Re: [PATCH crypto 2/2] crypto: safexcel - Map AEAD buffers with accurate DMA directions

[Herbert Xu]

On Tue, Sep 15, 2026 at 01:32:12PM +0200, Ralf Lici wrote:
>
> @@ -748,8 +833,17 @@ static int safexcel_send_req(struct crypto_async_request *base, int ring,
>  			return -EINVAL;
>  		}
>  		if (sreq->nr_src > 0) {
> -			src_mapped = dma_map_sg(priv->dev, src, sreq->nr_src,
> -						DMA_BIDIRECTIONAL);
> +			if (ctx->aead)
> +				src_mapped =
> +					safexcel_map_aead(priv->dev, src,
> +							  sreq->nr_src,
> +							  sreq->output_offset,
> +							  sreq->output_len,
> +							  true);
> +			else
> +				src_mapped = dma_map_sg(priv->dev, src,
> +							sreq->nr_src,
> +							DMA_BIDIRECTIONAL);

This path doesn't actualy matter because BIDIRECTIONAL is safe
even if you use it unnecessarily.

> @@ -775,8 +869,17 @@ static int safexcel_send_req(struct crypto_async_request *base, int ring,
>  		}
>  
>  		if (sreq->nr_dst > 0) {
> -			dst_mapped = dma_map_sg(priv->dev, dst, sreq->nr_dst,
> -						DMA_FROM_DEVICE);

So this is the really buggy path.  Why don't you just change it to
BIDRECTIONAL, perhaps with a comment referring to the assoc data?

Thanks,
-- 
Email: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt