public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: robin.murphy@arm•com (Robin Murphy)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH 2/3] iommu/dma: Use correct offset in map_sg
Date: Mon, 4 Jan 2016 16:08:23 +0000	[thread overview]
Message-ID: <568A98F7.3050902@arm.com> (raw)
In-Reply-To: <20151228160551.GA17673@8bytes.org>

On 28/12/15 16:05, Joerg Roedel wrote:
> On Fri, Dec 18, 2015 at 05:01:47PM +0000, Robin Murphy wrote:
>> When mapping a non-page-aligned scatterlist entry, we copy the original
>> offset to the output DMA address before aligning it to hand off to
>> iommu_map_sg(), then later adding the IOVA page address portion to get
>> the final mapped address. However, when the IOVA page size is smaller
>> than the CPU page size, it is the offset within the IOVA page we want,
>> not that within the CPU page, which can easily be larger than an IOVA
>> page and thus result in an incorrect final address.
>>
>> Fix the bug by taking only the IOVA-aligned part of the offset as the
>> basis of the DMA address, not the whole thing.
>>
>> Signed-off-by: Robin Murphy <robin.murphy@arm•com>
>> ---
>>   drivers/iommu/dma-iommu.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
>> index 982e716..03811e3 100644
>> --- a/drivers/iommu/dma-iommu.c
>> +++ b/drivers/iommu/dma-iommu.c
>> @@ -458,7 +458,7 @@ int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
>>   		size_t s_length = s->length;
>>   		size_t pad_len = (mask - iova_len + 1) & mask;
>>
>> -		sg_dma_address(s) = s->offset;
>> +		sg_dma_address(s) = s_offset;
>
> Does not apply on v4.4-rc7.

Ah, I did these on top of the map_sg tweak[1] locally, and failed to 
spot that that leaks into a context line here. There's no actual 
dependency so I'll rebase and repost this bugfix momentarily - as that 
other thread seems to have stalled, I'll probably have another crack at 
a proper scatterlist segment merging implementation sometime after the 
merge window.

Thanks,
Robin.

[1]:http://thread.gmane.org/gmane.linux.kernel.iommu/11512

  reply	other threads:[~2016-01-04 16:08 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-18 17:01 [PATCH 1/3] iommu/dma: Add some missing #includes Robin Murphy
2015-12-18 17:01 ` [PATCH 2/3] iommu/dma: Use correct offset in map_sg Robin Murphy
2015-12-28 16:05   ` Joerg Roedel
2016-01-04 16:08     ` Robin Murphy [this message]
2016-01-04 16:19   ` [PATCH RESEND] " Robin Murphy
2016-01-07 12:38     ` Joerg Roedel
2016-03-09  7:50   ` [PATCH 2/3] " Magnus Damm
2016-03-09 15:00     ` Robin Murphy
2016-03-10  7:47       ` Magnus Damm
2015-12-18 17:01 ` [PATCH 3/3] iommu/dma: Avoid unlikely high-order allocations Robin Murphy
2015-12-18 22:35   ` Doug Anderson
2016-01-19  2:55     ` Yong Wu
2015-12-28 16:08 ` [PATCH 1/3] iommu/dma: Add some missing #includes Joerg Roedel

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=568A98F7.3050902@arm.com \
    --to=robin.murphy@arm$(echo .)com \
    --cc=linux-arm-kernel@lists$(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