Re: [PATCH 3/4] cifs: Identify a connection by a conn_id.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



108         if (server->tcpStatus == CifsNeedReconnect
109             || server->tcpStatus == CifsExiting)
110                 return;

If you remove this check then the dmesg log may be populated by below
VFS logs. THe intent of this check was to avoid printing them because
they don't make sense in reconnect/umount situations and may confuse
users.

111
112         switch (rc) {
113         case -1:
114                 /* change_conf hasn't been executed */
115                 break;
116         case 0:
117                 cifs_server_dbg(VFS, "Possible client or server
bug - zero credits\n");
118                 break;
119         case 1:
120                 cifs_server_dbg(VFS, "disabling echoes and oplocks\n");
121                 break;
122         case 2:
123                 cifs_dbg(FYI, "disabling oplocks\n");
124                 break;
125         default:
126                 trace_smb3_add_credits(server->CurrentMid,
127                         server->hostname, rc, add);
128                 cifs_dbg(FYI, "%s: added %u credits total=%d\n",
__func__, add, rc);
129         }


I would also suggest to rename sin_flight to just "in_flight" :)

--
Best regards,
Pavel Shilovsky

сб, 6 февр. 2021 г. в 20:18, Shyam Prasad N <nspmangalore@xxxxxxxxx>:
>
> Forgot to attach again. :)
>
> On Sat, Feb 6, 2021 at 8:18 PM Shyam Prasad N <nspmangalore@xxxxxxxxx> wrote:
> >
> > Here's an updated version with some formatting changes per checkpatch.pl.
> > @Pavel Shilovsky @ronnie sahlberg Hoping that one of you can review
> > this. Particularly the above point.
> >
> > Regards,
> > Shyam
> >
> > On Thu, Feb 4, 2021 at 12:13 AM Shyam Prasad N <nspmangalore@xxxxxxxxx> wrote:
> > >
> > > @@ -97,17 +99,25 @@ smb2_add_credits(struct TCP_Server_Info *server,
> > > -       if (server->tcpStatus == CifsNeedReconnect
> > > -           || server->tcpStatus == CifsExiting)
> > > -               return;
> > >
> > > @Pavel Shilovsky This check prevented a tracepoint from getting
> > > printed. I do not see much value in these lines, since all we do is
> > > print the tracepoint and exit. Hence removing it. Please let me know
> > > if that is not okay.
> > >
> > > On Thu, Feb 4, 2021 at 12:09 AM Shyam Prasad N <nspmangalore@xxxxxxxxx> wrote:
> > > >
> > > > --
> > > > Regards,
> > > > Shyam
> > >
> > >
> > >
> > > --
> > > Regards,
> > > Shyam
> >
> >
> >
> > --
> > Regards,
> > Shyam
>
>
>
> --
> Regards,
> Shyam




[Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux