[PATCH v4 5/5] dma-buf: Clarify that dma-buf sg lists are page aligned

Daniel Vetter daniel at ffwll.ch
Thu Oct 15 16:03:44 UTC 2020


On Wed, Oct 14, 2020 at 09:16:01AM -0700, Jianxin Xiong wrote:
> The dma-buf API have been used under the assumption that the sg lists
> returned from dma_buf_map_attachment() are fully page aligned. Lots of
> stuff can break otherwise all over the place. Clarify this in the
> documentation and add a check when DMA API debug is enabled.
> 
> Signed-off-by: Jianxin Xiong <jianxin.xiong at intel.com>

lgtm, thanks for creating this and giving it a spin.

I'll queue this up in drm-misc-next for 5.11, should show up in linux-next
after the merge window is closed.

Cheers, Daniel

> ---
>  drivers/dma-buf/dma-buf.c | 21 +++++++++++++++++++++
>  include/linux/dma-buf.h   |  3 ++-
>  2 files changed, 23 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
> index 844967f..7309c83 100644
> --- a/drivers/dma-buf/dma-buf.c
> +++ b/drivers/dma-buf/dma-buf.c
> @@ -851,6 +851,9 @@ void dma_buf_unpin(struct dma_buf_attachment *attach)
>   * Returns sg_table containing the scatterlist to be returned; returns ERR_PTR
>   * on error. May return -EINTR if it is interrupted by a signal.
>   *
> + * On success, the DMA addresses and lengths in the returned scatterlist are
> + * PAGE_SIZE aligned.
> + *
>   * A mapping must be unmapped by using dma_buf_unmap_attachment(). Note that
>   * the underlying backing storage is pinned for as long as a mapping exists,
>   * therefore users/importers should not hold onto a mapping for undue amounts of
> @@ -904,6 +907,24 @@ struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach,
>  		attach->dir = direction;
>  	}
>  
> +#ifdef CONFIG_DMA_API_DEBUG
> +	{
> +		struct scatterlist *sg;
> +		u64 addr;
> +		int len;
> +		int i;
> +
> +		for_each_sgtable_dma_sg(sg_table, sg, i) {
> +			addr = sg_dma_address(sg);
> +			len = sg_dma_len(sg);
> +			if (!PAGE_ALIGNED(addr) || !PAGE_ALIGNED(len)) {
> +				pr_debug("%s: addr %llx or len %x is not page aligned!\n",
> +					 __func__, addr, len);
> +			}
> +		}
> +	}
> +#endif /* CONFIG_DMA_API_DEBUG */
> +
>  	return sg_table;
>  }
>  EXPORT_SYMBOL_GPL(dma_buf_map_attachment);
> diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h
> index a2ca294e..4a5fa70 100644
> --- a/include/linux/dma-buf.h
> +++ b/include/linux/dma-buf.h
> @@ -145,7 +145,8 @@ struct dma_buf_ops {
>  	 *
>  	 * A &sg_table scatter list of or the backing storage of the DMA buffer,
>  	 * already mapped into the device address space of the &device attached
> -	 * with the provided &dma_buf_attachment.
> +	 * with the provided &dma_buf_attachment. The addresses and lengths in
> +	 * the scatter list are PAGE_SIZE aligned.
>  	 *
>  	 * On failure, returns a negative error value wrapped into a pointer.
>  	 * May also return -EINTR when a signal was received while being
> -- 
> 1.8.3.1
> 
> _______________________________________________
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel

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


More information about the dri-devel mailing list