Am 05.10.2016 um 16:25 hat Paolo Bonzini geschrieben:
>
>
> On 05/10/2016 16:20, Kevin Wolf wrote:
> > Am 05.10.2016 um 15:55 hat Paolo Bonzini geschrieben:
> >> On 05/10/2016 15:13, Stefan Hajnoczi wrote:
> qemu_bh_delete is already clearing bh->scheduled at the same time
> as it's set
On 05/10/2016 16:20, Kevin Wolf wrote:
> Am 05.10.2016 um 15:55 hat Paolo Bonzini geschrieben:
>> On 05/10/2016 15:13, Stefan Hajnoczi wrote:
qemu_bh_delete is already clearing bh->scheduled at the same time
as it's setting bh->deleted. Since it's not using any memory
barriers, th
Am 05.10.2016 um 15:55 hat Paolo Bonzini geschrieben:
> On 05/10/2016 15:13, Stefan Hajnoczi wrote:
> > > qemu_bh_delete is already clearing bh->scheduled at the same time
> > > as it's setting bh->deleted. Since it's not using any memory
> > > barriers, there is no synchronization going on for bh
On 05/10/2016 15:13, Stefan Hajnoczi wrote:
> > qemu_bh_delete is already clearing bh->scheduled at the same time
> > as it's setting bh->deleted. Since it's not using any memory
> > barriers, there is no synchronization going on for bh->deleted,
> > and this makes the bh->deleted checks superfl
On Mon, Oct 03, 2016 at 06:14:15PM +0200, Paolo Bonzini wrote:
> qemu_bh_delete is already clearing bh->scheduled at the same time
> as it's setting bh->deleted. Since it's not using any memory
> barriers, there is no synchronization going on for bh->deleted,
> and this makes the bh->deleted check
qemu_bh_delete is already clearing bh->scheduled at the same time
as it's setting bh->deleted. Since it's not using any memory
barriers, there is no synchronization going on for bh->deleted,
and this makes the bh->deleted checks superfluous in aio_compute_timeout,
aio_bh_poll and aio_ctx_check.
J
