All of lore.kernel.org
 help / color / mirror / Atom feed
From: 'Joerg Roedel' <Joerg.Roedel@amd.com>
To: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: 'David Woodhouse' <dwmw2@infradead.org>,
	'Ohad Ben-Cohen' <ohad@wizery.com>,
	'KyongHo Cho' <pullip.cho@samsung.com>,
	'Kai Huang' <mail.kai.huang@gmail.com>,
	kvm@vger.kernel.org, 'Arnd Bergmann' <arnd@arndb.de>,
	linux-kernel@vger.kernel.org, iommu@lists.linux-foundation.org,
	'Laurent Pinchart' <laurent.pinchart@ideasonboard.com>,
	'David Brown' <davidb@codeaurora.org>,
	linux-omap@vger.kernel.org,
	'Stepan Moskovchenko' <stepanm@codeaurora.org>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: Changing IOMMU-API for generic DMA-mapping supported by the hardware
Date: Thu, 24 Nov 2011 16:27:47 +0100	[thread overview]
Message-ID: <20111124152747.GG11876@amd.com> (raw)
In-Reply-To: <023b01ccaaa7$ef16c910$cd445b30$%szyprowski@samsung.com>

On Thu, Nov 24, 2011 at 01:52:33PM +0100, Marek Szyprowski wrote:
> In my DMA-mapping IOMMU integration I've used a dma_iommu_mapping structure,
> which contains a pointer to iommu domain, a bitmap and a lock. Maybe we 
> should consider extending iommu domain with allocation bitmap (or other 
> structure that hold information about used/unused iova ranges)? From the
> DMA-mapping (as a IOMMU client) perspective we only need 2 more callbacks
> in IOMMU API: alloc_iova_range() and free_iova_range(). 
> 
> Each IOMMU implementation can provide these calls based on internal bitmap
> allocator which will also cover the issue with reserved ranges. What do you
> think about such solution?

Hmm, the main point of a generic DMA-mapping implementation is that a
common address-allocator will be used. Today every IOMMU driver that
implements the DMA-API has its own allocator, this is something to unify
between for all drivers.

The allocator information can be stored in the default iommu_domain. We
need a user-private pointer there, but that is easy to add.


	Joerg

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632

WARNING: multiple messages have this Message-ID (diff)
From: Joerg.Roedel@amd.com ('Joerg Roedel')
To: linux-arm-kernel@lists.infradead.org
Subject: Changing IOMMU-API for generic DMA-mapping supported by the hardware
Date: Thu, 24 Nov 2011 16:27:47 +0100	[thread overview]
Message-ID: <20111124152747.GG11876@amd.com> (raw)
In-Reply-To: <023b01ccaaa7$ef16c910$cd445b30$%szyprowski@samsung.com>

On Thu, Nov 24, 2011 at 01:52:33PM +0100, Marek Szyprowski wrote:
> In my DMA-mapping IOMMU integration I've used a dma_iommu_mapping structure,
> which contains a pointer to iommu domain, a bitmap and a lock. Maybe we 
> should consider extending iommu domain with allocation bitmap (or other 
> structure that hold information about used/unused iova ranges)? From the
> DMA-mapping (as a IOMMU client) perspective we only need 2 more callbacks
> in IOMMU API: alloc_iova_range() and free_iova_range(). 
> 
> Each IOMMU implementation can provide these calls based on internal bitmap
> allocator which will also cover the issue with reserved ranges. What do you
> think about such solution?

Hmm, the main point of a generic DMA-mapping implementation is that a
common address-allocator will be used. Today every IOMMU driver that
implements the DMA-API has its own allocator, this is something to unify
between for all drivers.

The allocator information can be stored in the default iommu_domain. We
need a user-private pointer there, but that is easy to add.


	Joerg

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632

WARNING: multiple messages have this Message-ID (diff)
From: 'Joerg Roedel' <Joerg.Roedel@amd.com>
To: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: 'David Woodhouse' <dwmw2@infradead.org>,
	'Ohad Ben-Cohen' <ohad@wizery.com>,
	'KyongHo Cho' <pullip.cho@samsung.com>,
	'Kai Huang' <mail.kai.huang@gmail.com>, <kvm@vger.kernel.org>,
	'Arnd Bergmann' <arnd@arndb.de>, <linux-kernel@vger.kernel.org>,
	<iommu@lists.linux-foundation.org>,
	'Laurent Pinchart' <laurent.pinchart@ideasonboard.com>,
	'David Brown' <davidb@codeaurora.org>,
	<linux-omap@vger.kernel.org>,
	'Stepan Moskovchenko' <stepanm@codeaurora.org>,
	<linux-arm-kernel@lists.infradead.org>
Subject: Re: Changing IOMMU-API for generic DMA-mapping supported by the hardware
Date: Thu, 24 Nov 2011 16:27:47 +0100	[thread overview]
Message-ID: <20111124152747.GG11876@amd.com> (raw)
In-Reply-To: <023b01ccaaa7$ef16c910$cd445b30$%szyprowski@samsung.com>

On Thu, Nov 24, 2011 at 01:52:33PM +0100, Marek Szyprowski wrote:
> In my DMA-mapping IOMMU integration I've used a dma_iommu_mapping structure,
> which contains a pointer to iommu domain, a bitmap and a lock. Maybe we 
> should consider extending iommu domain with allocation bitmap (or other 
> structure that hold information about used/unused iova ranges)? From the
> DMA-mapping (as a IOMMU client) perspective we only need 2 more callbacks
> in IOMMU API: alloc_iova_range() and free_iova_range(). 
> 
> Each IOMMU implementation can provide these calls based on internal bitmap
> allocator which will also cover the issue with reserved ranges. What do you
> think about such solution?

Hmm, the main point of a generic DMA-mapping implementation is that a
common address-allocator will be used. Today every IOMMU driver that
implements the DMA-API has its own allocator, this is something to unify
between for all drivers.

The allocator information can be stored in the default iommu_domain. We
need a user-private pointer there, but that is easy to add.


	Joerg

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632

  reply	other threads:[~2011-11-24 15:27 UTC|newest]

Thread overview: 81+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-17 11:27 [PATCH v4 0/7] iommu: split mapping to page sizes as supported by the hardware Ohad Ben-Cohen
2011-10-17 11:27 ` Ohad Ben-Cohen
2011-10-17 11:27 ` Ohad Ben-Cohen
2011-10-17 11:27 ` [PATCH v4 1/7] iommu/core: stop converting bytes to page order back and forth Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27 ` [PATCH v4 2/7] iommu/core: split mapping to page sizes as supported by the hardware Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-11-10  6:17   ` Kai Huang
2011-11-10  6:17     ` Kai Huang
2011-11-10  7:31     ` Ohad Ben-Cohen
2011-11-10  7:31       ` Ohad Ben-Cohen
2011-11-10 12:16       ` cody
2011-11-10 12:16         ` cody
2011-11-10 13:08         ` Joerg Roedel
2011-11-10 13:08           ` Joerg Roedel
2011-11-10 13:08           ` Joerg Roedel
2011-11-10 14:35           ` cody
2011-11-10 14:35             ` cody
2011-11-10 14:51             ` Joerg Roedel
2011-11-10 14:51               ` Joerg Roedel
2011-11-10 14:51               ` Joerg Roedel
2011-11-10 15:28     ` David Woodhouse
2011-11-10 15:28       ` David Woodhouse
2011-11-10 17:09       ` Joerg Roedel
2011-11-10 17:09         ` Joerg Roedel
2011-11-10 17:09         ` Joerg Roedel
2011-11-10 19:28         ` David Woodhouse
2011-11-10 19:28           ` David Woodhouse
2011-11-11 12:58           ` Joerg Roedel
2011-11-11 12:58             ` Joerg Roedel
2011-11-11 12:58             ` Joerg Roedel
2011-11-11 13:27             ` David Woodhouse
2011-11-11 13:27               ` David Woodhouse
2011-11-11 14:18               ` Joerg Roedel
2011-11-11 14:18                 ` Joerg Roedel
2011-11-11 14:18                 ` Joerg Roedel
2011-11-11 13:17           ` Changing IOMMU-API for generic DMA-mapping " Joerg Roedel
2011-11-11 13:17             ` Joerg Roedel
2011-11-24 12:52             ` Marek Szyprowski
2011-11-24 12:52               ` Marek Szyprowski
2011-11-24 15:27               ` 'Joerg Roedel' [this message]
2011-11-24 15:27                 ` 'Joerg Roedel'
2011-11-24 15:27                 ` 'Joerg Roedel'
2011-11-10 21:12         ` [PATCH v4 2/7] iommu/core: split mapping to page sizes as " Stepan Moskovchenko
2011-11-10 21:12           ` Stepan Moskovchenko
2011-11-11 13:24           ` Joerg Roedel
2011-11-11 13:24             ` Joerg Roedel
2011-11-11 13:24             ` Joerg Roedel
2011-11-12  2:04             ` Stepan Moskovchenko
2011-11-12  2:04               ` Stepan Moskovchenko
2011-11-13  1:43               ` KyongHo Cho
2011-11-13  1:43                 ` KyongHo Cho
2011-10-17 11:27 ` [PATCH v4 3/7] iommu/omap: announce supported page sizes Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27 ` [PATCH v4 4/7] iommu/msm: " Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27 ` [PATCH v4 5/7] iommu/amd: " Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27 ` [PATCH v4 6/7] iommu/intel: " Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27 ` [PATCH v4 7/7] iommu/core: remove the temporary pgsize settings Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-10-17 11:27   ` Ohad Ben-Cohen
2011-11-08 12:57 ` [PATCH v4 0/7] iommu: split mapping to page sizes as supported by the hardware Ohad Ben-Cohen
2011-11-08 12:57   ` Ohad Ben-Cohen
2011-11-08 14:01   ` Joerg Roedel
2011-11-08 14:01     ` Joerg Roedel
2011-11-08 14:01     ` Joerg Roedel
2011-11-08 14:03     ` Ohad Ben-Cohen
2011-11-08 14:03       ` Ohad Ben-Cohen
2011-11-08 16:23       ` Joerg Roedel
2011-11-08 16:23         ` Joerg Roedel
2011-11-08 16:23         ` Joerg Roedel
2011-11-08 16:43         ` Ohad Ben-Cohen
2011-11-08 16:43           ` Ohad Ben-Cohen

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=20111124152747.GG11876@amd.com \
    --to=joerg.roedel@amd.com \
    --cc=arnd@arndb.de \
    --cc=davidb@codeaurora.org \
    --cc=dwmw2@infradead.org \
    --cc=iommu@lists.linux-foundation.org \
    --cc=kvm@vger.kernel.org \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=m.szyprowski@samsung.com \
    --cc=mail.kai.huang@gmail.com \
    --cc=ohad@wizery.com \
    --cc=pullip.cho@samsung.com \
    --cc=stepanm@codeaurora.org \
    /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.