* [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
@ 2017-08-31 7:45 Christian König
[not found] ` <1504165548-2017-1-git-send-email-deathsimple-ANTagKRnAhcb1SvskN2V4Q@public.gmane.org>
0 siblings, 1 reply; 5+ messages in thread
From: Christian König @ 2017-08-31 7:45 UTC (permalink / raw)
To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW
From: Christian König <christian.koenig@amd.com>
The IB allocation sizes where correct.
This reverts commit d1551e7c7211fa135c4d79f09e237f5f9779e421.
Signed-off-by: Christian König <christian.koenig@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 473273d..b08f031 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1037,7 +1037,7 @@ static int amdgpu_vm_update_level(struct amdgpu_device *adev,
ndw = 64;
/* assume the worst case */
- ndw += parent->last_entry_used * 10;
+ ndw += parent->last_entry_used * 6;
pd_addr = amdgpu_bo_gpu_offset(parent->base.bo);
@@ -1542,7 +1542,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
ndw = 64;
/* one PDE write for each huge page */
- ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 10;
+ ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 6;
if (pages_addr) {
/* copy commands needed */
--
2.7.4
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
[not found] ` <1504165548-2017-1-git-send-email-deathsimple-ANTagKRnAhcb1SvskN2V4Q@public.gmane.org>
@ 2017-08-31 13:42 ` Kuehling, Felix
[not found] ` <DM5PR1201MB0235F7D990F5C465305A6CE0929D0-grEf7a3NxMBd8L2jMOIKKmrFom/aUZj6nBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
0 siblings, 1 reply; 5+ messages in thread
From: Kuehling, Felix @ 2017-08-31 13:42 UTC (permalink / raw)
To: Christian König, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Sorry, I didn't think this one through. I'll keep looking for the real problem.
________________________________________
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> on behalf of Christian König <deathsimple@vodafone.de>
Sent: Thursday, August 31, 2017 3:45:48 AM
To: amd-gfx@lists.freedesktop.org
Subject: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
From: Christian König <christian.koenig@amd.com>
The IB allocation sizes where correct.
This reverts commit d1551e7c7211fa135c4d79f09e237f5f9779e421.
Signed-off-by: Christian König <christian.koenig@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 473273d..b08f031 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1037,7 +1037,7 @@ static int amdgpu_vm_update_level(struct amdgpu_device *adev,
ndw = 64;
/* assume the worst case */
- ndw += parent->last_entry_used * 10;
+ ndw += parent->last_entry_used * 6;
pd_addr = amdgpu_bo_gpu_offset(parent->base.bo);
@@ -1542,7 +1542,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
ndw = 64;
/* one PDE write for each huge page */
- ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 10;
+ ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 6;
if (pages_addr) {
/* copy commands needed */
--
2.7.4
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
[not found] ` <DM5PR1201MB0235F7D990F5C465305A6CE0929D0-grEf7a3NxMBd8L2jMOIKKmrFom/aUZj6nBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
@ 2017-08-31 13:59 ` Christian König
[not found] ` <bb71139f-f95c-8f1c-ebd9-f67679ea9b8e-ANTagKRnAhcb1SvskN2V4Q@public.gmane.org>
0 siblings, 1 reply; 5+ messages in thread
From: Christian König @ 2017-08-31 13:59 UTC (permalink / raw)
To: Kuehling, Felix, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW
Am 31.08.2017 um 15:42 schrieb Kuehling, Felix:
> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Thanks.
> Sorry, I didn't think this one through. I'll keep looking for the real problem.
Is it possible that you merged my earlier patch to handle all fragment
sizes?
Cause earlier versions had exactly that problem.
Christian.
> ________________________________________
> From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> on behalf of Christian König <deathsimple@vodafone.de>
> Sent: Thursday, August 31, 2017 3:45:48 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
>
> From: Christian König <christian.koenig@amd.com>
>
> The IB allocation sizes where correct.
>
> This reverts commit d1551e7c7211fa135c4d79f09e237f5f9779e421.
>
> Signed-off-by: Christian König <christian.koenig@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> index 473273d..b08f031 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> @@ -1037,7 +1037,7 @@ static int amdgpu_vm_update_level(struct amdgpu_device *adev,
> ndw = 64;
>
> /* assume the worst case */
> - ndw += parent->last_entry_used * 10;
> + ndw += parent->last_entry_used * 6;
>
> pd_addr = amdgpu_bo_gpu_offset(parent->base.bo);
>
> @@ -1542,7 +1542,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
> ndw = 64;
>
> /* one PDE write for each huge page */
> - ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 10;
> + ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 6;
>
> if (pages_addr) {
> /* copy commands needed */
> --
> 2.7.4
>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
> _______________________________________________
> amd-gfx mailing list
> amd-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
[not found] ` <bb71139f-f95c-8f1c-ebd9-f67679ea9b8e-ANTagKRnAhcb1SvskN2V4Q@public.gmane.org>
@ 2017-08-31 15:50 ` Felix Kuehling
[not found] ` <711b8e15-8826-5a5e-f994-4ae4a0c6baa7-5C7GfCeVMHo@public.gmane.org>
0 siblings, 1 reply; 5+ messages in thread
From: Felix Kuehling @ 2017-08-31 15:50 UTC (permalink / raw)
To: Christian König, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW
On 2017-08-31 09:59 AM, Christian König wrote:
> Am 31.08.2017 um 15:42 schrieb Kuehling, Felix:
>> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
>
> Thanks.
>
>> Sorry, I didn't think this one through. I'll keep looking for the
>> real problem.
>
> Is it possible that you merged my earlier patch to handle all fragment
> sizes?
>
> Cause earlier versions had exactly that problem.
I don't think so. We merge from amd-staging-4.12. Was your earlier patch
already submitted to the branch?
Regards,
Felix
>
> Christian.
>
>> ________________________________________
>> From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> on behalf of
>> Christian König <deathsimple@vodafone.de>
>> Sent: Thursday, August 31, 2017 3:45:48 AM
>> To: amd-gfx@lists.freedesktop.org
>> Subject: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page
>> table updates"
>>
>> From: Christian König <christian.koenig@amd.com>
>>
>> The IB allocation sizes where correct.
>>
>> This reverts commit d1551e7c7211fa135c4d79f09e237f5f9779e421.
>>
>> Signed-off-by: Christian König <christian.koenig@amd.com>
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> index 473273d..b08f031 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> @@ -1037,7 +1037,7 @@ static int amdgpu_vm_update_level(struct
>> amdgpu_device *adev,
>> ndw = 64;
>>
>> /* assume the worst case */
>> - ndw += parent->last_entry_used * 10;
>> + ndw += parent->last_entry_used * 6;
>>
>> pd_addr = amdgpu_bo_gpu_offset(parent->base.bo);
>>
>> @@ -1542,7 +1542,7 @@ static int amdgpu_vm_bo_update_mapping(struct
>> amdgpu_device *adev,
>> ndw = 64;
>>
>> /* one PDE write for each huge page */
>> - ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 10;
>> + ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 6;
>>
>> if (pages_addr) {
>> /* copy commands needed */
>> --
>> 2.7.4
>>
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>
>
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates"
[not found] ` <711b8e15-8826-5a5e-f994-4ae4a0c6baa7-5C7GfCeVMHo@public.gmane.org>
@ 2017-08-31 15:54 ` Christian König
0 siblings, 0 replies; 5+ messages in thread
From: Christian König @ 2017-08-31 15:54 UTC (permalink / raw)
To: Felix Kuehling, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW
Am 31.08.2017 um 17:50 schrieb Felix Kuehling:
> On 2017-08-31 09:59 AM, Christian König wrote:
>> Am 31.08.2017 um 15:42 schrieb Kuehling, Felix:
>>> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
>> Thanks.
>>
>>> Sorry, I didn't think this one through. I'll keep looking for the
>>> real problem.
>> Is it possible that you merged my earlier patch to handle all fragment
>> sizes?
>>
>> Cause earlier versions had exactly that problem.
> I don't think so. We merge from amd-staging-4.12. Was your earlier patch
> already submitted to the branch?
No, I haven't committed it yet. Just though that Roger might have merged
that in the hybrid branch and you picked it up there.
Christian.
>
> Regards,
> Felix
>
>> Christian.
>>
>>> ________________________________________
>>> From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> on behalf of
>>> Christian König <deathsimple@vodafone.de>
>>> Sent: Thursday, August 31, 2017 3:45:48 AM
>>> To: amd-gfx@lists.freedesktop.org
>>> Subject: [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page
>>> table updates"
>>>
>>> From: Christian König <christian.koenig@amd.com>
>>>
>>> The IB allocation sizes where correct.
>>>
>>> This reverts commit d1551e7c7211fa135c4d79f09e237f5f9779e421.
>>>
>>> Signed-off-by: Christian König <christian.koenig@amd.com>
>>> ---
>>> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++--
>>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>>> index 473273d..b08f031 100644
>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>>> @@ -1037,7 +1037,7 @@ static int amdgpu_vm_update_level(struct
>>> amdgpu_device *adev,
>>> ndw = 64;
>>>
>>> /* assume the worst case */
>>> - ndw += parent->last_entry_used * 10;
>>> + ndw += parent->last_entry_used * 6;
>>>
>>> pd_addr = amdgpu_bo_gpu_offset(parent->base.bo);
>>>
>>> @@ -1542,7 +1542,7 @@ static int amdgpu_vm_bo_update_mapping(struct
>>> amdgpu_device *adev,
>>> ndw = 64;
>>>
>>> /* one PDE write for each huge page */
>>> - ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 10;
>>> + ndw += ((nptes >> adev->vm_manager.block_size) + 1) * 6;
>>>
>>> if (pages_addr) {
>>> /* copy commands needed */
>>> --
>>> 2.7.4
>>>
>>> _______________________________________________
>>> amd-gfx mailing list
>>> amd-gfx@lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>>> _______________________________________________
>>> amd-gfx mailing list
>>> amd-gfx@lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-08-31 15:54 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-08-31 7:45 [PATCH] drm/amdgpu: Revert "Fix IB allocation sizes for page table updates" Christian König
[not found] ` <1504165548-2017-1-git-send-email-deathsimple-ANTagKRnAhcb1SvskN2V4Q@public.gmane.org>
2017-08-31 13:42 ` Kuehling, Felix
[not found] ` <DM5PR1201MB0235F7D990F5C465305A6CE0929D0-grEf7a3NxMBd8L2jMOIKKmrFom/aUZj6nBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
2017-08-31 13:59 ` Christian König
[not found] ` <bb71139f-f95c-8f1c-ebd9-f67679ea9b8e-ANTagKRnAhcb1SvskN2V4Q@public.gmane.org>
2017-08-31 15:50 ` Felix Kuehling
[not found] ` <711b8e15-8826-5a5e-f994-4ae4a0c6baa7-5C7GfCeVMHo@public.gmane.org>
2017-08-31 15:54 ` 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.