* [PATCH] drm/amdgpu: Handle fault with same timestamp
@ 2021-12-08 20:16 Philip Yang
2021-12-08 20:21 ` Felix Kuehling
2021-12-08 20:25 ` Alex Deucher
0 siblings, 2 replies; 5+ messages in thread
From: Philip Yang @ 2021-12-08 20:16 UTC (permalink / raw)
To: amd-gfx; +Cc: Philip Yang, Felix.Kuehling
Remove not unique timestamp WARNING as same timestamp interrupt happens
on some chips,
Drain fault need to wait for the processed_timestamp to be truly greater
than the checkpoint or the ring to be empty to be sure no stale faults
are handled.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 4 ++--
drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ---
2 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index 8050f7ba93ad..3df146579ad9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -188,8 +188,8 @@ int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev,
checkpoint_ts = amdgpu_ih_decode_iv_ts(adev, ih, checkpoint_wptr, -1);
return wait_event_interruptible_timeout(ih->wait_process,
- !amdgpu_ih_ts_after(ih->processed_timestamp, checkpoint_ts),
- timeout);
+ amdgpu_ih_ts_after(checkpoint_ts, ih->processed_timestamp) ||
+ ih->rptr == amdgpu_ih_get_wptr(adev, ih), timeout);
}
/**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index e031f0cf93a2..571b11117992 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -522,9 +522,6 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
if (!handled)
amdgpu_amdkfd_interrupt(adev, entry.iv_entry);
- dev_WARN_ONCE(adev->dev, ih->processed_timestamp == entry.timestamp,
- "IH timestamps are not unique");
-
if (amdgpu_ih_ts_after(ih->processed_timestamp, entry.timestamp))
ih->processed_timestamp = entry.timestamp;
}
--
2.17.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Handle fault with same timestamp
2021-12-08 20:16 [PATCH] drm/amdgpu: Handle fault with same timestamp Philip Yang
@ 2021-12-08 20:21 ` Felix Kuehling
2021-12-08 20:25 ` Alex Deucher
1 sibling, 0 replies; 5+ messages in thread
From: Felix Kuehling @ 2021-12-08 20:21 UTC (permalink / raw)
To: Philip Yang, amd-gfx
Am 2021-12-08 um 3:16 p.m. schrieb Philip Yang:
> Remove not unique timestamp WARNING as same timestamp interrupt happens
> on some chips,
>
> Drain fault need to wait for the processed_timestamp to be truly greater
> than the checkpoint or the ring to be empty to be sure no stale faults
> are handled.
>
> Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 4 ++--
> drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ---
> 2 files changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> index 8050f7ba93ad..3df146579ad9 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> @@ -188,8 +188,8 @@ int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev,
> checkpoint_ts = amdgpu_ih_decode_iv_ts(adev, ih, checkpoint_wptr, -1);
>
> return wait_event_interruptible_timeout(ih->wait_process,
> - !amdgpu_ih_ts_after(ih->processed_timestamp, checkpoint_ts),
> - timeout);
> + amdgpu_ih_ts_after(checkpoint_ts, ih->processed_timestamp) ||
> + ih->rptr == amdgpu_ih_get_wptr(adev, ih), timeout);
> }
>
> /**
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> index e031f0cf93a2..571b11117992 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> @@ -522,9 +522,6 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
> if (!handled)
> amdgpu_amdkfd_interrupt(adev, entry.iv_entry);
>
> - dev_WARN_ONCE(adev->dev, ih->processed_timestamp == entry.timestamp,
> - "IH timestamps are not unique");
> -
> if (amdgpu_ih_ts_after(ih->processed_timestamp, entry.timestamp))
> ih->processed_timestamp = entry.timestamp;
> }
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Handle fault with same timestamp
2021-12-08 20:16 [PATCH] drm/amdgpu: Handle fault with same timestamp Philip Yang
2021-12-08 20:21 ` Felix Kuehling
@ 2021-12-08 20:25 ` Alex Deucher
2021-12-08 20:27 ` Alex Deucher
1 sibling, 1 reply; 5+ messages in thread
From: Alex Deucher @ 2021-12-08 20:25 UTC (permalink / raw)
To: Philip Yang; +Cc: Kuehling, Felix, amd-gfx list
On Wed, Dec 8, 2021 at 3:17 PM Philip Yang <Philip.Yang@amd.com> wrote:
>
> Remove not unique timestamp WARNING as same timestamp interrupt happens
> on some chips,
>
> Drain fault need to wait for the processed_timestamp to be truly greater
> than the checkpoint or the ring to be empty to be sure no stale faults
> are handled.
>
> Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Maybe add the link to the bug when you push this?
Alex
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 4 ++--
> drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ---
> 2 files changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> index 8050f7ba93ad..3df146579ad9 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> @@ -188,8 +188,8 @@ int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev,
> checkpoint_ts = amdgpu_ih_decode_iv_ts(adev, ih, checkpoint_wptr, -1);
>
> return wait_event_interruptible_timeout(ih->wait_process,
> - !amdgpu_ih_ts_after(ih->processed_timestamp, checkpoint_ts),
> - timeout);
> + amdgpu_ih_ts_after(checkpoint_ts, ih->processed_timestamp) ||
> + ih->rptr == amdgpu_ih_get_wptr(adev, ih), timeout);
> }
>
> /**
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> index e031f0cf93a2..571b11117992 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> @@ -522,9 +522,6 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
> if (!handled)
> amdgpu_amdkfd_interrupt(adev, entry.iv_entry);
>
> - dev_WARN_ONCE(adev->dev, ih->processed_timestamp == entry.timestamp,
> - "IH timestamps are not unique");
> -
> if (amdgpu_ih_ts_after(ih->processed_timestamp, entry.timestamp))
> ih->processed_timestamp = entry.timestamp;
> }
> --
> 2.17.1
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Handle fault with same timestamp
2021-12-08 20:25 ` Alex Deucher
@ 2021-12-08 20:27 ` Alex Deucher
2021-12-09 8:30 ` Christian König
0 siblings, 1 reply; 5+ messages in thread
From: Alex Deucher @ 2021-12-08 20:27 UTC (permalink / raw)
To: Philip Yang; +Cc: Kuehling, Felix, amd-gfx list
On Wed, Dec 8, 2021 at 3:25 PM Alex Deucher <alexdeucher@gmail.com> wrote:
>
> On Wed, Dec 8, 2021 at 3:17 PM Philip Yang <Philip.Yang@amd.com> wrote:
> >
> > Remove not unique timestamp WARNING as same timestamp interrupt happens
> > on some chips,
> >
> > Drain fault need to wait for the processed_timestamp to be truly greater
> > than the checkpoint or the ring to be empty to be sure no stale faults
> > are handled.
> >
> > Signed-off-by: Philip Yang <Philip.Yang@amd.com>
>
> Maybe add the link to the bug when you push this?
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1818
Alex
>
> Alex
>
> > ---
> > drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 4 ++--
> > drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ---
> > 2 files changed, 2 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> > index 8050f7ba93ad..3df146579ad9 100644
> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
> > @@ -188,8 +188,8 @@ int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev,
> > checkpoint_ts = amdgpu_ih_decode_iv_ts(adev, ih, checkpoint_wptr, -1);
> >
> > return wait_event_interruptible_timeout(ih->wait_process,
> > - !amdgpu_ih_ts_after(ih->processed_timestamp, checkpoint_ts),
> > - timeout);
> > + amdgpu_ih_ts_after(checkpoint_ts, ih->processed_timestamp) ||
> > + ih->rptr == amdgpu_ih_get_wptr(adev, ih), timeout);
> > }
> >
> > /**
> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> > index e031f0cf93a2..571b11117992 100644
> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> > @@ -522,9 +522,6 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
> > if (!handled)
> > amdgpu_amdkfd_interrupt(adev, entry.iv_entry);
> >
> > - dev_WARN_ONCE(adev->dev, ih->processed_timestamp == entry.timestamp,
> > - "IH timestamps are not unique");
> > -
> > if (amdgpu_ih_ts_after(ih->processed_timestamp, entry.timestamp))
> > ih->processed_timestamp = entry.timestamp;
> > }
> > --
> > 2.17.1
> >
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Handle fault with same timestamp
2021-12-08 20:27 ` Alex Deucher
@ 2021-12-09 8:30 ` Christian König
0 siblings, 0 replies; 5+ messages in thread
From: Christian König @ 2021-12-09 8:30 UTC (permalink / raw)
To: Alex Deucher, Philip Yang; +Cc: Kuehling, Felix, amd-gfx list
Am 08.12.21 um 21:27 schrieb Alex Deucher:
> On Wed, Dec 8, 2021 at 3:25 PM Alex Deucher <alexdeucher@gmail.com> wrote:
>> On Wed, Dec 8, 2021 at 3:17 PM Philip Yang <Philip.Yang@amd.com> wrote:
>>> Remove not unique timestamp WARNING as same timestamp interrupt happens
>>> on some chips,
>>>
>>> Drain fault need to wait for the processed_timestamp to be truly greater
>>> than the checkpoint or the ring to be empty to be sure no stale faults
>>> are handled.
>>>
>>> Signed-off-by: Philip Yang <Philip.Yang@amd.com>
>> Maybe add the link to the bug when you push this?
> Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1818
With that done Reviewed-by: Christian König <christian.koenig@amd.com>
>
> Alex
>
>> Alex
>>
>>> ---
>>> drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 4 ++--
>>> drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ---
>>> 2 files changed, 2 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
>>> index 8050f7ba93ad..3df146579ad9 100644
>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
>>> @@ -188,8 +188,8 @@ int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev,
>>> checkpoint_ts = amdgpu_ih_decode_iv_ts(adev, ih, checkpoint_wptr, -1);
>>>
>>> return wait_event_interruptible_timeout(ih->wait_process,
>>> - !amdgpu_ih_ts_after(ih->processed_timestamp, checkpoint_ts),
>>> - timeout);
>>> + amdgpu_ih_ts_after(checkpoint_ts, ih->processed_timestamp) ||
>>> + ih->rptr == amdgpu_ih_get_wptr(adev, ih), timeout);
>>> }
>>>
>>> /**
>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
>>> index e031f0cf93a2..571b11117992 100644
>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
>>> @@ -522,9 +522,6 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
>>> if (!handled)
>>> amdgpu_amdkfd_interrupt(adev, entry.iv_entry);
>>>
>>> - dev_WARN_ONCE(adev->dev, ih->processed_timestamp == entry.timestamp,
>>> - "IH timestamps are not unique");
>>> -
>>> if (amdgpu_ih_ts_after(ih->processed_timestamp, entry.timestamp))
>>> ih->processed_timestamp = entry.timestamp;
>>> }
>>> --
>>> 2.17.1
>>>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-12-09 16:58 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-08 20:16 [PATCH] drm/amdgpu: Handle fault with same timestamp Philip Yang
2021-12-08 20:21 ` Felix Kuehling
2021-12-08 20:25 ` Alex Deucher
2021-12-08 20:27 ` Alex Deucher
2021-12-09 8:30 ` Christian König
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.