Re: [PATCH] dma-buf: cma_heap: Fix mutex locking section

[Daniel Vetter <daniel@ffwll.ch>]

On Tue, Jan 04, 2022 at 03:35:45PM +0800, Weizhao Ouyang wrote:
> Fix cma_heap_buffer mutex locking critical section to protect vmap_cnt
> and vaddr.
> 
> Fixes: a5d2d29e24be ("dma-buf: heaps: Move heap-helper logic into the cma_heap implementation")
> Signed-off-by: Weizhao Ouyang <o451686892@gmail.com>
> ---
>  drivers/dma-buf/heaps/cma_heap.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/dma-buf/heaps/cma_heap.c b/drivers/dma-buf/heaps/cma_heap.c
> index 0c05b79870f9..83f02bd51dda 100644
> --- a/drivers/dma-buf/heaps/cma_heap.c
> +++ b/drivers/dma-buf/heaps/cma_heap.c
> @@ -124,10 +124,11 @@ static int cma_heap_dma_buf_begin_cpu_access(struct dma_buf *dmabuf,
>  	struct cma_heap_buffer *buffer = dmabuf->priv;
>  	struct dma_heap_attachment *a;
>  
> +	mutex_lock(&buffer->lock);
> +
>  	if (buffer->vmap_cnt)
>  		invalidate_kernel_vmap_range(buffer->vaddr, buffer->len);

Since this creates nesting with mm/, but optionally I think it'd be good
to prime lockdep so it knows about this. See e.g. dma_resv_lockdep() in
dma-resv.c, except I don't know offhand what the right lock for
invalidate_kernel_vmap_range is.
-Daniel


>  
> -	mutex_lock(&buffer->lock);
>  	list_for_each_entry(a, &buffer->attachments, list) {
>  		if (!a->mapped)
>  			continue;
> @@ -144,10 +145,11 @@ static int cma_heap_dma_buf_end_cpu_access(struct dma_buf *dmabuf,
>  	struct cma_heap_buffer *buffer = dmabuf->priv;
>  	struct dma_heap_attachment *a;
>  
> +	mutex_lock(&buffer->lock);
> +
>  	if (buffer->vmap_cnt)
>  		flush_kernel_vmap_range(buffer->vaddr, buffer->len);
>  
> -	mutex_lock(&buffer->lock);
>  	list_for_each_entry(a, &buffer->attachments, list) {
>  		if (!a->mapped)
>  			continue;
> -- 
> 2.32.0
> 

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch