All of lore.kernel.org
 help / color / mirror / Atom feed
From: Laurent Pinchart <laurent.pinchart-ryLnwIuWjnjg/C1BVhZhaw@public.gmane.org>
To: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
	Varun.Sethi-KZfg59tc24xl57MIdRCFDg@public.gmane.org,
	prem.mallappa-dY08KVG/lbpWk0Htik3J/w@public.gmane.org,
	Robin.Murphy-5wv7dgnIgG8@public.gmane.org,
	linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
Subject: Re: [PATCH 0/4] Generic IOMMU page table framework
Date: Mon, 01 Dec 2014 00:03:08 +0200	[thread overview]
Message-ID: <6034238.mfQ54vFFKj@avalon> (raw)
In-Reply-To: <1417089078-22900-1-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>

Hi Will,

Thank you for the patches.

On Thursday 27 November 2014 11:51:14 Will Deacon wrote:
> Hi all,
> 
> This series introduces a generic IOMMU page table allocation framework,
> implements support for ARM long-descriptors and then ports the arm-smmu
> driver over to the new code.
> 
> There are a few reasons for doing this:
> 
>   - Page table code is hard, and I don't enjoy shopping
> 
>   - A number of IOMMUs actually use the same table format, but currently
>     duplicate the code
> 
>   - It provides a CPU (and architecture) independent allocator, which
>     may be useful for some systems where the CPU is using a different
>     table format for its own mappings
> 
> As illustrated in the final patch, an IOMMU driver interacts with the
> allocator by passing in a configuration structure describing the
> input and output address ranges, the supported pages sizes and a set of
> ops for performing various TLB invalidation and PTE flushing routines.
> 
> The LPAE code implements support for 4k/2M/1G, 16k/32M and 64k/512M
> mappings, but I decided not to implement the contiguous bit in the
> interest of trying to keep the code semi-readable. This could always be
> added later, if needed.

Do you have any idea how much the contiguous bit can improve performances in 
real use cases ?

> I also included some self-tests for the LPAE implementation. Ideally
> we'd merge these, but I'm also happy to drop them if there are
> objections.
> 
> Tested with the self-tests, but also VFIO + MMU-500 at stage-1 and
> stage-2. Patches taken against my iommu/devel branch (queued by Joerg
> for 3.19).
> 
> All feedback welcome.
> 
> Will
> 
> --->8
> 
> Will Deacon (4):
>   iommu: introduce generic page table allocation framework
>   iommu: add ARM LPAE page table allocator
>   iommu: add self-consistency tests to ARM LPAE IO page table allocator
>   iommu/arm-smmu: make use of generic LPAE allocator
> 
>  MAINTAINERS                    |   1 +
>  arch/arm64/Kconfig             |   1 -
>  drivers/iommu/Kconfig          |  32 +-
>  drivers/iommu/Makefile         |   2 +
>  drivers/iommu/arm-smmu.c       | 872 +++++++++++---------------------------
>  drivers/iommu/io-pgtable-arm.c | 925 ++++++++++++++++++++++++++++++++++++++
>  drivers/iommu/io-pgtable.c     |  78 ++++
>  drivers/iommu/io-pgtable.h     |  77 ++++
>  8 files changed, 1361 insertions(+), 627 deletions(-)
>  create mode 100644 drivers/iommu/io-pgtable-arm.c
>  create mode 100644 drivers/iommu/io-pgtable.c
>  create mode 100644 drivers/iommu/io-pgtable.h

-- 
Regards,

Laurent Pinchart

WARNING: multiple messages have this Message-ID (diff)
From: laurent.pinchart@ideasonboard.com (Laurent Pinchart)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 0/4] Generic IOMMU page table framework
Date: Mon, 01 Dec 2014 00:03:08 +0200	[thread overview]
Message-ID: <6034238.mfQ54vFFKj@avalon> (raw)
In-Reply-To: <1417089078-22900-1-git-send-email-will.deacon@arm.com>

Hi Will,

Thank you for the patches.

On Thursday 27 November 2014 11:51:14 Will Deacon wrote:
> Hi all,
> 
> This series introduces a generic IOMMU page table allocation framework,
> implements support for ARM long-descriptors and then ports the arm-smmu
> driver over to the new code.
> 
> There are a few reasons for doing this:
> 
>   - Page table code is hard, and I don't enjoy shopping
> 
>   - A number of IOMMUs actually use the same table format, but currently
>     duplicate the code
> 
>   - It provides a CPU (and architecture) independent allocator, which
>     may be useful for some systems where the CPU is using a different
>     table format for its own mappings
> 
> As illustrated in the final patch, an IOMMU driver interacts with the
> allocator by passing in a configuration structure describing the
> input and output address ranges, the supported pages sizes and a set of
> ops for performing various TLB invalidation and PTE flushing routines.
> 
> The LPAE code implements support for 4k/2M/1G, 16k/32M and 64k/512M
> mappings, but I decided not to implement the contiguous bit in the
> interest of trying to keep the code semi-readable. This could always be
> added later, if needed.

Do you have any idea how much the contiguous bit can improve performances in 
real use cases ?

> I also included some self-tests for the LPAE implementation. Ideally
> we'd merge these, but I'm also happy to drop them if there are
> objections.
> 
> Tested with the self-tests, but also VFIO + MMU-500 at stage-1 and
> stage-2. Patches taken against my iommu/devel branch (queued by Joerg
> for 3.19).
> 
> All feedback welcome.
> 
> Will
> 
> --->8
> 
> Will Deacon (4):
>   iommu: introduce generic page table allocation framework
>   iommu: add ARM LPAE page table allocator
>   iommu: add self-consistency tests to ARM LPAE IO page table allocator
>   iommu/arm-smmu: make use of generic LPAE allocator
> 
>  MAINTAINERS                    |   1 +
>  arch/arm64/Kconfig             |   1 -
>  drivers/iommu/Kconfig          |  32 +-
>  drivers/iommu/Makefile         |   2 +
>  drivers/iommu/arm-smmu.c       | 872 +++++++++++---------------------------
>  drivers/iommu/io-pgtable-arm.c | 925 ++++++++++++++++++++++++++++++++++++++
>  drivers/iommu/io-pgtable.c     |  78 ++++
>  drivers/iommu/io-pgtable.h     |  77 ++++
>  8 files changed, 1361 insertions(+), 627 deletions(-)
>  create mode 100644 drivers/iommu/io-pgtable-arm.c
>  create mode 100644 drivers/iommu/io-pgtable.c
>  create mode 100644 drivers/iommu/io-pgtable.h

-- 
Regards,

Laurent Pinchart

  parent reply	other threads:[~2014-11-30 22:03 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-27 11:51 [PATCH 0/4] Generic IOMMU page table framework Will Deacon
2014-11-27 11:51 ` Will Deacon
     [not found] ` <1417089078-22900-1-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2014-11-27 11:51   ` [PATCH 1/4] iommu: introduce generic page table allocation framework Will Deacon
2014-11-27 11:51     ` Will Deacon
     [not found]     ` <1417089078-22900-2-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2014-11-30 22:00       ` Laurent Pinchart
2014-11-30 22:00         ` Laurent Pinchart
2014-12-01 12:13         ` Will Deacon
2014-12-01 12:13           ` Will Deacon
     [not found]           ` <20141201121338.GD18466-5wv7dgnIgG8@public.gmane.org>
2014-12-01 13:33             ` Laurent Pinchart
2014-12-01 13:33               ` Laurent Pinchart
2014-12-01 13:53               ` Will Deacon
2014-12-01 13:53                 ` Will Deacon
2014-12-14 23:46       ` Laurent Pinchart
2014-12-14 23:46         ` Laurent Pinchart
2014-12-15  9:45         ` Will Deacon
2014-12-15  9:45           ` Will Deacon
2014-11-27 11:51   ` [PATCH 2/4] iommu: add ARM LPAE page table allocator Will Deacon
2014-11-27 11:51     ` Will Deacon
     [not found]     ` <1417089078-22900-3-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2014-11-30 23:29       ` Laurent Pinchart
2014-11-30 23:29         ` Laurent Pinchart
2014-12-01 17:23         ` Will Deacon
2014-12-01 17:23           ` Will Deacon
     [not found]           ` <20141201172315.GI18466-5wv7dgnIgG8@public.gmane.org>
2014-12-01 20:21             ` Laurent Pinchart
2014-12-01 20:21               ` Laurent Pinchart
2014-12-02  9:41               ` Will Deacon
2014-12-02  9:41                 ` Will Deacon
     [not found]                 ` <20141202094156.GB9917-5wv7dgnIgG8@public.gmane.org>
2014-12-02 11:47                   ` Laurent Pinchart
2014-12-02 11:47                     ` Laurent Pinchart
2014-12-05 18:48                     ` Will Deacon
2014-12-05 18:48                       ` Will Deacon
2014-12-02 22:41       ` Mitchel Humpherys
2014-12-02 22:41         ` Mitchel Humpherys
     [not found]         ` <vnkw8uipznbj.fsf-Yf+dfxj6toJBVvN7MMdr1KRtKmQZhJ7pQQ4Iyu8u01E@public.gmane.org>
2014-12-03 11:11           ` Will Deacon
2014-12-03 11:11             ` Will Deacon
2014-12-05 10:55       ` Varun Sethi
2014-12-05 10:55         ` Varun Sethi
     [not found]         ` <BN3PR0301MB12198CE5D736CDC6A221EDC2EA790-CEkquS/Gb81uuip9JPHoc5wN6zqB+hSMnBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
2014-12-05 18:48           ` Will Deacon
2014-12-05 18:48             ` Will Deacon
     [not found]             ` <20141205184802.GH1203-5wv7dgnIgG8@public.gmane.org>
2014-12-14 17:45               ` Varun Sethi
2014-12-14 17:45                 ` Varun Sethi
     [not found]                 ` <BN3PR0301MB1219D3161E4E9DB314FDD8FAEA6E0-CEkquS/Gb81uuip9JPHoc5wN6zqB+hSMnBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
2014-12-15 13:30                   ` Will Deacon
2014-12-15 13:30                     ` Will Deacon
     [not found]                     ` <20141215133020.GJ20738-5wv7dgnIgG8@public.gmane.org>
2014-12-15 15:43                       ` Will Deacon
2014-12-15 15:43                         ` Will Deacon
2014-12-15 16:35                         ` Varun Sethi
2014-12-15 16:35                           ` Varun Sethi
     [not found]                           ` <BN3PR0301MB12194A8F5CFF870B7A124623EA6F0-CEkquS/Gb81uuip9JPHoc5wN6zqB+hSMnBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
2014-12-15 17:25                             ` Will Deacon
2014-12-15 17:25                               ` Will Deacon
2014-12-15 16:43                       ` Varun Sethi
2014-12-15 16:43                         ` Varun Sethi
     [not found]                         ` <BN3PR0301MB12199C5CAD33E745CC7E51F4EA6F0-CEkquS/Gb81uuip9JPHoc5wN6zqB+hSMnBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
2014-12-15 17:20                           ` Will Deacon
2014-12-15 17:20                             ` Will Deacon
2014-11-27 11:51   ` [PATCH 3/4] iommu: add self-consistency tests to ARM LPAE IO " Will Deacon
2014-11-27 11:51     ` Will Deacon
2014-11-27 11:51   ` [PATCH 4/4] iommu/arm-smmu: make use of generic LPAE allocator Will Deacon
2014-11-27 11:51     ` Will Deacon
2014-11-30 22:03   ` Laurent Pinchart [this message]
2014-11-30 22:03     ` [PATCH 0/4] Generic IOMMU page table framework Laurent Pinchart
2014-12-01 12:05     ` Will Deacon
2014-12-01 12:05       ` Will Deacon
     [not found]       ` <20141201120534.GC18466-5wv7dgnIgG8@public.gmane.org>
2014-12-02 13:47         ` Laurent Pinchart
2014-12-02 13:47           ` Laurent Pinchart
2014-12-02 13:53           ` Will Deacon
2014-12-02 13:53             ` Will Deacon
     [not found]             ` <20141202135356.GF9917-5wv7dgnIgG8@public.gmane.org>
2014-12-02 22:29               ` Laurent Pinchart
2014-12-02 22:29                 ` Laurent Pinchart
2014-12-14 23:49   ` Laurent Pinchart
2014-12-14 23:49     ` Laurent Pinchart
2014-12-15 16:10     ` Will Deacon
2014-12-15 16:10       ` Will Deacon
     [not found]       ` <20141215161052.GM20738-5wv7dgnIgG8@public.gmane.org>
2014-12-15 17:33         ` Laurent Pinchart
2014-12-15 17:33           ` Laurent Pinchart
2014-12-15 17:39           ` Will Deacon
2014-12-15 17:39             ` Will Deacon
     [not found]             ` <20141215173911.GT20738-5wv7dgnIgG8@public.gmane.org>
2014-12-15 17:46               ` Laurent Pinchart
2014-12-15 17:46                 ` Laurent Pinchart

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=6034238.mfQ54vFFKj@avalon \
    --to=laurent.pinchart-rylnwiuwjnjg/c1bvhzhaw@public.gmane.org \
    --cc=Robin.Murphy-5wv7dgnIgG8@public.gmane.org \
    --cc=Varun.Sethi-KZfg59tc24xl57MIdRCFDg@public.gmane.org \
    --cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
    --cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
    --cc=prem.mallappa-dY08KVG/lbpWk0Htik3J/w@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.