Re: [PATCH] drm/amdgpu: Add preferred_domain check when determine XGMI state

From: "Christian König" <ckoenig.leichtzumerken-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: "Kuehling, Felix" <Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>,
	"Liu, Shaoyun" <Shaoyun.Liu-5C7GfCeVMHo@public.gmane.org>,
	"amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
	<amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>
Subject: Re: [PATCH] drm/amdgpu: Add preferred_domain check when determine XGMI state
Date: Fri, 29 Mar 2019 11:50:37 +0100	[thread overview]
Message-ID: <96d87bd4-2344-8ef5-842d-806e59f2106a@gmail.com> (raw)
In-Reply-To: <a5df1503-edc9-f744-c340-4b45b243f29b-5C7GfCeVMHo@public.gmane.org>

Am 28.03.19 um 21:39 schrieb Kuehling, Felix:
> On 2019-03-28 4:38 p.m., Liu, Shaoyun wrote:
>> Avoid unnecessary XGMI hight pstate trigger when mapping none-vram memory for peer device
>>
>> Change-Id: I1881deff3da19f1f4b58d5765db03a590092a5b2
>> Signed-off-by: shaoyunl <shaoyun.liu@amd.com>
> This patch is Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
>
> Please also give Christian a chance to review this one before you submit.

Looks good to me as well, feel free to also add my rb.

Christian.

>
> Thanks,
>     Felix
>
>
>> ---
>>    drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 11 +++++++++++
>>    drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c  |  3 ++-
>>    2 files changed, 13 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
>> index 9ee8d7a..7b84036 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
>> @@ -31,6 +31,7 @@
>>    #include <drm/amdgpu_drm.h>
>>    #include "amdgpu.h"
>>    #include "amdgpu_display.h"
>> +#include "amdgpu_xgmi.h"
>>    
>>    void amdgpu_gem_object_free(struct drm_gem_object *gobj)
>>    {
>> @@ -666,6 +667,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
>>    	struct amdgpu_device *adev = dev->dev_private;
>>    	struct drm_amdgpu_gem_op *args = data;
>>    	struct drm_gem_object *gobj;
>> +	struct amdgpu_vm_bo_base *base;
>>    	struct amdgpu_bo *robj;
>>    	int r;
>>    
>> @@ -704,6 +706,15 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
>>    			amdgpu_bo_unreserve(robj);
>>    			break;
>>    		}
>> +		for (base = robj->vm_bo; base; base = base->next)
>> +			if (amdgpu_xgmi_same_hive(amdgpu_ttm_adev(robj->tbo.bdev),
>> +				amdgpu_ttm_adev(base->vm->root.base.bo->tbo.bdev))) {
>> +				r = -EINVAL;
>> +				amdgpu_bo_unreserve(robj);
>> +				goto out;
>> +			}
>> +
>> +
>>    		robj->preferred_domains = args->value & (AMDGPU_GEM_DOMAIN_VRAM |
>>    							AMDGPU_GEM_DOMAIN_GTT |
>>    							AMDGPU_GEM_DOMAIN_CPU);
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> index 3d221f0..eb242a1 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> @@ -2025,7 +2025,8 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
>>    	INIT_LIST_HEAD(&bo_va->valids);
>>    	INIT_LIST_HEAD(&bo_va->invalids);
>>    
>> -	if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev))) {
>> +	if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev)) &&
>> +	    (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM)) {
>>    		bo_va->is_xgmi = true;
>>    		mutex_lock(&adev->vm_manager.lock_pstate);
>>    		/* Power up XGMI if it can be potentially used */
> _______________________________________________
> 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