Agree with the patch. The original change was inspired by the existing
code returning -EINTR potentially after partially sending an SMB
packet. Returning -ERESTARTSYS seems like a right thing to do here.
Reviewed-by: Pavel Shilovsky <pshilov@xxxxxxxxxxxxx>
--
Best regards,
Pavel Shilovsky
пн, 30 нояб. 2020 г. в 08:35, Paulo Alcantara <pc@xxxxxx>:
>
> Steve French <smfrench@xxxxxxxxx> writes:
>
> > On Mon, Nov 30, 2020 at 7:02 AM Paulo Alcantara <pc@xxxxxx> wrote:
> >>
> >> Aurélien Aptel <aaptel@xxxxxxxx> writes:
> >>
> >> > Paulo Alcantara <pc@xxxxxx> writes:
> >> >> diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
> >> >> index e27e255d40dd..55853b9ed13d 100644
> >> >> --- a/fs/cifs/transport.c
> >> >> +++ b/fs/cifs/transport.c
> >> >> @@ -338,10 +338,8 @@ __smb_send_rqst(struct TCP_Server_Info *server, int num_rqst,
> >> >> if (ssocket == NULL)
> >> >> return -EAGAIN;
> >> >>
> >> >> - if (signal_pending(current)) {
> >> >> - cifs_dbg(FYI, "signal is pending before sending any data\n");
> >> >> - return -EINTR;
> >> >> - }
> >> >> + if (signal_pending(current))
> >> >> + return -ERESTARTSYS;
> >> >
> >> > Can we keep the debug message call?
> >>
> >> Yes.
> >>
> >> Steve, could you please update for-next with above debug msg?
> >
> > Done. Please check to see if my lightly modified debug message text is ok.
>
> OK for me. Thanks!
