From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Vivek Goyal <vgoyal@redhat.com>
Cc: virtio-fs@redhat.com, mszeredi@redhat.com, qemu-devel@nongnu.org,
stefanha@redhat.com, groug@kaod.org
Subject: Re: [PATCH v3 0/2] viritofsd: Add support for FUSE_HANDLE_KILLPRIV_V2
Date: Tue, 16 Feb 2021 16:58:33 +0000 [thread overview]
Message-ID: <YCv5uRWyJcMMLwjy@work-vm> (raw)
In-Reply-To: <20210208224024.43555-1-vgoyal@redhat.com>
* Vivek Goyal (vgoyal@redhat.com) wrote:
> Hi,
>
> This is V3 of the patches. V2 had been posted here.
>
> https://listman.redhat.com/archives/virtio-fs/2020-November/msg00084.html
>
> These patches add support for FUSE_HANDLE_KILLPRIV_V2. This is basically
> a feature file server can opt-in. And with xattr enabled, this improves
> write performance many fold. Ran following fio command while virtiofsd is
> running with options "-o cache=auto" and "-o xattr".
Queued
> fio libaio random writes
> ------------------------
> fio --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=/mnt/virtiofs/random_read_write.fio --bs=4k --iodepth=64 --size=4G --readwrite=randwrite
>
> Without patches:
> WRITE: bw=84.6MiB/s (88.7MB/s), 84.6MiB/s-84.6MiB/s (88.7MB/s-88.7MB/s), io=4096MiB (4295MB), run=48436-48436msec
>
> With patches:
> WRITE: bw=273MiB/s (286MB/s), 273MiB/s-273MiB/s (286MB/s-286MB/s), io=4096MiB (4295MB), run=15003-15003msec
>
> So with these patches, for this particular workload, there is roughly
> more than 200% improvement.
>
> fio psync random writes
> -----------------------
> Even, single queue depth writes with ioengine=psync, how significant
> improvement. Performance jumps from roughly 35MiB/s to 70MiB/s.
>
> fio --ioengine=psync --name=test --filename=/mnt/virtiofs/random_read_write.fio --bs=4k --size=4G --readwrite=randwrite
>
> Without patches:
> WRITE: bw=35.0MiB/s (36.7MB/s), 35.0MiB/s-35.0MiB/s (36.7MB/s-36.7MB/s), io=4096MiB (4295MB), run=116932-116932msec
>
> With patches:
> WRITE: bw=69.9MiB/s (73.3MB/s), 69.9MiB/s-69.9MiB/s (73.3MB/s-73.3MB/s), io=4096MiB (4295MB), run=58596-58596msec
>
> In V3 I rebased these patches to latest upstream. There were few
> conflicts due to recent security fixes from Stefan. Resolved these
> conflicts.
>
> Thanks
> Vivek
>
> Vivek Goyal (2):
> virtiofsd: Save error code early at the failure callsite
> viriofsd: Add support for FUSE_HANDLE_KILLPRIV_V2
>
> tools/virtiofsd/fuse_common.h | 15 ++++++
> tools/virtiofsd/fuse_lowlevel.c | 11 +++-
> tools/virtiofsd/fuse_lowlevel.h | 1 +
> tools/virtiofsd/passthrough_ll.c | 93 ++++++++++++++++++++++++++++----
> 4 files changed, 108 insertions(+), 12 deletions(-)
>
> --
> 2.25.4
>
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
prev parent reply other threads:[~2021-02-16 17:12 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-08 22:40 [PATCH v3 0/2] viritofsd: Add support for FUSE_HANDLE_KILLPRIV_V2 Vivek Goyal
2021-02-08 22:40 ` [PATCH v3 1/2] virtiofsd: Save error code early at the failure callsite Vivek Goyal
2021-02-08 22:40 ` [PATCH v3 2/2] viriofsd: Add support for FUSE_HANDLE_KILLPRIV_V2 Vivek Goyal
2021-02-16 16:58 ` Dr. David Alan Gilbert [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YCv5uRWyJcMMLwjy@work-vm \
--to=dgilbert@redhat.com \
--cc=groug@kaod.org \
--cc=mszeredi@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=vgoyal@redhat.com \
--cc=virtio-fs@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).