* [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2
@ 2021-01-05 11:45 Christian König
2021-01-05 18:12 ` Lyude Paul
2021-01-08 18:49 ` Mike Lothian
0 siblings, 2 replies; 5+ messages in thread
From: Christian König @ 2021-01-05 11:45 UTC (permalink / raw)
To: dri-devel
From: Lyude Paul <lyude@redhat.com>
Recently a regression was introduced which caused TTM's buffer eviction to
attempt to evict already-pinned BOs, causing issues with buffer eviction
under memory pressure along with suspend/resume:
nouveau 0000:1f:00.0: DRM: evicting buffers...
nouveau 0000:1f:00.0: DRM: Moving pinned object 00000000c428c3ff!
nouveau 0000:1f:00.0: fifo: fault 00 [READ] at 0000000000200000 engine 04
[BAR1] client 07 [HUB/HOST_CPU] reason 02 [PTE] on channel -1 [00ffeaa000
unknown]
nouveau 0000:1f:00.0: fifo: DROPPED_MMU_FAULT 00001000
nouveau 0000:1f:00.0: fifo: fault 01 [WRITE] at 0000000000020000 engine
0c [HOST6] client 07 [HUB/HOST_CPU] reason 02 [PTE] on channel 1
[00ffb28000 DRM]
nouveau 0000:1f:00.0: fifo: channel 1: killed
nouveau 0000:1f:00.0: fifo: runlist 0: scheduled for recovery
[TTM] Buffer eviction failed
nouveau 0000:1f:00.0: DRM: waiting for kernel channels to go idle...
nouveau 0000:1f:00.0: DRM: failed to idle channel 1 [DRM]
nouveau 0000:1f:00.0: DRM: resuming display...
After some bisection and investigation, it appears this resulted from the
recent changes to ttm_bo_move_to_lru_tail(). Previously when a buffer was
pinned, the buffer would be removed from the LRU once ttm_bo_unreserve
to maintain the LRU list when pinning or unpinning BOs. However, since:
commit 3d1a88e1051f ("drm/ttm: cleanup LRU handling further")
We've been exiting from ttm_bo_move_to_lru_tail() at the very beginning of
the function if the bo we're looking at is pinned, resulting in the pinned
BO never getting removed from the lru and as a result - causing issues when
it eventually becomes time for eviction.
So, let's fix this by calling ttm_bo_del_from_lru() from
ttm_bo_move_to_lru_tail() in the event that we're dealing with a pinned
buffer.
v2 (chk): reduce to only the fixing one liner since we always want to
call the callback whenever we would move on the LRU.
Signed-off-by: Lyude Paul <lyude@redhat.com>
Fixes: 3d1a88e1051f ("drm/ttm: cleanup LRU handling further")
Cc: Christian König <christian.koenig@amd.com>
Cc: Dave Airlie <airlied@redhat.com>
---
drivers/gpu/drm/ttm/ttm_bo.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 31e8b3da5563..b65f4b12f986 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -138,8 +138,10 @@ void ttm_bo_move_to_lru_tail(struct ttm_buffer_object *bo,
dma_resv_assert_held(bo->base.resv);
- if (bo->pin_count)
+ if (bo->pin_count) {
+ ttm_bo_del_from_lru(bo);
return;
+ }
man = ttm_manager_type(bdev, mem->mem_type);
list_move_tail(&bo->lru, &man->lru[bo->priority]);
--
2.25.1
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2
2021-01-05 11:45 [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2 Christian König
@ 2021-01-05 18:12 ` Lyude Paul
2021-01-05 18:19 ` Christian König
2021-01-08 18:49 ` Mike Lothian
1 sibling, 1 reply; 5+ messages in thread
From: Lyude Paul @ 2021-01-05 18:12 UTC (permalink / raw)
To: Christian König, dri-devel
Reviewed-by: Lyude Paul <lyude@redhat.com>
Guessing it's fine if I push this with your sob and review added?
On Tue, 2021-01-05 at 12:45 +0100, Christian König wrote:
> From: Lyude Paul <lyude@redhat.com>
>
> Recently a regression was introduced which caused TTM's buffer eviction to
> attempt to evict already-pinned BOs, causing issues with buffer eviction
> under memory pressure along with suspend/resume:
>
> nouveau 0000:1f:00.0: DRM: evicting buffers...
> nouveau 0000:1f:00.0: DRM: Moving pinned object 00000000c428c3ff!
> nouveau 0000:1f:00.0: fifo: fault 00 [READ] at 0000000000200000 engine 04
> [BAR1] client 07 [HUB/HOST_CPU] reason 02 [PTE] on channel -1 [00ffeaa000
> unknown]
> nouveau 0000:1f:00.0: fifo: DROPPED_MMU_FAULT 00001000
> nouveau 0000:1f:00.0: fifo: fault 01 [WRITE] at 0000000000020000 engine
> 0c [HOST6] client 07 [HUB/HOST_CPU] reason 02 [PTE] on channel 1
> [00ffb28000 DRM]
> nouveau 0000:1f:00.0: fifo: channel 1: killed
> nouveau 0000:1f:00.0: fifo: runlist 0: scheduled for recovery
> [TTM] Buffer eviction failed
> nouveau 0000:1f:00.0: DRM: waiting for kernel channels to go idle...
> nouveau 0000:1f:00.0: DRM: failed to idle channel 1 [DRM]
> nouveau 0000:1f:00.0: DRM: resuming display...
>
> After some bisection and investigation, it appears this resulted from the
> recent changes to ttm_bo_move_to_lru_tail(). Previously when a buffer was
> pinned, the buffer would be removed from the LRU once ttm_bo_unreserve
> to maintain the LRU list when pinning or unpinning BOs. However, since:
>
> commit 3d1a88e1051f ("drm/ttm: cleanup LRU handling further")
>
> We've been exiting from ttm_bo_move_to_lru_tail() at the very beginning of
> the function if the bo we're looking at is pinned, resulting in the pinned
> BO never getting removed from the lru and as a result - causing issues when
> it eventually becomes time for eviction.
>
> So, let's fix this by calling ttm_bo_del_from_lru() from
> ttm_bo_move_to_lru_tail() in the event that we're dealing with a pinned
> buffer.
>
> v2 (chk): reduce to only the fixing one liner since we always want to
> call the callback whenever we would move on the LRU.
>
> Signed-off-by: Lyude Paul <lyude@redhat.com>
> Fixes: 3d1a88e1051f ("drm/ttm: cleanup LRU handling further")
> Cc: Christian König <christian.koenig@amd.com>
> Cc: Dave Airlie <airlied@redhat.com>
> ---
> drivers/gpu/drm/ttm/ttm_bo.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
> index 31e8b3da5563..b65f4b12f986 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
> @@ -138,8 +138,10 @@ void ttm_bo_move_to_lru_tail(struct ttm_buffer_object
> *bo,
>
> dma_resv_assert_held(bo->base.resv);
>
> - if (bo->pin_count)
> + if (bo->pin_count) {
> + ttm_bo_del_from_lru(bo);
> return;
> + }
>
> man = ttm_manager_type(bdev, mem->mem_type);
> list_move_tail(&bo->lru, &man->lru[bo->priority]);
--
Cheers,
Lyude Paul (she/her)
Software Engineer at Red Hat
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2
2021-01-05 18:12 ` Lyude Paul
@ 2021-01-05 18:19 ` Christian König
0 siblings, 0 replies; 5+ messages in thread
From: Christian König @ 2021-01-05 18:19 UTC (permalink / raw)
To: Lyude Paul, dri-devel
Am 05.01.21 um 19:12 schrieb Lyude Paul:
> Reviewed-by: Lyude Paul <lyude@redhat.com>
>
> Guessing it's fine if I push this with your sob and review added?
Works for me, just take care that you pick the right branch.
I always seem to push my stuff into the wrong one.
Christian.
>
> On Tue, 2021-01-05 at 12:45 +0100, Christian König wrote:
>> From: Lyude Paul <lyude@redhat.com>
>>
>> Recently a regression was introduced which caused TTM's buffer eviction to
>> attempt to evict already-pinned BOs, causing issues with buffer eviction
>> under memory pressure along with suspend/resume:
>>
>> nouveau 0000:1f:00.0: DRM: evicting buffers...
>> nouveau 0000:1f:00.0: DRM: Moving pinned object 00000000c428c3ff!
>> nouveau 0000:1f:00.0: fifo: fault 00 [READ] at 0000000000200000 engine 04
>> [BAR1] client 07 [HUB/HOST_CPU] reason 02 [PTE] on channel -1 [00ffeaa000
>> unknown]
>> nouveau 0000:1f:00.0: fifo: DROPPED_MMU_FAULT 00001000
>> nouveau 0000:1f:00.0: fifo: fault 01 [WRITE] at 0000000000020000 engine
>> 0c [HOST6] client 07 [HUB/HOST_CPU] reason 02 [PTE] on channel 1
>> [00ffb28000 DRM]
>> nouveau 0000:1f:00.0: fifo: channel 1: killed
>> nouveau 0000:1f:00.0: fifo: runlist 0: scheduled for recovery
>> [TTM] Buffer eviction failed
>> nouveau 0000:1f:00.0: DRM: waiting for kernel channels to go idle...
>> nouveau 0000:1f:00.0: DRM: failed to idle channel 1 [DRM]
>> nouveau 0000:1f:00.0: DRM: resuming display...
>>
>> After some bisection and investigation, it appears this resulted from the
>> recent changes to ttm_bo_move_to_lru_tail(). Previously when a buffer was
>> pinned, the buffer would be removed from the LRU once ttm_bo_unreserve
>> to maintain the LRU list when pinning or unpinning BOs. However, since:
>>
>> commit 3d1a88e1051f ("drm/ttm: cleanup LRU handling further")
>>
>> We've been exiting from ttm_bo_move_to_lru_tail() at the very beginning of
>> the function if the bo we're looking at is pinned, resulting in the pinned
>> BO never getting removed from the lru and as a result - causing issues when
>> it eventually becomes time for eviction.
>>
>> So, let's fix this by calling ttm_bo_del_from_lru() from
>> ttm_bo_move_to_lru_tail() in the event that we're dealing with a pinned
>> buffer.
>>
>> v2 (chk): reduce to only the fixing one liner since we always want to
>> call the callback whenever we would move on the LRU.
>>
>> Signed-off-by: Lyude Paul <lyude@redhat.com>
>> Fixes: 3d1a88e1051f ("drm/ttm: cleanup LRU handling further")
>> Cc: Christian König <christian.koenig@amd.com>
>> Cc: Dave Airlie <airlied@redhat.com>
>> ---
>> drivers/gpu/drm/ttm/ttm_bo.c | 4 +++-
>> 1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
>> index 31e8b3da5563..b65f4b12f986 100644
>> --- a/drivers/gpu/drm/ttm/ttm_bo.c
>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
>> @@ -138,8 +138,10 @@ void ttm_bo_move_to_lru_tail(struct ttm_buffer_object
>> *bo,
>>
>> dma_resv_assert_held(bo->base.resv);
>>
>> - if (bo->pin_count)
>> + if (bo->pin_count) {
>> + ttm_bo_del_from_lru(bo);
>> return;
>> + }
>>
>> man = ttm_manager_type(bdev, mem->mem_type);
>> list_move_tail(&bo->lru, &man->lru[bo->priority]);
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2
2021-01-05 11:45 [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2 Christian König
2021-01-05 18:12 ` Lyude Paul
@ 2021-01-08 18:49 ` Mike Lothian
2021-01-11 10:13 ` Christian König
1 sibling, 1 reply; 5+ messages in thread
From: Mike Lothian @ 2021-01-08 18:49 UTC (permalink / raw)
To: Christian König, Lyude; +Cc: Maling list - DRI developers
[-- Attachment #1.1: Type: text/plain, Size: 116 bytes --]
Hi
This breaks things for me on my Prime system
https://gitlab.freedesktop.org/drm/misc/-/issues/23
Cheers
Mike
[-- Attachment #1.2: Type: text/html, Size: 304 bytes --]
[-- Attachment #2: Type: text/plain, Size: 160 bytes --]
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2
2021-01-08 18:49 ` Mike Lothian
@ 2021-01-11 10:13 ` Christian König
0 siblings, 0 replies; 5+ messages in thread
From: Christian König @ 2021-01-11 10:13 UTC (permalink / raw)
To: Mike Lothian, Lyude; +Cc: Maling list - DRI developers
Am 08.01.21 um 19:49 schrieb Mike Lothian:
> Hi
>
> This breaks things for me on my Prime system
>
> https://gitlab.freedesktop.org/drm/misc/-/issues/23
This is most likely not correct.
The patch just fixes another bug which would break prime before it hits
the issue in the i915 driver.
Christian.
>
> Cheers
>
> Mike
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-01-11 10:13 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-01-05 11:45 [PATCH] drm/ttm: Remove pinned bos from LRU in ttm_bo_move_to_lru_tail() v2 Christian König
2021-01-05 18:12 ` Lyude Paul
2021-01-05 18:19 ` Christian König
2021-01-08 18:49 ` Mike Lothian
2021-01-11 10:13 ` 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.