Re: [PATCH v4 3/7] smb: client: sync retrans on remount

[Meetakshi Setiya]

Reviewed-by: Meetakshi Setiya <msetiya@xxxxxxxxxxxxx>

On Mon, Jun 15, 2026 at 10:54 PM <rajasimandalos@xxxxxxxxx> wrote:
>
> From: Rajasi Mandal <rajasimandal@xxxxxxxxxxxxx>
>
> The retrans mount option controls how many times the client retries
> sending a request before giving up.  Although remount already stored
> the new value in cifs_sb->ctx, it was never propagated to
> server->retrans, so the running connection kept using the old count.
>
> Introduce smb3_sync_server_opts() to push ctx options that live on
> TCP_Server_Info into the live server struct after a successful
> remount.  For now it handles retrans; subsequent patches will extend
> it to other server-level knobs.
>
> The previous patch in this series ("smb: client: sync runtime state
> into ctx on reconfigure") duplicates the existing context on remount,
> so a bare 'mount -o remount' carries the old retrans value forward and
> the parser only overwrites it when the user explicitly passes
> retrans=N.  No special-casing for the bare-remount /proc/mounts replay
> is therefore needed; in particular, an explicit retrans=0 from the
> user now correctly resets the value to the compile-time default.
>
> Writes to server->retrans are unsynchronized to match the existing
> convention: every other reader of the field (smb2ops.c retry path,
> cifs_show_options(), the original cifs_get_tcp_session() writer) does
> not take server->srv_lock, and match_server() only reads it under
> srv_lock as part of the larger match function.
>
> Note that retrans lives on TCP_Server_Info and is therefore shared
> across all mounts to the same server; reconfiguring one mount updates
> the value seen by every other mount sharing the connection.  This is
> consistent with the existing semantics: cifs_show_options() already
> reads server->retrans for /proc/mounts on every mount, and the
> runtime retry path always uses the server's value rather than any
> per-mount copy.
>
> Signed-off-by: Rajasi Mandal <rajasimandal@xxxxxxxxxxxxx>
> Reviewed-by: Meetakshi Setiya <msetiya@xxxxxxxxxxxxx>
> ---
>  fs/smb/client/fs_context.c | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
>
> diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c
> index d5a7d4dd402b..bfbea0648f6a 100644
> --- a/fs/smb/client/fs_context.c
> +++ b/fs/smb/client/fs_context.c
> @@ -1278,6 +1278,24 @@ static void smb3_sync_ses_chan_max(struct cifs_ses *ses, size_t max_channels)
>         spin_unlock(&ses->chan_lock);
>  }
>
> +/*
> + * Synchronize server-level options that are stored on TCP_Server_Info
> + * at mount time.  These fields are consulted at runtime (retry logic)
> + * so remount needs to update the live server struct in addition to
> + * cifs_sb->ctx.  Note these live on TCP_Server_Info and are therefore
> + * shared across all mounts to the same server; reconfiguring one mount
> + * updates the value seen by every other mount sharing the connection,
> + * matching how cifs_show_options() and the runtime retry path already
> + * read them unsynchronized from the server struct.
> + */
> +static void smb3_sync_server_opts(struct cifs_sb_info *cifs_sb)
> +{
> +       struct TCP_Server_Info *server = cifs_sb_master_tcon(cifs_sb)->ses->server;
> +       struct smb3_fs_context *ctx = cifs_sb->ctx;
> +
> +       server->retrans = ctx->retrans;
> +}
> +
>  static int smb3_reconfigure(struct fs_context *fc)
>  {
>         struct smb3_fs_context *ctx = smb3_fc2context(fc);
> @@ -1447,6 +1465,8 @@ static int smb3_reconfigure(struct fs_context *fc)
>         if (!rc)
>                 rc = dfs_cache_remount_fs(cifs_sb);
>  #endif
> +       if (!rc)
> +               smb3_sync_server_opts(cifs_sb);
>
>         return rc;
>
> --
> 2.43.0
>