[PATCH v4 0/5] iommu: Add Allwinner H6 IOMMU driver

From: Maxime Ripard <maxime@cerno.tech>
To: Joerg Roedel <joro@8bytes.org>, Chen-Yu Tsai <wens@csie.org>,
	Maxime Ripard <mripard@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Rob Herring <robh+dt@kernel.org>,
	Frank Rowand <frowand.list@gmail.com>
Cc: devicetree@vger.kernel.org, iommu@lists.linux-foundation.org,
	Maxime Ripard <maxime@cerno.tech>,
	linux-arm-kernel@lists.infradead.org
Subject: [PATCH v4 0/5] iommu: Add Allwinner H6 IOMMU driver
Date: Wed, 13 May 2020 16:07:19 +0200	[thread overview]
Message-ID: <cover.b27dedd61e008ffcf55a028ccddda3bb4d21dfc8.1589378833.git-series.maxime@cerno.tech> (raw)

Hi,

Here's a series adding support for the IOMMU introduced in the Allwinner
H6. The driver from Allwinner hints at more SoCs using it in the future
(with more masters), so we can bet on that IOMMU becoming pretty much
standard in new SoCs from Allwinner.

One thing I wasn't really sure about was how to expose the statistics
reported by the IOMMU PMU (TLB hit rates, latencies, and so on). The
Allwinner driver exposes them through custom sysfs files, while they would
be best represented through perf I guess? Anyway, I'm planning to support
them later on.

Patches 1 and 3 should go through the IOMMU tree, 2 and 5 through drm-misc
and 4 through arm-soc.

Let me know what you think,
Maxime

Changes from v3:
  - Rebased on next
  - Removed PTW flush on map / unmap
  - Added Rob's reviewed-by

Changes from v2:
  - Rebased on 5.7
  - Add dt bindings patch
  - Allow the identity domain to be allocated
  - Add an unlikely to the check on whether a PTE already exists in map
  - Remove locking and adjust the PT installation to use an atomic
    operation instead
  - Switch to iotlb_sync / flush_iotlb_all callback instead of flushing by
    ourselves.

Changes from v1:
  - Add a patch to configure the IOMMU on the virtual DRM device
  - Rework the domain allocation / freeing
  - Remove the runtime_pm handling to power up the device and rely on
    refcounting
  - use map gfp argument for kmem cache allocation
  - Removed unused macros
  - Switched from BIT(0) to 1 for the page table entry valid flag to make
    it more obvious that it's over multiple bits.
  - Switch to module_initcall
  - Make accesses to the fwspec more consistant
  - Removed dev_info logs
  - Reworked invalidation / flushing
  - Allow for compilation with COMPILE_TEST

Maxime Ripard (5):
  dt-bindings: iommu: Add Allwinner H6 IOMMU bindings
  dt-bindings: display: sun8i-mixer: Allow for an iommu property
  iommu: Add Allwinner H6 IOMMU driver
  arm64: dts: allwinner: h6: Add IOMMU
  drm/sun4i: mixer: Call of_dma_configure if there's an IOMMU

 Documentation/devicetree/bindings/display/allwinner,sun8i-a83t-de2-mixer.yaml |    3 +-
 Documentation/devicetree/bindings/iommu/allwinner,sun50i-h6-iommu.yaml        |   61 ++++-
 arch/arm64/boot/dts/allwinner/sun50i-h6.dtsi                                  |   10 +-
 drivers/gpu/drm/sun4i/sun8i_mixer.c                                           |   13 +-
 drivers/iommu/Kconfig                                                         |    9 +-
 drivers/iommu/Makefile                                                        |    1 +-
 drivers/iommu/sun50i-iommu.c                                                  | 1027 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
 7 files changed, 1124 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/iommu/allwinner,sun50i-h6-iommu.yaml
 create mode 100644 drivers/iommu/sun50i-iommu.c

base-commit: e098d7762d602be640c53565ceca342f81e55ad2
-- 
git-series 0.9.1
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu