AMD-Vi support for IOMMU sysfs. This allows us to associate devices
with a specific IOMMU device and examine the capabilities and features
of that IOMMU. The AMD IOMMU is hosted on and actual PCI device, so
we make that device the parent for the IOMMU class device. This
initial implementaiton exposes only the capability header and extended
features register for the IOMMU.
# find /sys | grep ivhd
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:00.0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:02.0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:04.0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:09.0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:11.0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:12.0
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:12.2
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/devices/0000:00:13.0
...
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/power
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/power/control
...
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/device
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/subsystem
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/amd-iommu
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/amd-iommu/cap
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/amd-iommu/features
/sys/devices/pci0000:00/0000:00:00.2/iommu/ivhd0/uevent
/sys/class/iommu/ivhd0
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The common iommu_group_get_for_dev() allows us to greatly simplify
our group lookup for a new device. Also, since we insert IVRS
aliases into the PCI DMA alias quirks, we should alway come up with
the same results as the existing code.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
AMD-Vi already has a concept of an alias provided via the IVRS table.
Now that PCI-core also understands aliases, we need to incorporate
both aspects when programming the IOMMU. IVRS is generally quite
reliable, so we continue to prefer it when an alias is present. For
cases where we have an IVRS alias that does not match the PCI alias
or where PCI does not report an alias, report the mismatch to allow
us to collect more quirks and dynamically incorporate the alias into
the device alias quirks where possible.
This should allow AMD-Vi to work with devices like Marvell and Ricoh
with DMA function alias quirks unknown to the BIOS.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Fix two compile warnings about unused variables introduced
by commit ecef115.
Reported-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
amd_iommu_rlookup_table[devid] != NULL is already guaranteed
by check_device called before, it's fine to attach device at
this point.
Signed-off-by: Vaughan Cao <vaughan.cao@oracle.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
An apparent cut and paste error prevents the correct flags from being
set on the alias device resulting in MSI on conventional PCI devices
failing to work. This also produces error events from the IOMMU like:
AMD-Vi: Event logged [INVALID_DEVICE_REQUEST device=00:14.4 address=0x000000fdf8000000 flags=0x0a00]
Where 14.4 is a PCIe-to-PCI bridge with a device behind it trying to
use MSI interrupts.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Joerg Roedel <joro@8bytes.org>
In reality, the spec can only support 16-bit PASID since
INVALIDATE_IOTLB_PAGES and COMPLETE_PPR_REQUEST commands only allow 16-bit
PASID. So, we updated the PASID_MASK accordingly and invoke BUG_ON
if the hardware is reporting PASmax more than 16-bit.
Besides, max PASID is defined as ((2^(PASmax+1)) - 1). The current does not
determine this correctly.
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Tested-by: Jay Cornwall <Jay.Cornwall@amd.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
This patch corrects the PASID format in the INVALIDATE_IOTLB_PAGES
command, which was caused by incorrect information in
the AMD IOMMU Architectural Specification v2.01 document.
Incorrect format:
cmd->data[0][16:23] = PASID[7:0]
cmd->data[1][16:27] = PASID[19:8]
Correct format:
cmd->data[0][16:23] = PASID[15:8]
cmd->data[1][16:23] = PASID[7:0]
However, this does not affect the IOMMUv2 hardware implementation,
and has been corrected since version 2.02 of the specification
(available through AMD NDA).
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Use PCI standard marco dev_is_pci() instead of directly compare
pci_bus_type to check whether it is pci device.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
If we use a large mapping, the expectation is that only unmaps from
the first pte in the superpage are supported. Unmaps from offsets
into the superpage should fail (ie. return zero sized unmap). In the
current code, unmapping from an offset clears the size of the full
mapping starting from an offset. For instance, if we map a 16k
physically contiguous range at IOVA 0x0 with a large page, then
attempt to unmap 4k at offset 12k, 4 ptes are cleared (12k - 28k) and
the unmap returns 16k unmapped. This potentially incorrectly clears
valid mappings and confuses drivers like VFIO that use the unmap size
to release pinned pages.
Fix by refusing to unmap from offsets into the page.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Joerg Roedel <joro@8bytes.org>
The IOMMU pagetables can have up to 6 levels, but the code
in free_pagetable() only releases the first 3 levels. Fix
this leak by releasing all levels.
Reported-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
If a device is multifunction and does not have ACS enabled then we
assume that the entire package lacks ACS and use function 0 as the
base of the group. The PCIe spec however states that components are
permitted to implement ACS on some, none, or all of their applicable
functions. It's therefore conceivable that function 0 may be fully
independent and support ACS while other functions do not. Instead
use the lowest function of the slot that does not have ACS enabled
as the base of the group. This may be the current device, which is
intentional. So long as we use a consistent algorithm, all the
non-ACS functions will be grouped together and ACS functions will
get separate groups.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
The updates are mostly about the x86 IOMMUs this time. Exceptions are
the groundwork for the PAMU IOMMU from Freescale (for a PPC platform)
and an extension to the IOMMU group interface. On the x86 side this
includes a workaround for VT-d to disable interrupt remapping on broken
chipsets. On the AMD-Vi side the most important new feature is a kernel
command-line interface to override broken information in IVRS ACPI
tables and get interrupt remapping working this way. Besides that there
are small fixes all over the place.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=MVc/
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU updates from Joerg Roedel:
"The updates are mostly about the x86 IOMMUs this time.
Exceptions are the groundwork for the PAMU IOMMU from Freescale (for a
PPC platform) and an extension to the IOMMU group interface.
On the x86 side this includes a workaround for VT-d to disable
interrupt remapping on broken chipsets. On the AMD-Vi side the most
important new feature is a kernel command-line interface to override
broken information in IVRS ACPI tables and get interrupt remapping
working this way.
Besides that there are small fixes all over the place."
* tag 'iommu-updates-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (24 commits)
iommu/tegra: Fix printk formats for dma_addr_t
iommu: Add a function to find an iommu group by id
iommu/vt-d: Remove warning for HPET scope type
iommu: Move swap_pci_ref function to drivers/iommu/pci.h.
iommu/vt-d: Disable translation if already enabled
iommu/amd: fix error return code in early_amd_iommu_init()
iommu/AMD: Per-thread IOMMU Interrupt Handling
iommu: Include linux/err.h
iommu/amd: Workaround for ERBT1312
iommu/amd: Document ivrs_ioapic and ivrs_hpet parameters
iommu/amd: Don't report firmware bugs with cmd-line ivrs overrides
iommu/amd: Add ioapic and hpet ivrs override
iommu/amd: Add early maps for ioapic and hpet
iommu/amd: Extend IVRS special device data structure
iommu/amd: Move add_special_device() to __init
iommu: Fix compile warnings with forward declarations
iommu/amd: Properly initialize irq-table lock
iommu/amd: Use AMD specific data structure for irq remapping
iommu/amd: Remove map_sg_no_iommu()
iommu/vt-d: add quirk for broken interrupt remapping on 55XX chipsets
...
The swap_pci_ref function is used by the IOMMU API code for
swapping pci device pointers, while determining the iommu
group for the device.
Currently this function was being implemented for different
IOMMU drivers. This patch moves the function to a new file,
drivers/iommu/pci.h so that the implementation can be
shared across various IOMMU drivers.
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
In the current interrupt handling scheme, there are as many threads as
the number of IOMMUs. Each thread is created and assigned to an IOMMU at
the time of registering interrupt handlers (request_threaded_irq).
When an IOMMU HW generates an interrupt, the irq handler (top half) wakes up
the corresponding thread to process event and PPR logs of all IOMMUs
starting from the 1st IOMMU.
In the system with multiple IOMMU,this handling scheme complicates the
synchronization of the IOMMU data structures and status registers as
there could be multiple threads competing for the same IOMMU while
the other IOMMU could be left unhandled.
To simplify, this patch is proposing a different interrupt handling scheme
by having each thread only managing interrupts of the corresponding IOMMU.
This can be achieved by passing the struct amd_iommu when registering the
interrupt handlers. This structure is unique for each IOMMU and can be used
by the bottom half thread to identify the IOMMU to be handled instead
of calling for_each_iommu. Besides this also eliminate the needs to lock
the IOMMU for processing event and PPR logs.
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Work around an IOMMU hardware bug where clearing the
EVT_INT or PPR_INT bit in the status register may race with
the hardware trying to set it again. When not handled the
bit might not be cleared and we lose all future event or ppr
interrupts.
Reported-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Joerg Roedel <joro@8bytes.org>
For compatibility reasons the irq remapping code for the AMD
IOMMU used the same per-irq data structure as the Intel
implementation. Now that support for the AMD specific data
structure is upstream we can use this one instead.
Reviewed-by: Shuah Khan <shuahkhan@gmail.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
This function was intended as a fall-back if the map_sg
function is called for a device not mapped by the IOMMU.
Since the AMD IOMMU driver uses per-device dma_ops this can
never happen. So this function isn't needed anymore.
Reviewed-by: Shuah Khan <shuahkhan@gmail.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
This is required in case of PAMU, as it can support a window size of up
to 64G (even on 32bit).
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Current driver does not clear the IOMMU event log interrupt bit
in the IOMMU status register after processing an interrupt.
This causes the IOMMU hardware to generate event log interrupt only once.
This has been observed in both IOMMU v1 and V2 hardware.
This patch clears the bit by writing 1 to bit 1 of the IOMMU
status register (MMIO Offset 2020h)
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
There is a bug introduced with commit 27c2127 that causes
devices which are hot unplugged and then hot-replugged to
not have per-device dma_ops set. This causes these devices
to not function correctly. Fixed with this patch.
Cc: stable@vger.kernel.org
Reported-by: Andreas Degert <andreas.degert@googlemail.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Change to remove local PCI_BUS() define and use the new PCI_BUS_NUM()
interface from PCI.
Signed-off-by: Shuah Khan <shuah.khan@hp.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Joerg Roedel <joro@8bytes.org>
Besides some fixes and cleanups in the code there are three more
important changes to point out this time:
* New IOMMU driver for the ARM SHMOBILE platform
* An IOMMU-API extension for non-paging IOMMUs (required for
upcoming PAMU driver)
* Rework of the way the Tegra IOMMU driver accesses its
registetrs - register windows are easier to extend now.
There are also a few changes to non-iommu code, but that is acked by the
respective maintainers.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=G4Ot
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v3.9' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU Updates from Joerg Roedel:
"Besides some fixes and cleanups in the code there are three more
important changes to point out this time:
* New IOMMU driver for the ARM SHMOBILE platform
* An IOMMU-API extension for non-paging IOMMUs (required for
upcoming PAMU driver)
* Rework of the way the Tegra IOMMU driver accesses its
registetrs - register windows are easier to extend now.
There are also a few changes to non-iommu code, but that is acked by
the respective maintainers."
* tag 'iommu-updates-v3.9' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (23 commits)
iommu/tegra: assume CONFIG_OF in SMMU driver
iommu/tegra: assume CONFIG_OF in gart driver
iommu/amd: Remove redundant NULL check before dma_ops_domain_free().
iommu/amd: Initialize device table after dma_ops
iommu/vt-d: Zero out allocated memory in dmar_enable_qi
iommu/tegra: smmu: Fix incorrect mask for regbase
iommu/exynos: Make exynos_sysmmu_disable static
ARM: mach-shmobile: r8a7740: Add IPMMU device
ARM: mach-shmobile: sh73a0: Add IPMMU device
ARM: mach-shmobile: sh7372: Add IPMMU device
iommu/shmobile: Add iommu driver for Renesas IPMMU modules
iommu: Add DOMAIN_ATTR_WINDOWS domain attribute
iommu: Add domain window handling functions
iommu: Implement DOMAIN_ATTR_PAGING attribute
iommu: Check for valid pgsize_bitmap in iommu_map/unmap
iommu: Make sure DOMAIN_ATTR_MAX is really the maximum
iommu/tegra: smmu: Change SMMU's dependency on ARCH_TEGRA
iommu/tegra: smmu: Use helper function to check for valid register offset
iommu/tegra: smmu: Support variable MMIO ranges/blocks
iommu/tegra: Add missing spinlock initialization
...
dma_ops_domain_free on a NULL pointer is a no-op, so the NULL check in
amd_iommu_init_dma_ops() can be removed.
Signed-off-by: Cyril Roelandt <tipecaml@gmail.com>
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Move all the code to either to the header file
asm/irq_remapping.h or to drivers/iommu/.
Signed-off-by: Joerg Roedel <joro@8bytes.org>
Acked-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The AMD IOMMU driver only uses the page-sizes it gets from
IOMMU core and uses the appropriate page-size. So this
comment is not necessary.
Signed-off-by: Joerg Roedel <joro@8bytes.org>
There is a bug in the hardware that will be triggered when
this page size is used. Make sure this does not happen.
Signed-off-by: Joerg Roedel <joro@8bytes.org>
An alias doesn't always point to a physical device. When this
happens we must first verify that the IOMMU group isn't rooted in
a device above the alias. In this case the alias is effectively
just another quirk for the devices aliased to it. Alternatively,
the virtual alias itself may be the root of the IOMMU group. To
support this, allow a group to be hosted on the alias dev_data
for use by anything that might have the same alias.
Signed-off-by: Alex williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Add a WARN_ON to make it clear why we don't add dma_pdev->dev to the
group we're allocating.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This needs to be broken apart, start with pulling all the IOMMU
group init code into a new function.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
We should return NULL on error instead of the freed pointer.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Report the availability of irq remapping through the
IOMMU-API to allow KVM device passthrough again without
additional module parameter overrides.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Add the six routines required to setup interrupt remapping
with the AMD IOMMU. Also put it all together into the AMD
specific irq_remap_ops.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Add the routine to setup interrupt remapping for ioapic
interrupts. Also add a routine to change the affinity of an
irq and to free an irq allocation for interrupt remapping.
The last two functions will also be used for MSI interrupts.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Add routines to:
* Alloc remapping tables and single entries from these
tables
* Change entries in the tables
* Free entries in the table
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Add routine to invalidate the IOMMU cache for interupt
translations. Also include the IRTE caches when flushing all
IOMMU caches.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The irq remapping tables for the AMD IOMMU need to be
aligned on a 128 byte boundary. Create a seperate slab-cache
to guarantee this alignment.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The IVRS ACPI table provides information about the IOAPICs
and the HPETs available in the system and which PCI device
ID they use in transactions. Save that information for later
usage in interrupt remapping.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The new IOMMU groups code in the AMD IOMMU driver makes the
assumption that there is a pci_dev struct available for all
device-ids listed in the IVRS ACPI table. Unfortunatly this
assumption is not true and so this code causes a NULL
pointer dereference at boot on some systems.
Fix it by making sure the given pointer is never NULL when
passed to the group specific code. The real fix is larger
and will be queued for v3.7.
Reported-by: Florian Dazinger <florian@dazinger.net>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Fix some typos in comments and user-visible messages. No
functional changes.
Signed-off-by: Frank Arnold <frank.arnold@amd.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
SR-IOV can create buses without a bridge. There may be other cases
where this happens as well. In these cases skip to the parent bus
and continue testing devices there.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This did not work because devices are not put into the
pt_domain. Fix this.
Cc: stable@vger.kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
A few sparse warnings fire in drivers/iommu/amd_iommu_init.c.
Fix most of them with this patch. Also fix the sparse
warnings in drivers/iommu/irq_remapping.c while at it.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
write_file_bool() modifies 32 bits of data, so "amd_iommu_unmap_flush"
needs to be 32 bits as well or we'll corrupt memory. Fortunately it
looks like the data is aligned with a gap after the declaration so this
is harmless in production.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Work around broken devices and adhere to ACS support when determining
IOMMU grouping.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Add IOMMU group support to AMD-Vi device init and uninit code.
Existing notifiers make sure this gets called for each device.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
IOMMU device groups are currently a rather vague associative notion
with assembly required by the user or user level driver provider to
do anything useful. This patch intends to grow the IOMMU group concept
into something a bit more consumable.
To do this, we first create an object representing the group, struct
iommu_group. This structure is allocated (iommu_group_alloc) and
filled (iommu_group_add_device) by the iommu driver. The iommu driver
is free to add devices to the group using it's own set of policies.
This allows inclusion of devices based on physical hardware or topology
limitations of the platform, as well as soft requirements, such as
multi-function trust levels or peer-to-peer protection of the
interconnects. Each device may only belong to a single iommu group,
which is linked from struct device.iommu_group. IOMMU groups are
maintained using kobject reference counting, allowing for automatic
removal of empty, unreferenced groups. It is the responsibility of
the iommu driver to remove devices from the group
(iommu_group_remove_device).
IOMMU groups also include a userspace representation in sysfs under
/sys/kernel/iommu_groups. When allocated, each group is given a
dynamically assign ID (int). The ID is managed by the core IOMMU group
code to support multiple heterogeneous iommu drivers, which could
potentially collide in group naming/numbering. This also keeps group
IDs to small, easily managed values. A directory is created under
/sys/kernel/iommu_groups for each group. A further subdirectory named
"devices" contains links to each device within the group. The iommu_group
file in the device's sysfs directory, which formerly contained a group
number when read, is now a link to the iommu group. Example:
$ ls -l /sys/kernel/iommu_groups/26/devices/
total 0
lrwxrwxrwx. 1 root root 0 Apr 17 12:57 0000:00:1e.0 ->
../../../../devices/pci0000:00/0000:00:1e.0
lrwxrwxrwx. 1 root root 0 Apr 17 12:57 0000:06:0d.0 ->
../../../../devices/pci0000:00/0000:00:1e.0/0000:06:0d.0
lrwxrwxrwx. 1 root root 0 Apr 17 12:57 0000:06:0d.1 ->
../../../../devices/pci0000:00/0000:00:1e.0/0000:06:0d.1
$ ls -l /sys/kernel/iommu_groups/26/devices/*/iommu_group
[truncating perms/owner/timestamp]
/sys/kernel/iommu_groups/26/devices/0000:00:1e.0/iommu_group ->
../../../kernel/iommu_groups/26
/sys/kernel/iommu_groups/26/devices/0000:06:0d.0/iommu_group ->
../../../../kernel/iommu_groups/26
/sys/kernel/iommu_groups/26/devices/0000:06:0d.1/iommu_group ->
../../../../kernel/iommu_groups/26
Groups also include several exported functions for use by user level
driver providers, for example VFIO. These include:
iommu_group_get(): Acquires a reference to a group from a device
iommu_group_put(): Releases reference
iommu_group_for_each_dev(): Iterates over group devices using callback
iommu_group_[un]register_notifier(): Allows notification of device add
and remove operations relevant to the group
iommu_group_id(): Return the group number
This patch also extends the IOMMU API to allow attaching groups to
domains. This is currently a simple wrapper for iterating through
devices within a group, but it's expected that the IOMMU API may
eventually make groups a more integral part of domains.
Groups intentionally do not try to manage group ownership. A user
level driver provider must independently acquire ownership for each
device within a group before making use of the group as a whole.
This may change in the future if group usage becomes more pervasive
across both DMA and IOMMU ops.
Groups intentionally do not provide a mechanism for driver locking
or otherwise manipulating driver matching/probing of devices within
the group. Such interfaces are generic to devices and beyond the
scope of IOMMU groups. If implemented, user level providers have
ready access via iommu_group_for_each_dev and group notifiers.
iommu_device_group() is removed here as it has no users. The
replacement is:
group = iommu_group_get(dev);
id = iommu_group_id(group);
iommu_group_put(group);
AMD-Vi & Intel VT-d support re-added in following patches.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
When a device is added to the system at runtime the AMD
IOMMU driver initializes the necessary data structures to
handle translation for it. But it forgets to change the
per-device dma_ops to point to the AMD IOMMU driver. So
mapping actually never happens and all DMA accesses end in
an IO_PAGE_FAULT. Fix this.
Reported-by: Stefan Assmann <sassmann@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
In the error path of the ppr_notifer it can happen that the
iommu->lock is taken recursivly. This patch fixes the
problem by releasing the iommu->lock before any notifier is
invoked. This also requires to move the erratum workaround
for the ppr-log (interrupt may be faster than data in the log)
one function up.
Cc: stable@vger.kernel.org # v3.3, v3.4
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Due to a recent erratum it can happen that the head pointer
of the event-log is updated before the actual event-log
entry is written. This patch implements the recommended
workaround.
Cc: stable@vger.kernel.org # all stable kernels
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Unfortunatly the PRI spec changed and moved the
TLP-prefix-required bit to a different location. This patch
makes the necessary change in the AMD IOMMU driver.
Regressions are not expected because all hardware
implementing the PRI capability sets this bit to zero
anyway.
Cc: stable@vger.kernel.org # v3.3
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Pull DMA mapping branch from Marek Szyprowski:
"Short summary for the whole series:
A few limitations have been identified in the current dma-mapping
design and its implementations for various architectures. There exist
more than one function for allocating and freeing the buffers:
currently these 3 are used dma_{alloc, free}_coherent,
dma_{alloc,free}_writecombine, dma_{alloc,free}_noncoherent.
For most of the systems these calls are almost equivalent and can be
interchanged. For others, especially the truly non-coherent ones
(like ARM), the difference can be easily noticed in overall driver
performance. Sadly not all architectures provide implementations for
all of them, so the drivers might need to be adapted and cannot be
easily shared between different architectures. The provided patches
unify all these functions and hide the differences under the already
existing dma attributes concept. The thread with more references is
available here:
http://www.spinics.net/lists/linux-sh/msg09777.html
These patches are also a prerequisite for unifying DMA-mapping
implementation on ARM architecture with the common one provided by
dma_map_ops structure and extending it with IOMMU support. More
information is available in the following thread:
http://thread.gmane.org/gmane.linux.kernel.cross-arch/12819
More works on dma-mapping framework are planned, especially in the
area of buffer sharing and managing the shared mappings (together with
the recently introduced dma_buf interface: commit d15bd7ee44
"dma-buf: Introduce dma buffer sharing mechanism").
The patches in the current set introduce a new alloc/free methods
(with support for memory attributes) in dma_map_ops structure, which
will later replace dma_alloc_coherent and dma_alloc_writecombine
functions."
People finally started piping up with support for merging this, so I'm
merging it as the last of the pending stuff from the merge window.
Looks like pohmelfs is going to wait for 3.5 and more external support
for merging.
* 'for-linus' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping:
common: DMA-mapping: add NON-CONSISTENT attribute
common: DMA-mapping: add WRITE_COMBINE attribute
common: dma-mapping: introduce mmap method
common: dma-mapping: remove old alloc_coherent and free_coherent methods
Hexagon: adapt for dma_map_ops changes
Unicore32: adapt for dma_map_ops changes
Microblaze: adapt for dma_map_ops changes
SH: adapt for dma_map_ops changes
Alpha: adapt for dma_map_ops changes
SPARC: adapt for dma_map_ops changes
PowerPC: adapt for dma_map_ops changes
MIPS: adapt for dma_map_ops changes
X86 & IA64: adapt for dma_map_ops changes
common: dma-mapping: introduce generic alloc() and free() methods
Adapt core x86 and IA64 architecture code for dma_map_ops changes: replace
alloc/free_coherent with generic alloc/free methods.
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
[removed swiotlb related changes and replaced it with wrappers,
merged with IA64 patch to avoid inter-patch dependences in intel-iommu code]
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Tony Luck <tony.luck@intel.com>
Fix the following section warnings :
WARNING: vmlinux.o(.text+0x49dbc): Section mismatch in reference
from the function acpi_map_cpu2node() to the variable
.cpuinit.data:__apicid_to_node The function acpi_map_cpu2node()
references the variable __cpuinitdata __apicid_to_node. This is
often because acpi_map_cpu2node lacks a __cpuinitdata
annotation or the annotation of __apicid_to_node is wrong.
WARNING: vmlinux.o(.text+0x49dc1): Section mismatch in reference
from the function acpi_map_cpu2node() to the function
.cpuinit.text:numa_set_node() The function acpi_map_cpu2node()
references the function __cpuinit numa_set_node(). This is often
because acpi_map_cpu2node lacks a __cpuinit annotation or the
annotation of numa_set_node is wrong.
WARNING: vmlinux.o(.text+0x526e77): Section mismatch in
reference from the function prealloc_protection_domains() to the
function .init.text:alloc_passthrough_domain() The function
prealloc_protection_domains() references the function __init
alloc_passthrough_domain(). This is often because
prealloc_protection_domains lacks a __init annotation or the annotation of alloc_passthrough_domain is wrong.
Signed-off-by: Steffen Persvold <sp@numascale.com>
Link: http://lkml.kernel.org/r/1331810188-24785-1-git-send-email-sp@numascale.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On some systems the IVRS table does not contain all PCI
devices present in the system. In case a device not present
in the IVRS table is translated by the IOMMU no DMA is
possible from that device by default.
This patch fixes this by removing the DTE entry for every
PCI device present in the system and not covered by IVRS.
Cc: stable@vger.kernel.org # >= 3.0
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The IOMMUv2 driver added a few statistic counter which are
interesting in the iommu=pt mode too. So initialize the
statistic counter for that mode too.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This function can be used to find out which features
necessary for IOMMUv2 usage are available on a given device.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The symbolic register names for PCI and PASID changed in
PCI code. This patch adapts the AMD IOMMU driver to these
changes.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The AMD IOMMUv2 driver needs to get the IOMMUv2 domain
associated with a particular device. This patch adds a
function to get this information.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
To send completions for PPR requests this patch adds a
function which can be used by the IOMMUv2 driver.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch adds functions necessary to set and clear the
GCR3 values associated with a particular PASID in an IOMMUv2
domain.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The functions added with this patch allow to manage the
IOMMU and the device TLBs for all devices in an IOMMUv2
domain.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This function can be used to switch a domain into
paging-mode 0. In this mode all devices can access physical
system memory directly without any remapping.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
If the device starts to use IOMMUv2 features the dma handles
need to stay valid. The only sane way to do this is to use a
identity mapping for the device and not translate it by the
iommu. This is implemented with this patch. Since this lifts
the device-isolation there is also a new kernel parameter
which allows to disable that feature.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Convert the contents of 'struct dev_table_entry' to u64 to
allow updating the DTE wit 64bit writes as required by the
spec.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The option iommu=group_mf indicates the that the iommu driver should
expose all functions of a multi-function PCI device as the same
iommu_device_group. This is useful for disallowing individual functions
being exposed as independent devices to userspace as there are often
hidden dependencies. Virtual functions are not affected by this option.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Just use the amd_iommu_alias_table directly.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Let the IOMMU core know we support arbitrary page sizes (as long as
they're an order of 4KiB).
This way the IOMMU core will retain the existing behavior we're used to;
it will let us map regions that:
- their size is an order of 4KiB
- they are naturally aligned
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Joerg Roedel <Joerg.Roedel@amd.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Express sizes in bytes rather than in page order, to eliminate the
size->order->size conversions we have whenever the IOMMU API is calling
the low level drivers' map/unmap methods.
Adopt all existing drivers.
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: David Brown <davidb@codeaurora.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Joerg Roedel <Joerg.Roedel@amd.com>
Cc: Stepan Moskovchenko <stepanm@codeaurora.org>
Cc: KyongHo Cho <pullip.cho@samsung.com>
Cc: Hiroshi DOYU <hdoyu@nvidia.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The shift direction was wrong because the function takes a
page number and i is the address is the loop.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The domain_flush_devices() function takes the domain->lock.
But this function is only called from update_domain() which
itself is already called unter the domain->lock. This causes
a deadlock situation when the dma-address-space of a domain
grows larger than 1GB.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The value is only set to true but never set back to false,
which causes to many completion-wait commands to be sent to
hardware. Fix it with this patch.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Reserve the MSI address range in the address allocator so
that MSI addresses are not handed out as dma handles.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
A few parts of the driver were missing in drivers/iommu.
Move them there to have the complete driver in that
directory.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This should ease finding similarities with different platforms,
with the intention of solving problems once in a generic framework
which everyone can use.
Compile-tested on x86_64.
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Alex Williamson