Re: [Qemu-devel] [PATCH 2/2] s390x/css: fix bits must be zero check for TIC

[Dong Jia Shi]

* Halil Pasic <address@hidden> [2017-07-26 13:38:33 +0200]:

> 
> 
> On 07/26/2017 05:01 AM, Dong Jia Shi wrote:
> > Hello Halil,
> > 
> > * Halil Pasic <address@hidden> [2017-07-26 00:44:42 +0200]:
> > 
> >> According to the PoP bit positions 0-3 and 8-32 of the format-1 CCW must
> >> contain zeros.  Bits 0-3 are already covered by cmd_code validity
> >> checking, and bit 32 is covered by the CCW address checking.
> >>
> >> Bits 8-31 correspond to CCW1.flags and CCW1.count.  Currently we only
> >> check for the absence of certain flags.  Let's fix this.
> >>
> >> Signed-off-by: Halil Pasic <address@hidden>
> >> ---
> >>  hw/s390x/css.c | 3 ++-
> >>  1 file changed, 2 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/hw/s390x/css.c b/hw/s390x/css.c
> >> index d17e21b7af..1f04ce4a1b 100644
> >> --- a/hw/s390x/css.c
> >> +++ b/hw/s390x/css.c
> >> @@ -884,7 +884,8 @@ static int css_interpret_ccw(SubchDev *sch, hwaddr 
> >> ccw_addr,
> >>              ret = -EINVAL;
> >>              break;
> >>          }
> >> -        if (ccw.flags & (CCW_FLAG_CC | CCW_FLAG_DC)) {
> >> +        if (ccw.flags || ccw.count) {
> >> +            /* We have already sanitized these if fmt 0. */
> > ccw0 does not have the same restrictions as ccw1. We don't sanitize
> > these for ccw0.
> > 
> 
> Yes you have misunderstood. For fmt 1 these bits have to be zero
> otherwise a program-check condition is to be recognized. Thus we don't
> sanitize for fmt 1.
> 
> For fmt 0 these bits are ignored. We sanitize them in
> for some time now by setting them to zero when making a CCW1
> out of an CCW0. If we would recognize a program-check for
> fmt 0 that would be wrong.
Yup, I know this.

> 
> The comment tells us why this code is good for CCW0 too,
> and why can we omit sch->ccw_fmt_1 from the conditon.
Ahh, I see the point now. Yes, I misunderstood.

Another point is we have translated ccw0 to ccw1. So here we only focus
on handling ccw1 stuff. Mentioning ccw0 seems a little redundant.

Anyway, I will leave this to you to decide. No problem from my side now.

> 
> Regards,
> Halil
> 
> > (This comment is still here. Did I misunderstand things? :)
> > 
> >>              ret = -EINVAL;
> >>              break;
> >>          }
> >> -- 
> >> 2.11.2
> >>
> > 
> > With the comment removed:
> > Reviewed-by: Dong Jia Shi <address@hidden>
> > 
> 
> 

-- 
Dong Jia Shi