From: David Howells <dhowells@redhat.com>
To: nspmangalore@gmail.com
Cc: dhowells@redhat.com, linux-cifs@vger.kernel.org,
smfrench@gmail.com, pc@manguebit.com, bharathsm@microsoft.com,
Shyam Prasad N <sprasad@microsoft.com>
Subject: Re: [PATCH 4/4] cifs: make retry logic in read/write path consistent with other paths
Date: Wed, 21 Jan 2026 23:36:23 +0000 [thread overview]
Message-ID: <1653031.1769038583@warthog.procyon.org.uk> (raw)
In-Reply-To: <20260120062152.628822-4-sprasad@microsoft.com>
nspmangalore@gmail.com wrote:
> @@ -4994,6 +5014,14 @@ smb2_writev_callback(struct TCP_Server_Info *server, struct mid_q_entry *mid)
> server->credits, server->in_flight,
> 0, cifs_trace_rw_credits_write_response_clear);
> wdata->credits.value = 0;
> +
> + /* see if we need to retry */
> + if (is_replayable_error(wdata->result) &&
> + smb2_should_replay(tcon,
> + &wdata->retries,
> + &wdata->cur_sleep))
> + wdata->replay = true;
> +
This is really, really going to suck performance wise. ->callback() is called
in the context of the I/O thread for that socket. smb2_should_replay() does
an msleep() of up to 2 seconds. That means you aren't going to be processing
*any* messages coming down the socket from the server for the duration.
Better to offload the pause to netfslib if we can.
David
next prev parent reply other threads:[~2026-01-21 23:36 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-20 6:21 [PATCH 1/4] netfs: when subreq is marked for retry, do not check if it faced an error nspmangalore
2026-01-20 6:21 ` [PATCH 2/4] netfs: avoid double increment of retry_count in subreq nspmangalore
2026-01-21 23:23 ` David Howells
2026-01-20 6:21 ` [PATCH 3/4] cifs: Initialize cur_sleep value if not already done nspmangalore
2026-01-21 23:28 ` David Howells
2026-01-20 6:21 ` [PATCH 4/4] cifs: make retry logic in read/write path consistent with other paths nspmangalore
2026-01-20 11:32 ` kernel test robot
2026-01-20 17:03 ` David Howells
2026-01-21 4:45 ` Shyam Prasad N
2026-01-21 1:56 ` Steve French
2026-01-21 4:46 ` Shyam Prasad N
2026-01-21 23:36 ` David Howells [this message]
2026-01-21 23:50 ` David Howells
2026-01-29 11:38 ` Shyam Prasad N
2026-01-21 23:08 ` [PATCH 1/4] netfs: when subreq is marked for retry, do not check if it faced an error David Howells
2026-01-21 23:18 ` David Howells
2026-01-29 12:13 ` Shyam Prasad N
2026-01-29 13:43 ` David Howells
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=1653031.1769038583@warthog.procyon.org.uk \
--to=dhowells@redhat.com \
--cc=bharathsm@microsoft.com \
--cc=linux-cifs@vger.kernel.org \
--cc=nspmangalore@gmail.com \
--cc=pc@manguebit.com \
--cc=smfrench@gmail.com \
--cc=sprasad@microsoft.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