Re: [PATCH 13/24] xfs: make inode reclaim almost non-blocking

From: Amir Goldstein <amir73il@gmail.com>
To: Dave Chinner <david@fromorbit.com>
Cc: linux-xfs <linux-xfs@vger.kernel.org>
Subject: Re: [PATCH 13/24] xfs: make inode reclaim almost non-blocking
Date: Fri, 22 May 2020 15:19:16 +0300	[thread overview]
Message-ID: <CAOQ4uxiit=kANPpQBDuT5kU4uuzO1Oi2p8HwU4wuoF1k7viXuw@mail.gmail.com> (raw)
In-Reply-To: <20200522035029.3022405-14-david@fromorbit.com>

On Fri, May 22, 2020 at 6:51 AM Dave Chinner <david@fromorbit.com> wrote:
>
> From: Dave Chinner <dchinner@redhat.com>
>
> Now that dirty inode writeback doesn't cause read-modify-write
> cycles on the inode cluster buffer under memory pressure, the need
> to throttle memory reclaim to the rate at which we can clean dirty
> inodes goes away. That is due to the fact that we no longer thrash
> inode cluster buffers under memory pressure to clean dirty inodes.
>
> This means inode writeback no longer stalls on memory allocation
> or read IO, and hence can be done asynchrnously without generating
> memory pressure. As a result, blocking inode writeback in reclaim is
> no longer necessary to prevent reclaim priority windup as cleaning
> dirty inodes is no longer dependent on having memory reserves
> available for the filesystem to make progress reclaiming inodes.
>
> Hence we can convert inode reclaim to be non-blocking for shrinker
> callouts, both for direct reclaim and kswapd.
>
> On a vanilla kernel, running a 16-way fsmark create workload on a
> 4 node/16p/16GB RAM machine, I can reliably pin 14.75GB of RAM via
> userspace mlock(). The OOM killer gets invoked at 15GB of
> pinned RAM.
>
> With this patch alone, pinning memory triggers premature OOM
> killer invocation, sometimes with as much as 45% of RAM being free.
> It's trivially easy to trigger the OOM killer when reclaim does not
> block.
>
> With pinning inode clusters in RAM adn then adding this patch, I can

typo: adn

> reliably pin 14.5GB of RAM and still have the fsmark workload run to
> completion. The OOM killer gets invoked 14.75GB of pinned RAM, which
> is only a small amount of memory less than the vanilla kernel. It is
> much more reliable than just with async reclaim alone.
>
> simoops shows that allocation stalls go away when async reclaim is
> used. Vanilla kernel:
>
> Run time: 1924 seconds
> Read latency (p50: 3,305,472) (p95: 3,723,264) (p99: 4,001,792)
> Write latency (p50: 184,064) (p95: 553,984) (p99: 807,936)
> Allocation latency (p50: 2,641,920) (p95: 3,911,680) (p99: 4,464,640)
> work rate = 13.45/sec (avg 13.44/sec) (p50: 13.46) (p95: 13.58) (p99: 13.70)
> alloc stall rate = 3.80/sec (avg: 2.59) (p50: 2.54) (p95: 2.96) (p99: 3.02)
>
> With inode cluster pinning and async reclaim:
>
> Run time: 1924 seconds
> Read latency (p50: 3,305,472) (p95: 3,715,072) (p99: 3,977,216)
> Write latency (p50: 187,648) (p95: 553,984) (p99: 789,504)
> Allocation latency (p50: 2,748,416) (p95: 3,919,872) (p99: 4,448,256)
> work rate = 13.28/sec (avg 13.32/sec) (p50: 13.26) (p95: 13.34) (p99: 13.34)
> alloc stall rate = 0.02/sec (avg: 0.02) (p50: 0.01) (p95: 0.03) (p99: 0.03)
>
> Latencies don't really change much, nor does the work rate. However,
> allocation almost never stalls with these changes, whilst the
> vanilla kernel is sometimes reporting 20 stalls/s over a 60s sample
> period. This difference is due to inode reclaim being largely
> non-blocking now.
>
> IOWs, once we have pinned inode cluster buffers, we can make inode
> reclaim non-blocking without a major risk of premature and/or
> spurious OOM killer invocation, and without any changes to memory
> reclaim infrastructure.
>
> Signed-off-by: Dave Chinner <dchinner@redhat.com>
> ---

I can confirm observing the stalls on production servers and that
this patch alone fixes the stalls.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>

Thanks,
Amir.