From: will.deacon@arm.com (Will Deacon)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 0/3] iommu/arm-smmu: Add driver for ARM SMMUv3 devices
Date: Fri, 8 May 2015 19:00:43 +0100 [thread overview]
Message-ID: <1431108046-9675-1-git-send-email-will.deacon@arm.com> (raw)
Hi all,
SMMUv3 is a significant departure from previous ARM SMMU designs in that
the majority of the configuration data has been moved from MMIO registers
to in-memory data structures, with communication between the CPU and the
SMMU being mediated via in-memory circular queues. It also has native
support for MSI generation, PCI masters (e.g. ATS, PRI, PASIDs), ARM 8.1
architectural extensions (e.g. hardware translation table updates) and
optional RAS extensions. It is designed for large numbers of masters and
consequently uses a 32-bit StreamID.
This patch series adds an initial driver for ARM SMMUv3 devices, which
enables dma-mapping and VFIO for PCI masters. Unfortunately, there is
not yet any documentation available for the hardware, but I have tested
this successfully on internal simulation. Having the driver available at
this early stage is useful for ARM partners building implementations and
provides a common upstream base on which the driver can be developed.
Future work will look at:
- DMA to paged memory using PRI (for now we send a PRI_DENY response
and dump the transaction to the console)
- MSI generation from the SMMU (requires generic changes to the
irqdomain code)
- Support for non-PCI masters (pending core of_xlate rework as
discussed on linux-arm-kernel)
There's also the nested translation angle, but we should work out what
we're doing on SMMUv2 before implementing anything here.
All feedback welcome,
Will
--->8
Will Deacon (3):
Documentation: dt-bindings: Add device-tree binding for ARM SMMUv3
IOMMU
iommu/arm-smmu: Add initial driver support for ARM SMMUv3 devices
drivers/vfio: Allow type-1 IOMMU instantiation on top of an ARM SMMUv3
.../devicetree/bindings/iommu/arm,smmu-v3.txt | 37 +
MAINTAINERS | 3 +-
drivers/iommu/Kconfig | 13 +
drivers/iommu/Makefile | 1 +
drivers/iommu/arm-smmu-v3.c | 2599 ++++++++++++++++++++
drivers/vfio/Kconfig | 2 +-
6 files changed, 2653 insertions(+), 2 deletions(-)
create mode 100644 Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt
create mode 100644 drivers/iommu/arm-smmu-v3.c
--
2.1.4
next reply other threads:[~2015-05-08 18:00 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-08 18:00 Will Deacon [this message]
2015-05-08 18:00 ` [PATCH 1/3] Documentation: dt-bindings: Add device-tree binding for ARM SMMUv3 IOMMU Will Deacon
2015-05-08 18:00 ` [PATCH 2/3] iommu/arm-smmu: Add initial driver support for ARM SMMUv3 devices Will Deacon
2015-05-12 7:40 ` leizhen
2015-05-12 16:55 ` Will Deacon
2015-05-13 8:33 ` leizhen
2015-05-21 11:25 ` Will Deacon
2015-05-25 2:07 ` leizhen
2015-05-26 16:12 ` Will Deacon
2015-05-27 9:12 ` leizhen
2015-05-19 15:24 ` Joerg Roedel
2015-05-20 17:09 ` Will Deacon
2015-05-29 6:43 ` Joerg Roedel
2015-05-29 11:35 ` Robin Murphy
2015-05-29 14:40 ` Joerg Roedel
2015-06-01 9:40 ` Will Deacon
2015-06-02 7:39 ` Joerg Roedel
2015-06-02 9:47 ` Will Deacon
2015-06-02 18:43 ` Joerg Roedel
2015-05-08 18:00 ` [PATCH 3/3] drivers/vfio: Allow type-1 IOMMU instantiation on top of an ARM SMMUv3 Will Deacon
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=1431108046-9675-1-git-send-email-will.deacon@arm.com \
--to=will.deacon@arm.com \
--cc=linux-arm-kernel@lists.infradead.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).