On Wednesday 13 May 2009 12:10:34 Alex Williamson wrote:
> On Tue, 2009-05-12 at 21:42 -0600, Alex Williamson wrote:
> > On Wed, 2009-05-13 at 11:30 +0800, Yang, Sheng wrote:
> > > > + kvm->used_gsi_bitmap = malloc(gsi_bytes);
> > > > + if (!kvm->used_gsi_bitmap) {
> > > > + pthread_mutex_unlock(&kvm->gsi_mutex);
> > > > + goto out_close;
> > > > + }
> > > > + memset(kvm->used_gsi_bitmap, 0, gsi_bytes);
> > > > + kvm->max_gsi = gsi_bytes * 8;
> > >
> > > So max_gsi = gsi_count / 4?
>
> kvm->max_gsi actually becomes the number of GSIs available in the
> bitmap, which may be more than gsi_count if we rounded up. We
> preallocate GSIs between gsi_count and max_gsi to avoid using them.
> This just lets us not need to special case testing whether a bit in the
> last index is < gsi_count. Am I overlooking anything here? Thanks,
>
Oh, I understand that, and I just follow the logic of last comment here(which
"gsi_bytes = (gsi_count + 31) / 32" )... Sorry for confusion...
--
regards
Yang, Sheng
--
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
