Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220420195425.34911-1-logang@deltatee.com>
In-Reply-To: <20220420195425.34911-1-logang@deltatee.com>
From:   Xiao Ni <xni@redhat.com>
Date:   Thu, 21 Apr 2022 16:45:20 +0800
Message-ID: <CALTww28fwNpm0O_jc7-2Xr0JSX9i6F1kgoUQ8m_k6ZgPa1XxXw@mail.gmail.com>
Subject: Re: [PATCH v2 00/12] Improve Raid5 Lock Contention
To:     Logan Gunthorpe <logang@deltatee.com>
Cc:     open list <linux-kernel@vger.kernel.org>,
        linux-raid <linux-raid@vger.kernel.org>,
        Song Liu <song@kernel.org>,
        Christoph Hellwig <hch@infradead.org>,
        Guoqing Jiang <guoqing.jiang@linux.dev>,
        Stephen Bates <sbates@raithlin.com>,
        Martin Oliveira <Martin.Oliveira@eideticom.com>,
        David Sloan <David.Sloan@eideticom.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Apr 21, 2022 at 3:55 AM Logan Gunthorpe <logang@deltatee.com> wrote:
>
> Hi,
>
> This is v2 of this series which addresses Christoph's feedback and
> fixes some bugs. The first posting is at [1]. A git branch is
> available at [2].
>
> --
>
> I've been doing some work trying to improve the bulk write performance
> of raid5 on large systems with fast NVMe drives. The bottleneck appears
> largely to be lock contention on the hash_lock and device_lock. This
> series improves the situation slightly by addressing a couple of low
> hanging fruit ways to take the lock fewer times in the request path.
>
> Patch 9 adjusts how batching works by keeping a reference to the
> previous stripe_head in raid5_make_request(). Under most situtations,
> this removes the need to take the hash_lock in stripe_add_to_batch_list()
> which should reduce the number of times the lock is taken by a factor of
> about 2.
>
> Patch 12 pivots the way raid5_make_request() works. Before the patch, the
> code must find the stripe_head for every 4KB page in the request, so each
> stripe head must be found once for every data disk. The patch changes this
> so that all the data disks can be added to a stripe_head at once and the
> number of times the stripe_head must be found (and thus the number of
> times the hash_lock is taken) should be reduced by a factor roughly equal
> to the number of data disks.
>
> The remaining patches are just cleanup and prep patches for those two
> patches.
>
> Doing apples to apples testing this series on a small VM with 5 ram
> disks, I saw a bandwidth increase of roughly 14% and lock contentions
> on the hash_lock (as reported by lock stat) reduced by more than a factor
> of 5 (though it is still significantly contended).
>
> Testing on larger systems with NVMe drives saw similar small bandwidth
> increases from 3% to 20% depending on the parameters. Oddly small arrays
> had larger gains, likely due to them having lower starting bandwidths; I
> would have expected larger gains with larger arrays (seeing there
> should have been even fewer locks taken in raid5_make_request()).


Hi Logan

Could you share the commands to get the test result (lock contention
and performance)?

Regards
Xiao