All of lore.kernel.org
 help / color / mirror / Atom feed
From: nspmangalore@gmail.com
To: linux-cifs@vger.kernel.org, smfrench@gmail.com, pc@manguebit.com,
	bharathsm@microsoft.com
Cc: Shyam Prasad N <sprasad@microsoft.com>
Subject: [PATCH 3/5] cifs: do not search for channel if server is terminating
Date: Thu,  1 Feb 2024 11:15:28 +0000	[thread overview]
Message-ID: <20240201111530.17194-3-sprasad@microsoft.com> (raw)
In-Reply-To: <20240201111530.17194-1-sprasad@microsoft.com>

From: Shyam Prasad N <sprasad@microsoft.com>

In order to scale down the channels, the following sequence
of operations happen:
1. server struct is marked for terminate
2. the channel is deallocated in the ses->chans array
3. at a later point the cifsd thread actually terminates the server

Between 2 and 3, there can be calls to find the channel for
a server struct. When that happens, there can be an ugly warning
that's logged. But this is expected.

So this change does two things:
1. in cifs_ses_get_chan_index, if server->terminate is set, return
2. always make sure server->terminate is set with chan_lock held

Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
---
 fs/smb/client/sess.c    | 4 ++++
 fs/smb/client/smb2pdu.c | 2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/fs/smb/client/sess.c b/fs/smb/client/sess.c
index cde81042bebd..3d2548c35c9d 100644
--- a/fs/smb/client/sess.c
+++ b/fs/smb/client/sess.c
@@ -75,6 +75,10 @@ cifs_ses_get_chan_index(struct cifs_ses *ses,
 {
 	unsigned int i;
 
+	/* if the channel is waiting for termination */
+	if (server->terminate)
+		return CIFS_INVAL_CHAN_INDEX;
+
 	for (i = 0; i < ses->chan_count; i++) {
 		if (ses->chans[i].server == server)
 			return i;
diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c
index 75713559e69f..cf5665f6f220 100644
--- a/fs/smb/client/smb2pdu.c
+++ b/fs/smb/client/smb2pdu.c
@@ -178,6 +178,7 @@ cifs_chan_skip_or_disable(struct cifs_ses *ses,
 		}
 
 		ses->chans[chan_index].server = NULL;
+		server->terminate = true;
 		spin_unlock(&ses->chan_lock);
 
 		/*
@@ -188,7 +189,6 @@ cifs_chan_skip_or_disable(struct cifs_ses *ses,
 		 */
 		cifs_put_tcp_session(server, from_reconnect);
 
-		server->terminate = true;
 		cifs_signal_cifsd_for_reconnect(server, false);
 
 		/* mark primary server as needing reconnect */
-- 
2.34.1


  parent reply	other threads:[~2024-02-01 11:15 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-01 11:15 [PATCH 1/5] cifs: avoid redundant calls to disable multichannel nspmangalore
2024-02-01 11:15 ` [PATCH 2/5] cifs: change tcon status when need_reconnect is set on it nspmangalore
2024-02-01 11:15 ` nspmangalore [this message]
2024-02-01 11:15 ` [PATCH 4/5] cifs: failure to add channel on iface should bump up weight nspmangalore
2024-02-01 11:15 ` [PATCH 5/5] cifs: enforce nosharesock when multichannel is used nspmangalore
2024-02-06  6:13   ` Shyam Prasad N

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=20240201111530.17194-3-sprasad@microsoft.com \
    --to=nspmangalore@gmail.com \
    --cc=bharathsm@microsoft.com \
    --cc=linux-cifs@vger.kernel.org \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.