depopulate_slot]);
> }
> @@ -581,7 +582,8 @@ static void pcpu_reintegrate_chunk(struct pcpu_chunk *chunk)
>
> if (chunk->isolated) {
> chunk->isolated = false;
> - pcpu_nr_empty_pop_pages += chunk->nr_empty_pop_pages;
> + WRITE_ONCE(pcpu_nr_empty_pop_pages,
> + pcpu_nr_empty_pop_pages + chunk->nr_empty_pop_pages);
> pcpu_chunk_relocate(chunk, -1);
> }
> }
> @@ -599,7 +601,8 @@ static inline void pcpu_update_empty_pages(struct pcpu_chunk *chunk, int nr)
> {
> chunk->nr_empty_pop_pages += nr;
> if (chunk != pcpu_reserved_chunk && !chunk->isolated)
> - pcpu_nr_empty_pop_pages += nr;
> + WRITE_ONCE(pcpu_nr_empty_pop_pages,
> + pcpu_nr_empty_pop_pages + nr);
> }
>
> /*
> @@ -1891,7 +1894,7 @@ void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved,
> mutex_unlock(&pcpu_alloc_mutex);
> }
>
> - if (pcpu_nr_empty_pop_pages < PCPU_EMPTY_POP_PAGES_LOW)
> + if (READ_ONCE(pcpu_nr_empty_pop_pages) < PCPU_EMPTY_POP_PAGES_LOW)
> pcpu_schedule_balance_work();
>
> /* clear the areas and return address relative to base address */
> @@ -2754,7 +2757,7 @@ void __init pcpu_setup_first_chunk(const struct pcpu_alloc_info *ai,
> tmp_addr = (unsigned long)base_addr + static_size + ai->reserved_size;
> pcpu_first_chunk = pcpu_alloc_first_chunk(tmp_addr, dyn_size);
>
> - pcpu_nr_empty_pop_pages = pcpu_first_chunk->nr_empty_pop_pages;
> + WRITE_ONCE(pcpu_nr_empty_pop_pages, pcpu_first_chunk->nr_empty_pop_pages);
> pcpu_chunk_relocate(pcpu_first_chunk, -1);
>
> /* include all regions of the first chunk */
>
