public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: m.szyprowski@samsung•com (Marek Szyprowski)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH 3/9] ARM: dma-mapping: Always pass proper prot flags to iommu_map()
Date: Mon, 30 Sep 2013 15:40:58 +0200	[thread overview]
Message-ID: <52497F6A.2090505@samsung.com> (raw)
In-Reply-To: <1380234982-1677-4-git-send-email-andreas.herrmann@calxeda.com>

Hello,

On 2013-09-27 00:36, Andreas Herrmann wrote:
> ... otherwise it is impossible for the low level iommu driver to
> figure out which pte flags should be used.
>
> In __map_sg_chunk we can derive the flags from dma_data_direction.
>
> In __iommu_create_mapping we should treat the memory like
> DMA_BIDIRECTIONAL and pass both IOMMU_READ and IOMMU_WRITE to
> iommu_map.
> __iommu_create_mapping is used during dma_alloc_coherent (via
> arm_iommu_alloc_attrs).  AFAIK dma_alloc_coherent is responsible for
> allocation _and_ mapping.  I think this implies that access to the
> mapped pages should be allowed.
>
> Cc: Marek Szyprowski <m.szyprowski@samsung•com>
> Signed-off-by: Andreas Herrmann <andreas.herrmann@calxeda•com>

Thanks pointing the issue and preparing the patch. I will push it to the 
dma-mapping fixes branch.

> ---
>   arch/arm/mm/dma-mapping.c |   43 ++++++++++++++++++++++++++++---------------
>   1 file changed, 28 insertions(+), 15 deletions(-)
>
> diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
> index f5e1a84..1272ed2 100644
> --- a/arch/arm/mm/dma-mapping.c
> +++ b/arch/arm/mm/dma-mapping.c
> @@ -1232,7 +1232,8 @@ __iommu_create_mapping(struct device *dev, struct page **pages, size_t size)
>   				break;
>   
>   		len = (j - i) << PAGE_SHIFT;
> -		ret = iommu_map(mapping->domain, iova, phys, len, 0);
> +		ret = iommu_map(mapping->domain, iova, phys, len,
> +				IOMMU_READ|IOMMU_WRITE);
>   		if (ret < 0)
>   			goto fail;
>   		iova += len;
> @@ -1431,6 +1432,27 @@ static int arm_iommu_get_sgtable(struct device *dev, struct sg_table *sgt,
>   					 GFP_KERNEL);
>   }
>   
> +static int __dma_direction_to_prot(enum dma_data_direction dir)
> +{
> +	int prot;
> +
> +	switch (dir) {
> +	case DMA_BIDIRECTIONAL:
> +		prot = IOMMU_READ | IOMMU_WRITE;
> +		break;
> +	case DMA_TO_DEVICE:
> +		prot = IOMMU_READ;
> +		break;
> +	case DMA_FROM_DEVICE:
> +		prot = IOMMU_WRITE;
> +		break;
> +	default:
> +		prot = 0;
> +	}
> +
> +	return prot;
> +}
> +
>   /*
>    * Map a part of the scatter-gather list into contiguous io address space
>    */
> @@ -1444,6 +1466,7 @@ static int __map_sg_chunk(struct device *dev, struct scatterlist *sg,
>   	int ret = 0;
>   	unsigned int count;
>   	struct scatterlist *s;
> +	int prot;
>   
>   	size = PAGE_ALIGN(size);
>   	*handle = DMA_ERROR_CODE;
> @@ -1460,7 +1483,9 @@ static int __map_sg_chunk(struct device *dev, struct scatterlist *sg,
>   			!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs))
>   			__dma_page_cpu_to_dev(sg_page(s), s->offset, s->length, dir);
>   
> -		ret = iommu_map(mapping->domain, iova, phys, len, 0);
> +		prot = __dma_direction_to_prot(dir);
> +
> +		ret = iommu_map(mapping->domain, iova, phys, len, prot);
>   		if (ret < 0)
>   			goto fail;
>   		count += len >> PAGE_SHIFT;
> @@ -1665,19 +1690,7 @@ static dma_addr_t arm_coherent_iommu_map_page(struct device *dev, struct page *p
>   	if (dma_addr == DMA_ERROR_CODE)
>   		return dma_addr;
>   
> -	switch (dir) {
> -	case DMA_BIDIRECTIONAL:
> -		prot = IOMMU_READ | IOMMU_WRITE;
> -		break;
> -	case DMA_TO_DEVICE:
> -		prot = IOMMU_READ;
> -		break;
> -	case DMA_FROM_DEVICE:
> -		prot = IOMMU_WRITE;
> -		break;
> -	default:
> -		prot = 0;
> -	}
> +	prot = __dma_direction_to_prot(dir);
>   
>   	ret = iommu_map(mapping->domain, dma_addr, page_to_phys(page), len, prot);
>   	if (ret < 0)

Best regards
-- 
Marek Szyprowski
Samsung R&D Institute Poland

  parent reply	other threads:[~2013-09-30 13:40 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-26 22:36 [PATCH 0/9] arm-smmu: Misc changes/Calxeda ECX-2000 support Andreas Herrmann
2013-09-26 22:36 ` [PATCH 1/9] iommu/arm-smmu: Switch to arch_initcall for driver registration Andreas Herrmann
2013-09-27  8:58   ` Will Deacon
2013-09-27  9:24     ` Andreas Herrmann
2013-09-27 10:02       ` [PATCH] iommu/arm-smmu: Switch to subsys_initcall " Andreas Herrmann
2013-09-26 22:36 ` [PATCH 2/9] iommu/arm-smmu: Calculate SMMU_CB_BASE from smmu register values Andreas Herrmann
2013-09-27  9:51   ` Will Deacon
2013-09-27 10:23     ` Andreas Herrmann
2013-09-27 10:51       ` Will Deacon
2013-09-27 11:05         ` Andreas Herrmann
2013-09-27 11:08           ` Will Deacon
2013-09-27 14:33             ` [PATCH] iommu/arm-smmu: Refine check for proper size of mapped region Andreas Herrmann
2013-09-26 22:36 ` [PATCH 3/9] ARM: dma-mapping: Always pass proper prot flags to iommu_map() Andreas Herrmann
2013-09-27  8:35   ` Will Deacon
2013-09-30 13:40   ` Marek Szyprowski [this message]
2013-09-26 22:36 ` [PATCH 4/9] iommu/arm-smmu: Check for num_context_irqs > 0 to avoid divide by zero exception Andreas Herrmann
2013-09-27  8:41   ` Will Deacon
2013-09-27  9:03     ` Andreas Herrmann
2013-09-27 10:23       ` Will Deacon
2013-09-27 10:39         ` Andreas Herrmann
2013-09-27 10:48           ` Will Deacon
2013-09-27 11:07             ` Andreas Herrmann
2013-09-27 14:30               ` [PATCH] " Andreas Herrmann
2013-09-26 22:36 ` [PATCH 5/9] iommu/arm-smmu: Clear global and context bank fault status registers Andreas Herrmann
2013-09-27  8:52   ` Will Deacon
2013-09-30 13:54     ` Andreas Herrmann
2013-09-30 13:56       ` [PATCH] " Andreas Herrmann
2013-09-30 16:06         ` Will Deacon
2013-09-30 17:17           ` Andreas Herrmann
2013-09-30 18:30             ` Will Deacon
2013-09-30 21:06               ` Andreas Herrmann
2013-09-26 22:36 ` [PATCH 6/9] iommu/arm-smmu: Support buggy implemenations where all config accesses are secure Andreas Herrmann
2013-09-27 13:05   ` Will Deacon
2013-09-27 13:48     ` Andreas Herrmann
2013-09-26 22:36 ` [PATCH 7/9] iommu/arm-smmu: Add function that conditionally isolates all masters of all SMMUs Andreas Herrmann
2013-09-27 13:00   ` Will Deacon
2013-10-07 15:42     ` Andreas Herrmann
2013-10-08 10:43       ` Will Deacon
2013-09-26 22:36 ` [PATCH 8/9] iommu/arm-smmu: Introduce a default fault handler Andreas Herrmann
2013-09-27 10:09   ` Will Deacon
2013-09-27 10:45     ` Andreas Herrmann
2013-09-27 21:22       ` [PATCH] iommu/arm-smmu: Print context fault information Andreas Herrmann
2013-09-26 22:36 ` [PATCH 9/9] ARM: dts: Add nodes for SMMUs on Calxeda ECX-2000 Andreas Herrmann

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=52497F6A.2090505@samsung.com \
    --to=m.szyprowski@samsung$(echo .)com \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox