If device has not been registered, or there is a pending work,
we should reschedule a sch_event call again.
Signed-off-by: Dong Jia Shi <[email protected]>
---
drivers/s390/cio/vfio_ccw_drv.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/s390/cio/vfio_ccw_drv.c b/drivers/s390/cio/vfio_ccw_drv.c
index ea6a2d0b2894..770fa9cfc310 100644
--- a/drivers/s390/cio/vfio_ccw_drv.c
+++ b/drivers/s390/cio/vfio_ccw_drv.c
@@ -177,6 +177,7 @@ static int vfio_ccw_sch_event(struct subchannel *sch, int process)
{
struct vfio_ccw_private *private = dev_get_drvdata(&sch->dev);
unsigned long flags;
+ int rc = -EAGAIN;
spin_lock_irqsave(sch->lock, flags);
if (!device_is_registered(&sch->dev))
@@ -187,6 +188,7 @@ static int vfio_ccw_sch_event(struct subchannel *sch, int process)
if (cio_update_schib(sch)) {
vfio_ccw_fsm_event(private, VFIO_CCW_EVENT_NOT_OPER);
+ rc = 0;
goto out_unlock;
}
@@ -195,11 +197,12 @@ static int vfio_ccw_sch_event(struct subchannel *sch, int process)
private->state = private->mdev ? VFIO_CCW_STATE_IDLE :
VFIO_CCW_STATE_STANDBY;
}
+ rc = 0;
out_unlock:
spin_unlock_irqrestore(sch->lock, flags);
- return 0;
+ return rc;
}
static struct css_device_id vfio_ccw_sch_ids[] = {
--
2.13.5
On Wed, 2 May 2018 09:25:59 +0200
Dong Jia Shi <[email protected]> wrote:
> If device has not been registered, or there is a pending work,
s/a pending work/work pending/
> we should reschedule a sch_event call again.
>
> Signed-off-by: Dong Jia Shi <[email protected]>
Slightly off-topic question: Is the mail address with or without the
'vnet' the one to go for?
> ---
> drivers/s390/cio/vfio_ccw_drv.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/s390/cio/vfio_ccw_drv.c b/drivers/s390/cio/vfio_ccw_drv.c
> index ea6a2d0b2894..770fa9cfc310 100644
> --- a/drivers/s390/cio/vfio_ccw_drv.c
> +++ b/drivers/s390/cio/vfio_ccw_drv.c
> @@ -177,6 +177,7 @@ static int vfio_ccw_sch_event(struct subchannel *sch, int process)
> {
> struct vfio_ccw_private *private = dev_get_drvdata(&sch->dev);
> unsigned long flags;
> + int rc = -EAGAIN;
>
> spin_lock_irqsave(sch->lock, flags);
> if (!device_is_registered(&sch->dev))
> @@ -187,6 +188,7 @@ static int vfio_ccw_sch_event(struct subchannel *sch, int process)
>
> if (cio_update_schib(sch)) {
> vfio_ccw_fsm_event(private, VFIO_CCW_EVENT_NOT_OPER);
> + rc = 0;
> goto out_unlock;
> }
>
> @@ -195,11 +197,12 @@ static int vfio_ccw_sch_event(struct subchannel *sch, int process)
> private->state = private->mdev ? VFIO_CCW_STATE_IDLE :
> VFIO_CCW_STATE_STANDBY;
> }
> + rc = 0;
>
> out_unlock:
> spin_unlock_irqrestore(sch->lock, flags);
>
> - return 0;
> + return rc;
> }
>
> static struct css_device_id vfio_ccw_sch_ids[] = {
Patch looks good to me.
Reviewed-by: Cornelia Huck <[email protected]>
I guess I should pick this for 4.17?
On Thu, 3 May 2018 10:06:40 +0800
Dong Jia Shi <[email protected]> wrote:
> * Cornelia Huck <[email protected]> [2018-05-02 16:42:51 +0200]:
>
> > On Wed, 2 May 2018 09:25:59 +0200
> > Dong Jia Shi <[email protected]> wrote:
> >
> > > If device has not been registered, or there is a pending work,
> >
> > s/a pending work/work pending/
Also, s/device/the device/
> >
> > > we should reschedule a sch_event call again.
> > >
> > > Signed-off-by: Dong Jia Shi <[email protected]>
(...)
> >
> > I guess I should pick this for 4.17?
> >
> I think there is no pressure for this one. So please feel free to decide
> as your convenience.
>
OK, I've queued this for the next release. Will wait for more patches
(e.g. the ida check pushdown) before I send a pull request.