Re: [PATCH v2 5/8] drm/panfrost: Handle page mapping failure

[Boris Brezillon <boris.brezillon@collabora.com>]

On Thu, 28 Nov 2024 21:06:20 +0000
Adrián Larumbe <adrian.larumbe@collabora.com> wrote:

> When mapping the pages of a BO, either a heap type at page fault time or
> else a non-heap BO at object creation time, if the ARM page table mapping
> function fails, we unmap what had been mapped so far and bail out.
> 
> Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com>
> ---
>  drivers/gpu/drm/panfrost/panfrost_mmu.c | 44 ++++++++++++++++++++++---
>  1 file changed, 39 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c
> index 03ac527b35e7..5e30888bea0e 100644
> --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c
> +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c
> @@ -290,13 +290,31 @@ static void panfrost_mmu_flush_range(struct panfrost_device *pfdev,
>  	pm_runtime_put_autosuspend(pfdev->base.dev);
>  }
>  
> +static void mmu_unmap_range(size_t len, u64 iova, struct io_pgtable_ops *ops)

Can we use the following prototype instead?

static void mmu_unmap_range(struct panfrost_mmu *mmu,
			    u64 iova, size_t len)

> +{
> +	size_t pgsize, unmapped_len = 0;
> +	size_t unmapped_page, pgcount;
> +
> +	while (unmapped_len < len) {
> +		pgsize = get_pgsize(iova, len - unmapped_len, &pgcount);
> +
> +		unmapped_page = ops->unmap_pages(ops, iova, pgsize, pgcount, NULL);
> +		WARN_ON(unmapped_page != pgsize * pgcount);
> +
> +		iova += pgsize * pgcount;
> +		unmapped_len += pgsize * pgcount;
> +	}
> +}
> +
>  static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu,
>  		      u64 iova, int prot, struct sg_table *sgt)
>  {
>  	unsigned int count;
>  	struct scatterlist *sgl;
>  	struct io_pgtable_ops *ops = mmu->pgtbl_ops;
> +	size_t total_mapped = 0;
>  	u64 start_iova = iova;
> +	int ret = 0;
>  
>  	for_each_sgtable_dma_sg(sgt, sgl, count) {
>  		unsigned long paddr = sg_dma_address(sgl);
> @@ -310,10 +328,15 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu,
>  			size_t pgcount, mapped = 0;
>  			size_t pgsize = get_pgsize(iova | paddr, len, &pgcount);
>  
> -			ops->map_pages(ops, iova, paddr, pgsize, pgcount, prot,
> +			ret = ops->map_pages(ops, iova, paddr, pgsize, pgcount, prot,
>  				       GFP_KERNEL, &mapped);
> +			if (ret) {
> +				mmu_unmap_range(total_mapped, start_iova, ops);
> +				return ret;

I general, I'm not a huge fan of those error paths where things are
undone manually, because they tend to be overlooked when new steps are
added. I'd rather have a  "goto error_unmap_pages;" here, and the
according label defined at the end of the function.

> +			}
>  			/* Don't get stuck if things have gone wrong */
>  			mapped = max(mapped, pgsize);
> +			total_mapped += mapped;
>  			iova += mapped;
>  			paddr += mapped;
>  			len -= mapped;
> @@ -333,6 +356,7 @@ int panfrost_mmu_map(struct panfrost_gem_mapping *mapping)
>  	struct panfrost_device *pfdev = to_panfrost_device(obj->dev);
>  	struct sg_table *sgt;
>  	int prot = IOMMU_READ | IOMMU_WRITE;
> +	int ret;
>  
>  	if (WARN_ON(mapping->active))
>  		return 0;
> @@ -344,8 +368,13 @@ int panfrost_mmu_map(struct panfrost_gem_mapping *mapping)
>  	if (WARN_ON(IS_ERR(sgt)))
>  		return PTR_ERR(sgt);
>  
> -	mmu_map_sg(pfdev, mapping->mmu, mapping->mmnode.start << PAGE_SHIFT,
> -		   prot, sgt);
> +	ret = mmu_map_sg(pfdev, mapping->mmu, mapping->mmnode.start << PAGE_SHIFT,
> +			 prot, sgt);
> +	if (ret) {
> +		drm_gem_shmem_put_pages(shmem);

Same here.

> +		return ret;
> +	}
> +
>  	mapping->active = true;
>  
>  	return 0;
> @@ -532,8 +561,10 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as,
>  	if (ret)
>  		goto err_map;
>  
> -	mmu_map_sg(pfdev, bomapping->mmu, addr,
> -		   IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt);
> +	ret = mmu_map_sg(pfdev, bomapping->mmu, addr,
> +			 IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt);
> +	if (ret)
> +		goto err_mmu_map_sg;
>  
>  	bomapping->active = true;
>  	bo->heap_rss_size += SZ_2M;
> @@ -547,6 +578,9 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as,
>  
>  	return 0;
>  
> +err_mmu_map_sg:
> +	dma_unmap_sgtable(pfdev->base.dev, sgt,
> +			  DMA_BIDIRECTIONAL, 0);
>  err_map:
>  	sg_free_table(sgt);
>  err_unlock: