Re: [RFC PATCH v3 0/7] Introduce automatic DMA configuration for IOMMU masters

2014-09-17 Thread Robin Murphy
Hi Will, On 12/09/14 17:34, Will Deacon wrote: Hi all, Here is version three of the RFC I've previously posted here: RFCv1: http://lists.infradead.org/pipermail/linux-arm-kernel/2014-August/283023.html RFCv2:

Re: [RFC PATCH v3 1/7] iommu: provide early initialisation hook for IOMMU drivers

2014-09-18 Thread Robin Murphy
Hi Will, After some fun times wondering why on Earth of_iommu_init was trying to instantiate a GIC, I think we may need one of these as part of this patch, too ;) Robin. ---8--- diff --git a/drivers/iommu/of_iommu.c b/drivers/iommu/of_iommu.c index 8656b63..1927691 100644 ---

Re: [RFC PATCH v4 8/8] arm: dma-mapping: plumb our iommu mapping ops into arch_setup_dma_ops

2014-11-17 Thread Robin Murphy
Hi Will, On 14/11/14 18:56, Will Deacon wrote: This patch plumbs the existing ARM IOMMU DMA infrastructure (which isn't actually called outside of a few drivers) into arch_setup_dma_ops, so that we can use IOMMUs for DMA transfers in a more generic fashion. Since this significantly complicates

Re: [RFC PATCH v4 6/8] dma-mapping: set dma segment properties in of_dma_configure

2014-11-25 Thread Robin Murphy
Hi Will, On 14/11/14 18:56, Will Deacon wrote: of_dma_configure determines the size of the DMA range for a device by either parsing the dma-ranges property or inspecting the coherent DMA mask. This same information can be used to initialise the max segment size and boundary_mask to a default

Re: [RFC PATCH v4 0/8] Introduce automatic DMA configuration for IOMMU masters

2014-11-25 Thread Robin Murphy
mapping running on arm64, so with the proviso that patch 6 definitely causes problems, and patches 7 and 8 ported to arm64 rather than used directly, for the series: Tested-by: Robin Murphy robin.mur...@arm.com Hopefully it's now stable enough that I can finish cleaning up my hacked-up SMMU

[RFC PATCH 2/4] iommu: consolidate IOVA allocator code

2014-11-25 Thread Robin Murphy
In order to share the IOVA allocator with other architectures, break the unnecssary dependency on the Intel IOMMU driver and move the remaining IOVA internals to iova.c Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/intel-iommu.c | 33

[RFC PATCH 0/4] Genericise the IOVA allocator

2014-11-25 Thread Robin Murphy
], that frankly looks nicer ;) I've merely left them in as context here. [1]:http://thread.gmane.org/gmane.linux.ports.arm.kernel/331299 [2]:http://article.gmane.org/gmane.linux.kernel.iommu/7436 Robin Murphy (4): iommu: build iova.c for any IOMMU iommu: consolidate IOVA allocator code iommu

[RFC PATCH 4/4] iommu: make IOVA domain page size explicit

2014-11-25 Thread Robin Murphy
page granularity from an implicit compile-time constant to a per-domain property so we can make use of it in IOVA domain context at runtime. To keep the abstraction tidy, extend the little API of inline iova_* helpers to parallel some of the equivalent PAGE_* macros. Signed-off-by: Robin Murphy

[RFC PATCH 1/4] iommu: build iova.c for any IOMMU

2014-11-25 Thread Robin Murphy
In preparation for sharing the IOVA allocator, build it for all IOMMU API users. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile index 16edef7

[RFC PATCH 3/4] iommu: make IOVA domain low limit flexible

2014-11-25 Thread Robin Murphy
description of alloc_iova since we're touching it anyway. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/intel-iommu.c | 9 ++--- drivers/iommu/iova.c| 10 ++ include/linux/iova.h| 7 +++ 3 files changed, 15 insertions(+), 11 deletions(-) diff --git

[PATCH] iommu: Decouple iommu_map_sg from CPU page size

2014-11-25 Thread Robin Murphy
be mapped and therefore exposed to the device, and brings the default iommu_map_sg implementation in line with iommu_map/unmap with respect to alignment. Signed-off-by: Robin Murphy robin.mur...@arm.com --- Hi Joerg, I noticed this whilst wiring up DMA mapping to this new API - on arm64 we anticipate

Re: [RFC PATCH 4/4] iommu: make IOVA domain page size explicit

2014-11-26 Thread Robin Murphy
Hi, On 26/11/14 07:17, leizhen wrote: [...] diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index a3dbba8..9e50625 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -55,12 +55,16 @@ void free_iova_mem(struct iova *iova) } void -init_iova_domain(struct iova_domain

Re: [PATCH v6 1/8] iommu: provide early initialisation hook for IOMMU drivers

2014-12-04 Thread Robin Murphy
, but it didn't blow up or regress anything. Regards, Robin. ---8--- From 1f3d2612682c239e53f2c20e2ac5d19ef3f5387c Mon Sep 17 00:00:00 2001 Message-Id: 1f3d2612682c239e53f2c20e2ac5d19ef3f5387c.1417695078.git.robin.mur...@arm.com From: Robin Murphy robin.mur...@arm.com Date: Thu, 4 Dec 2014 11:53:13

Re: [PATCH v6 1/8] iommu: provide early initialisation hook for IOMMU drivers

2014-12-04 Thread Robin Murphy
: b2e8c91ac49bef4008661e4628cd6b7249d84af5.1417698001.git.robin.mur...@arm.com From: Robin Murphy robin.mur...@arm.com Date: Thu, 4 Dec 2014 11:53:13 + Subject: [PATCH v2] iommu: store DT-probed IOMMU data privately Since the data pointer in the DT node is public and may be overwritten by conflicting code

Re: [PATCH v6 1/8] iommu: provide early initialisation hook for IOMMU drivers

2014-12-04 Thread Robin Murphy
Hi Thierry, On 04/12/14 14:49, Thierry Reding wrote: On Thu, Dec 04, 2014 at 01:43:32PM +, Robin Murphy wrote: Hi Grant, thanks for the advice - silly micro-optimisations removed, and I'll make a note to do so from my in-development code, too ;) I didn't much like the casting either, so

Re: [PATCH v6 1/8] iommu: provide early initialisation hook for IOMMU drivers

2014-12-04 Thread Robin Murphy
for this fix - I'm just keen to get the merge-blocker out of the way. Regards, Robin. ---8--- From 9eba5081aaf4fa8ed5158675a6e622be11a64ae2 Mon Sep 17 00:00:00 2001 Message-Id: 9eba5081aaf4fa8ed5158675a6e622be11a64ae2.1417719305.git.robin.mur...@arm.com From: Robin Murphy robin.mur...@arm.com Date

Re: [PATCH v6 1/8] iommu: provide early initialisation hook for IOMMU drivers

2014-12-05 Thread Robin Murphy
Hi Will, On 05/12/14 12:10, Will Deacon wrote: [...] Do you expect drivers to modify that *priv pointer after the ops structure is registered? I'd be very surprised if that was the use case. It's fine for the driver to register a non-const version, but once it is registered, the infrastructure

Re: [PATCH v6 6/8] dma-mapping: detect and configure IOMMU in of_dma_configure

2014-12-10 Thread Robin Murphy
mapping ops. Acked-by: Arnd Bergmann a...@arndb.de Acked-by: Marek Szyprowski m.szyprow...@samsung.com Tested-by: Robin Murphy robin.mur...@arm.com Signed-off-by: Will Deacon will.dea...@arm.com --- arch/arm/include/asm/dma-mapping.h | 4 +++- drivers/of/platform.c | 21

Re: [PATCH v6 6/8] dma-mapping: detect and configure IOMMU in of_dma_configure

2014-12-17 Thread Robin Murphy
Hi Will, On 17/12/14 12:09, Will Deacon wrote: On Tue, Dec 16, 2014 at 12:08:15PM +, Arnd Bergmann wrote: On Monday 15 December 2014 18:09:33 Will Deacon wrote: Using a single domain is a bit of a waste of resources in my case, so an evolution would be to create four domains and assign

Re: [RFC PATCH 3/5] iommu: implement common IOMMU ops for DMA mapping

2015-01-23 Thread Robin Murphy
Hi Laura, Thanks for looking On 23/01/15 17:42, Laura Abbott wrote: On 1/12/2015 12:48 PM, Robin Murphy wrote: Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API. This will do the bulk of the heavy lifting for IOMMU

Re: [RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-01-23 Thread Robin Murphy
On 23/01/15 16:47, Catalin Marinas wrote: On Mon, Jan 12, 2015 at 08:48:52PM +, Robin Murphy wrote: Catalin Marinas (1): arm64: Combine coherent and non-coherent swiotlb dma_ops Robin Murphy (4): arm64: implement generic IOMMU configuration iommu: implement common IOMMU ops

Re: [RFC PATCH 4/5] arm64: add IOMMU dma_ops

2015-01-23 Thread Robin Murphy
Hi Will, Thanks for taking a look On 23/01/15 15:26, Will Deacon wrote: On Mon, Jan 12, 2015 at 08:48:56PM +, Robin Murphy wrote: Taking some inspiration from the arch/arm code, implement the arch-specific side of the DMA mapping ops using the new IOMMU-DMA layer. Signed-off-by: Robin

[RFC PATCH v2 2/3] arm64: add IOMMU dma_ops

2015-02-06 Thread Robin Murphy
Taking some inspiration from the arch/arm code, implement the arch-specific side of the DMA mapping ops using the new IOMMU-DMA layer. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/include/asm/device.h | 3 + arch/arm64/include/asm/dma-mapping.h | 17 ++ arch/arm64/mm

[RFC PATCH v2 1/3] iommu: implement common IOMMU ops for DMA mapping

2015-02-06 Thread Robin Murphy
in the common code are addressed we can complete the refactoring by porting arch/arm over for a merge-worthy series. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/Kconfig | 7 + drivers/iommu/Makefile| 1 + drivers/iommu/dma-iommu.c | 552

[RFC PATCH v2 0/3] arm64: IOMMU-backed DMA mapping

2015-02-06 Thread Robin Murphy
://thread.gmane.org/gmane.linux.ports.tegra/20907 [3]:http://thread.gmane.org/gmane.linux.kernel.iommu/8492 Robin Murphy (3): iommu: implement common IOMMU ops for DMA mapping arm64: add IOMMU dma_ops arm64: hook up IOMMU dma_ops arch/arm64/Kconfig | 1 + arch/arm64/include/asm

[RFC PATCH v2 3/3] arm64: hook up IOMMU dma_ops

2015-02-06 Thread Robin Murphy
With iommu_dma_ops in place, hook them up to the configuration code, so IOMMU-fronted devices will get them automatically. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/Kconfig | 1 + arch/arm64/include/asm/dma-mapping.h | 11 ++- arch/arm64/mm/dma

Re: [PATCH/RFC 4/4] iommu/arm-smmu: Support deferred probing

2015-02-06 Thread Robin Murphy
Hi Laura, As a heads up, I'm still vainly hoping to move the ARM SMMU driver entirely over to the generic framework - there's an iommu/dev branch on top of the iommu/dma branch I pushed earlier[1] which you might want to take a peek at to check if we're likely to end up pulling in different

Re: [RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-01-15 Thread Robin Murphy
On 13/01/15 08:02, Yingjoe Chen wrote: On Mon, 2015-01-12 at 20:48 +, Robin Murphy wrote: Hi all, Whilst it's a long way off perfect, this has reached the point of being functional and stable enough to be useful, so here it is. The core consists of the meat of the arch/arm implementation

Re: [RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-01-16 Thread Robin Murphy
On 16/01/15 07:21, Yong Wu wrote: [...] Now that the server seems to be properly accessible again, I've made a branch with both series available here: git://linux-arm.org/linux-rm iommu/dma Note that in the current state it also depends on having the ARM SMMU driver selected in the config,

Re: [RFC PATCH v2 1/3] iommu: implement common IOMMU ops for DMA mapping

2015-02-10 Thread Robin Murphy
Hi Will, Thanks for the comments, On 09/02/15 04:05, Will Deacon wrote: Hi Robin, On Fri, Feb 06, 2015 at 02:55:13PM +, Robin Murphy wrote: Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API. This will do

Re: [PATCH v4 3/6] of: fix size when dma-range is not used

2015-01-27 Thread Robin Murphy
Hi Murali, On 23/01/15 22:32, Murali Karicheri wrote: Fix the dma-range size when the DT attribute is missing. i.e set size to dev-coherent_dma_mask + 1 instead of dev-coherent_dma_mask. To detect overflow when mask is set to max of u64, add a check, log error and return. Some platform use

Re: [PATCH v4 6/6] arm: dma-mapping: updates to limit dma_mask and iommu mapping size

2015-01-27 Thread Robin Murphy
Hi Murali, On 23/01/15 22:32, Murali Karicheri wrote: Limit the dma_mask to minimum of dma_mask and dma_base + size - 1. Also arm_iommu_create_mapping() has size parameter of size_t and arm_setup_iommu_dma_ops() can take a value higher than that. So limit the size to SIZE_MAX. Signed-off-by:

Re: [PATCH v5 3/8] of: fix size when dma-range is not used

2015-01-28 Thread Robin Murphy
Hi Murali, [sorry, missed replying to yesterday's version] On 27/01/15 21:00, Murali Karicheri wrote: Fix the dma-range size when the DT attribute is missing. i.e set size to dev-coherent_dma_mask + 1 instead of dev-coherent_dma_mask. Also add code to check invalid values of size configured

Re: [PATCH v4 3/6] of: fix size when dma-range is not used

2015-01-28 Thread Robin Murphy
On 28/01/15 11:05, Catalin Marinas wrote: On Tue, Jan 27, 2015 at 06:55:15PM +, Murali Karicheri wrote: On 01/27/2015 06:27 AM, Robin Murphy wrote: On 23/01/15 22:32, Murali Karicheri wrote: Fix the dma-range size when the DT attribute is missing. i.e set size to dev-coherent_dma_mask + 1

Re: [PATCH 2/5] iommu: Allocate a default domain for iommu groups

2015-01-28 Thread Robin Murphy
On 28/01/15 14:30, Will Deacon wrote: On Tue, Jan 27, 2015 at 12:08:56AM +, Joerg Roedel wrote: From: Joerg Roedel jroe...@suse.de The default domain will be used (if supported by the iommu driver) when the devices in the iommu group are not attached to any other domain. Signed-off-by:

Re: [RFC PATCH 3/5] iommu: implement common IOMMU ops for DMA mapping

2015-01-27 Thread Robin Murphy
Hi Joerg, On 27/01/15 00:21, Joerg Roedel wrote: Hi Robin, thanks for the patch, I think it is good start to move forward. See my comments below. On Mon, Jan 12, 2015 at 08:48:55PM +, Robin Murphy wrote: Taking inspiration from the existing arch/arm code, break out some generic functions

Re: [RFC PATCH 4/5] arm64: add IOMMU dma_ops

2015-01-27 Thread Robin Murphy
Hi Joseph, Thanks for giving it a spin, On 26/01/15 03:25, Joseph Lo wrote: On 01/13/2015 04:48 AM, Robin Murphy wrote: Taking some inspiration from the arch/arm code, implement the arch-specific side of the DMA mapping ops using the new IOMMU-DMA layer. Signed-off-by: Robin Murphy robin.mur

Re: [PATCH RESEND] dma-mapping: tidy up dma_parms default handling

2015-01-12 Thread Robin Murphy
On 09/01/15 19:45, Arnd Bergmann wrote: On Friday 09 January 2015 16:56:03 Robin Murphy wrote: This one's a bit tricky to find a home for - I think technically it's probably an IOMMU patch, but then the long-underlying problem doesn't seem to have blown up anything until arm64, and my

Re: [RFC PATCH 0/4] Genericise the IOVA allocator

2015-01-12 Thread Robin Murphy
Hi Joerg, On 12/01/15 15:52, Joerg Roedel wrote: [...] Thanks for doing this, I like this patch-set. I would also appreciate if someone from Intel could have a look at it, David? Besides, can you please re-post this patch-set rebased to latest upstream with the better versions of patch 1 and

[RFC PATCH 5/5] arm64: hook up IOMMU dma_ops

2015-01-12 Thread Robin Murphy
With iommu_dma_ops in place, hook them up to the configuration code, so IOMMU-fronted devices will get them automatically. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/Kconfig | 1 + arch/arm64/include/asm/dma-mapping.h | 10 +- arch/arm64/mm/dma

[RFC PATCH 3/5] iommu: implement common IOMMU ops for DMA mapping

2015-01-12 Thread Robin Murphy
this in common code as the first step towards consolidating the numerous versions spread around between architecture code and IOMMU drivers. Signed-off-by: Robin Murphy robin.mur...@arm.com --- include/linux/dma-iommu.h | 78 lib/Kconfig | 8 + lib/Makefile | 1

[RFC PATCH 1/5] arm64: Combine coherent and non-coherent swiotlb dma_ops

2015-01-12 Thread Robin Murphy
From: Catalin Marinas catalin.mari...@arm.com Since dev_archdata now has a dma_coherent state, combine the two coherent and non-coherent operations and remove their declaration, together with set_dma_ops, from the arch dma-mapping.h file. Signed-off-by: Catalin Marinas catalin.mari...@arm.com

[RFC PATCH 2/5] arm64: implement generic IOMMU configuration

2015-01-12 Thread Robin Murphy
Add the necessary call to of_iommu_init. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/kernel/setup.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index b809911..8304141 100644 --- a/arch/arm64/kernel/setup.c +++ b

[RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-01-12 Thread Robin Murphy
/gmane.linux.kernel.iommu/8208 Catalin Marinas (1): arm64: Combine coherent and non-coherent swiotlb dma_ops Robin Murphy (4): arm64: implement generic IOMMU configuration iommu: implement common IOMMU ops for DMA mapping arm64: add IOMMU dma_ops arm64: hook up IOMMU dma_ops arch/arm64/Kconfig

[RFC PATCH 4/5] arm64: add IOMMU dma_ops

2015-01-12 Thread Robin Murphy
Taking some inspiration from the arch/arm code, implement the arch-specific side of the DMA mapping ops using the new IOMMU-DMA layer. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/include/asm/device.h | 3 + arch/arm64/include/asm/dma-mapping.h | 12 ++ arch/arm64/mm

[PATCH RESEND] dma-mapping: tidy up dma_parms default handling

2015-01-09 Thread Robin Murphy
a bit by replacing the bare constants with slightly more meaningful macros and removing the superfluous else statements. Signed-off-by: Robin Murphy robin.mur...@arm.com --- Hi, various maintainers from Git logs ;) This one's a bit tricky to find a home for - I think technically it's probably

Re: [RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-01-13 Thread Robin Murphy
On 13/01/15 08:02, Yingjoe Chen wrote: On Mon, 2015-01-12 at 20:48 +, Robin Murphy wrote: Hi all, Whilst it's a long way off perfect, this has reached the point of being functional and stable enough to be useful, so here it is. The core consists of the meat of the arch/arm implementation

Re: [RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-01-13 Thread Robin Murphy
On 13/01/15 11:08, Stefano Stabellini wrote: On Mon, 12 Jan 2015, Robin Murphy wrote: Hi all, Whilst it's a long way off perfect, this has reached the point of being functional and stable enough to be useful, so here it is. The core consists of the meat of the arch/arm implementation modified

[PATCH 1/4] iommu: allow building iova.c independently

2015-01-12 Thread Robin Murphy
In preparation for sharing the IOVA allocator, split it out under its own Kconfig symbol. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/Kconfig | 4 drivers/iommu/Makefile | 3 ++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/Kconfig b

[PATCH 0/4] Genericise the IOVA allocator

2015-01-12 Thread Robin Murphy
Kconfig symbol rather than a hack Patch 4: sanity check for powers of two also, and clarify the comment [1]:http://thread.gmane.org/gmane.linux.kernel.iommu/7480 [2]:http://thread.gmane.org/gmane.linux.kernel.iommu/7436 Robin Murphy (4): iommu: allow building iova.c independently iommu

[PATCH 3/4] iommu: make IOVA domain low limit flexible

2015-01-12 Thread Robin Murphy
description of alloc_iova since we're touching it anyway. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/intel-iommu.c | 9 ++--- drivers/iommu/iova.c| 10 ++ include/linux/iova.h| 7 +++ 3 files changed, 15 insertions(+), 11 deletions(-) diff --git

[PATCH 4/4] iommu: make IOVA domain page size explicit

2015-01-12 Thread Robin Murphy
page granularity from an implicit compile-time constant to a per-domain property so we can make use of it in IOVA domain context at runtime. To keep the abstraction tidy, extend the little API of inline iova_* helpers to parallel some of the equivalent PAGE_* macros. Signed-off-by: Robin Murphy

[PATCH 2/4] iommu: consolidate IOVA allocator code

2015-01-12 Thread Robin Murphy
In order to share the IOVA allocator with other architectures, break the unnecssary dependency on the Intel IOMMU driver and move the remaining IOVA internals to iova.c Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/intel-iommu.c | 33

Re: [PATCH 2/5] iommu/mediatek: Add mt8173 IOMMU driver

2015-03-20 Thread Robin Murphy
On 18/03/15 11:22, Yong Wu wrote: Hi Tomasz, Thanks very much for your review. please help check below. The others I will fix in the next version. Hi Robin, There are some place I would like you can have a look and give me some suggestion. On Wed, 2015-03-11 at 19:53 +0900, Tomasz Figa

Re: [RFC PATCH v2 2/3] arm64: add IOMMU dma_ops

2015-03-09 Thread Robin Murphy
Hi Russell, On 09/03/15 17:59, Russell King - ARM Linux wrote: [...] For a noncoherent device, dma_map_single() will end up calling __dma_map_area() with the page offset and size of the original request, so the updated part gets flushed by VA, and the rest of the page isn't touched if it

Re: [RFC PATCH v2 2/3] arm64: add IOMMU dma_ops

2015-03-10 Thread Robin Murphy
On 09/03/15 20:09, Robin Murphy wrote: [...] I think ideally you'd call dma_map_page when you first create the page table, dma_sync_single_for_device on any update, and dma_unmap_page when you tear it down, and you'd also use the appropriate DMA addresses everywhere instead of physical addresses

Re: [RFC PATCH 0/5] arm64: IOMMU-backed DMA mapping

2015-03-05 Thread Robin Murphy
Hi Marek, On 05/03/15 14:31, Marek Szyprowski wrote: Hello, On 2015-01-12 21:48, Robin Murphy wrote: Hi all, Whilst it's a long way off perfect, this has reached the point of being functional and stable enough to be useful, so here it is. The core consists of the meat of the arch/arm

Re: [PATCH] iommu/arm-smmu: fix leak in arm_smmu_flush_pgtable

2015-03-05 Thread Robin Murphy
On 05/03/15 17:28, Mitchel Humpherys wrote: On Thu, Mar 05 2015 at 02:38:45 AM, Robin Murphy robin.mur...@arm.com wrote: Hi Mitch, On 05/03/15 00:18, Mitchel Humpherys wrote: We're currently mapping a page in arm_smmu_flush_pgtable without ever unmapping it. Fix this by calling

[PATCH] iommu: arm-smmu: set a more appropriate DMA mask

2015-03-05 Thread Robin Murphy
outside the default 32-bit mask. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/arm-smmu.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c index fc13dd5..dc0ae62 100644 --- a/drivers/iommu/arm-smmu.c +++ b/drivers/iommu

Re: [RFC PATCH v2 2/3] arm64: add IOMMU dma_ops

2015-03-05 Thread Robin Murphy
Hi Laura, On 05/03/15 00:19, Laura Abbott wrote: [...] Consider that the IOMMU's page table walker is a DMA master in its own right, and that is the device you're mapping the page tables for. Therefore your IOMMU driver needs to have access to the struct device of the IOMMU itself to use for

Re: [PATCH] iommu/arm-smmu: fix leak in arm_smmu_flush_pgtable

2015-03-05 Thread Robin Murphy
Hi Mitch, On 05/03/15 00:18, Mitchel Humpherys wrote: We're currently mapping a page in arm_smmu_flush_pgtable without ever unmapping it. Fix this by calling dma_unmap_page on the returned dma address. Since the only reason we're calling dma_map_page is to make sure it actually gets flushed

Re: [PATCH] arm: dma-mapping: fix off-by-one check in arm_setup_iommu_dma_ops

2015-04-28 Thread Robin Murphy
Hi Marek, On 28/04/15 10:08, Marek Szyprowski wrote: Patch 22b3c181c6c324a46f71aae806d8ddbe61d25761 (arm: dma-mapping: limit IOMMU mapping size) added a check for IO address space size. However this patch broke IOMMU initialization for typical platforms initialized from device tree, which get

Re: [PATCH] arm: dma-mapping: fix off-by-one check in arm_setup_iommu_dma_ops

2015-04-28 Thread Robin Murphy
On 28/04/15 15:52, Robin Murphy wrote: Hi Marek, On 28/04/15 10:08, Marek Szyprowski wrote: Patch 22b3c181c6c324a46f71aae806d8ddbe61d25761 (arm: dma-mapping: limit IOMMU mapping size) added a check for IO address space size. However this patch broke IOMMU initialization for typical platforms

Re: [PATCH v6 01/25] arm: dma-mapping: add support for creating reserved mappings in iova space

2015-05-06 Thread Robin Murphy
Hi Marek, On 04/05/15 09:15, Marek Szyprowski wrote: Some devices (like frame buffers) are enabled by bootloader and configured to perform DMA operations automatically (like displaying boot logo or splash screen). Such devices operate and perform DMA operation usually until the proper driver

Re: [PATCH v2 3/6] iommu: add ARM short descriptor page table allocator.

2015-05-15 Thread Robin Murphy
Oops, seems I'm rather behind on things - I started this review on the RFC, but I'll finish it here... On 15/05/15 10:43, Yong Wu wrote: This patch is for ARM Short Descriptor Format.It has 2-levels pagetable and the allocator supports 4K/64K/1M/16M. From the look of the code, this doesn't

Re: [RFC/PATCH 2/9] arm: dma-mapping: Support IOMMU mappings spanning the full 32 bits range

2015-05-19 Thread Robin Murphy
Hi Laurent, On 15/05/15 00:00, Laurent Pinchart wrote: The arm_iommu_create_mapping() function takes the IOMMU mapping size as a size_t, limiting the size to 4GB - 1 on 32 bit platforms while most bus masters would support the full 32 bits range. Fix this by passing the size as a 64 bits

Re: [RFC/PATCH 6/9] drivers: platform: Configure dma operations at probe time

2015-05-19 Thread Robin Murphy
Hi Laurent, On 15/05/15 00:00, Laurent Pinchart wrote: Configuring DMA ops at probe time will allow deferring device probe when the IOMMU isn't available yet. This is great, as I think it also subtly solves the ordering problem the current domain allocation has with platform devices. WRT to

Re: [RFC v2 PATCH 0/6] MT8173 IOMMU SUPPORT

2015-05-18 Thread Robin Murphy
On 16/05/15 08:08, Daniel Kurtz wrote: On Fri, May 15, 2015 at 5:43 PM, Yong Wu yong...@mediatek.com wrote: This patch adds support for m4u(Multimedia Memory Management Unit), Currently it only support the m4u with 2 levels of page table on mt8173. It is based on Robin Murphy's arm64:

Re: [PATCH v2 2/4] iommu: Implement common IOMMU ops for DMA mapping

2015-06-18 Thread Robin Murphy
On 18/06/15 16:00, Yong Wu wrote: Hi Robin, while our drm test, we meet a problem while dma-mmap. On Thu, 2015-06-11 at 16:54 +0100, Robin Murphy wrote: Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API

[PATCH v2 4/4] arm64: Hook up IOMMU dma_ops

2015-06-11 Thread Robin Murphy
With iommu_dma_ops in place, hook them up to the configuration code, so IOMMU-fronted devices will get them automatically. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/Kconfig | 1 + arch/arm64/include/asm/dma-mapping.h | 11 ++- arch/arm64/mm/dma

[PATCH v2 0/4] arm64: IOMMU-backed DMA mapping

2015-06-11 Thread Robin Murphy
[2]:git://linux-arm.org/linux-rm iommu/dma Robin Murphy (4): iommu/iova: Avoid over-allocating when size-aligned iommu: Implement common IOMMU ops for DMA mapping arm64: Add IOMMU dma_ops arm64: Hook up IOMMU dma_ops arch/arm64/Kconfig | 1 + arch/arm64/include/asm

[PATCH v2 1/4] iommu/iova: Avoid over-allocating when size-aligned

2015-06-11 Thread Robin Murphy
padding makes no sense. CC: David Woodhouse dw...@infradead.org Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/intel-iommu.c | 2 ++ drivers/iommu/iova.c| 23 ++- 2 files changed, 8 insertions(+), 17 deletions(-) diff --git a/drivers/iommu/intel-iommu.c

[PATCH v2 2/4] iommu: Implement common IOMMU ops for DMA mapping

2015-06-11 Thread Robin Murphy
Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API. This will do the bulk of the heavy lifting for IOMMU-backed dma-mapping. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/Kconfig | 7 + drivers

Re: [PATCH 06/22] iommu: Allocate a default domain for iommu groups

2015-06-11 Thread Robin Murphy
Hi Joerg, On 28/05/15 17:41, Joerg Roedel wrote: From: Joerg Roedel jroe...@suse.de The default domain will be used (if supported by the iommu driver) when the devices in the iommu group are not attached to any other domain. Signed-off-by: Joerg Roedel jroe...@suse.de ---

Re: [PATCH 06/22] iommu: Allocate a default domain for iommu groups

2015-06-11 Thread Robin Murphy
Hi Joerg, On 28/05/15 17:41, Joerg Roedel wrote: From: Joerg Roedel jroe...@suse.de The default domain will be used (if supported by the iommu driver) when the devices in the iommu group are not attached to any other domain. Signed-off-by: Joerg Roedel jroe...@suse.de ---

Re: [PATCH 08/22] iommu: Make sure a device is always attached to a domain

2015-06-11 Thread Robin Murphy
On 28/05/15 17:41, Joerg Roedel wrote: From: Joerg Roedel jroe...@suse.de Make use of the default domain and re-attach a device to it when it is detached from another domain. Also enforce that a device has to be in the default domain before it can be attached to a different domain.

Re: [PATCH 10/22] iommu: Introduce direct mapped region handling

2015-06-11 Thread Robin Murphy
On 28/05/15 17:41, Joerg Roedel wrote: From: Joerg Roedel jroe...@suse.de Add two new functions to the IOMMU-API to allow the IOMMU drivers to export the requirements for direct mapped regions per device. This is useful for exporting the information in Intel VT-d's RMRR entries or AMD-Vi's

Re: [PATCH 2/3] iommu/arm-smmu: Add initial driver support for ARM SMMUv3 devices

2015-05-29 Thread Robin Murphy
with the pgsize_bitmap to cheat the system, I'm sure we don't want to add any more of that. How about something like the below diff as a first step? Robin. ---8--- From: Robin Murphy robin.mur...@arm.com Date: Fri, 29 May 2015 12:05:58 +0100 Subject: [PATCH] iommu: Allow drivers to support

[PATCH 1/4] iommu/iova: Avoid over-allocating when size-aligned

2015-05-27 Thread Robin Murphy
padding makes no sense. CC: David Woodhouse dw...@infradead.org Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/intel-iommu.c | 2 ++ drivers/iommu/iova.c| 23 ++- 2 files changed, 8 insertions(+), 17 deletions(-) diff --git a/drivers/iommu/intel-iommu.c

[PATCH 2/4] iommu: Implement common IOMMU ops for DMA mapping

2015-05-27 Thread Robin Murphy
Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API. This will do the bulk of the heavy lifting for IOMMU-backed dma-mapping. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/Kconfig | 7 + drivers

[PATCH 3/4] arm64: Add IOMMU dma_ops

2015-05-27 Thread Robin Murphy
the kind of problematic system topologies in the current IOMMU API anyway. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/include/asm/device.h | 3 + arch/arm64/include/asm/dma-mapping.h | 14 ++ arch/arm64/mm/dma-mapping.c | 342

[PATCH 0/4] arm64: IOMMU-backed DMA mapping

2015-05-27 Thread Robin Murphy
for platform devices, which should hopefully tie in with default domains to make the add_device callback mostly redundant. [1]:http://thread.gmane.org/gmane.linux.kernel.iommu/8773 [2]:http://thread.gmane.org/gmane.linux.kernel.iommu/9552 Robin Murphy (4): iommu/iova: Avoid over-allocating when

[PATCH 4/4] arm64: Hook up IOMMU dma_ops

2015-05-27 Thread Robin Murphy
With iommu_dma_ops in place, hook them up to the configuration code, so IOMMU-fronted devices will get them automatically. Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/Kconfig | 1 + arch/arm64/include/asm/dma-mapping.h | 11 ++- arch/arm64/mm/dma

Re: [PATCH 2/4] iommu: Implement common IOMMU ops for DMA mapping

2015-05-29 Thread Robin Murphy
-dma-contig.c#L70 On Wed, 2015-05-27 at 15:09 +0100, Robin Murphy wrote: Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API. This will do the bulk of the heavy lifting for IOMMU-backed dma-mapping. Signed-off-by: Robin

Re: [PATCH v2 2/4] iommu: Implement common IOMMU ops for DMA mapping

2015-07-03 Thread Robin Murphy
On 03/07/15 10:27, Yong Wu wrote: [...] +/** + * iommu_dma_alloc - Allocate and map a buffer contiguous in IOVA space + * @dev: Device to allocate memory for. Must be a real device + * attached to an iommu_dma_domain + * @size: Size of buffer in bytes + * @gfp: Allocation flags + * @prot:

Re: [PATCH 6/8] iommu/arm-smmu: add support for non-pci devices

2015-06-30 Thread Robin Murphy
On 26/06/15 09:33, Zhen Lei wrote: Now, we only support a master with only one stream id. It will cover most hardware platforms and coding so easy. Please refer Documentation\devicetree\bindings\iommu\iommu.txt on how to bind device tree. Signed-off-by: Zhen Lei thunder.leiz...@huawei.com ---

[PATCH v2 4/9] iommu/arm-smmu-v3: Clean up DMA API usage

2015-07-29 Thread Robin Murphy
With the correct DMA API calls now integrated into the io-pgtable code, let that handle the flushing of non-coherent page table updates. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/arm-smmu-v3.c | 18 +++--- 1 file changed, 3 insertions(+), 15 deletions

[PATCH v2 3/9] iommu/arm-smmu: Clean up DMA API usage

2015-07-29 Thread Robin Murphy
With the correct DMA API calls now integrated into the io-pgtable code, let that handle the flushing of non-coherent page table updates. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/arm-smmu.c | 22 ++ 1 file changed, 6 insertions(+), 16 deletions

[PATCH v2 8/9] iommu/arm-smmu-v3: Remove arm_smmu_flush_pgtable()

2015-07-29 Thread Robin Murphy
With the io-pgtable code now enforcing its own appropriate sync points, the vestigial flush_pgtable callback becomes entirely redundant, so remove it altogether. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/arm-smmu-v3.c | 10 -- 1 file changed, 10 deletions

[PATCH v2 6/9] iommu/io-pgtable-arm: Centralise sync points

2015-07-29 Thread Robin Murphy
of the operation, and obviate the need to call flush_pgtable on every individual update for synchronisation. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/io-pgtable-arm.c | 43 +++--- drivers/iommu/io-pgtable.h | 3 ++- 2 files changed, 26

[PATCH v2 7/9] iommu/arm-smmu: Remove arm_smmu_flush_pgtable()

2015-07-29 Thread Robin Murphy
With the io-pgtable code now enforcing its own appropriate sync points, the vestigial flush_pgtable callback becomes entirely redundant, so remove it altogether. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/arm-smmu.c | 13 - 1 file changed, 13 deletions

[PATCH v2 2/9] iommu/arm-smmu: Sort out coherency

2015-07-29 Thread Robin Murphy
to the dma-coherent property in the device tree for the final say. As an added bonus, since systems exist where an external CTTW signal has been tied off incorrectly at integration, allowing DT to override it offers a neat workaround for coherency issues with such SMMUs. Signed-off-by: Robin Murphy

[PATCH v2 9/9] iommu/io-pgtable: Remove flush_pgtable callback

2015-07-29 Thread Robin Murphy
With the users fully converted to DMA API operations, it's dead, Jim. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/io-pgtable-arm.c | 6 -- drivers/iommu/io-pgtable.h | 2 -- 2 files changed, 8 deletions(-) diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers

[PATCH v2 5/9] iommu/ipmmu-vmsa: Clean up DMA API usage

2015-07-29 Thread Robin Murphy
With the correct DMA API calls now integrated into the io-pgtable code, let that handle the flushing of non-coherent page table updates. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/ipmmu-vmsa.c | 19 +-- 1 file changed, 5 insertions(+), 14 deletions

[PATCH v2 1/9] iommu/io-pgtable-arm: Allow appropriate DMA API use

2015-07-29 Thread Robin Murphy
pressure on ZONE_DMA, and treat any DMA translation as a warning sign. Signed-off-by: Robin Murphy robin.mur...@arm.com --- Changes since v1[1]: - Make device pointer mandatory and use DMA API unconditionally - Remove flush_pgtable callback entirely - Style, consistency and typo fixes [1]:http

Re: [PATCH 05/13] iommu/io-pgtable-arm: Allow appropriate DMA API use

2015-08-04 Thread Robin Murphy
Hi Laurent, [ +RMK, as his patch is indirectly involved here ] On 04/08/15 14:16, Laurent Pinchart wrote: Hi Will and Robin, Thank you for the patch. On Monday 03 August 2015 14:25:47 Will Deacon wrote: From: Robin Murphy robin.mur...@arm.com Currently, users of the LPAE page table code

[PATCH v5 3/3] arm64: Hook up IOMMU dma_ops

2015-07-31 Thread Robin Murphy
With iommu_dma_ops in place, hook them up to the configuration code, so IOMMU-fronted devices will get them automatically. Acked-by: Catalin Marinas catalin.mari...@arm.com Signed-off-by: Robin Murphy robin.mur...@arm.com --- arch/arm64/Kconfig | 1 + arch/arm64/include/asm

[PATCH v5 1/3] iommu: Implement common IOMMU ops for DMA mapping

2015-07-31 Thread Robin Murphy
Taking inspiration from the existing arch/arm code, break out some generic functions to interface the DMA-API to the IOMMU-API. This will do the bulk of the heavy lifting for IOMMU-backed dma-mapping. Signed-off-by: Robin Murphy robin.mur...@arm.com --- drivers/iommu/Kconfig | 7 + drivers

[PATCH v5 0/3] arm64: IOMMU-backed DMA mapping

2015-07-31 Thread Robin Murphy
merged first as a stable base to work with (and un-block arm64 in the meantime). Have we decided yet whether this should go via the IOMMU tree or the arm64 tree? Thanks, Robin. [1]:http://thread.gmane.org/gmane.linux.kernel.iommu/10181 Robin Murphy (3): iommu: Implement common IOMMU ops for DMA

Re: [PATCH 3/5] iommu/arm-smmu: Clean up DMA API usage

2015-07-28 Thread Robin Murphy
On 28/07/15 11:18, Will Deacon wrote: On Mon, Jul 27, 2015 at 07:18:10PM +0100, Robin Murphy wrote: With the correct DMA API calls now integrated into the io-pgtable code, let that handle the flushing of non-coherent page table updates. Signed-off-by: Robin Murphy robin.mur...@arm.com

  1   2   3   4   5   6   7   8   9   10   >