aste
you the crash log.
/Monk
-Original Message-
From: Koenig, Christian
Sent: 2018年2月28日 20:39
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Chris Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
Ok, I've figured out what is going on here. The
gfx@lists.freedesktop.org; Chris Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
Am 27.02.2018 um 04:36 schrieb Liu, Monk:
Well then there is something else broken and I actually think that this is the
root cause here.
What if this VM submits four different job
2018年2月27日 16:34
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Chris Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
Am 27.02.2018 um 04:36 schrieb Liu, Monk:
>> Well then there is something else broken and I actually think that this is
>> th
7;t think there
is other reason.
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
Of Christian K?nig
Sent: 2018年2月26日 18:45
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Chris
Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared
al Message-
From: Koenig, Christian
Sent: 2018年2月26日 19:19
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Chris Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
> What if all those context are not equal ? and accumulated finally to 4 ?
Well then there is s
: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of
Christian K?nig
Sent: 2018年2月26日 18:45
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Chris Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
We must call reserve_shared before amdgpu_bo_
Christian K?nig
Sent: 2018年2月26日 18:45
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Chris Wilson
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
> We must call reserve_shared before amdgpu_bo_fence
Actually that's not correct. See reservation_object_add_s
Quoting Christian König (2018-02-26 10:44:40)
> > We must call reserve_shared before amdgpu_bo_fence
> Actually that's not correct. See reservation_object_add_shared_fence()
> should replace the fence when it has the same context as a previously
> added fence.
>
> So we call reserve_shared o
v obj, so the
BUG() in
Reservation.c file will hit,
We must call reserve_shared before amdgpu_bo_fence
-Original Message-
From: Christian König [mailto:ckoenig.leichtzumer...@gmail.com]
Sent: 2018年2月26日 17:48
To: Liu, Monk ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH 22/2
k ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared count bug
That fix is incorrect.
amdgpu_vm_get_pd_bo() should already reserves a shared slot for the fence:
> entry->tv.shared = true;
So you must run into this issue because of somethi
vation.c's
reservation_object_add_shared_fence()
This is a clear fix !
-Original Message-
From: Christian König [mailto:ckoenig.leichtzumer...@gmail.com]
Sent: 2018年2月26日 17:48
To: Liu, Monk ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH 22/22] drm/amdgpu: fix reservation obj shared
That fix is incorrect.
amdgpu_vm_get_pd_bo() should already reserves a shared slot for the fence:
entry->tv.shared = true;
So you must run into this issue because of something else.
Regards,
Christian.
Am 26.02.2018 um 06:35 schrieb Monk Liu:
should call reservation_object_reserve_s
should call reservation_object_reserve_shared before
amdgpu_bo_fence(), otherwise there are odds kernel
hit BUG in reversation.c
bug:
[12622.076435] [ cut here ]
[12622.076438] kernel BUG at drivers/dma-buf/reservation.c:233!
[12622.078046] invalid opcode: [#1] SMP KAS
13 matches
Mail list logo
