Re: [PATCH v2 1/1] s390x: css: report errors from ccw_dstream_read/write

[Cornelia Huck <cohuck@redhat.com>]

On Fri, 9 Apr 2021 11:55:56 +0200
Pierre Morel <pmorel@linux.ibm.com> wrote:

> On 4/9/21 10:49 AM, Cornelia Huck wrote:
> > On Fri, 9 Apr 2021 10:38:37 +0200
> > Halil Pasic <pasic@linux.ibm.com> wrote:
> >   
> >> On Thu,  8 Apr 2021 18:32:09 +0200
> >> Pierre Morel <pmorel@linux.ibm.com> wrote:
> >>  
> >>> ccw_dstream_read/write functions returned values are sometime
> >>> not taking into account and reported back to the upper level
> >>> of interpretation of CCW instructions.
> >>>
> >>> It follows that accessing an invalid address does not trigger
> >>> a subchannel status program check to the guest as it should.
> >>>
> >>> Let's test the return values of ccw_dstream_write[_buf] and
> >>> ccw_dstream_read[_buf] and report it to the caller.
> >>>
> >>> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>  
> >>
> >> Acked-by: Halil Pasic <pasic@linux.ibm.com>
> >>
> >> I did not look into the whole scsw.count stuff or into wether
> >> your changes to 3270 (look form <mark></mark> in the diff part) affect
> >> more than just ccw_dstream_*.
> >>
> >> I would have preferred this patch split up based on the intended effect
> >> and thus also subsystem (css, virtio-ccw, 3270), but I've alluded to
> >> that before, and since we are in a hurry I can live with it as is.
> >>
> >> Regards,
> >> Halil
> >>  
> >>> ---
> >>>   hw/char/terminal3270.c | 11 +++++--
> >>>   hw/s390x/3270-ccw.c    |  5 +++-
> >>>   hw/s390x/css.c         | 14 +++++----
> >>>   hw/s390x/virtio-ccw.c  | 66 ++++++++++++++++++++++++++++++------------
> >>>   4 files changed, 69 insertions(+), 27 deletions(-)
> >>>
> >>> diff --git a/hw/char/terminal3270.c b/hw/char/terminal3270.c
> >>> index a9a46c8ed3..82e85fac2e 100644
> >>> --- a/hw/char/terminal3270.c
> >>> +++ b/hw/char/terminal3270.c
> >>> @@ -200,9 +200,13 @@ static int read_payload_3270(EmulatedCcw3270Device *dev)
> >>>   {
> >>>       Terminal3270 *t = TERMINAL_3270(dev);
> >>>       int len;
> >>> +    int ret;
> >>>   
> >>>       len = MIN(ccw_dstream_avail(get_cds(t)), t->in_len);
> >>> -    ccw_dstream_write_buf(get_cds(t), t->inv, len);
> >>> +    ret = ccw_dstream_write_buf(get_cds(t), t->inv, len);
> >>> +    if (ret < 0) {
> >>> +        return ret;
> >>> +    }
> >>>       t->in_len -= len;
> >>>   
> >>>       return len;
> >>> @@ -260,7 +264,10 @@ static int write_payload_3270(EmulatedCcw3270Device *dev, uint8_t cmd)
> >>>   
> >>>       t->outv[out_len++] = cmd;
> >>>       do {
> >>> -        ccw_dstream_read_buf(get_cds(t), &t->outv[out_len], len);
> >>> +        retval = ccw_dstream_read_buf(get_cds(t), &t->outv[out_len], len);
> >>> +        if (retval < 0) {
> >>> +            return retval;
> >>> +        }
> >>>           count = ccw_dstream_avail(get_cds(t));
> >>>           out_len += len;
> >>>   
> >>> diff --git a/hw/s390x/3270-ccw.c b/hw/s390x/3270-ccw.c
> >>> index 821319eee6..f3e7342b1e 100644
> >>> --- a/hw/s390x/3270-ccw.c
> >>> +++ b/hw/s390x/3270-ccw.c
> >>> @@ -31,6 +31,9 @@ static int handle_payload_3270_read(EmulatedCcw3270Device *dev, CCW1 *ccw)
> >>>       }
> >>>   
> >>>       len = ck->read_payload_3270(dev);  
> >>
> >> <mark>
> >>  
> >>> +    if (len < 0) {
> >>> +        return len;
> >>> +    }
> >>>       ccw_dev->sch->curr_status.scsw.count = ccw->count - len;
> >>>       
> >>
> >> </mark>
> >>
> >> Do we eventually update scsw.count?  
> > 
> > I think we can consider the contents of scsw.count 'unpredictable', no?  
> 
> I think so, the (len < 0) here will trigger a program check and the POP 
> specifies the count as "not meaningful" in case of a program check.

Yes, that's what I meant.

> 
> 
> >   
> >>  
> >>>       return 0;
> >>> @@ -50,7 +53,7 @@ static int handle_payload_3270_write(EmulatedCcw3270Device *dev, CCW1 *ccw)
> >>>       len = ck->write_payload_3270(dev, ccw->cmd_code);
> >>>   
> >>>       if (len <= 0) {  
> >>
> >> <mark>  
> >>> -        return -EIO;
> >>> +        return len ? len : -EIO;  
> >>
> >> </mark>  
> 
> Here we do not change the previous behavior.
> This problem, if it is one, is not related to not checking the dstream 
> read/write functions.

I agree.

> 
> >>  
> >>>       }
> >>>   
> >>>       ccw_dev->sch->curr_status.scsw.count = ccw->count - len;
> >>> diff --git a/hw/s390x/css.c b/hw/s390x/css.c
> >>> index fe47751df4..4149b8e5a7 100644  
> >>  
> >   
>