Re: [PATCH 3/8] mm/vmscan: Throttle reclaim when no progress is being made

[Vlastimil Babka]

On 11/24/21 11:32, Mel Gorman wrote:
> On Tue, Nov 23, 2021 at 05:19:12PM -0800, Darrick J. Wong wrote:
>> On Fri, Oct 22, 2021 at 03:46:46PM +0100, Mel Gorman wrote:
>> > Memcg reclaim throttles on congestion if no reclaim progress is made.
>> > This makes little sense, it might be due to writeback or a host of
>> > other factors.
>> > 
>> > For !memcg reclaim, it's messy. Direct reclaim primarily is throttled
>> > in the page allocator if it is failing to make progress. Kswapd
>> > throttles if too many pages are under writeback and marked for
>> > immediate reclaim.
>> > 
>> > This patch explicitly throttles if reclaim is failing to make progress.
>> 
>> Hi Mel,
>> 
>> Ever since Christoph broke swapfiles, I've been carrying around a little
>> fstest in my dev tree[1] that tries to exercise paging things in and out
>> of a swapfile.  Sadly I've been trapped in about three dozen customer
>> escalations for over a month, which means I haven't been able to do much
>> upstream in weeks.  Like submit this test upstream. :(
>> 
>> Now that I've finally gotten around to trying out a 5.16-rc2 build, I
>> notice that the runtime of this test has gone from ~5s to 2 hours.
>> Among other things that it does, the test sets up a cgroup with a memory
>> controller limiting the memory usage to 25MB, then runs a program that
>> tries to dirty 50MB of memory.  There's 2GB of memory in the VM, so
>> we're not running reclaim globally, but the cgroup gets throttled very
>> severely.
>> 
> 
> Ok, so this test cannot make progress until some of the cgroup pages get
> cleaned. What is the expectation for the test? Should it OOM or do you
> expect it to have spin-like behaviour until some writeback completes?
> I'm guessing you'd prefer it to spin and right now it's sleeping far
> too much.
> 
>> AFAICT the system is mostly idle, but it's difficult to tell because ps
>> and top also get stuck waiting for this cgroup for whatever reason. 
> 
> But this is surprising because I expect that ps and top are not running
> within the cgroup. Was /proc/PID/stack readable? 
> 
>> My
>> uninformed spculation is that usemem_and_swapoff takes a page fault
>> while dirtying the 50MB memory buffer, prepares to pull a page in from
>> swap, tries to evict another page to stay under the memcg limit, but
>> that decides that it's making no progress and calls
>> reclaim_throttle(..., VMSCAN_THROTTLE_NOPROGRESS).
>> 
>> The sleep is uninterruptible, so I can't even kill -9 fstests to shut it
>> down.  Eventually we either finish the test or (for the mlock part) the
>> OOM killer actually kills the process, but this takes a very long time.
>> 
> 
> The sleep can be interruptible.
> 
>> Any thoughts?  For now I can just hack around this by skipping
>> reclaim_throttle if cgroup_reclaim() == true, but that's probably not
>> the correct fix. :)
>> 
> 
> No, it wouldn't be but a possibility is throttling for only 1 jiffy if
> reclaiming within a memcg and the zone is balanced overall.
> 
> The interruptible part should just be the patch below. I need to poke at
> the cgroup limit part a bit

As the throttle timeout is short anyway, will the TASK_UNINTERRUPTIBLE vs
TASK_INTERRUPTIBLE make a difference for the (ability to kill? AFAIU
typically this inability to kill is because of a loop that doesn't check for
fatal_signal_pending().

> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index fb9584641ac7..07db03883062 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1068,7 +1068,7 @@ void reclaim_throttle(pg_data_t *pgdat, enum vmscan_throttle_state reason)
>  		break;
>  	}
>  
> -	prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
> +	prepare_to_wait(wqh, &wait, TASK_INTERRUPTIBLE);
>  	ret = schedule_timeout(timeout);
>  	finish_wait(wqh, &wait);
>  
>