From: "Roedel, Joerg" <Joerg.Roedel@amd.com>
To: Stepan Moskovchenko <stepanm@codeaurora.org>
Cc: Zach Pfeffer <zpfeffer@codeaurora.org>,
FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>,
"arnd@arndb.de" <arnd@arndb.de>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-arm-msm@vger.kernel.org" <linux-arm-msm@vger.kernel.org>,
"dwalker@codeaurora.org" <dwalker@codeaurora.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/2] arm: msm: Add System MMU support.
Date: Wed, 4 Aug 2010 11:52:37 +0200 [thread overview]
Message-ID: <20100804095237.GH18307@amd.com> (raw)
In-Reply-To: <4C586348.2050209@codeaurora.org>
On Tue, Aug 03, 2010 at 02:43:20PM -0400, Stepan Moskovchenko wrote:
> On 8/3/2010 2:23 AM, Roedel, Joerg wrote:
> > A domain is defined by a single page-table which can be modified using
> > the iommu_map/iommu_unmap function calls. I am not completly sure what
> > you mean by an iommu_context. Can you describe what it means in your
> > context?
>
> Joerg,
> I think with some rework, all my use cases can be handled by your
> existing iommu API. If the domain is treated basically a page table,
> there will be some changes, but I think it can be done. I will push a
> new version of my driver in a few days.
>
> One thing that may be helpful for the future, however, is maybe
> something like adding iommu_tlb_flush to the ops. I suppose this would
> either have to take a device, or the domain would need to keep a list of
> devices it had been attached to (so that their TLBs can be invalidated).
> But I suppose on the other hand, iommu_map/unmap may be able to just
> implicitly invalidate the TLB also, since TLB invalidation often follows
> map/unmap. What are your thoughts?
Sounds good. I am curious for your patches :-)
For the TLB-flush question, I think it would make sense to add iommu
tlb flushing functions to the IOMMU-API. We currently flush the TLB
implicitly in the map/unmap calls but thats very inefficient. It would
be better to have a seperate function for it in the API. The right
parameter for such a function is a domain. The IOMMU driver knows which
devices are attached to a domain an could easily flush all TLBs.
One alternative I can think of: An iommu_domain_commit() function which
syncs software changes of a domain to the hardware. The map/unmap calls
have to save which parts of the tlb need to be flushed and commit does
flush those parts then (or flush everything).
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 (Roedel, Joerg)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 1/2] arm: msm: Add System MMU support.
Date: Wed, 4 Aug 2010 11:52:37 +0200 [thread overview]
Message-ID: <20100804095237.GH18307@amd.com> (raw)
In-Reply-To: <4C586348.2050209@codeaurora.org>
On Tue, Aug 03, 2010 at 02:43:20PM -0400, Stepan Moskovchenko wrote:
> On 8/3/2010 2:23 AM, Roedel, Joerg wrote:
> > A domain is defined by a single page-table which can be modified using
> > the iommu_map/iommu_unmap function calls. I am not completly sure what
> > you mean by an iommu_context. Can you describe what it means in your
> > context?
>
> Joerg,
> I think with some rework, all my use cases can be handled by your
> existing iommu API. If the domain is treated basically a page table,
> there will be some changes, but I think it can be done. I will push a
> new version of my driver in a few days.
>
> One thing that may be helpful for the future, however, is maybe
> something like adding iommu_tlb_flush to the ops. I suppose this would
> either have to take a device, or the domain would need to keep a list of
> devices it had been attached to (so that their TLBs can be invalidated).
> But I suppose on the other hand, iommu_map/unmap may be able to just
> implicitly invalidate the TLB also, since TLB invalidation often follows
> map/unmap. What are your thoughts?
Sounds good. I am curious for your patches :-)
For the TLB-flush question, I think it would make sense to add iommu
tlb flushing functions to the IOMMU-API. We currently flush the TLB
implicitly in the map/unmap calls but thats very inefficient. It would
be better to have a seperate function for it in the API. The right
parameter for such a function is a domain. The IOMMU driver knows which
devices are attached to a domain an could easily flush all TLBs.
One alternative I can think of: An iommu_domain_commit() function which
syncs software changes of a domain to the hardware. The map/unmap calls
have to save which parts of the tlb need to be flushed and commit does
flush those parts then (or flush everything).
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
next prev parent reply other threads:[~2010-08-04 9:49 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-27 22:41 [PATCH 1/2] arm: msm: Add System MMU support Stepan Moskovchenko
2010-07-27 22:41 ` Stepan Moskovchenko
2010-07-27 22:43 ` Daniel Walker
2010-07-27 22:43 ` Daniel Walker
2010-07-28 8:39 ` Arnd Bergmann
2010-07-28 8:39 ` Arnd Bergmann
2010-07-28 17:39 ` stepanm
2010-07-28 17:39 ` stepanm at codeaurora.org
2010-07-28 17:50 ` Arnd Bergmann
2010-07-28 17:50 ` Arnd Bergmann
2010-07-28 21:21 ` Russell King - ARM Linux
2010-07-28 21:21 ` Russell King - ARM Linux
2010-07-29 4:15 ` FUJITA Tomonori
2010-07-29 4:15 ` FUJITA Tomonori
2010-07-29 8:12 ` Arnd Bergmann
2010-07-29 8:12 ` Arnd Bergmann
2010-07-29 11:47 ` Russell King - ARM Linux
2010-07-29 11:47 ` Russell King - ARM Linux
2010-07-30 6:14 ` FUJITA Tomonori
2010-07-30 6:14 ` FUJITA Tomonori
2010-07-29 0:58 ` stepanm
2010-07-29 0:58 ` stepanm at codeaurora.org
2010-07-29 3:35 ` FUJITA Tomonori
2010-07-29 3:35 ` FUJITA Tomonori
2010-07-29 8:26 ` Arnd Bergmann
2010-07-29 8:26 ` Arnd Bergmann
2010-07-29 8:35 ` FUJITA Tomonori
2010-07-29 8:35 ` FUJITA Tomonori
2010-07-29 8:40 ` Roedel, Joerg
2010-07-29 8:40 ` Roedel, Joerg
2010-07-29 8:46 ` FUJITA Tomonori
2010-07-29 8:46 ` FUJITA Tomonori
2010-07-29 9:06 ` Roedel, Joerg
2010-07-29 9:06 ` Roedel, Joerg
2010-07-29 9:14 ` FUJITA Tomonori
2010-07-29 9:14 ` FUJITA Tomonori
2010-07-29 9:25 ` Roedel, Joerg
2010-07-29 9:25 ` Roedel, Joerg
2010-07-29 9:28 ` Roedel, Joerg
2010-07-29 9:28 ` Roedel, Joerg
2010-07-29 9:44 ` FUJITA Tomonori
2010-07-29 9:44 ` FUJITA Tomonori
2010-07-29 10:01 ` Roedel, Joerg
2010-07-29 10:01 ` Roedel, Joerg
2010-07-29 11:25 ` Arnd Bergmann
2010-07-29 11:25 ` Arnd Bergmann
2010-07-29 12:12 ` Roedel, Joerg
2010-07-29 12:12 ` Roedel, Joerg
2010-07-29 13:01 ` Arnd Bergmann
2010-07-29 13:01 ` Arnd Bergmann
2010-07-30 5:19 ` stepanm
2010-07-30 5:19 ` stepanm at codeaurora.org
2010-07-30 8:01 ` Arnd Bergmann
2010-07-30 8:01 ` Arnd Bergmann
2010-07-30 16:25 ` stepanm
2010-07-30 16:25 ` stepanm at codeaurora.org
2010-07-30 21:59 ` Arnd Bergmann
2010-07-30 21:59 ` Arnd Bergmann
2010-07-30 22:58 ` stepanm
2010-07-30 22:58 ` stepanm at codeaurora.org
2010-07-31 9:37 ` Arnd Bergmann
2010-07-31 9:37 ` Arnd Bergmann
2010-08-02 7:58 ` Roedel, Joerg
2010-08-02 7:58 ` Roedel, Joerg
2010-08-02 20:29 ` Zach Pfeffer
2010-08-02 20:29 ` Zach Pfeffer
2010-08-03 9:23 ` Roedel, Joerg
2010-08-03 9:23 ` Roedel, Joerg
2010-08-03 18:43 ` Stepan Moskovchenko
2010-08-03 18:43 ` Stepan Moskovchenko
2010-08-04 9:52 ` Roedel, Joerg [this message]
2010-08-04 9:52 ` Roedel, Joerg
2010-07-31 3:15 ` Benjamin Herrenschmidt
2010-07-31 3:15 ` Benjamin Herrenschmidt
2010-08-02 7:48 ` Roedel, Joerg
2010-08-02 7:48 ` Roedel, Joerg
2010-08-02 8:03 ` Benjamin Herrenschmidt
2010-08-02 8:03 ` Benjamin Herrenschmidt
2010-08-02 8:10 ` Roedel, Joerg
2010-08-02 8:10 ` Roedel, Joerg
2010-08-02 8:30 ` FUJITA Tomonori
2010-08-02 8:30 ` FUJITA Tomonori
2010-08-02 9:03 ` Russell King - ARM Linux
2010-08-02 9:03 ` Russell King - ARM Linux
2010-08-02 9:20 ` FUJITA Tomonori
2010-08-02 9:20 ` FUJITA Tomonori
2010-08-02 10:04 ` Russell King - ARM Linux
2010-08-02 10:04 ` Russell King - ARM Linux
2010-08-02 15:26 ` FUJITA Tomonori
2010-08-02 15:26 ` FUJITA Tomonori
2010-08-02 9:45 ` Roedel, Joerg
2010-08-02 9:45 ` Roedel, Joerg
2010-08-02 8:35 ` Roedel, Joerg
2010-08-02 8:35 ` Roedel, Joerg
2010-07-31 2:30 ` Benjamin Herrenschmidt
2010-07-31 2:30 ` Benjamin Herrenschmidt
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=20100804095237.GH18307@amd.com \
--to=joerg.roedel@amd.com \
--cc=arnd@arndb.de \
--cc=dwalker@codeaurora.org \
--cc=fujita.tomonori@lab.ntt.co.jp \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=stepanm@codeaurora.org \
--cc=zpfeffer@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.