[PATCH] drm/amdkfd: Only apply heavy-weight TLB flush on Aldebaran

[Felix Kuehling]

Am 2021-07-08 um 3:53 p.m. schrieb Eric Huang:
> It is to workaround HW bug on other Asics and based on
> reverting two commits:
>   drm/amdkfd: Add heavy-weight TLB flush after unmapping
>   drm/amdkfd: Add memory sync before TLB flush on unmap
>
> Signed-off-by: Eric Huang <jinhuieric.huang at amd.com>

Reviewed-by: Felix Kuehling <Felix.Kuehling at amd.com>


> ---
>  drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 37 +++++++++++++-----------
>  1 file changed, 20 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> index ebb4872c5a9d..5f2655cf0162 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> @@ -1773,26 +1773,29 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep,
>  	}
>  	mutex_unlock(&p->mutex);
>  
> -	err = amdgpu_amdkfd_gpuvm_sync_memory(dev->kgd, (struct kgd_mem *) mem, true);
> -	if (err) {
> -		pr_debug("Sync memory failed, wait interrupted by user signal\n");
> -		goto sync_memory_failed;
> -	}
> +	if (dev->device_info->asic_family == CHIP_ALDEBARAN) {
> +		err = amdgpu_amdkfd_gpuvm_sync_memory(dev->kgd,
> +				(struct kgd_mem *) mem, true);
>  
> -	/* Flush TLBs after waiting for the page table updates to complete */
> -	for (i = 0; i < args->n_devices; i++) {
> -		peer = kfd_device_by_id(devices_arr[i]);
> -		if (WARN_ON_ONCE(!peer))
> -			continue;
> -		peer_pdd = kfd_get_process_device_data(peer, p);
> -		if (WARN_ON_ONCE(!peer_pdd))
> -			continue;
> -		if (!amdgpu_read_lock(peer->ddev, true)) {
> -			kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT);
> -			amdgpu_read_unlock(peer->ddev);
> +		if (err) {
> +			pr_debug("Sync memory failed, wait interrupted by user signal\n");
> +			goto sync_memory_failed;
>  		}
> -	}
>  
> +		/* Flush TLBs after waiting for the page table updates to complete */
> +		for (i = 0; i < args->n_devices; i++) {
> +			peer = kfd_device_by_id(devices_arr[i]);
> +			if (WARN_ON_ONCE(!peer))
> +				continue;
> +			peer_pdd = kfd_get_process_device_data(peer, p);
> +			if (WARN_ON_ONCE(!peer_pdd))
> +				continue;
> +			if (!amdgpu_read_lock(peer->ddev, true)) {
> +				kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT);
> +				amdgpu_read_unlock(peer->ddev);
> +			}
> +		}
> +	}
>  	kfree(devices_arr);
>  
>  	return 0;