From: cel@kernel.org
To: NeilBrown <neil@brown.name>, Jeff Layton <jlayton@kernel.org>,
Olga Kornievskaia <okorniev@redhat.com>,
Dai Ngo <dai.ngo@oracle.com>, Tom Talpey <tom@talpey.com>
Cc: <linux-nfs@vger.kernel.org>, <linux-rdma@vger.kernel.org>,
Chuck Lever <chuck.lever@oracle.com>
Subject: [PATCH v5 00/19] Allocate payload arrays dynamically
Date: Fri, 9 May 2025 15:03:34 -0400 [thread overview]
Message-ID: <20250509190354.5393-1-cel@kernel.org> (raw)
From: Chuck Lever <chuck.lever@oracle.com>
In order to make RPCSVC_MAXPAYLOAD larger (or variable in size), we
need to do something clever with the payload arrays embedded in
struct svc_rqst and elsewhere.
My preference is to keep these arrays allocated all the time because
allocating them on demand increases the risk of a memory allocation
failure during a large I/O. This is a quick-and-dirty approach that
might be replaced once NFSD is converted to use large folios.
The downside of this design choice is that it pins a few pages per
NFSD thread (and that's the current situation already). But note
that because RPCSVC_MAXPAGES is 259, each array is just over a page
in size, making the allocation waste quite a bit of memory beyond
the end of the array due to power-of-2 allocator round up. This gets
worse as the MAXPAGES value is doubled or quadrupled.
This series also addresses similar issues in the socket and RDMA
transports.
Changes since v4:
* Replace the use of rq_vec instead of allocating it dynamically
Changes since v3:
* Improved the rdma_rw context count estimate
* Dropped "NFSD: Remove NFSSVC_MAXBLKSIZE from .pc_xdrressize"
* Cleaned up the max size macros a bit
* Completed the implementation of adjustable max_block_size
Changes since v2:
* Address Jeff's review comments
* Address Neil's review comments
* Start removing a few uses of NFSSVC_MAXBLKSIZE
Chuck Lever (19):
svcrdma: Reduce the number of rdma_rw contexts per-QP
sunrpc: Add a helper to derive maxpages from sv_max_mesg
sunrpc: Remove backchannel check in svc_init_buffer()
sunrpc: Replace the rq_pages array with dynamically-allocated memory
sunrpc: Replace the rq_bvec array with dynamically-allocated memory
NFSD: Use rqstp->rq_bvec in nfsd_iter_read()
NFSD: De-duplicate the svc_fill_write_vector() call sites
SUNRPC: Export xdr_buf_to_bvec()
NFSD: Use rqstp->rq_bvec in nfsd_iter_write()
SUNRPC: Remove svc_fill_write_vector()
SUNRPC: Remove svc_rqst :: rq_vec
sunrpc: Adjust size of socket's receive page array dynamically
svcrdma: Adjust the number of entries in svc_rdma_recv_ctxt::rc_pages
svcrdma: Adjust the number of entries in svc_rdma_send_ctxt::sc_pages
sunrpc: Remove the RPCSVC_MAXPAGES macro
NFSD: Remove NFSD_BUFSIZE
NFSD: Remove NFSSVC_MAXBLKSIZE_V2 macro
NFSD: Add a "default" block size
SUNRPC: Bump the maximum payload size for the server
fs/nfsd/nfs3proc.c | 5 +-
fs/nfsd/nfs4proc.c | 10 +--
fs/nfsd/nfs4state.c | 2 +-
fs/nfsd/nfs4xdr.c | 2 +-
fs/nfsd/nfsd.h | 24 +++----
fs/nfsd/nfsproc.c | 13 ++--
fs/nfsd/nfssvc.c | 2 +-
fs/nfsd/nfsxdr.c | 4 +-
fs/nfsd/vfs.c | 67 ++++++++++++++------
fs/nfsd/vfs.h | 10 +--
include/linux/sunrpc/svc.h | 46 ++++++++------
include/linux/sunrpc/svc_rdma.h | 6 +-
include/linux/sunrpc/svcsock.h | 4 +-
net/sunrpc/svc.c | 80 +++++++-----------------
net/sunrpc/svc_xprt.c | 10 +--
net/sunrpc/svcsock.c | 15 +++--
net/sunrpc/xdr.c | 1 +
net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 8 ++-
net/sunrpc/xprtrdma/svc_rdma_rw.c | 2 +-
net/sunrpc/xprtrdma/svc_rdma_sendto.c | 16 ++++-
net/sunrpc/xprtrdma/svc_rdma_transport.c | 14 +++--
21 files changed, 170 insertions(+), 171 deletions(-)
--
2.49.0
next reply other threads:[~2025-05-09 19:03 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-09 19:03 cel [this message]
2025-05-09 19:03 ` [PATCH v5 01/19] svcrdma: Reduce the number of rdma_rw contexts per-QP cel
2025-05-13 12:00 ` Jeff Layton
2025-05-09 19:03 ` [PATCH v5 02/19] sunrpc: Add a helper to derive maxpages from sv_max_mesg cel
2025-05-09 19:03 ` [PATCH v5 03/19] sunrpc: Remove backchannel check in svc_init_buffer() cel
2025-05-09 19:03 ` [PATCH v5 04/19] sunrpc: Replace the rq_pages array with dynamically-allocated memory cel
2025-05-09 19:03 ` [PATCH v5 05/19] sunrpc: Replace the rq_bvec " cel
2025-05-09 19:03 ` [PATCH v5 06/19] NFSD: Use rqstp->rq_bvec in nfsd_iter_read() cel
2025-05-09 19:03 ` [PATCH v5 07/19] NFSD: De-duplicate the svc_fill_write_vector() call sites cel
2025-05-13 12:02 ` Jeff Layton
2025-05-09 19:03 ` [PATCH v5 08/19] SUNRPC: Export xdr_buf_to_bvec() cel
2025-05-09 19:03 ` [PATCH v5 09/19] NFSD: Use rqstp->rq_bvec in nfsd_iter_write() cel
2025-05-09 19:03 ` [PATCH v5 10/19] SUNRPC: Remove svc_fill_write_vector() cel
2025-05-09 19:03 ` [PATCH v5 11/19] SUNRPC: Remove svc_rqst :: rq_vec cel
2025-05-09 19:03 ` [PATCH v5 12/19] sunrpc: Adjust size of socket's receive page array dynamically cel
2025-05-09 19:03 ` [PATCH v5 13/19] svcrdma: Adjust the number of entries in svc_rdma_recv_ctxt::rc_pages cel
2025-05-09 19:03 ` [PATCH v5 14/19] svcrdma: Adjust the number of entries in svc_rdma_send_ctxt::sc_pages cel
2025-05-09 19:03 ` [PATCH v5 15/19] sunrpc: Remove the RPCSVC_MAXPAGES macro cel
2025-05-09 19:03 ` [PATCH v5 16/19] NFSD: Remove NFSD_BUFSIZE cel
2025-05-09 19:03 ` [PATCH v5 17/19] NFSD: Remove NFSSVC_MAXBLKSIZE_V2 macro cel
2025-05-09 19:03 ` [PATCH v5 18/19] NFSD: Add a "default" block size cel
2025-05-09 19:03 ` [PATCH v5 19/19] SUNRPC: Bump the maximum payload size for the server cel
2025-05-12 16:44 ` Aurélien Couderc
2025-05-12 18:09 ` Chuck Lever
2025-05-13 8:42 ` Aurélien Couderc
2025-05-13 12:08 ` Chuck Lever
2025-05-14 0:11 ` NeilBrown
2025-05-13 12:05 ` [PATCH v5 00/19] Allocate payload arrays dynamically 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=20250509190354.5393-1-cel@kernel.org \
--to=cel@kernel.org \
--cc=chuck.lever@oracle.com \
--cc=dai.ngo@oracle.com \
--cc=jlayton@kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=neil@brown.name \
--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