Re: [PATCH v1 06/14] drm/amdkfd: add SPM support for SVM

[Felix Kuehling <felix.kuehling@amd.com>]

Am 2021-08-24 um 11:48 p.m. schrieb Alex Sierra:
> When CPU is connected throug XGMI, it has coherent
> access to VRAM resource. In this case that resource
> is taken from a table in the device gmc aperture base.
> This resource is used along with the device type, which could
> be DEVICE_PRIVATE or DEVICE_PUBLIC to create the device
> page map region.
>
> Signed-off-by: Alex Sierra <alex.sierra@amd.com>
> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
> ---
> v7:
> Remove lookup_resource call, so export symbol for this function
> is not longer required. Patch dropped "kernel: resource:
> lookup_resource as exported symbol"
> ---
>  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 29 +++++++++++++++---------
>  1 file changed, 18 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> index 47ee9a895cd2..dd245699479f 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> @@ -865,7 +865,7 @@ int svm_migrate_init(struct amdgpu_device *adev)
>  {
>  	struct kfd_dev *kfddev = adev->kfd.dev;
>  	struct dev_pagemap *pgmap;
> -	struct resource *res;
> +	struct resource *res = NULL;
>  	unsigned long size;
>  	void *r;
>  
> @@ -880,19 +880,25 @@ int svm_migrate_init(struct amdgpu_device *adev)
>  	 * should remove reserved size
>  	 */
>  	size = ALIGN(adev->gmc.real_vram_size, 2ULL << 20);
> -	res = devm_request_free_mem_region(adev->dev, &iomem_resource, size);
> -	if (IS_ERR(res))
> -		return -ENOMEM;
> +	if (adev->gmc.xgmi.connected_to_cpu) {
> +		pgmap->range.start = adev->gmc.aper_base;
> +		pgmap->range.end = adev->gmc.aper_base + adev->gmc.aper_size - 1;
> +		pgmap->type = MEMORY_DEVICE_PUBLIC;
> +	} else {
> +		res = devm_request_free_mem_region(adev->dev, &iomem_resource, size);
> +		if (IS_ERR(res))
> +			return -ENOMEM;
> +		pgmap->range.start = res->start;
> +		pgmap->range.end = res->end;
> +		pgmap->type = MEMORY_DEVICE_PRIVATE;
> +	}
>  
> -	pgmap->type = MEMORY_DEVICE_PRIVATE;
>  	pgmap->nr_range = 1;
> -	pgmap->range.start = res->start;
> -	pgmap->range.end = res->end;
>  	pgmap->ops = &svm_migrate_pgmap_ops;
>  	pgmap->owner = SVM_ADEV_PGMAP_OWNER(adev);
> -	pgmap->flags = MIGRATE_VMA_SELECT_DEVICE_PRIVATE;
> +	pgmap->flags = 0;
>  	r = devm_memremap_pages(adev->dev, pgmap);
> -	if (IS_ERR(r)) {
> +	if (res && IS_ERR(r)) {

I think the (res && ...) condition means you only detect failures for
DEVICE_PRIVATE memory. Why are you ignoring failures for DEVICE_PUBLIC?

For DEVICE_PUBLIC you can skip devm_release_mem_region, but you still
need to detect and return the error. Also, using res as an indicator is
a bit obscure. I'd put an if (pgmap->type == MEMORY_DEVICE_PRIVATE)
before the devm_release_mem_region call.

Regards,
  Felix


>  		pr_err("failed to register HMM device memory\n");
>  		devm_release_mem_region(adev->dev, res->start,
>  					res->end - res->start + 1);
> @@ -914,6 +920,7 @@ void svm_migrate_fini(struct amdgpu_device *adev)
>  	struct dev_pagemap *pgmap = &adev->kfd.dev->pgmap;
>  
>  	devm_memunmap_pages(adev->dev, pgmap);
> -	devm_release_mem_region(adev->dev, pgmap->range.start,
> -				pgmap->range.end - pgmap->range.start + 1);
> +	if (pgmap->type == MEMORY_DEVICE_PRIVATE)
> +		devm_release_mem_region(adev->dev, pgmap->range.start,
> +					pgmap->range.end - pgmap->range.start + 1);
>  }