All of lore.kernel.org
 help / color / mirror / Atom feed
From: Olav Haugan <ohaugan@codeaurora.org>
To: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: joro@8bytes.org, laurent.pinchart+renesas@ideasonboard.com,
	robdclark@gmail.com, will.deacon@arm.com,
	thierry.reding@gmail.com, kgene.kim@samsung.com,
	dwmw2@infradead.org, hdoyu@nvidia.com, Varun.Sethi@freescale.com,
	konrad.wilk@oracle.com, s-anna@ti.com, mitchelh@codeaurora.org,
	iommu@lists.linux-foundation.org,
	linux-arm-kernel@lists.infradead.org,
	linux-arm-msm@vger.kernel.org
Subject: Re: [PATCH v5 1/1] iommu-api: Add map_sg/unmap_sg functions
Date: Tue, 12 Aug 2014 10:10:33 -0700	[thread overview]
Message-ID: <53EA4A89.30108@codeaurora.org> (raw)
In-Reply-To: <4605275.P2SLc4fsBd@avalon>

Hi Laurent,

On 8/12/2014 9:55 AM, Laurent Pinchart wrote:
> Hi Olav,
> 
> Thank you for the patch.
> 
> On Monday 11 August 2014 15:45:50 Olav Haugan wrote:
>> Mapping and unmapping are more often than not in the critical path.
>> map_sg and unmap_sg allows IOMMU driver implementations to optimize
>> the process of mapping and unmapping buffers into the IOMMU page tables.
>>
>> Instead of mapping a buffer one page at a time and requiring potentially
>> expensive TLB operations for each page, this function allows the driver
>> to map all pages in one go and defer TLB maintenance until after all
>> pages have been mapped.
>>
>> Additionally, the mapping operation would be faster in general since
>> clients does not have to keep calling map API over and over again for
>> each physically contiguous chunk of memory that needs to be mapped to a
>> virtually contiguous region.
>>
>> Signed-off-by: Olav Haugan <ohaugan@codeaurora.org>
>> ---
>>  drivers/iommu/amd_iommu.c      |  2 ++
>>  drivers/iommu/arm-smmu.c       |  2 ++
>>  drivers/iommu/exynos-iommu.c   |  2 ++
>>  drivers/iommu/intel-iommu.c    |  2 ++
>>  drivers/iommu/iommu.c          | 33 +++++++++++++++++++++++++++++++
>>  drivers/iommu/ipmmu-vmsa.c     |  2 ++
>>  drivers/iommu/msm_iommu.c      |  2 ++
>>  drivers/iommu/omap-iommu.c     |  2 ++
>>  drivers/iommu/shmobile-iommu.c |  2 ++
>>  drivers/iommu/tegra-smmu.c     |  2 ++
>>  include/linux/iommu.h          | 44 +++++++++++++++++++++++++++++++++++++++
>>  11 files changed, 95 insertions(+)
> 
> [snip]
> 
>> diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
>> index 1698360..24cf727 100644
>> --- a/drivers/iommu/iommu.c
>> +++ b/drivers/iommu/iommu.c
>> @@ -1088,6 +1088,39 @@ size_t iommu_unmap(struct iommu_domain *domain,
>> unsigned long iova, size_t size)
> 
> [snip]
> 
>> +int default_iommu_unmap_sg(struct iommu_domain *domain, unsigned long iova,
>> +			       size_t size, unsigned long flags)
> 
> I would have called this iommu_default_unmap_sg (and same comment for 
> default_iommu_map_sg) to keep the iommu_ prefix, but that's up to you.
> 
>> +{
>> +	return iommu_unmap(domain, iova, size);
>> +}
>> +EXPORT_SYMBOL_GPL(default_iommu_unmap_sg);
> 
> Do you expect drivers to need to override this ? What are the use cases for 
> non-default implementation of unmap_sg different than this ?

Good question. Yes, maybe some drivers does not need or want to override
this but a use case is to provide a more optimized version of the
map_sg/unmap_sg functions. For example a very simple way to optimize
this would be to have an implementation that unmaps everything and then
does a TLB invalidate instead of doing a TLB invalidate after every
single unmap (which happens with the default implementation if your
driver does TLB invalidate after unmapping).

> [snip]
> 
>> diff --git a/include/linux/iommu.h b/include/linux/iommu.h
>> index 20f9a52..ee106ce 100644
>> --- a/include/linux/iommu.h
>> +++ b/include/linux/iommu.h
> 
> [snip]
> 
>> @@ -240,6 +256,20 @@ static inline int report_iommu_fault(struct
>> iommu_domain *domain, return ret;
>>  }
>>
>> +static inline int iommu_map_sg(struct iommu_domain *domain, unsigned long
>> iova,
>> +			       struct scatterlist *sg, unsigned int nents,
>> +			       int prot, unsigned long flags)
>> +{
>> +	return domain->ops->map_sg(domain, iova, sg, nents, prot, flags);
> 
> Instead of having to modify all IOMMU drivers to set the map_sg operation to 
> default_iommu_map_sg, how about calling it automatically as a fallback when 
> map_sg is NULL ? Something like
> 
> 	if (domain->ops->map_sg)
> 		return domain->ops->map_sg(domain, iova, sg, nents, prot, flags);
> 
> 	return default_iommu_map_sg(domain, iova, sg, nents, prot, flags);

This was my original proposal but after some discussion on the list we
ended up with what we have now.

> 
>> +}
>> +
>> +static inline int iommu_unmap_sg(struct iommu_domain *domain,
>> +				 unsigned long iova, size_t size,
>> +				 unsigned long flags)
>> +{
>> +	return domain->ops->unmap_sg(domain, iova, size, flags);
>> +}
>> +
> 

Thanks,

Olav

-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation

WARNING: multiple messages have this Message-ID (diff)
From: ohaugan@codeaurora.org (Olav Haugan)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v5 1/1] iommu-api: Add map_sg/unmap_sg functions
Date: Tue, 12 Aug 2014 10:10:33 -0700	[thread overview]
Message-ID: <53EA4A89.30108@codeaurora.org> (raw)
In-Reply-To: <4605275.P2SLc4fsBd@avalon>

Hi Laurent,

On 8/12/2014 9:55 AM, Laurent Pinchart wrote:
> Hi Olav,
> 
> Thank you for the patch.
> 
> On Monday 11 August 2014 15:45:50 Olav Haugan wrote:
>> Mapping and unmapping are more often than not in the critical path.
>> map_sg and unmap_sg allows IOMMU driver implementations to optimize
>> the process of mapping and unmapping buffers into the IOMMU page tables.
>>
>> Instead of mapping a buffer one page at a time and requiring potentially
>> expensive TLB operations for each page, this function allows the driver
>> to map all pages in one go and defer TLB maintenance until after all
>> pages have been mapped.
>>
>> Additionally, the mapping operation would be faster in general since
>> clients does not have to keep calling map API over and over again for
>> each physically contiguous chunk of memory that needs to be mapped to a
>> virtually contiguous region.
>>
>> Signed-off-by: Olav Haugan <ohaugan@codeaurora.org>
>> ---
>>  drivers/iommu/amd_iommu.c      |  2 ++
>>  drivers/iommu/arm-smmu.c       |  2 ++
>>  drivers/iommu/exynos-iommu.c   |  2 ++
>>  drivers/iommu/intel-iommu.c    |  2 ++
>>  drivers/iommu/iommu.c          | 33 +++++++++++++++++++++++++++++++
>>  drivers/iommu/ipmmu-vmsa.c     |  2 ++
>>  drivers/iommu/msm_iommu.c      |  2 ++
>>  drivers/iommu/omap-iommu.c     |  2 ++
>>  drivers/iommu/shmobile-iommu.c |  2 ++
>>  drivers/iommu/tegra-smmu.c     |  2 ++
>>  include/linux/iommu.h          | 44 +++++++++++++++++++++++++++++++++++++++
>>  11 files changed, 95 insertions(+)
> 
> [snip]
> 
>> diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
>> index 1698360..24cf727 100644
>> --- a/drivers/iommu/iommu.c
>> +++ b/drivers/iommu/iommu.c
>> @@ -1088,6 +1088,39 @@ size_t iommu_unmap(struct iommu_domain *domain,
>> unsigned long iova, size_t size)
> 
> [snip]
> 
>> +int default_iommu_unmap_sg(struct iommu_domain *domain, unsigned long iova,
>> +			       size_t size, unsigned long flags)
> 
> I would have called this iommu_default_unmap_sg (and same comment for 
> default_iommu_map_sg) to keep the iommu_ prefix, but that's up to you.
> 
>> +{
>> +	return iommu_unmap(domain, iova, size);
>> +}
>> +EXPORT_SYMBOL_GPL(default_iommu_unmap_sg);
> 
> Do you expect drivers to need to override this ? What are the use cases for 
> non-default implementation of unmap_sg different than this ?

Good question. Yes, maybe some drivers does not need or want to override
this but a use case is to provide a more optimized version of the
map_sg/unmap_sg functions. For example a very simple way to optimize
this would be to have an implementation that unmaps everything and then
does a TLB invalidate instead of doing a TLB invalidate after every
single unmap (which happens with the default implementation if your
driver does TLB invalidate after unmapping).

> [snip]
> 
>> diff --git a/include/linux/iommu.h b/include/linux/iommu.h
>> index 20f9a52..ee106ce 100644
>> --- a/include/linux/iommu.h
>> +++ b/include/linux/iommu.h
> 
> [snip]
> 
>> @@ -240,6 +256,20 @@ static inline int report_iommu_fault(struct
>> iommu_domain *domain, return ret;
>>  }
>>
>> +static inline int iommu_map_sg(struct iommu_domain *domain, unsigned long
>> iova,
>> +			       struct scatterlist *sg, unsigned int nents,
>> +			       int prot, unsigned long flags)
>> +{
>> +	return domain->ops->map_sg(domain, iova, sg, nents, prot, flags);
> 
> Instead of having to modify all IOMMU drivers to set the map_sg operation to 
> default_iommu_map_sg, how about calling it automatically as a fallback when 
> map_sg is NULL ? Something like
> 
> 	if (domain->ops->map_sg)
> 		return domain->ops->map_sg(domain, iova, sg, nents, prot, flags);
> 
> 	return default_iommu_map_sg(domain, iova, sg, nents, prot, flags);

This was my original proposal but after some discussion on the list we
ended up with what we have now.

> 
>> +}
>> +
>> +static inline int iommu_unmap_sg(struct iommu_domain *domain,
>> +				 unsigned long iova, size_t size,
>> +				 unsigned long flags)
>> +{
>> +	return domain->ops->unmap_sg(domain, iova, size, flags);
>> +}
>> +
> 

Thanks,

Olav

-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation

  reply	other threads:[~2014-08-12 17:10 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-11 22:45 [PATCH v5 0/1] Add iommu map_sg/unmap_sg API Olav Haugan
2014-08-11 22:45 ` Olav Haugan
2014-08-11 22:45 ` [PATCH v5 1/1] iommu-api: Add map_sg/unmap_sg functions Olav Haugan
2014-08-11 22:45   ` Olav Haugan
     [not found]   ` <1407797150-515-2-git-send-email-ohaugan-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2014-08-12  1:35     ` Konrad Rzeszutek Wilk
2014-08-12  1:35       ` Konrad Rzeszutek Wilk
     [not found]       ` <20140812013559.GA25121-0iZWjJA6G8GSPmnEAIUT9EEOCMrvLtNR@public.gmane.org>
2014-08-12 16:53         ` Olav Haugan
2014-08-12 16:53           ` Olav Haugan
2014-08-12  1:51     ` Hiroshi Doyu
2014-08-12  1:51       ` Hiroshi Doyu
2014-08-12 10:48       ` Rob Clark
2014-08-12 10:48         ` Rob Clark
2014-08-12 16:56         ` Olav Haugan
2014-08-12 16:56           ` Olav Haugan
     [not found]           ` <53EA472B.3020900-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2014-08-18 14:07             ` joro-zLv9SwRftAIdnm+yROfE0A
2014-08-18 14:07               ` joro at 8bytes.org
     [not found]               ` <20140818140730.GC9809-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2014-08-18 18:32                 ` Rob Clark
2014-08-18 18:32                   ` Rob Clark
     [not found]                   ` <CAF6AEGtGmAQjMc5W5a6x5QiyUc8v0XHcsNAFe65Znc=DDKPS+w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-08-18 20:48                     ` Olav Haugan
2014-08-18 20:48                       ` Olav Haugan
     [not found]                       ` <53F266AE.40303-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2014-08-18 21:26                         ` joro-zLv9SwRftAIdnm+yROfE0A
2014-08-18 21:26                           ` joro at 8bytes.org
     [not found]                           ` <20140818212627.GI9809-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2014-08-18 21:32                             ` Olav Haugan
2014-08-18 21:32                               ` Olav Haugan
2014-08-12 16:55     ` Laurent Pinchart
2014-08-12 16:55       ` Laurent Pinchart
2014-08-12 17:10       ` Olav Haugan [this message]
2014-08-12 17:10         ` Olav Haugan
2014-08-18 21:55     ` Joerg Roedel
2014-08-18 21:55       ` Joerg Roedel
2014-08-18 22:47       ` Olav Haugan
2014-08-18 22:47         ` Olav Haugan
     [not found]         ` <53F2829C.2040809-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2014-08-19 11:59           ` Joerg Roedel
2014-08-19 11:59             ` Joerg Roedel
     [not found]             ` <20140819115954.GC16329-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2014-08-19 16:11               ` Laurent Pinchart
2014-08-19 16:11                 ` Laurent Pinchart
2014-08-19 18:40                 ` Olav Haugan
2014-08-19 18:40                   ` Olav Haugan
     [not found]                   ` <53F39A18.70409-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2014-08-19 20:52                     ` Laurent Pinchart
2014-08-19 20:52                       ` Laurent Pinchart
2014-08-20  5:21                       ` Thierry Reding
2014-08-20  5:21                         ` Thierry Reding
2014-08-20 13:02                       ` Konrad Rzeszutek Wilk
2014-08-20 13:02                         ` Konrad Rzeszutek Wilk
     [not found]                         ` <20140820130250.GA3120-0iZWjJA6G8GSPmnEAIUT9EEOCMrvLtNR@public.gmane.org>
2014-08-20 14:15                           ` Laurent Pinchart
2014-08-20 14:15                             ` Laurent Pinchart
2014-08-19 18:37               ` Olav Haugan
2014-08-19 18:37                 ` Olav Haugan
2014-09-25 17:01     ` Joerg Roedel
2014-09-25 17:01       ` Joerg Roedel
     [not found]       ` <20140925170108.GE8306-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2014-10-06 19:02         ` Olav Haugan
2014-10-06 19:02           ` Olav Haugan
2014-10-15  9:16           ` Thierry Reding
2014-10-15  9:16             ` Thierry Reding
2014-10-16 17:23             ` Olav Haugan
2014-10-16 17:23               ` Olav Haugan
2014-10-17  9:09               ` Joerg Roedel
2014-10-17  9:09                 ` 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=53EA4A89.30108@codeaurora.org \
    --to=ohaugan@codeaurora.org \
    --cc=Varun.Sethi@freescale.com \
    --cc=dwmw2@infradead.org \
    --cc=hdoyu@nvidia.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=joro@8bytes.org \
    --cc=kgene.kim@samsung.com \
    --cc=konrad.wilk@oracle.com \
    --cc=laurent.pinchart+renesas@ideasonboard.com \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=mitchelh@codeaurora.org \
    --cc=robdclark@gmail.com \
    --cc=s-anna@ti.com \
    --cc=thierry.reding@gmail.com \
    --cc=will.deacon@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.