Re: [RFC 2/2] drivers/iommu: Unroll unsuccessful mapping

[Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>]

Hi Krzysztof,

On Wed, 2026-05-27 at 14:08 +0000, Krzysztof Karas wrote:
> Currently, if iommu maps fewer bytes than requested (iova_len),
> it proceeds to free the iova, but never tries to unmap already
> touched bytes. This behavior may cause memory hogging down the
> line.
> 
> Correct that by unmapping before exiting.
> 
> Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
> ---
>  drivers/iommu/dma-iommu.c | 10 ++++++++--

You have to submit this patch to IOMMU DMA-API LAYER maintainers (see
MAINTAINERS file for list and maintainer addresses) in order to get binding
comments, I believe.

Thanks,
Janusz

>  1 file changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index 54d96e847f16..64ac2bdfc574 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -1515,8 +1515,14 @@ int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
>  	 * implementation - it knows better than we do.
>  	 */
>  	ret = iommu_map_sg(domain, iova, sg, nents, prot, GFP_ATOMIC);
> -	if (ret < 0 || ret < iova_len)
> +	if (ret < 0 || ret < iova_len) {
> +		if (ret > 0) {
> +			/* Unmap partially mapped bytes before freeing IOVA */
> +			if (iommu_unmap(domain, iova, ret) != ret)
> +				ret = -EIO;
> +		}
>  		goto out_free_iova;
> +	}
>  
>  	return __finalise_sg(dev, sg, nents, iova);
>  
> @@ -1525,7 +1531,7 @@ int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
>  out_restore_sg:
>  	__invalidate_sg(sg, nents);
>  out:
> -	if (ret != -ENOMEM && ret != -EREMOTEIO)
> +	if (ret != -ENOMEM && ret != -EREMOTEIO && ret != -EIO)
>  		return -EINVAL;
>  	return ret;
>  }