This patch fixes the following soft lockup:
BUG: soft lockup - CPU#0 stuck for 23s! [weston:307]
On weston idle-timeout the IP is powered down and reset
asserted. On weston resume we get a massive vblank
IRQ storm due to the LDI registers having lost some state.
This state loss is caused by ade_crtc_atomic_begin() not
calling ade_ldi_set_mode(). With this patch applied
resuming from Weston idle-timeout works well.
Signed-off-by: Peter Griffin <[email protected]>
Tested-by: John Stultz <[email protected]>
Cc: [email protected]
---
drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
index c96c228..72c6357 100644
--- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
+++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
@@ -519,9 +519,12 @@ static void ade_crtc_atomic_begin(struct drm_crtc *crtc,
{
struct ade_crtc *acrtc = to_ade_crtc(crtc);
struct ade_hw_ctx *ctx = acrtc->ctx;
+ struct drm_display_mode *mode = &crtc->state->mode;
+ struct drm_display_mode *adj_mode = &crtc->state->adjusted_mode;
if (!ctx->power_on)
(void)ade_power_up(ctx);
+ ade_ldi_set_mode(acrtc, mode, adj_mode);
}
static void ade_crtc_atomic_flush(struct drm_crtc *crtc,
--
2.7.4
On Mon, Aug 28, 2017 at 04:44:30PM +0800, Xinliang Liu wrote:
> Hi,
>
> On 15 August 2017 at 22:14, Peter Griffin <[email protected]> wrote:
>
> > This patch fixes the following soft lockup:
> > BUG: soft lockup - CPU#0 stuck for 23s! [weston:307]
> >
> > On weston idle-timeout the IP is powered down and reset
> > asserted. On weston resume we get a massive vblank
> > IRQ storm due to the LDI registers having lost some state.
> >
> > This state loss is caused by ade_crtc_atomic_begin() not
> > calling ade_ldi_set_mode(). With this patch applied
> > resuming from Weston idle-timeout works well.
> >
> > Signed-off-by: Peter Griffin <[email protected]>
> > Tested-by: John Stultz <[email protected]>
> >
>
> Thanks Peter,
> This patch looks good to me.
> Reviewed-by: Xinliang Liu <[email protected]>
>
> @Sean, could you please help to apply to drm-misc if others has no more
> comments, thanks.
hisilicon isn't maintained in drm-misc, and you're the maintainer. This is
not how it works. So either
a) pick up the patch and send out a pull request to Dave Airlie
b) move hisilicon over to drm-misc and become a drm-misc maintainer
yourself. This needs a MAINTAINERS update to point the git tree at
drm-misc.
drm-misc maintainers don't maintain everyone else's driver as a service,
that simply doesn't scale.
Thanks, Daniel
>
> Thanks,
> Xinliang
>
>
> > Cc: [email protected]
> > ---
> > drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
> > b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
> > index c96c228..72c6357 100644
> > --- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
> > +++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
> > @@ -519,9 +519,12 @@ static void ade_crtc_atomic_begin(struct drm_crtc
> > *crtc,
> > {
> > struct ade_crtc *acrtc = to_ade_crtc(crtc);
> > struct ade_hw_ctx *ctx = acrtc->ctx;
> > + struct drm_display_mode *mode = &crtc->state->mode;
> > + struct drm_display_mode *adj_mode = &crtc->state->adjusted_mode;
> >
> > if (!ctx->power_on)
> > (void)ade_power_up(ctx);
> > + ade_ldi_set_mode(acrtc, mode, adj_mode);
> > }
> >
> > static void ade_crtc_atomic_flush(struct drm_crtc *crtc,
> > --
> > 2.7.4
> >
> > _______________________________________________
> > dri-devel mailing list
> > [email protected]
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> >
> _______________________________________________
> dri-devel mailing list
> [email protected]
> https://lists.freedesktop.org/mailman/listinfo/dri-devel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
Hi Daniel,
On 28 August 2017 at 16:51, Daniel Vetter <[email protected]> wrote:
> On Mon, Aug 28, 2017 at 04:44:30PM +0800, Xinliang Liu wrote:
>> Hi,
>>
>> On 15 August 2017 at 22:14, Peter Griffin <[email protected]> wrote:
>>
>> > This patch fixes the following soft lockup:
>> > BUG: soft lockup - CPU#0 stuck for 23s! [weston:307]
>> >
>> > On weston idle-timeout the IP is powered down and reset
>> > asserted. On weston resume we get a massive vblank
>> > IRQ storm due to the LDI registers having lost some state.
>> >
>> > This state loss is caused by ade_crtc_atomic_begin() not
>> > calling ade_ldi_set_mode(). With this patch applied
>> > resuming from Weston idle-timeout works well.
>> >
>> > Signed-off-by: Peter Griffin <[email protected]>
>> > Tested-by: John Stultz <[email protected]>
>> >
>>
>> Thanks Peter,
>> This patch looks good to me.
>> Reviewed-by: Xinliang Liu <[email protected]>
>>
>> @Sean, could you please help to apply to drm-misc if others has no more
>> comments, thanks.
>
> hisilicon isn't maintained in drm-misc, and you're the maintainer. This is
> not how it works. So either
> a) pick up the patch and send out a pull request to Dave Airlie
> b) move hisilicon over to drm-misc and become a drm-misc maintainer
> yourself. This needs a MAINTAINERS update to point the git tree at
> drm-misc.
>
> drm-misc maintainers don't maintain everyone else's driver as a service,
> that simply doesn't scale.
Sorry for my misunderstanding and thanks for pointing out that how
drm-misc works.
So I will pick up the patch and send a pull request.
Thanks,
Xinliang
>
> Thanks, Daniel
>
>>
>> Thanks,
>> Xinliang
>>
>>
>> > Cc: [email protected]
>> > ---
>> > drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c | 3 +++
>> > 1 file changed, 3 insertions(+)
>> >
>> > diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
>> > b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
>> > index c96c228..72c6357 100644
>> > --- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
>> > +++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
>> > @@ -519,9 +519,12 @@ static void ade_crtc_atomic_begin(struct drm_crtc
>> > *crtc,
>> > {
>> > struct ade_crtc *acrtc = to_ade_crtc(crtc);
>> > struct ade_hw_ctx *ctx = acrtc->ctx;
>> > + struct drm_display_mode *mode = &crtc->state->mode;
>> > + struct drm_display_mode *adj_mode = &crtc->state->adjusted_mode;
>> >
>> > if (!ctx->power_on)
>> > (void)ade_power_up(ctx);
>> > + ade_ldi_set_mode(acrtc, mode, adj_mode);
>> > }
>> >
>> > static void ade_crtc_atomic_flush(struct drm_crtc *crtc,
>> > --
>> > 2.7.4
>> >
>> > _______________________________________________
>> > dri-devel mailing list
>> > [email protected]
>> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
>> >
>
>> _______________________________________________
>> dri-devel mailing list
>> [email protected]
>> https://lists.freedesktop.org/mailman/listinfo/dri-devel
>
>
> --
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch