From: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
To: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org
Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org,
Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
Subject: [PATCH 0/9] Add support for ARM SMMU architectures 1 and 2
Date: Mon, 10 Jun 2013 19:34:36 +0100 [thread overview]
Message-ID: <1370889285-22799-1-git-send-email-will.deacon@arm.com> (raw)
Hi all,
This patch series adds support for ARM's SMMU architectures 1 and 2 to
Linux. It has been tested with models of MMU-400 (AArch32) and MMU-500
(AArch32 and AArch64) coupled with four PL330s doing memory-to-memory
DMA.
The first six patches are preparatory and fix problems that I
encountered whilst trying to use both the ARM IOMMU API and the PL330
driver. The remaining patches add the device-tree binding, previously
discussed at:
http://lists.infradead.org/pipermail/linux-arm-kernel/2013-April/160663.html
http://lists.infradead.org/pipermail/linux-arm-kernel/2013-April/162945.html
as well as the driver implementation and an update to the MAINTAINERS
file.
The driver is written only with single-stage (possibly stage 2), LPAE,
non-secure translation in mind. Nested translation may of course be added
later via KVM, but that will require some changes to the core IOMMU API
in Linux. Both arm and arm64 are supported, as well as chained SMMUs,
although we only install our translation in the SMMU furthest from the
device. We're also currently limited in our IPA size, due to input size
restrictions at stage 2. This can be resolved later by munging about in
the page table allocation code, where we currently re-use parts of the
CPU page table helpers.
Read and write protection is supported as far as the pte formats allow
(i.e. no write-only at stage-1) and memory attributes are either normal,
non-cacheable (default) or normal, cacheable, write-back, write-allocate.
TLB broadcasting is not used for either ASIDs (since we don't support
user tables for DMA in Linux) or VMIDs (which are instead used to tag
address spaces in order to limit the scope of TLB invalidation
operations).
Both 4k and 64k (AArch64 only) pages are supported, with the contiguous
hint bit being used in the pte entries for mappings that allow it.
Support for stream-indexing and non-coherent table walking is provided,
but untested.
All comments welcome,
Will
Will Deacon (9):
dma: pl330: rip out broken, redundant ID probing
dma: pl330: use dma_addr_t for describing bus addresses
ARM: dma-mapping: convert DMA direction into IOMMU protection
attributes
ARM: dma-mapping: NULLify dev->archdata.mapping pointer on detach
arm64: pgtable: use pte_index instead of __pte_index
arm64: device: add iommu pointer to device archdata
documentation: iommu: add description of ARM System MMU binding
iommu: add support for ARM Ltd. System MMU architecture
MAINTAINERS: add entry for ARM system MMU driver
.../devicetree/bindings/iommu/arm,smmu.txt | 70 +
MAINTAINERS | 6 +
arch/arm/mm/dma-mapping.c | 20 +-
arch/arm64/include/asm/device.h | 3 +
arch/arm64/include/asm/pgtable.h | 4 +-
drivers/dma/pl330.c | 29 +-
drivers/iommu/Kconfig | 13 +
drivers/iommu/Makefile | 1 +
drivers/iommu/arm-smmu.c | 1965 ++++++++++++++++++++
9 files changed, 2081 insertions(+), 30 deletions(-)
create mode 100644 Documentation/devicetree/bindings/iommu/arm,smmu.txt
create mode 100644 drivers/iommu/arm-smmu.c
--
1.8.2.2
next reply other threads:[~2013-06-10 18:34 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-10 18:34 Will Deacon [this message]
[not found] ` <1370889285-22799-1-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-10 18:34 ` [PATCH 1/9] dma: pl330: rip out broken, redundant ID probing Will Deacon
2013-06-11 4:40 ` Jassi Brar
[not found] ` <CAJe_Zhc1UoTC4q4oaW=dzyi_10Q7EoezoT=G8_v+yCmBxV75+A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-06-11 8:45 ` Will Deacon
[not found] ` <1370889285-22799-2-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-11 4:37 ` Jassi Brar
2013-06-11 22:31 ` Grant Likely
2013-06-12 5:31 ` Vinod Koul
2013-06-10 18:34 ` [PATCH 2/9] dma: pl330: use dma_addr_t for describing bus addresses Will Deacon
[not found] ` <1370889285-22799-3-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-11 4:37 ` Jassi Brar
[not found] ` <CAJe_ZheKMVQgq42Vx5N1TXXdgFJ2sp50ixU30A7beXhmSVHnZQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-06-12 5:31 ` Vinod Koul
2013-06-11 22:32 ` Grant Likely
2013-06-11 4:39 ` Jassi Brar
2013-06-10 18:34 ` [PATCH 3/9] ARM: dma-mapping: convert DMA direction into IOMMU protection attributes Will Deacon
2013-06-19 8:37 ` Marek Szyprowski
[not found] ` <51C16DAF.1090205-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org>
2013-06-19 8:52 ` Will Deacon
[not found] ` <20130619085202.GC20351-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-19 8:57 ` Marek Szyprowski
[not found] ` <1370889285-22799-4-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-25 10:12 ` Hiroshi Doyu
[not found] ` <20130625131215.d3cea2a5668a3d41dbbeb064-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2013-06-25 11:37 ` Will Deacon
[not found] ` <20130625113714.GF31838-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-25 11:52 ` Hiroshi Doyu
[not found] ` <20130625.145226.1632119404634300971.hdoyu-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2013-06-25 12:34 ` Will Deacon
2013-06-10 18:34 ` [PATCH 4/9] ARM: dma-mapping: NULLify dev->archdata.mapping pointer on detach Will Deacon
[not found] ` <1370889285-22799-5-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-11 5:34 ` Hiroshi Doyu
[not found] ` <20130611.083455.1500863288897785600.hdoyu-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2013-06-11 8:50 ` Will Deacon
[not found] ` <20130611085015.GC24729-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-11 9:39 ` Hiroshi Doyu
[not found] ` <20130611123933.4d278ff4e056f395788ad060-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2013-06-19 8:59 ` Marek Szyprowski
2013-06-10 18:34 ` [PATCH 5/9] arm64: pgtable: use pte_index instead of __pte_index Will Deacon
2013-06-10 18:34 ` [PATCH 6/9] arm64: device: add iommu pointer to device archdata Will Deacon
2013-06-10 18:34 ` [PATCH 7/9] documentation: iommu: add description of ARM System MMU binding Will Deacon
[not found] ` <1370889285-22799-8-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-12 8:44 ` Grant Likely
2013-06-20 20:08 ` Joerg Roedel
[not found] ` <20130620200845.GF11309-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2013-06-21 9:57 ` Will Deacon
[not found] ` <20130621095729.GA7766-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-21 13:55 ` Joerg Roedel
[not found] ` <20130621135507.GI11309-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2013-06-21 16:41 ` Will Deacon
2013-06-25 19:18 ` Stuart Yoder
[not found] ` <CALRxmdBxFWoRKv+bUu8VEwNNcAJUej9jM2V8N0rrqrr_Vpe8fQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-06-26 13:39 ` Will Deacon
[not found] ` <20130626133941.GD7417-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-26 16:19 ` Stuart Yoder
[not found] ` <CALRxmdCycFK2wW=C4aU79mudSaT+2vU8nzXxepdstubg+YSdQg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-06-26 17:42 ` Will Deacon
[not found] ` <20130626174231.GH10333-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-27 18:22 ` Stuart Yoder
[not found] ` <CALRxmdD5fyp06xW+z=rWagJc_bcJmpr1H9Zbdf=xbg9cCzvVfw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-06-28 9:06 ` Will Deacon
[not found] ` <20130628090635.GB29002-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-28 16:03 ` Stuart Yoder
2013-06-10 18:34 ` [PATCH 8/9] iommu: add support for ARM Ltd. System MMU architecture Will Deacon
[not found] ` <1370889285-22799-9-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-20 21:26 ` Joerg Roedel
[not found] ` <20130620212646.GG11309-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2013-06-21 10:23 ` Will Deacon
[not found] ` <20130621102318.GB7766-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-21 14:13 ` Joerg Roedel
2013-06-21 15:00 ` Will Deacon
[not found] ` <20130621150006.GG7766-MRww78TxoiP5vMa5CHWGZ34zcgK1vI+I0E9HWUfgJXw@public.gmane.org>
2013-06-21 15:30 ` Joerg Roedel
[not found] ` <20130621153044.GL11309-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2013-06-21 16:40 ` Will Deacon
2013-06-10 18:34 ` [PATCH 9/9] MAINTAINERS: add entry for ARM system MMU driver Will Deacon
[not found] ` <1370889285-22799-10-git-send-email-will.deacon-5wv7dgnIgG8@public.gmane.org>
2013-06-12 8:45 ` Grant Likely
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=1370889285-22799-1-git-send-email-will.deacon@arm.com \
--to=will.deacon-5wv7dgnigg8@public.gmane.org \
--cc=devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).