Re: [linus:master] [mm] 24e44cc22a: BUG:KCSAN:data-race_in_pcpu_alloc_noprof/pcpu_block_update_hint_alloc

[Dennis Zhou]

Hi Oliver,

On Tue, Jul 23, 2024 at 02:09:38PM +0800, Oliver Sang wrote:
> hi, Dennis Zhou,
> 
> On Mon, Jul 22, 2024 at 10:50:53PM -0700, Dennis Zhou wrote:
> > On Mon, Jul 22, 2024 at 01:53:52PM -0700, Boqun Feng wrote:
> > > On Mon, Jul 22, 2024 at 11:27:48AM -0700, Dennis Zhou wrote:
> > > > 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.
> > > > 
> > > 
> > > Not sure I followed what exactly you suggested here because I'm not
> > > familiar with the logic, but a simpler version would be:
> > > 
> > > 
> > 
> > I believe that's the only naked access of pcpu_nr_empty_pop_pages. So
> > I was thinking this'll fix this problem.
> > 
> > I also don't know how to rerun this CI tho..
> 
> we could test this patch. what's the base? could we apply it directly upon
> 24e44cc22a?
> 
> BTW, our bot is not so clever so far to auto test fix patches, so this is kind
> of manual effort. due to resource constraint, it will be hard for us to test
> each patch (we saw several patches in this thread already) or test very fast.
> 

Ah yeah that makes sense. If you don't mind testing the last one I sent,
the one below, that applies cleanly to 24e44cc22a.

Thanks,
Dennis

> > 
> > ---
> > diff --git a/mm/percpu.c b/mm/percpu.c
> > index 20d91af8c033..325fb8412e90 100644
> > --- a/mm/percpu.c
> > +++ b/mm/percpu.c
> > @@ -1864,6 +1864,10 @@ void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved,
> >  
> >  area_found:
> >  	pcpu_stats_area_alloc(chunk, size);
> > +
> > +	if (pcpu_nr_empty_pop_pages < PCPU_EMPTY_POP_PAGES_LOW)
> > +		pcpu_schedule_balance_work();
> > +
> >  	spin_unlock_irqrestore(&pcpu_lock, flags);
> >  
> >  	/* populate if not all pages are already there */
> > @@ -1891,9 +1895,6 @@ 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)
> > -		pcpu_schedule_balance_work();
> > -
> >  	/* clear the areas and return address relative to base address */
> >  	for_each_possible_cpu(cpu)
> >  		memset((void *)pcpu_chunk_addr(chunk, cpu, 0) + off, 0, size);