Hi Christian,
Are you going to submit this change to amd-staging-drm-next?
amd-kfd-staging would pick it up from there automatically.
Regards,
Felix
On 2018-08-15 01:57 PM, Felix Kuehling wrote:
> I applied your change to my local KFD staging branch and it through a
> presubmission build/test
Am 16.08.2018 um 20:43 schrieb Felix Kuehling:
On 2018-08-16 02:26 PM, Christian König wrote:
Am 16.08.2018 um 20:23 schrieb Felix Kuehling:
On 2018-08-16 02:18 PM, Christian König wrote:
Am 16.08.2018 um 18:50 schrieb Felix Kuehling:
On 2018-08-16 02:43 AM, Christian König wrote:
[SNIP]
I
On 2018-08-16 02:26 PM, Christian König wrote:
> Am 16.08.2018 um 20:23 schrieb Felix Kuehling:
>> On 2018-08-16 02:18 PM, Christian König wrote:
>>> Am 16.08.2018 um 18:50 schrieb Felix Kuehling:
On 2018-08-16 02:43 AM, Christian König wrote:
[SNIP]
> I mean it could be that in the
Am 16.08.2018 um 20:23 schrieb Felix Kuehling:
On 2018-08-16 02:18 PM, Christian König wrote:
Am 16.08.2018 um 18:50 schrieb Felix Kuehling:
On 2018-08-16 02:43 AM, Christian König wrote:
[SNIP]
I mean it could be that in the worst case we race and stop a KFD
process for no good reason.
Right
On 2018-08-16 02:18 PM, Christian König wrote:
> Am 16.08.2018 um 18:50 schrieb Felix Kuehling:
>> On 2018-08-16 02:43 AM, Christian König wrote:
>> [SNIP]
>>> I mean it could be that in the worst case we race and stop a KFD
>>> process for no good reason.
>> Right. For a more practical example, a
Am 16.08.2018 um 18:50 schrieb Felix Kuehling:
On 2018-08-16 02:43 AM, Christian König wrote:
[SNIP]
I mean it could be that in the worst case we race and stop a KFD
process for no good reason.
Right. For a more practical example, a KFD BO can get evicted just
before the application decides to
On 2018-08-16 02:43 AM, Christian König wrote:
> Am 15.08.2018 um 20:49 schrieb Felix Kuehling:
>> On 2018-08-15 02:27 PM, Christian König wrote:
>>> Am 15.08.2018 um 20:17 schrieb Felix Kuehling:
On 2018-08-15 03:02 AM, Christian König wrote:
> Hi Felix,
>
> yeah, you pretty much
Am 15.08.2018 um 20:49 schrieb Felix Kuehling:
On 2018-08-15 02:27 PM, Christian König wrote:
Am 15.08.2018 um 20:17 schrieb Felix Kuehling:
On 2018-08-15 03:02 AM, Christian König wrote:
Hi Felix,
yeah, you pretty much nailed it.
The problem is that the array itself is RCU protected. This m
On 2018-08-15 02:27 PM, Christian König wrote:
> Am 15.08.2018 um 20:17 schrieb Felix Kuehling:
>> On 2018-08-15 03:02 AM, Christian König wrote:
>>> Hi Felix,
>>>
>>> yeah, you pretty much nailed it.
>>>
>>> The problem is that the array itself is RCU protected. This means that
>>> you can only c
Am 15.08.2018 um 20:17 schrieb Felix Kuehling:
On 2018-08-15 03:02 AM, Christian König wrote:
Hi Felix,
yeah, you pretty much nailed it.
The problem is that the array itself is RCU protected. This means that
you can only copy the whole structure when you want to update it.
The exception is re
On 2018-08-15 03:02 AM, Christian König wrote:
> Hi Felix,
>
> yeah, you pretty much nailed it.
>
> The problem is that the array itself is RCU protected. This means that
> you can only copy the whole structure when you want to update it.
>
> The exception is reservation_object_add_shared() which o
I applied your change to my local KFD staging branch and it through a
presubmission build/test (sorry, only accessible from inside AMD):
http://git.amd.com:8080/#/c/167906/
It passed, but checkpatch pointed out one issue:
http://git.amd.com:8080/#/c/167906/1/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkf
Fix quite a number of bugs here. Unfortunately only compile tested.
v2: fix copy&paste error
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 103 ++-
1 file changed, 46 insertions(+), 57 deletions(-)
diff --git a/drivers/gpu/drm/amd/amd
Hi Felix,
yeah, you pretty much nailed it.
The problem is that the array itself is RCU protected. This means that
you can only copy the whole structure when you want to update it.
The exception is reservation_object_add_shared() which only works
because we replace an either signaled fence or
[+Harish]
I think this looks good for the most part. See one comment inline below.
But bear with me while I'm trying to understand what was wrong with the
old code. Please correct me if I get this wrong or point out anything
I'm missing.
The reservation_object_list looks to be protected by a com
Fix quite a number of bugs here. Unfortunately only compile tested.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 103 ++-
1 file changed, 46 insertions(+), 57 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
16 matches
Mail list logo
