On Sat, Jan 19, 2013 at 10:58:33AM +0200, Pekka Enberg wrote:
> On Wed, Jan 16, 2013 at 6:52 PM, Cong Ding <dinggnu@xxxxxxxxx> wrote:
> > After we check (state.kcount != 0), state.kcount has to be 0 in all the "else"
> > branchs.
> >
> > Signed-off-by: Cong Ding <dinggnu@xxxxxxxxx>
> > ---
> > tools/kvm/hw/i8042.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/tools/kvm/hw/i8042.c b/tools/kvm/hw/i8042.c
> > index 9f8be6a..9035732 100644
> > --- a/tools/kvm/hw/i8042.c
> > +++ b/tools/kvm/hw/i8042.c
> > @@ -189,7 +189,7 @@ static u32 kbd_read_data(void)
> > state.mcount--;
> > kvm__irq_line(state.kvm, AUX_IRQ, 0);
> > kbd_update_irq();
> > - } else if (state.kcount == 0) {
> > + } else {
> > i = state.kread - 1;
> > if (i < 0)
> > i = QUEUE_SIZE;
>
> This doesn't look right. The 'kcount' field is an int so the value can
> be negative.
But the former check is "state.kcount != 0" as I described in the commit
message. Notice the difference between variable names in the "if" condition: the
first one is kcount, the second one is mcount, and the third one is same as the
first one kcount.
Ok, the original code is
if (state.kcount != 0) {
/* do something when (state.kcount != 0) */
} else if (state.mcount > 0) {
/* do something when (state.kcount == 0 && state.mount > 0) */
} else if (state.kcount == 0) {
/* do something when (state.kcount == 0 && state.mount <= 0) */
}
For the third branch, it runs when state.kcount == 0 and state.mount <= 0,
it's not necessary to ensure state.kcount == 0 again.
- cong
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
