On Wed, Dec 10, 2014 at 5:21 PM, Ryota Ozaki <[email protected]> wrote: > On Wed, Dec 10, 2014 at 1:11 AM, Ryota Ozaki <[email protected]> wrote: >> On Wed, Dec 10, 2014 at 12:13 AM, Masao Uebayashi <[email protected]> wrote: >>> On Tue, Dec 9, 2014 at 5:41 PM, Ryota Ozaki <[email protected]> wrote: >>>> In the case of softint-ed interrupts on, I think we can safely >>>> remove the indirection. OTOH, in the case of off, I'm not sure >>>> we can do the same thing. >>> >>> As far as you don't break other virtio(4) drivers than vioif(4)... >> >> The patch doesn't change other drivers' behavior though, I'll check if >> the change affects the other virtio drivers. (I hope they're well isolated.) > > My code inspection and tests under disk/net io loads don't show > any problems at this point. > > Though, I fixed a bug in the case of SOFTINT_INTR && MPSAFE_NET :-/ > > Here is a updated patch: > http://www.netbsd.org/~ozaki-r/vioif-softint-intr.diff
I forgot to take care of SOFTINT_INTR && !NET_MPSAFE case :-/ Anyway fixed and updated the patch. ozaki-r
