All of lore.kernel.org
 help / color / mirror / Atom feed
From: Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
To: Sricharan <sricharan-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
	'Will Deacon' <will.deacon-5wv7dgnIgG8@public.gmane.org>,
	linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
Subject: Re: [PATCH 0/5] Implement SMMU passthrough using the default domain
Date: Thu, 26 Jan 2017 18:18:58 +0100	[thread overview]
Message-ID: <20170126171857.GE27598@8bytes.org> (raw)
In-Reply-To: <010801d27654$47b7e450$d727acf0$@codeaurora.org>

On Tue, Jan 24, 2017 at 08:42:23PM +0530, Sricharan wrote:
> Thanks for this series. We had a case with the GPU.
> The GPU's iommu was setup by kernel and the GPU
> also does dynamic updates for on-the-fly switching between
> process pagetables.  GPU driver was not using DMA domain and
> the GPU's firmware was always expecting to run out  of contextbank
>  '0' (although not correct) , which was not the case after the DMA domain
> was made default  as '0' was getting allocated for DMA domain and
> there were concerns about reusing the DMA domain as well.
> Now with this series, looks there is an way out of that that can be tried.
> 
> So should the default domain not be per device specific selectable ?

Note that iommu-drivers can request direct-mapping for any given device
on its initializtion. This is used on x86 for devices that need a 1-1
mapping for some reason.

Also device drivers can use the iommu-api and assign their own domain to
a device, which allows them to manage the dma address space on their
own.


	Joerg

WARNING: multiple messages have this Message-ID (diff)
From: joro@8bytes.org (Joerg Roedel)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 0/5] Implement SMMU passthrough using the default domain
Date: Thu, 26 Jan 2017 18:18:58 +0100	[thread overview]
Message-ID: <20170126171857.GE27598@8bytes.org> (raw)
In-Reply-To: <010801d27654$47b7e450$d727acf0$@codeaurora.org>

On Tue, Jan 24, 2017 at 08:42:23PM +0530, Sricharan wrote:
> Thanks for this series. We had a case with the GPU.
> The GPU's iommu was setup by kernel and the GPU
> also does dynamic updates for on-the-fly switching between
> process pagetables.  GPU driver was not using DMA domain and
> the GPU's firmware was always expecting to run out  of contextbank
>  '0' (although not correct) , which was not the case after the DMA domain
> was made default  as '0' was getting allocated for DMA domain and
> there were concerns about reusing the DMA domain as well.
> Now with this series, looks there is an way out of that that can be tried.
> 
> So should the default domain not be per device specific selectable ?

Note that iommu-drivers can request direct-mapping for any given device
on its initializtion. This is used on x86 for devices that need a 1-1
mapping for some reason.

Also device drivers can use the iommu-api and assign their own domain to
a device, which allows them to manage the dma address space on their
own.


	Joerg

  reply	other threads:[~2017-01-26 17:18 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-19 18:19 [PATCH 0/5] Implement SMMU passthrough using the default domain Will Deacon
2017-01-19 18:19 ` Will Deacon
     [not found] ` <1484849955-1871-1-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2017-01-19 18:19   ` [PATCH 1/5] iommu/arm-smmu: Restrict domain attributes to UNMANAGED domains Will Deacon
2017-01-19 18:19     ` Will Deacon
     [not found]     ` <1484849955-1871-2-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2017-01-19 18:41       ` Robin Murphy
2017-01-19 18:41         ` Robin Murphy
     [not found]         ` <89d795b9-a6d7-7f41-705c-b918ee66ebf0-5wv7dgnIgG8@public.gmane.org>
2017-01-26 17:06           ` Joerg Roedel
2017-01-26 17:06             ` Joerg Roedel
2017-01-26 17:03       ` Joerg Roedel
2017-01-26 17:03         ` Joerg Roedel
     [not found]         ` <20170126170329.GB27598-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2017-01-26 17:57           ` Will Deacon
2017-01-26 17:57             ` Will Deacon
2017-01-19 18:19   ` [PATCH 2/5] iommu/arm-smmu: Install bypass S2CRs for IOMMU_DOMAIN_IDENTITY domains Will Deacon
2017-01-19 18:19     ` Will Deacon
2017-01-19 18:50     ` Robin Murphy
2017-01-19 18:50       ` Robin Murphy
2017-01-19 18:19   ` [PATCH 3/5] iommu/arm-smmu-v3: Install bypass STEs " Will Deacon
2017-01-19 18:19     ` Will Deacon
     [not found]     ` <1484849955-1871-4-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2017-01-19 18:56       ` Robin Murphy
2017-01-19 18:56         ` Robin Murphy
2017-01-19 18:19   ` [PATCH 4/5] arm64: dma-mapping: Only swizzle DMA ops for IOMMU_DOMAIN_DMA Will Deacon
2017-01-19 18:19     ` Will Deacon
     [not found]     ` <1484849955-1871-5-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2017-01-19 19:00       ` Robin Murphy
2017-01-19 19:00         ` Robin Murphy
2017-01-26 17:57         ` Will Deacon
2017-01-26 17:57           ` Will Deacon
2017-01-19 18:19   ` [PATCH 5/5] iommu: Allow default domain type to be set on the kernel command line Will Deacon
2017-01-19 18:19     ` Will Deacon
     [not found]     ` <1484849955-1871-6-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2017-01-26 17:15       ` Joerg Roedel
2017-01-26 17:15         ` Joerg Roedel
     [not found]         ` <20170126171555.GD27598-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2017-01-26 17:26           ` Robin Murphy
2017-01-26 17:26             ` Robin Murphy
2017-01-26 17:48           ` Will Deacon
2017-01-26 17:48             ` Will Deacon
2017-01-24 15:12   ` [PATCH 0/5] Implement SMMU passthrough using the default domain Sricharan
2017-01-24 15:12     ` Sricharan
2017-01-26 17:18     ` Joerg Roedel [this message]
2017-01-26 17:18       ` Joerg Roedel
     [not found]       ` <20170126171857.GE27598-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2017-02-02 15:02         ` Rob Clark
2017-02-02 15:02           ` Rob Clark
     [not found]           ` <CAF6AEGtfrrwVZ_W86_HqVsiqfcM=XFa+g2LECFYQJmvv7-JSqQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-02-02 15:12             ` Will Deacon
2017-02-02 15:12               ` Will Deacon
     [not found]               ` <20170202151241.GE13839-5wv7dgnIgG8@public.gmane.org>
2017-02-03 12:20                 ` Rob Clark
2017-02-03 12:20                   ` Rob Clark
2017-02-02 15:45             ` Sricharan
2017-02-02 15:45               ` Sricharan
2017-02-02 16:10               ` Will Deacon
2017-02-02 16:10                 ` Will Deacon
     [not found]                 ` <20170202161047.GG13839-5wv7dgnIgG8@public.gmane.org>
2017-02-03 16:33                   ` Sricharan
2017-02-03 16:33                     ` Sricharan

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=20170126171857.GE27598@8bytes.org \
    --to=joro-zlv9swrftaidnm+yrofe0a@public.gmane.org \
    --cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
    --cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
    --cc=sricharan-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org \
    --cc=will.deacon-5wv7dgnIgG8@public.gmane.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.