Re: Ping: [PATCH 2/2] x86/vpic: also execute dpci callback for non-specific EOI

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Mon, Sep 21, 2020 at 12:05:51PM +0200, Jan Beulich wrote:
> On 21.08.2020 09:45, Jan Beulich wrote:
> > On 20.08.2020 18:28, Andrew Cooper wrote:
> >> On 20/08/2020 16:34, Roger Pau Monne wrote:
> >>> Currently the dpci EOI callback is only executed for specific EOIs.
> >>> This is wrong as non-specific EOIs will also clear the ISR bit and
> >>> thus end the interrupt. Re-arrange the code a bit so that the common
> >>> EOI handling path can be shared between all EOI modes.
> >>>
> >>> Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
> >>> ---
> >>>  xen/arch/x86/hvm/vpic.c | 10 +++++-----
> >>>  1 file changed, 5 insertions(+), 5 deletions(-)
> >>>
> >>> diff --git a/xen/arch/x86/hvm/vpic.c b/xen/arch/x86/hvm/vpic.c
> >>> index feb1db2ee3..3cf12581e9 100644
> >>> --- a/xen/arch/x86/hvm/vpic.c
> >>> +++ b/xen/arch/x86/hvm/vpic.c
> >>> @@ -249,15 +249,15 @@ static void vpic_ioport_write(
> >>>                  if ( priority == VPIC_PRIO_NONE )
> >>>                      break;
> >>>                  pin = (priority + vpic->priority_add) & 7;
> >>> -                vpic->isr &= ~(1 << pin);
> >>> -                if ( cmd == 5 )
> >>> -                    vpic->priority_add = (pin + 1) & 7;
> >>> -                break;
> >>> +                goto common_eoi;
> >>> +
> >>>              case 3: /* Specific EOI                */
> >>>              case 7: /* Specific EOI & Rotate       */
> >>>                  pin = val & 7;
> >>
> >> You'll need a /* Fallthrough */ here to keep various things happy.
> > 
> > Are you sure? There's ...
> > 
> >> Otherwise, Acked-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> >>
> >> Can fix on commit if you're happy.
> >>
> >>> +
> >>> +            common_eoi:
> > 
> > ... an ordinary label here, not a case one.
> 
> I would have wanted to commit this, but it's still not clear to me
> whether the adjustment you ask for is really needed.

Hello,

Was about to send a further series I have on top of this and saw this
is still on my patch queue. I'm happy with either way, but I would
like to get this committed if possible (as I think from a technical
PoV we all agree it's correct).

Thanks, Roger.