On 10/13/20 7:12 PM, Paolo Bonzini wrote:
> On 13/10/20 16:38, Claudio Fontana wrote:
>> +void bdrv_bh_schedule_oneshot(AioContext *ctx, QEMUBHFunc *cb, void *opaque)
>> +{
>> + if (replay_events_enabled()) {
>> + replay_bh_schedule_oneshot_event(ctx, cb, opaque);
>> + } else {
>> + aio_bh_schedule_oneshot(ctx, cb, opaque);
>> + }
>> +}
>> +
>> +void bdrv_bh_schedule(QEMUBH *bh)
>> +{
>> + if (replay_events_enabled()) {
>> + replay_bh_schedule_event(bh);
>> + } else {
>> + qemu_bh_schedule(bh);
>> + }
>> +}
>
> This is definitely better, but I'll defer to Kevin with respect to the
> naming of the function; having a bdrv_* function that has nothing to do
> with the block layer is still smelly of a sub-optimal API, and I'm not
> sure why the API change belongs in the series.
>
> Paolo
>
Hi Paolo,
I am not attached to the specific name, if someone has a better naming /
proposes a better prefix I will replace of course.
Thank you,
Claudio
