Re: [PATCH v3 1/1] virtio-blk: avoid preallocating big SGL for data

From: Max Gurtovoy <mgurtovoy@nvidia.com>
To: Stefan Hajnoczi <stefanha@redhat.com>,
	Jens Axboe <axboe@kernel.dk>, "mst@redhat.com" <mst@redhat.com>
Cc: <hch@infradead.org>, <mst@redhat.com>,
	<virtualization@lists.linux-foundation.org>,
	<kvm@vger.kernel.org>, <israelr@nvidia.com>, <nitzanc@nvidia.com>,
	<oren@nvidia.com>, <linux-block@vger.kernel.org>
Subject: Re: [PATCH v3 1/1] virtio-blk: avoid preallocating big SGL for data
Date: Thu, 23 Sep 2021 16:40:56 +0300	[thread overview]
Message-ID: <56cf84e2-fec0-08e8-0a47-24bb1df71883@nvidia.com> (raw)
In-Reply-To: <YUCUF7co94CRGkGU@stefanha-x1.localdomain>

Hi MST/Jens,

Do we need more review here or are we ok with the code and the test matrix ?

If we're ok, we need to decide if this goes through virtio PR or block PR.

Cheers,

-Max.

On 9/14/2021 3:22 PM, Stefan Hajnoczi wrote:
> On Mon, Sep 13, 2021 at 05:50:21PM +0300, Max Gurtovoy wrote:
>> On 9/6/2021 6:09 PM, Stefan Hajnoczi wrote:
>>> On Wed, Sep 01, 2021 at 04:14:34PM +0300, Max Gurtovoy wrote:
>>>> No need to pre-allocate a big buffer for the IO SGL anymore. If a device
>>>> has lots of deep queues, preallocation for the sg list can consume
>>>> substantial amounts of memory. For HW virtio-blk device, nr_hw_queues
>>>> can be 64 or 128 and each queue's depth might be 128. This means the
>>>> resulting preallocation for the data SGLs is big.
>>>>
>>>> Switch to runtime allocation for SGL for lists longer than 2 entries.
>>>> This is the approach used by NVMe drivers so it should be reasonable for
>>>> virtio block as well. Runtime SGL allocation has always been the case
>>>> for the legacy I/O path so this is nothing new.
>>>>
>>>> The preallocated small SGL depends on SG_CHAIN so if the ARCH doesn't
>>>> support SG_CHAIN, use only runtime allocation for the SGL.
>>>>
>>>> Re-organize the setup of the IO request to fit the new sg chain
>>>> mechanism.
>>>>
>>>> No performance degradation was seen (fio libaio engine with 16 jobs and
>>>> 128 iodepth):
>>>>
>>>> IO size      IOPs Rand Read (before/after)         IOPs Rand Write (before/after)
>>>> --------     ---------------------------------    ----------------------------------
>>>> 512B          318K/316K                                    329K/325K
>>>>
>>>> 4KB           323K/321K                                    353K/349K
>>>>
>>>> 16KB          199K/208K                                    250K/275K
>>>>
>>>> 128KB         36K/36.1K                                    39.2K/41.7K
>>> I ran fio randread benchmarks with 4k, 16k, 64k, and 128k at iodepth 1,
>>> 8, and 64 on two vCPUs. The results look fine, there is no significant
>>> regression.
>>>
>>> iodepth=1 and iodepth=64 are very consistent. For some reason the
>>> iodepth=8 has significant variance but I don't think it's the fault of
>>> this patch.
>>>
>>> Fio results and the Jupyter notebook export are available here (check
>>> out benchmark.html to see the graphs):
>>>
>>> https://gitlab.com/stefanha/virt-playbooks/-/tree/virtio-blk-sgl-allocation-benchmark/notebook
>>>
>>> Guest:
>>> - Fedora 34
>>> - Linux v5.14
>>> - 2 vCPUs (pinned), 4 GB RAM (single host NUMA node)
>>> - 1 IOThread (pinned)
>>> - virtio-blk aio=native,cache=none,format=raw
>>> - QEMU 6.1.0
>>>
>>> Host:
>>> - RHEL 8.3
>>> - Linux 4.18.0-240.22.1.el8_3.x86_64
>>> - Intel(R) Xeon(R) Silver 4214 CPU @ 2.20GHz
>>> - Intel Optane DC P4800X
>>>
>>> Stefan
>> Thanks, Stefan.
>>
>> Would you like me to add some of the results in my commit msg ? or Tested-By
>> sign ?
> Thanks, there's no need to change the commit description.
>
> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
> Tested-by: Stefan Hajnoczi <stefanha@redhat.com>