From: Jeff Layton <jlayton@kernel.org>
To: Chuck Lever <chuck.lever@oracle.com>, Neil Brown <neilb@suse.de>,
Dai Ngo <Dai.Ngo@oracle.com>, Tom Talpey <tom@talpey.com>
Cc: Olga Kornievskaia <okorniev@redhat.com>,
linux-nfs@vger.kernel.org, linux-kernel@vger.kernel.org,
Jeff Layton <jlayton@kernel.org>
Subject: [PATCH 0/2] nfsd: allow the use of multiple backchannel slots
Date: Mon, 28 Oct 2024 10:26:25 -0400 [thread overview]
Message-ID: <20241028-bcwide-v1-0-0e75a8219dc0@kernel.org> (raw)
The NFSv4.1 server code has always only supported a single callback
slot. This has the effect of serializing all the callbacks on the
server, which can cause a bottleneck when there is a lot of callback
activity. This patchset allows the NFS server to use up to 32
backchannel slots when communicating with clients.
Note that so far, we don't have a great way to drive a lot of concurrent
backchannel activity with standard testsuites. I did set up a server
with two clients and had one client open a bunch of files, and a second
client fork off a bunch of processes that statx'ed those files
repeatedly. With that I was able to see the parallelization in action
via CB_GETATTR activity. I'm still thinking about how best to test this
with pynfs.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
Jeff Layton (2):
nfsd: remove nfsd4_session->se_bchannel
nfsd: allow for more callback session slots
fs/nfsd/nfs4callback.c | 107 +++++++++++++++++++++++++++++++++++--------------
fs/nfsd/nfs4state.c | 9 +++--
fs/nfsd/state.h | 13 +++---
fs/nfsd/trace.h | 2 +-
4 files changed, 89 insertions(+), 42 deletions(-)
---
base-commit: c7b8826b41906db1c930cbb10abb94eb24247f20
change-id: 20241025-bcwide-6bd7e4b63db2
Best regards,
--
Jeff Layton <jlayton@kernel.org>
next reply other threads:[~2024-10-28 14:26 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-28 14:26 Jeff Layton [this message]
2024-10-28 14:26 ` [PATCH 1/2] nfsd: remove nfsd4_session->se_bchannel Jeff Layton
2024-10-28 14:26 ` [PATCH 2/2] nfsd: allow for more callback session slots Jeff Layton
2024-10-28 14:46 ` Jeff Layton
2024-10-28 19:05 ` Chuck Lever
2024-10-29 10:28 ` Jeff Layton
2024-10-29 13:09 ` Chuck Lever III
2024-10-29 17:54 ` Tom Talpey
2024-10-29 18:24 ` Jeff Layton
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=20241028-bcwide-v1-0-0e75a8219dc0@kernel.org \
--to=jlayton@kernel.org \
--cc=Dai.Ngo@oracle.com \
--cc=chuck.lever@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=neilb@suse.de \
--cc=okorniev@redhat.com \
--cc=tom@talpey.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