Including (in no particular order):
- Page table code for AMD IOMMU now supports large pages where
smaller page-sizes were mapped before. VFIO had to work around
that in the past and I included a patch to remove it (acked by
Alex Williamson)
- Patches to unmodularize a couple of IOMMU drivers that would
never work as modules anyway.
- Work to unify the the iommu-related pointers in
'struct device' into one pointer. This work is not finished
yet, but will probably be in the next cycle.
- NUMA aware allocation in iommu-dma code
- Support for r8a774a1 and r8a774c0 in the Renesas IOMMU driver
- Scalable mode support for the Intel VT-d driver
- PM runtime improvements for the ARM-SMMU driver
- Support for the QCOM-SMMUv2 IOMMU hardware from Qualcom
- Various smaller fixes and improvements
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=DT9A
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v4.21' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU updates from Joerg Roedel:
- Page table code for AMD IOMMU now supports large pages where smaller
page-sizes were mapped before. VFIO had to work around that in the
past and I included a patch to remove it (acked by Alex Williamson)
- Patches to unmodularize a couple of IOMMU drivers that would never
work as modules anyway.
- Work to unify the the iommu-related pointers in 'struct device' into
one pointer. This work is not finished yet, but will probably be in
the next cycle.
- NUMA aware allocation in iommu-dma code
- Support for r8a774a1 and r8a774c0 in the Renesas IOMMU driver
- Scalable mode support for the Intel VT-d driver
- PM runtime improvements for the ARM-SMMU driver
- Support for the QCOM-SMMUv2 IOMMU hardware from Qualcom
- Various smaller fixes and improvements
* tag 'iommu-updates-v4.21' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (78 commits)
iommu: Check for iommu_ops == NULL in iommu_probe_device()
ACPI/IORT: Don't call iommu_ops->add_device directly
iommu/of: Don't call iommu_ops->add_device directly
iommu: Consolitate ->add/remove_device() calls
iommu/sysfs: Rename iommu_release_device()
dmaengine: sh: rcar-dmac: Use device_iommu_mapped()
xhci: Use device_iommu_mapped()
powerpc/iommu: Use device_iommu_mapped()
ACPI/IORT: Use device_iommu_mapped()
iommu/of: Use device_iommu_mapped()
driver core: Introduce device_iommu_mapped() function
iommu/tegra: Use helper functions to access dev->iommu_fwspec
iommu/qcom: Use helper functions to access dev->iommu_fwspec
iommu/of: Use helper functions to access dev->iommu_fwspec
iommu/mediatek: Use helper functions to access dev->iommu_fwspec
iommu/ipmmu-vmsa: Use helper functions to access dev->iommu_fwspec
iommu/dma: Use helper functions to access dev->iommu_fwspec
iommu/arm-smmu: Use helper functions to access dev->iommu_fwspec
ACPI/IORT: Use helper functions to access dev->iommu_fwspec
iommu: Introduce wrappers around dev->iommu_fwspec
...
Use the new helpers dev_iommu_fwspec_get()/set() to access
the dev->iommu_fwspec pointer. This makes it easier to move
that pointer later into another struct.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Change function __iommu_dma_alloc_pages() to allocate pages for DMA from
respective device NUMA node. The ternary operator which would be for
alloc_pages_node() is tidied along with this.
The motivation for this change is to have a policy for page allocation
consistent with direct DMA mapping, which attempts to allocate pages local
to the device, as mentioned in [1].
In addition, for certain workloads it has been observed a marginal
performance improvement. The patch caused an observation of 0.9% average
throughput improvement for running tcrypt with HiSilicon crypto engine.
We also include a modification to use kvzalloc() for kzalloc()/vzalloc()
combination.
[1] https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1692998.html
Signed-off-by: Ganapatrao Kulkarni <ganapatrao.kulkarni@cavium.com>
[JPG: Added kvzalloc(), drop pages ** being device local, remove ternary operator, update message]
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Return DMA_MAPPING_ERROR instead of 0 on a dma mapping failure and let
the core dma-mapping code handle the rest.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
With the flush queue infrastructure already abstracted into IOVA
domains, hooking it up in iommu-dma is pretty simple. Since there is a
degree of dependency on the IOMMU driver knowing what to do to play
along, we key the whole thing off a domain attribute which will be set
on default DMA ops domains to request non-strict invalidation. That way,
drivers can indicate the appropriate support by acknowledging the
attribute, and we can easily fall back to strict invalidation otherwise.
The flush queue callback needs a handle on the iommu_domain which owns
our cookie, so we have to add a pointer back to that, but neatly, that's
also sufficient to indicate whether we're using a flush queue or not,
and thus which way to release IOVAs. The only slight subtlety is
switching __iommu_dma_unmap() from calling iommu_unmap() to explicit
iommu_unmap_fast()/iommu_tlb_sync() so that we can elide the sync
entirely in non-strict mode.
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
[rm: convert to domain attribute, tweak comments and commit message]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Most parts of iommu-dma already assume they are operating on a default
domain set up by iommu_dma_init_domain(), and can be converted straight
over to avoid the refcounting bottleneck. MSI page mappings may be in
an unmanaged domain with an explicit MSI-only cookie, so retain the
non-specific lookup, but that's OK since they're far from a contended
fast path either way.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Take the new bus limit into account (when present) for IOVA allocations,
to accommodate those SoCs which integrate off-the-shelf IP blocks with
narrower interconnects such that the link between a device output and an
IOMMU input can truncate DMA addresses to even fewer bits than the
native size of either block's interface would imply.
Eventually it might make sense for the DMA core to apply this constraint
up-front in dma_set_mask() and friends, but for now this seems like the
least risky approach.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
This pretty much reverts commit 273df96353 ("iommu/dma: Make PCI
window reservation generic") by moving the PCI window region
reservation back into the dma specific path so that these regions
doesn't get exposed via the IOMMU API interface. With this change,
the vfio interface will report only iommu specific reserved regions
to the user space.
Cc: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Fixes: 273df96353 ('iommu/dma: Make PCI window reservation generic')
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Modified iommu_dma_get_resv_regions() to include GICv3 ITS
region on ACPI based ARM platfiorms which may require HW MSI
reservations.
Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Since IOVA allocation failure is not unusual case we need to flush
CPUs' rcache in hope we will succeed in next round.
However, it is useful to decide whether we need rcache flush step because
of two reasons:
- Not scalability. On large system with ~100 CPUs iterating and flushing
rcache for each CPU becomes serious bottleneck so we may want to defer it.
- free_cpu_cached_iovas() does not care about max PFN we are interested in.
Thus we may flush our rcaches and still get no new IOVA like in the
commonly used scenario:
if (dma_limit > DMA_BIT_MASK(32) && dev_is_pci(dev))
iova = alloc_iova_fast(iovad, iova_len, DMA_BIT_MASK(32) >> shift);
if (!iova)
iova = alloc_iova_fast(iovad, iova_len, dma_limit >> shift);
1. First alloc_iova_fast() call is limited to DMA_BIT_MASK(32) to get
PCI devices a SAC address
2. alloc_iova() fails due to full 32-bit space
3. rcaches contain PFNs out of 32-bit space so free_cpu_cached_iovas()
throws entries away for nothing and alloc_iova() fails again
4. Next alloc_iova_fast() call cannot take advantage of rcache since we
have just defeated caches. In this case we pick the slowest option
to proceed.
This patch reworks flushed_rcache local flag to be additional function
argument instead and control rcache flush step. Also, it updates all users
to do the flush as the last chance.
Signed-off-by: Tomasz Nowicki <Tomasz.Nowicki@caviumnetworks.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Now that the cached node optimisation can apply to all allocations, the
couple of users which were playing tricks with dma_32bit_pfn in order to
benefit from it can stop doing so. Conversely, there is also no need for
all the other users to explicitly calculate a 'real' 32-bit PFN, when
init_iova_domain() can happily do that itself from the page granularity.
CC: Thierry Reding <thierry.reding@gmail.com>
CC: Jonathan Hunter <jonathanh@nvidia.com>
CC: David Airlie <airlied@linux.ie>
CC: Sudeep Dutt <sudeep.dutt@intel.com>
CC: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Zhen Lei <thunder.leizhen@huawei.com>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
[rm: use iova_shift(), rewrote commit message]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This update comes with:
* Support for lockless operation in the ARM io-pgtable code.
This is an important step to solve the scalability problems in
the common dma-iommu code for ARM
* Some Errata workarounds for ARM SMMU implemenations
* Rewrite of the deferred IO/TLB flush code in the AMD IOMMU
driver. The code suffered from very high flush rates, with the
new implementation the flush rate is down to ~1% of what it
was before
* Support for amd_iommu=off when booting with kexec. Problem
here was that the IOMMU driver bailed out early without
disabling the iommu hardware, if it was enabled in the old
kernel
* The Rockchip IOMMU driver is now available on ARM64
* Align the return value of the iommu_ops->device_group
call-backs to not miss error values
* Preempt-disable optimizations in the Intel VT-d and common
IOVA code to help Linux-RT
* Various other small cleanups and fixes
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=pmBf
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU updates from Joerg Roedel:
"This update comes with:
- Support for lockless operation in the ARM io-pgtable code.
This is an important step to solve the scalability problems in the
common dma-iommu code for ARM
- Some Errata workarounds for ARM SMMU implemenations
- Rewrite of the deferred IO/TLB flush code in the AMD IOMMU driver.
The code suffered from very high flush rates, with the new
implementation the flush rate is down to ~1% of what it was before
- Support for amd_iommu=off when booting with kexec.
The problem here was that the IOMMU driver bailed out early without
disabling the iommu hardware, if it was enabled in the old kernel
- The Rockchip IOMMU driver is now available on ARM64
- Align the return value of the iommu_ops->device_group call-backs to
not miss error values
- Preempt-disable optimizations in the Intel VT-d and common IOVA
code to help Linux-RT
- Various other small cleanups and fixes"
* tag 'iommu-updates-v4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (60 commits)
iommu/vt-d: Constify intel_dma_ops
iommu: Warn once when device_group callback returns NULL
iommu/omap: Return ERR_PTR in device_group call-back
iommu: Return ERR_PTR() values from device_group call-backs
iommu/s390: Use iommu_group_get_for_dev() in s390_iommu_add_device()
iommu/vt-d: Don't disable preemption while accessing deferred_flush()
iommu/iova: Don't disable preempt around this_cpu_ptr()
iommu/arm-smmu-v3: Add workaround for Cavium ThunderX2 erratum #126
iommu/arm-smmu-v3: Enable ACPI based HiSilicon CMD_PREFETCH quirk(erratum 161010701)
iommu/arm-smmu-v3: Add workaround for Cavium ThunderX2 erratum #74
ACPI/IORT: Fixup SMMUv3 resource size for Cavium ThunderX2 SMMUv3 model
iommu/arm-smmu-v3, acpi: Add temporary Cavium SMMU-V3 IORT model number definitions
iommu/io-pgtable-arm: Use dma_wmb() instead of wmb() when publishing table
iommu/io-pgtable: depend on !GENERIC_ATOMIC64 when using COMPILE_TEST with LPAE
iommu/arm-smmu-v3: Remove io-pgtable spinlock
iommu/arm-smmu: Remove io-pgtable spinlock
iommu/io-pgtable-arm-v7s: Support lockless operation
iommu/io-pgtable-arm: Support lockless operation
iommu/io-pgtable: Introduce explicit coherency
iommu/io-pgtable-arm-v7s: Refactor split_blk_unmap
...
DMA_ERROR_CODE is not a public API and will go away soon. dma dma-iommu
driver already implements a proper ->mapping_error method, so it's only
using the value internally. Add a new local define using the value
that arm64 which is the only current user of dma-iommu.
Signed-off-by: Christoph Hellwig <hch@lst.de>
When walking the rbtree, the fact that iovad->start_pfn and limit_pfn
are both inclusive limits creates an ambiguity once limit_pfn reaches
the bottom of the address space and they overlap. Commit 5016bdb796
("iommu/iova: Fix underflow bug in __alloc_and_insert_iova_range") fixed
the worst side-effect of this, that of underflow wraparound leading to
bogus allocations, but the remaining fallout is that any attempt to
allocate start_pfn itself erroneously fails.
The cleanest way to resolve the ambiguity is to simply make limit_pfn an
exclusive limit when inside the guts of the rbtree. Since we're working
with PFNs, representing one past the top of the address space is always
possible without fear of overflow, and elsewhere it just makes life a
little more straightforward.
Reported-by: Aaron Sierra <asierra@xes-inc.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When __iommu_dma_map() and iommu_dma_free_iova() are called from
iommu_dma_get_msi_page(), various iova_*() helpers are still invoked in
the process, whcih is unwise since they access a different member of the
union (the iova_domain) from that which was last written, and there's no
guarantee that sensible values will result anyway.
CLean up the code paths that are valid for an MSI cookie to ensure we
only do iova_domain-specific things when we're actually dealing with one.
Fixes: a44e665758 ("iommu/dma: Clean up MSI IOVA allocation")
Reported-by: Nate Watterson <nwatters@codeaurora.org>
Tested-by: Shanker Donthineni <shankerd@codeaurora.org>
Tested-by: Bharat Bhushan <bharat.bhushan@nxp.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
With IOVA allocation suitably tidied up, we are finally free to opt in
to the per-CPU caching mechanism. The caching alone can provide a modest
improvement over walking the rbtree for weedier systems (iperf3 shows
~10% more ethernet throughput on an ARM Juno r1 constrained to a single
650MHz Cortex-A53), but the real gain will be in sidestepping the rbtree
lock contention which larger ARM-based systems with lots of parallel I/O
are starting to feel the pain of.
Reviewed-by: Nate Watterson <nwatters@codeaurora.org>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Now that allocation is suitably abstracted, our private alloc/free
helpers can drive the trivial MSI cookie allocator directly as well,
which lets us clean up its exposed guts from iommu_dma_map_msi_msg() and
simplify things quite a bit.
Reviewed-by: Nate Watterson <nwatters@codeaurora.org>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
In preparation for some IOVA allocation improvements, clean up all the
explicit struct iova usage such that all our mapping, unmapping and
cleanup paths deal exclusively with addresses rather than implementation
details. In the process, a few of the things we're touching get renamed
for the sake of internal consistency.
Reviewed-by: Nate Watterson <nwatters@codeaurora.org>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Now that we're applying the IOMMU API reserved regions to our IOVA
domains, we shouldn't need to privately special-case PCI windows, or
indeed anything else which isn't specific to our iommu-dma layer.
However, since those aren't IOMMU-specific either, rather than start
duplicating code into IOMMU drivers let's transform the existing
function into an iommu_get_resv_regions() helper that they can share.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Now that it's simple to discover the necessary reservations for a given
device/IOMMU combination, let's wire up the appropriate handling. Basic
reserved regions and direct-mapped regions we simply have to carve out
of IOVA space (the IOMMU core having already mapped the latter before
attaching the device). For hardware MSI regions, we also pre-populate
the cookie with matching msi_pages. That way, irqchip drivers which
normally assume MSIs to require mapping at the IOMMU can keep working
without having to special-case their iommu_dma_map_msi_msg() hook, or
indeed be aware at all of quirks preventing the IOMMU from translating
certain addresses.
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Even if a host controller's CPU-side MMIO windows into PCI I/O space do
happen to leak into PCI memory space such that it might treat them as
peer addresses, trying to reserve the corresponding I/O space addresses
doesn't do anything to help solve that problem. Stop doing a silly thing.
Fixes: fade1ec055 ("iommu/dma: Avoid PCI host bridge windows")
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Back when this was first written, dma_supported() was somewhat of a
murky mess, with subtly different interpretations being relied upon in
various places. The "does device X support DMA to address range Y?"
uses assuming Y to be physical addresses, which motivated the current
iommu_dma_supported() implementation and are alluded to in the comment
therein, have since been cleaned up, leaving only the far less ambiguous
"can device X drive address bits Y" usage internal to DMA API mask
setting. As such, there is no reason to keep a slightly misleading
callback which does nothing but duplicate the current default behaviour;
we already constrain IOVA allocations to the iommu_domain aperture where
necessary, so let's leave DMA mask business to architecture-specific
code where it belongs.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Whilst PCI devices may have 64-bit DMA masks, they still benefit from
using 32-bit addresses wherever possible in order to avoid DAC (PCI) or
longer address packets (PCIe), which may incur a performance overhead.
Implement the same optimisation as other allocators by trying to get a
32-bit address first, only falling back to the full mask if that fails.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
iommu_dma_init_domain() was originally written under the misconception
that dma_32bit_pfn represented some sort of size limit for IOVA domains.
Since the truth is almost the exact opposite of that, rework the logic
and comments to reflect its real purpose of optimising lookups when
allocating from a subset of the available 64-bit space.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
IOMMU domain users such as VFIO face a similar problem to DMA API ops
with regard to mapping MSI messages in systems where the MSI write is
subject to IOMMU translation. With the relevant infrastructure now in
place for managed DMA domains, it's actually really simple for other
users to piggyback off that and reap the benefits without giving up
their own IOVA management, and without having to reinvent their own
wheel in the MSI layer.
Allow such users to opt into automatic MSI remapping by dedicating a
region of their IOVA space to a managed cookie, and extend the mapping
routine to implement a trivial linear allocator in such cases, to avoid
the needless overhead of a full-blown IOVA domain.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Tomasz Nowicki <tomasz.nowicki@caviumnetworks.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Tomasz Nowicki <tomasz.nowicki@caviumnetworks.com>
Tested-by: Bharat Bhushan <bharat.bhushan@nxp.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
The newly added DMA_ATTR_PRIVILEGED is useful for creating mappings that
are only accessible to privileged DMA engines. Implement it in
dma-iommu.c so that the ARM64 DMA IOMMU mapper can make use of it.
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Mitchel Humpherys <mitchelh@codeaurora.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
With the new dma_{map,unmap}_resource() functions added to the DMA API
for the benefit of cases like slave DMA, add suitable implementations to
the arsenal of our generic layer. Since cache maintenance should not be
a concern, these can both be standalone callback implementations without
the need for arch code wrappers.
CC: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
With our DMA ops enabled for PCI devices, we should avoid allocating
IOVAs which a host bridge might misinterpret as peer-to-peer DMA and
lead to faults, corruption or other badness. To be safe, punch out holes
for all of the relevant host bridge's windows when initialising a DMA
domain for a PCI device.
CC: Marek Szyprowski <m.szyprowski@samsung.com>
CC: Inki Dae <inki.dae@samsung.com>
Reported-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
When an MSI doorbell is located downstream of an IOMMU, attaching
devices to a DMA ops domain and switching on translation leads to a rude
shock when their attempt to write to the physical address returned by
the irqchip driver faults (or worse, writes into some already-mapped
buffer) and no interrupt is forthcoming.
Address this by adding a hook for relevant irqchip drivers to call from
their compose_msi_msg() callback, to swizzle the physical address with
an appropriatly-mapped IOVA for any device attached to one of our DMA
ops domains.
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Where a device driver has set a 64-bit DMA mask to indicate the absence
of addressing limitations, we still need to ensure that we don't
allocate IOVAs beyond the actual input size of the IOMMU. The reported
aperture is the most reliable way we have of inferring that input
address size, so use that to enforce a hard upper limit where available.
Fixes: 0db2e5d18f ("iommu: Implement common IOMMU ops for DMA mapping")
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Due to the limitations of having to wait until we see a device's DMA
restrictions before we know how we want an IOVA domain initialised,
there is a window for error if a DMA ops domain is allocated but later
freed without ever being used. In that case, init_iova_domain() was
never called, so calling put_iova_domain() from iommu_put_dma_cookie()
ends up trying to take an uninitialised lock and crashing.
Make things robust by skipping the call unless the IOVA domain actually
has been initialised, as we probably should have done from the start.
Fixes: 0db2e5d18f ("iommu: Implement common IOMMU ops for DMA mapping")
Cc: stable@vger.kernel.org
Reported-by: Nate Watterson <nwatters@codeaurora.org>
Reviewed-by: Nate Watterson <nwatters@codeaurora.org>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The dma-mapping core and the implementations do not change the DMA
attributes passed by pointer. Thus the pointer can point to const data.
However the attributes do not have to be a bitfield. Instead unsigned
long will do fine:
1. This is just simpler. Both in terms of reading the code and setting
attributes. Instead of initializing local attributes on the stack
and passing pointer to it to dma_set_attr(), just set the bits.
2. It brings safeness and checking for const correctness because the
attributes are passed by value.
Semantic patches for this change (at least most of them):
virtual patch
virtual context
@r@
identifier f, attrs;
@@
f(...,
- struct dma_attrs *attrs
+ unsigned long attrs
, ...)
{
...
}
@@
identifier r.f;
@@
f(...,
- NULL
+ 0
)
and
// Options: --all-includes
virtual patch
virtual context
@r@
identifier f, attrs;
type t;
@@
t f(..., struct dma_attrs *attrs);
@@
identifier r.f;
@@
f(...,
- NULL
+ 0
)
Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Acked-by: Vineet Gupta <vgupta@synopsys.com>
Acked-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>
Acked-by: Mark Salter <msalter@redhat.com> [c6x]
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris]
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm]
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
Acked-by: Joerg Roedel <jroedel@suse.de> [iommu]
Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp]
Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core]
Acked-by: David Vrabel <david.vrabel@citrix.com> [xen]
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb]
Acked-by: Joerg Roedel <jroedel@suse.de> [iommu]
Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon]
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k]
Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390]
Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32]
Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc]
Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now that we know exactly which page sizes our caller wants to use in the
given domain, we can restrict higher-order allocation attempts to just
those sizes, if any, and avoid wasting any time or effort on other sizes
which offer no benefit. In the same vein, this also lets us accommodate
a minimum order greater than 0 for special cases.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Tested-by: Yong Wu <yong.wu@mediatek.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Many IOMMUs support multiple page table formats, meaning that any given
domain may only support a subset of the hardware page sizes presented in
iommu_ops->pgsize_bitmap. There are also certain use-cases where the
creator of a domain may want to control which page sizes are used, for
example to force the use of hugepage mappings to reduce pagetable walk
depth.
To this end, add a per-domain pgsize_bitmap to represent the subset of
page sizes actually in use, to make it possible for domains with
different requirements to coexist.
Signed-off-by: Will Deacon <will.deacon@arm.com>
[rm: hijacked and rebased original patch with new commit message]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Stop wasting IOVA space by over-aligning scatterlist segments for a
theoretical worst-case segment boundary mask, and instead take the real
limits into account to merge consecutive segments wherever appropriate,
so our callers can benefit from getting back nicely simplified lists.
This also represents the last piece of functionality wanted by users of
the current arch/arm implementation, thus brings us a small step closer
to converting that over to the common code.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
With the change to stashing just the IOVA-page-aligned remainder of the
CPU-page offset rather than the whole thing, the failure path in
__invalidate_sg() also needs tweaking to account for that in the case of
differing page sizes where the two offsets may not be equivalent.
Similarly in __finalise_sg(), lest the architecture-specific wrappers
later get the wrong address for cache maintenance on sync or unmap.
Fixes: 164afb1d85 ("iommu/dma: Use correct offset in map_sg")
Reported-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Cc: stable@ver.kernel.org # v4.4+
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When mapping a non-page-aligned scatterlist entry, we copy the original
offset to the output DMA address before aligning it to hand off to
iommu_map_sg(), then later adding the IOVA page address portion to get
the final mapped address. However, when the IOVA page size is smaller
than the CPU page size, it is the offset within the IOVA page we want,
not that within the CPU page, which can easily be larger than an IOVA
page and thus result in an incorrect final address.
Fix the bug by taking only the IOVA-aligned part of the offset as the
basis of the DMA address, not the whole thing.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Doug reports that the equivalent page allocator on 32-bit ARM exhibits
particularly pathalogical behaviour under memory pressure when
fragmentation is high, where allocating a 4MB buffer takes tens of
seconds and the number of calls to alloc_pages() is over 9000![1]
We can drastically improve that situation without losing the other
benefits of high-order allocations when they would succeed, by assuming
memory pressure is relatively constant over the course of an allocation,
and not retrying allocations at orders we know to have failed before.
This way, the best-case behaviour remains unchanged, and in the worst
case we should see at most a dozen or so (MAX_ORDER - 1) failed attempts
before falling back to single pages for the remainder of the buffer.
[1]:http://lists.infradead.org/pipermail/linux-arm-kernel/2015-December/394660.html
Reported-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
dma-iommu.c was naughtily relying on an implicit transitive #include of
linux/vmalloc.h, which is apparently not present on some architectures.
Add that, plus a couple more headers for other functions which are used
similarly.
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
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.
Since associating an IOVA allocator with an IOMMU domain is a fairly
common need, rather than introduce yet another private structure just to
do this for ourselves, extend the top-level struct iommu_domain with the
notion. A simple opaque cookie allows reuse by other IOMMU API users
with their various different incompatible allocator types.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>