get_irq_table() previously acquired amd_iommu_devtable_lock which is not
a raw lock, and thus cannot be acquired from atomic context on
PREEMPT_RT. Many calls to modify_irte*() come from atomic context due to
the IRQ desc->lock, as does amd_iommu_update_ga() due to the preemption
disabling in vcpu_load/put().
The only difference between calling get_irq_table() and reading from
irq_lookup_table[] directly, other than the lock acquisition and
amd_iommu_rlookup_table[] check, is if the table entry is unpopulated,
which should never happen when looking up a devid that came from an
irq_2_irte struct, as get_irq_table() would have already been called on
that devid during irq_remapping_alloc().
The lock acquisition is not needed in these cases because entries in
irq_lookup_table[] never change once non-NULL -- nor would the
amd_iommu_devtable_lock usage in get_irq_table() provide meaningful
protection if they did, since it's released before using the looked up
table in the get_irq_table() caller.
Rename the old get_irq_table() to alloc_irq_table(), and create a new
lockless get_irq_table() to be used in non-allocating contexts that WARNs
if it doesn't find what it's looking for.
Signed-off-by: Scott Wood <swood@redhat.com>
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>
Unmap returns a size_t all throughout the IOMMU framework.
Make io-pgtable match this convention.
Moreover, there isn't a need to have a signed int return type
as we return 0 in case of failures.
Signed-off-by: Vivek Gautam <vivek.gautam@codeaurora.org>
Acked-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Currently, iommu_unmap, iommu_unmap_fast and iommu_map_sg return
size_t. However, some of the return values are error codes (< 0),
which can be misinterpreted as large size. Therefore, returning size 0
instead to signify failure to map/unmap.
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
It's called only from intel_iommu_init(), which is init function.
Signed-off-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
In the commit 05f80300dc ("iommu: Finish making iommu_group support
mandatory"), the iommu framework has supposed all the iommu drivers have
their owner iommu-group, it get rid of the FIXME workarounds while the
group is NULL. But the flow of Mediatek M4U gen1 looks a bit trick that
it will hang at this case:
==========================================
Unable to handle kernel NULL pointer dereference at virtual address 00000030
pgd = c0004000
[00000030] *pgd=00000000
PC is at mutex_lock+0x28/0x54
LR is at iommu_attach_device+0xa4/0xd4
pc : [<c07632e8>] lr : [<c04736fc>] psr: 60000013
sp : df0edbb8 ip : df0edbc8 fp : df0edbc4
r10: c114da14 r9 : df2a3e40 r8 : 00000003
r7 : df27a210 r6 : df2a90c4 r5 : 00000030 r4 : 00000000
r3 : df0f8000 r2 : fffff000 r1 : df29c610 r0 : 00000030
Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
xxx
(mutex_lock) from [<c04736fc>] (iommu_attach_device+0xa4/0xd4)
(iommu_attach_device) from [<c011b9dc>] (__arm_iommu_attach_device+0x28/0x90)
(__arm_iommu_attach_device) from [<c011ba60>] (arm_iommu_attach_device+0x1c/0x30)
(arm_iommu_attach_device) from [<c04759ac>] (mtk_iommu_add_device+0xfc/0x214)
(mtk_iommu_add_device) from [<c0472aa4>] (add_iommu_group+0x3c/0x68)
(add_iommu_group) from [<c047d044>] (bus_for_each_dev+0x78/0xac)
(bus_for_each_dev) from [<c04734a4>] (bus_set_iommu+0xb0/0xec)
(bus_set_iommu) from [<c0476310>] (mtk_iommu_probe+0x328/0x368)
(mtk_iommu_probe) from [<c048189c>] (platform_drv_probe+0x5c/0xc0)
(platform_drv_probe) from [<c047f510>] (driver_probe_device+0x2f4/0x4d8)
(driver_probe_device) from [<c047f800>] (__driver_attach+0x10c/0x128)
(__driver_attach) from [<c047d044>] (bus_for_each_dev+0x78/0xac)
(bus_for_each_dev) from [<c047ec78>] (driver_attach+0x2c/0x30)
(driver_attach) from [<c047e640>] (bus_add_driver+0x1e0/0x278)
(bus_add_driver) from [<c048052c>] (driver_register+0x88/0x108)
(driver_register) from [<c04817ec>] (__platform_driver_register+0x50/0x58)
(__platform_driver_register) from [<c0b31380>] (m4u_init+0x24/0x28)
(m4u_init) from [<c0101c38>] (do_one_initcall+0xf0/0x17c)
=========================
The root cause is that the device's iommu-group is NULL while
arm_iommu_attach_device is called. This patch prepare a new iommu-group
for the iommu consumer devices to fix this issue.
CC: Robin Murphy <robin.murphy@arm.com>
CC: Honghui Zhang <honghui.zhang@mediatek.com>
Fixes: 05f80300dc ("iommu: Finish making iommu_group support mandatory")
Reported-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Yong Wu <yong.wu@mediatek.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Since iommu_group_get_for_dev() already tries iommu_group_get() and will
not call ops->device_group if the group is already non-NULL, the check
in get_device_iommu_group() is always redundant and it reduces to a
duplicate of the generic version; let's just use that one instead.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
search_dev_data() acquires a non-raw lock, which can't be done
from atomic context on PREEMPT_RT. There is no need to look at
dev_data because guest_mode should never be set if use_vapic is
not set.
Signed-off-by: Scott Wood <swood@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Several functions in this driver are called from atomic context,
and thus raw locks must be used in order to be safe on PREEMPT_RT.
This includes paths that must wait for command completion, which is
a potential PREEMPT_RT latency concern but not easily avoidable.
Signed-off-by: Scott Wood <swood@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Including:
- 5-level page-table support for the Intel IOMMU.
- Error reporting improvements for the AMD IOMMU driver
- Additional DT bindings for ipmmu-vmsa (Renesas)
- Smaller fixes and cleanups
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=+LFh
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v4.16' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU updates from Joerg Roedel:
"This time there are not a lot of changes coming from the IOMMU side.
That is partly because I returned from my parental leave late in the
development process and probably partly because everyone was busy with
Spectre and Meltdown mitigation work and didn't find the time for
IOMMU work. So here are the few changes that queued up for this merge
window:
- 5-level page-table support for the Intel IOMMU.
- error reporting improvements for the AMD IOMMU driver
- additional DT bindings for ipmmu-vmsa (Renesas)
- small fixes and cleanups"
* tag 'iommu-updates-v4.16' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu: Clean up of_iommu_init_fn
iommu/ipmmu-vmsa: Remove redundant of_iommu_init_fn hook
iommu/msm: Claim bus ops on probe
iommu/vt-d: Enable 5-level paging mode in the PASID entry
iommu/vt-d: Add a check for 5-level paging support
iommu/vt-d: Add a check for 1GB page support
iommu/vt-d: Enable upto 57 bits of domain address width
iommu/vt-d: Use domain instead of cache fetching
iommu/exynos: Don't unconditionally steal bus ops
iommu/omap: Fix debugfs_create_*() usage
iommu/vt-d: clean up pr_irq if request_threaded_irq fails
iommu: Check the result of iommu_group_get() for NULL
iommu/ipmmu-vmsa: Add r8a779(70|95) DT bindings
iommu/ipmmu-vmsa: Add r8a7796 DT binding
iommu/amd: Set the device table entry PPR bit for IOMMU V2 devices
iommu/amd - Record more information about unknown events
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=ttq9
-----END PGP SIGNATURE-----
Merge tag 'pci-v4.16-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI updates from Bjorn Helgaas:
- skip AER driver error recovery callbacks for correctable errors
reported via ACPI APEI, as we already do for errors reported via the
native path (Tyler Baicar)
- fix DPC shared interrupt handling (Alex Williamson)
- print full DPC interrupt number (Keith Busch)
- enable DPC only if AER is available (Keith Busch)
- simplify DPC code (Bjorn Helgaas)
- calculate ASPM L1 substate parameter instead of hardcoding it (Bjorn
Helgaas)
- enable Latency Tolerance Reporting for ASPM L1 substates (Bjorn
Helgaas)
- move ASPM internal interfaces out of public header (Bjorn Helgaas)
- allow hot-removal of VGA devices (Mika Westerberg)
- speed up unplug and shutdown by assuming Thunderbolt controllers
don't support Command Completed events (Lukas Wunner)
- add AtomicOps support for GPU and Infiniband drivers (Felix Kuehling,
Jay Cornwall)
- expose "ari_enabled" in sysfs to help NIC naming (Stuart Hayes)
- clean up PCI DMA interface usage (Christoph Hellwig)
- remove PCI pool API (replaced with DMA pool) (Romain Perier)
- deprecate pci_get_bus_and_slot(), which assumed PCI domain 0 (Sinan
Kaya)
- move DT PCI code from drivers/of/ to drivers/pci/ (Rob Herring)
- add PCI-specific wrappers for dev_info(), etc (Frederick Lawler)
- remove warnings on sysfs mmap failure (Bjorn Helgaas)
- quiet ROM validation messages (Alex Deucher)
- remove redundant memory alloc failure messages (Markus Elfring)
- fill in types for compile-time VGA and other I/O port resources
(Bjorn Helgaas)
- make "pci=pcie_scan_all" work for Root Ports as well as Downstream
Ports to help AmigaOne X1000 (Bjorn Helgaas)
- add SPDX tags to all PCI files (Bjorn Helgaas)
- quirk Marvell 9128 DMA aliases (Alex Williamson)
- quirk broken INTx disable on Ceton InfiniTV4 (Bjorn Helgaas)
- fix CONFIG_PCI=n build by adding dummy pci_irqd_intx_xlate() (Niklas
Cassel)
- use DMA API to get MSI address for DesignWare IP (Niklas Cassel)
- fix endpoint-mode DMA mask configuration (Kishon Vijay Abraham I)
- fix ARTPEC-6 incorrect IS_ERR() usage (Wei Yongjun)
- add support for ARTPEC-7 SoC (Niklas Cassel)
- add endpoint-mode support for ARTPEC (Niklas Cassel)
- add Cadence PCIe host and endpoint controller driver (Cyrille
Pitchen)
- handle multiple INTx status bits being set in dra7xx (Vignesh R)
- translate dra7xx hwirq range to fix INTD handling (Vignesh R)
- remove deprecated Exynos PHY initialization code (Jaehoon Chung)
- fix MSI erratum workaround for HiSilicon Hip06/Hip07 (Dongdong Liu)
- fix NULL pointer dereference in iProc BCMA driver (Ray Jui)
- fix Keystone interrupt-controller-node lookup (Johan Hovold)
- constify qcom driver structures (Julia Lawall)
- rework Tegra config space mapping to increase space available for
endpoints (Vidya Sagar)
- simplify Tegra driver by using bus->sysdata (Manikanta Maddireddy)
- remove PCI_REASSIGN_ALL_BUS usage on Tegra (Manikanta Maddireddy)
- add support for Global Fabric Manager Server (GFMS) event to
Microsemi Switchtec switch driver (Logan Gunthorpe)
- add IDs for Switchtec PSX 24xG3 and PSX 48xG3 (Kelvin Cao)
* tag 'pci-v4.16-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci: (140 commits)
PCI: cadence: Add EndPoint Controller driver for Cadence PCIe controller
dt-bindings: PCI: cadence: Add DT bindings for Cadence PCIe endpoint controller
PCI: endpoint: Fix EPF device name to support multi-function devices
PCI: endpoint: Add the function number as argument to EPC ops
PCI: cadence: Add host driver for Cadence PCIe controller
dt-bindings: PCI: cadence: Add DT bindings for Cadence PCIe host controller
PCI: Add vendor ID for Cadence
PCI: Add generic function to probe PCI host controllers
PCI: generic: fix missing call of pci_free_resource_list()
PCI: OF: Add generic function to parse and allocate PCI resources
PCI: Regroup all PCI related entries into drivers/pci/Makefile
PCI/DPC: Reformat DPC register definitions
PCI/DPC: Add and use DPC Status register field definitions
PCI/DPC: Squash dpc_rp_pio_get_info() into dpc_process_rp_pio_error()
PCI/DPC: Remove unnecessary RP PIO register structs
PCI/DPC: Push dpc->rp_pio_status assignment into dpc_rp_pio_get_info()
PCI/DPC: Squash dpc_rp_pio_print_error() into dpc_rp_pio_get_info()
PCI/DPC: Make RP PIO log size check more generic
PCI/DPC: Rename local "status" to "dpc_status"
PCI/DPC: Squash dpc_rp_pio_print_tlp_header() into dpc_rp_pio_print_error()
...
A number of new drivers get added this time, along with many low-priority
bugfixes. The most interesting changes by subsystem are:
bus drivers:
- Updates to the Broadcom bus interface driver to support newer SoC types
- The TI OMAP sysc driver now supports updated DT bindings
memory controllers:
- A new driver for Tegra186 gets added
- A new driver for the ti-emif sram, to allow relocating
suspend/resume handlers there
SoC specific:
- A new driver for Qualcomm QMI, the interface to the modem on MSM SoCs
- A new driver for power domains on the actions S700 SoC
- A driver for the Xilinx Zynq VCU logicoreIP
reset controllers:
- A new driver for Amlogic Meson-AGX
- various bug fixes
tee subsystem:
- A new user interface got added to enable asynchronous communication
with the TEE supplicant.
- A new method of using user space memory for communication with
the TEE is added
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=yz0k
-----END PGP SIGNATURE-----
Merge tag 'armsoc-drivers' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC driver updates from Arnd Bergmann:
"A number of new drivers get added this time, along with many
low-priority bugfixes. The most interesting changes by subsystem are:
bus drivers:
- Updates to the Broadcom bus interface driver to support newer SoC
types
- The TI OMAP sysc driver now supports updated DT bindings
memory controllers:
- A new driver for Tegra186 gets added
- A new driver for the ti-emif sram, to allow relocating
suspend/resume handlers there
SoC specific:
- A new driver for Qualcomm QMI, the interface to the modem on MSM
SoCs
- A new driver for power domains on the actions S700 SoC
- A driver for the Xilinx Zynq VCU logicoreIP
reset controllers:
- A new driver for Amlogic Meson-AGX
- various bug fixes
tee subsystem:
- A new user interface got added to enable asynchronous communication
with the TEE supplicant.
- A new method of using user space memory for communication with the
TEE is added"
* tag 'armsoc-drivers' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (84 commits)
of: platform: fix OF node refcount leak
soc: fsl: guts: Add a NULL check for devm_kasprintf()
bus: ti-sysc: Fix smartreflex sysc mask
psci: add CPU_IDLE dependency
soc: xilinx: Fix Kconfig alignment
soc: xilinx: xlnx_vcu: Use bitwise & rather than logical && on clkoutdiv
soc: xilinx: xlnx_vcu: Depends on HAS_IOMEM for xlnx_vcu
soc: bcm: brcmstb: Be multi-platform compatible
soc: brcmstb: biuctrl: exit without warning on non brcmstb platforms
Revert "soc: brcmstb: Only register SoC device on STB platforms"
bus: omap: add MODULE_LICENSE tags
soc: brcmstb: Only register SoC device on STB platforms
tee: shm: Potential NULL dereference calling tee_shm_register()
soc: xilinx: xlnx_vcu: Add Xilinx ZYNQMP VCU logicoreIP init driver
dt-bindings: soc: xilinx: Add DT bindings to xlnx_vcu driver
soc: xilinx: Create folder structure for soc specific drivers
of: platform: populate /firmware/ node from of_platform_default_populate_init()
soc: samsung: Add SPDX license identifiers
soc: qcom: smp2p: Use common error handling code in qcom_smp2p_probe()
tee: shm: don't put_page on null shm->pages
...
Commit 4d4bbd8526 ("mm, oom_reaper: skip mm structs with mmu
notifiers") prevented the oom reaper from unmapping private anonymous
memory with the oom reaper when the oom victim mm had mmu notifiers
registered.
The rationale is that doing mmu_notifier_invalidate_range_{start,end}()
around the unmap_page_range(), which is needed, can block and the oom
killer will stall forever waiting for the victim to exit, which may not
be possible without reaping.
That concern is real, but only true for mmu notifiers that have
blockable invalidate_range_{start,end}() callbacks. This patch adds a
"flags" field to mmu notifier ops that can set a bit to indicate that
these callbacks do not block.
The implementation is steered toward an expensive slowpath, such as
after the oom reaper has grabbed mm->mmap_sem of a still alive oom
victim.
[rientjes@google.com: mmu_notifier_invalidate_range_end() can also call the invalidate_range() must not block, fix comment]
Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1801091339570.240101@chino.kir.corp.google.com
[akpm@linux-foundation.org: make mm_has_blockable_invalidate_notifiers() return bool, use rwsem_is_locked()]
Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1712141329500.74052@chino.kir.corp.google.com
Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Dimitri Sivanich <sivanich@hpe.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Oded Gabbay <oded.gabbay@gmail.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Doug Ledford <dledford@redhat.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Mike Marciniszyn <mike.marciniszyn@intel.com>
Cc: Sean Hefty <sean.hefty@intel.com>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Jérôme Glisse <jglisse@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now that no more drivers rely on arbitrary early initialisation via an
of_iommu_init_fn hook, let's clean up the redundant remnants. The
IOMMU_OF_DECLARE() macro needs to remain for now, as the probe-deferral
mechanism has no other nice way to detect built-in drivers before they
have registered themselves, such that it can make the right decision.
Reviewed-by: Sricharan R <sricharan@codeaurora.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Having of_iommu_init() call ipmmu_init() via ipmmu_vmsa_iommu_of_setup()
does nothing that the subsys_initcall wouldn't do slightly later anyway,
since probe-deferral of masters means it is no longer critical to
register the driver super-early. Clean it up.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Since the MSM IOMMU driver now probes via DT exclusively rather than
platform data, dependent masters should be deferred until the IOMMU
itself is ready. Thus we can do away with the early initialisation
hook to unconditionally claim the bus ops, and instead do that only
once an IOMMU is actually probed. Furthermore, this should also make
the driver safe for multiplatform kernels on non-MSM SoCs.
Reviewed-by: Sricharan R <sricharan@codeaurora.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
If the CPU has support for 5-level paging enabled and the IOMMU also
supports 5-level paging then enable the 5-level paging mode for first-
level translations - used when SVM is enabled.
Signed-off-by: Sohil Mehta <sohil.mehta@intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Add a check to verify IOMMU 5-level paging support. If the CPU supports
supports 5-level paging but the IOMMU does not support it then disable
SVM by not allocating PASID tables.
Signed-off-by: Sohil Mehta <sohil.mehta@intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Add a check to verify IOMMU 1GB page support. If the CPU supports 1GB
pages but the IOMMU does not support it then disable SVM by not
allocating PASID tables.
Signed-off-by: Sohil Mehta <sohil.mehta@intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Update the IOMMU default domain address width to 57 bits. This would
enable the IOMMU to do upto 5-levels of paging for second level
translations - IOVA translation requests without PASID.
Even though the maximum supported address width is being increased to
57, __iommu_calculate_agaw() would set the actual supported address
width to the maximum support available in IOMMU hardware.
Signed-off-by: Sohil Mehta <sohil.mehta@intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Removing the early device registration hook overlooked the fact that
it only ran conditionally on a compatible device being present in the
DT. With exynos_iommu_init() now running as an unconditional initcall,
problems arise on non-Exynos systems when other IOMMU drivers find
themselves unable to install their ops on the platform bus, or at worst
the Exynos ops get called with someone else's domain and all hell breaks
loose.
The global ops/cache setup could probably all now be triggered from the
first IOMMU probe, as with dma_dev assigment, but for the time being the
simplest fix is to resurrect the logic from commit a7b67cd5d9
("iommu/exynos: Play nice in multi-platform builds") to explicitly check
the DT for the presence of an Exynos IOMMU before trying anything.
Fixes: 928055a01b ("iommu/exynos: Remove custom platform device registration code")
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When exposing data access through debugfs, the correct
debugfs_create_*() functions must be used, depending on data type.
Remove all casts from data pointers passed to debugfs_create_*()
functions, as such casts prevent the compiler from flagging bugs.
omap_iommu.nr_tlb_entries is "int", hence casting to "u8 *" exposes only
a part of it. Fix this by using debugfs_create_u32() instead.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Move the few remaining bits of swiotlb glue towards their callers,
and remove the pointless on ia64 swiotlb variable.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
pci_get_bus_and_slot() is restrictive such that it assumes domain=0 as
where a PCI device is present. This restricts the device drivers to be
reused for other domain numbers.
Getting ready to remove pci_get_bus_and_slot() function in favor of
pci_get_domain_bus_and_slot().
Hard-code the domain number as 0 for the AMD IOMMU driver.
Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Signed-off-by: Bjorn Helgaas <helgaas@kernel.org>
Reviewed-by: Gary R Hook <gary.hook@amd.com>
Acked-by: Joerg Roedel <jroedel@suse.de>
For PCI devices behind an aliasing PCIe-to-PCI/X bridge, the bridge
alias to DevFn 0.0 on the subordinate bus may match the original RID of
the device, resulting in the same SID being present in the device's
fwspec twice. This causes trouble later in arm_smmu_write_strtab_ent()
when we wind up visiting the STE a second time and find it already live.
Avoid the issue by giving arm_smmu_install_ste_for_dev() the cleverness
to skip over duplicates. It seems mildly counterintuitive compared to
preventing the duplicates from existing in the first place, but since
the DT and ACPI probe paths build their fwspecs differently, this is
actually the cleanest and most self-contained way to deal with it.
Cc: <stable@vger.kernel.org>
Fixes: 8f78515425 ("iommu/arm-smmu: Implement of_xlate() for SMMUv3")
Reported-by: Tomasz Nowicki <tomasz.nowicki@caviumnetworks.com>
Tested-by: Tomasz Nowicki <Tomasz.Nowicki@cavium.com>
Tested-by: Jayachandran C. <jnair@caviumnetworks.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Kasan reports a double free when finalise_stage_fn fails: the io_pgtable
ops are freed by arm_smmu_domain_finalise and then again by
arm_smmu_domain_free. Prevent this by leaving pgtbl_ops empty on failure.
Cc: <stable@vger.kernel.org>
Fixes: 48ec83bcbc ("iommu/arm-smmu: Add initial driver support for ARM SMMUv3 devices")
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe.brucker@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
The 'early' argument of irq_domain_activate_irq() is actually used to
denote reservation mode. To avoid confusion, rename it before abuse
happens.
No functional change.
Fixes: 7249164346 ("genirq/irqdomain: Update irq_domain_ops.activate() signature")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Alexandru Chirvasitu <achirvasub@gmail.com>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Dou Liyang <douly.fnst@cn.fujitsu.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Mikael Pettersson <mikpelinux@gmail.com>
Cc: Josh Poulson <jopoulso@microsoft.com>
Cc: Mihai Costache <v-micos@microsoft.com>
Cc: Stephen Hemminger <sthemmin@microsoft.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: linux-pci@vger.kernel.org
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Dexuan Cui <decui@microsoft.com>
Cc: Simon Xiao <sixiao@microsoft.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Cc: Jork Loeser <Jork.Loeser@microsoft.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: devel@linuxdriverproject.org
Cc: KY Srinivasan <kys@microsoft.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Sakari Ailus <sakari.ailus@intel.com>,
Cc: linux-media@vger.kernel.org
The Tegra memory controller driver will now instruct the SMMU driver to
create groups, which will make it easier for device drivers to share an
IOMMU domain between multiple devices.
Initial Tegra186 support is also added in a separate driver.
-----BEGIN PGP SIGNATURE-----
iQJHBAABCAAxFiEEiOrDCAFJzPfAjcif3SOs138+s6EFAlo6tlMTHHRyZWRpbmdA
bnZpZGlhLmNvbQAKCRDdI6zXfz6zoR7MEACpoyPkV5fGz3z8rdsYaVZSA+upYEHw
CC4v5XslNCucmk57XIYTbAZRu7W5MuzP3hq+DoYM5B5EHxgt1a2YHNtjjg3eLx1T
U+8hjoOPOYu7FBVmUf4V6Buvmws90DxtfvqcfIScJJEAPgKsT/wfGXNstc88kDnO
c+huQxk5QZsyGBgt1sS61zlymaLpDlSoE6kScXnjM1K6XtUdZ+8zmoMdZ7Ka8DUv
kwBoYXgUpVminWpSrfuV7f3/8BtTSJ/b2rUPz7Ren0JJTFrhIi/ckeQ5YhTa46Dn
oqR0f6MuLRGADT+E5DfoF7otvE2Fvq0Uq/YlypIf9gtiit5mhPM7ZXb3LOwDGiCp
0mwljW9Uov7ZpFnb8b9SlVf1b+zDeeLxtARq4qzgky5OkWZp4hXZw+Os3ebXQYW3
d/MUfwdqgNKOccGRsRkk9OEdUiFGpGtInb0xM9OWBCHvdrXhbK9llZ3wnrE+M3SY
BELM+94YorqVPx5JTmiM7kTozVwpkvZQG6kU2J81Y55FI1ELQF7Mk0yzFs562tG5
2abc07xkSUCdC6zsrgNOy+AtuRH8oHccUfZRACLUhzHNCrvfr3MrEcEv8l4VRJp4
ZLfR8UjmtTI8YllBKdSdUtFB4nN5oyoIPCNsgwzXcTnKV5SzAyA/KuDW6IWoNU1L
6ekAxpeiHVnLKA==
=ncAl
-----END PGP SIGNATURE-----
Merge tag 'tegra-for-4.16-memory' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tegra/linux into next/drivers
Pull "memory: tegra: Changes for v4.16-rc1" from Thierry Reding:
The Tegra memory controller driver will now instruct the SMMU driver to
create groups, which will make it easier for device drivers to share an
IOMMU domain between multiple devices.
Initial Tegra186 support is also added in a separate driver.
* tag 'tegra-for-4.16-memory' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
iommu/tegra-smmu: Fix return value check in tegra_smmu_group_get()
iommu/tegra: Allow devices to be grouped
memory: tegra: Create SMMU display groups
memory: tegra: Add Tegra186 support
dt-bindings: memory: Add Tegra186 support
dt-bindings: misc: Add Tegra186 MISC registers bindings
In case of error, the function iommu_group_alloc() returns ERR_PTR() and
never returns NULL. The NULL test in the return value check should be
replaced with IS_ERR().
Fixes: 7f4c9176f7 ("iommu/tegra: Allow devices to be grouped")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
The result of iommu_group_get() was being blindly used in both
attach and detach which results in a dereference when trying
to work with an unknown device.
Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The AMD IOMMU specification Rev 3.00 (December 2016) introduces a
new Enhanced PPR Handling Support (EPHSup) bit in the MMIO register
offset 0030h (IOMMU Extended Feature Register).
When EPHSup=1, the IOMMU hardware requires the PPR bit of the
device table entry (DTE) to be set in order to support PPR for a
particular endpoint device.
Please see https://support.amd.com/TechDocs/48882_IOMMU.pdf for
this revision of the AMD IOMMU specification.
Signed-off-by: Gary R Hook <gary.hook@amd.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
When an unknown type event occurs, the default information written to
the syslog should dump raw event data. This could provide insight into
the event that occurred.
Signed-off-by: Gary R Hook <gary.hook@amd.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Implement the ->device_group() and ->of_xlate() callbacks which are used
in order to group devices. Each group can then share a single domain.
This is implemented primarily in order to achieve the same semantics on
Tegra210 and earlier as on Tegra186 where the Tegra SMMU was replaced by
an ARM SMMU. Users of the IOMMU API can now use the same code to share
domains between devices, whereas previously they used to attach each
device individually.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
The intel-iommu DMA ops fail to correctly handle scatterlists where
sg->offset is greater than PAGE_SIZE - the IOVA allocation is computed
appropriately based on the page-aligned portion of the offset, but the
mapping is set up relative to sg->page, which means it fails to actually
cover the whole buffer (and in the worst case doesn't cover it at all):
(sg->dma_address + sg->dma_len) ----+
sg->dma_address ---------+ |
iov_pfn------+ | |
| | |
v v v
iova: a b c d e f
|--------|--------|--------|--------|--------|
<...calculated....>
[_____mapped______]
pfn: 0 1 2 3 4 5
|--------|--------|--------|--------|--------|
^ ^ ^
| | |
sg->page ----+ | |
sg->offset --------------+ |
(sg->offset + sg->length) ----------+
As a result, the caller ends up overrunning the mapping into whatever
lies beyond, which usually goes badly:
[ 429.645492] DMAR: DRHD: handling fault status reg 2
[ 429.650847] DMAR: [DMA Write] Request device [02:00.4] fault addr f2682000 ...
Whilst this is a fairly rare occurrence, it can happen from the result
of intermediate scatterlist processing such as scatterwalk_ffwd() in the
crypto layer. Whilst that particular site could be fixed up, it still
seems worthwhile to bring intel-iommu in line with other DMA API
implementations in handling this robustly.
To that end, fix the intel_map_sg() path to line up the mapping
correctly (in units of MM pages rather than VT-d pages to match the
aligned_nrpages() calculation) regardless of the offset, and use
sg_phys() consistently for clarity.
Reported-by: Harsh Jain <Harsh@chelsio.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed by: Ashok Raj <ashok.raj@intel.com>
Tested by: Jacob Pan <jacob.jun.pan@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
* Enforce MSI multiple IRQ alignment in AMD IOMMU
* VT-d PASID error handling fixes
* Add r8a7795 IPMMU support
* Manage runtime PM links on exynos at {add,remove}_device callbacks
* Fix Mediatek driver name to avoid conflict
* Add terminate support to qcom fault handler
* 64-bit IOVA optimizations
* Simplfy IOVA domain destruction, better use of rcache, and
skip anchor nodes on copy
* Convert to IOMMU TLB sync API in io-pgtable-arm{-v7s}
* Drop command queue lock when waiting for CMD_SYNC completion on
ARM SMMU implementations supporting MSI to cacheable memory
* iomu-vmsa cleanup inspired by missed IOTLB sync callbacks
* Fix sleeping lock with preemption disabled for RT
* Dual MMU support for TI DRA7xx DSPs
* Optional flush option on IOVA allocation avoiding overhead when
caller can try other options
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)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=o4qF
-----END PGP SIGNATURE-----
Merge tag 'iommu-v4.15-rc1' of git://github.com/awilliam/linux-vfio
Pull IOMMU updates from Alex Williamson:
"As Joerg mentioned[1], he's out on paternity leave through the end of
the year and I'm filling in for him in the interim:
- Enforce MSI multiple IRQ alignment in AMD IOMMU
- VT-d PASID error handling fixes
- Add r8a7795 IPMMU support
- Manage runtime PM links on exynos at {add,remove}_device callbacks
- Fix Mediatek driver name to avoid conflict
- Add terminate support to qcom fault handler
- 64-bit IOVA optimizations
- Simplfy IOVA domain destruction, better use of rcache, and skip
anchor nodes on copy
- Convert to IOMMU TLB sync API in io-pgtable-arm{-v7s}
- Drop command queue lock when waiting for CMD_SYNC completion on ARM
SMMU implementations supporting MSI to cacheable memory
- iomu-vmsa cleanup inspired by missed IOTLB sync callbacks
- Fix sleeping lock with preemption disabled for RT
- Dual MMU support for TI DRA7xx DSPs
- Optional flush option on IOVA allocation avoiding overhead when
caller can try other options
[1] https://lkml.org/lkml/2017/10/22/72"
* tag 'iommu-v4.15-rc1' of git://github.com/awilliam/linux-vfio: (54 commits)
iommu/iova: Use raw_cpu_ptr() instead of get_cpu_ptr() for ->fq
iommu/mediatek: Fix driver name
iommu/ipmmu-vmsa: Hook up r8a7795 DT matching code
iommu/ipmmu-vmsa: Allow two bit SL0
iommu/ipmmu-vmsa: Make IMBUSCTR setup optional
iommu/ipmmu-vmsa: Write IMCTR twice
iommu/ipmmu-vmsa: IPMMU device is 40-bit bus master
iommu/ipmmu-vmsa: Make use of IOMMU_OF_DECLARE()
iommu/ipmmu-vmsa: Enable multi context support
iommu/ipmmu-vmsa: Add optional root device feature
iommu/ipmmu-vmsa: Introduce features, break out alias
iommu/ipmmu-vmsa: Unify ipmmu_ops
iommu/ipmmu-vmsa: Clean up struct ipmmu_vmsa_iommu_priv
iommu/ipmmu-vmsa: Simplify group allocation
iommu/ipmmu-vmsa: Unify domain alloc/free
iommu/ipmmu-vmsa: Fix return value check in ipmmu_find_group_dma()
iommu/vt-d: Clear pasid table entry when memory unbound
iommu/vt-d: Clear Page Request Overflow fault bit
iommu/vt-d: Missing checks for pasid tables if allocation fails
iommu/amd: Limit the IOVA page range to the specified addresses
...
get_cpu_ptr() disabled preemption and returns the ->fq object of the
current CPU. raw_cpu_ptr() does the same except that it not disable
preemption which means the scheduler can move it to another CPU after it
obtained the per-CPU object.
In this case this is not bad because the data structure itself is
protected with a spin_lock. This change shouldn't matter however on RT
it does because the sleeping lock can't be accessed with disabled
preemption.
Cc: Joerg Roedel <joro@8bytes.org>
Cc: iommu@lists.linux-foundation.org
Reported-by: vinadhy@gmail.com
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
There exist two Mediatek iommu drivers for the two different
generations of the device. But both drivers have the same name
"mtk-iommu". This breaks the registration of the second driver:
Error: Driver 'mtk-iommu' is already registered, aborting...
Fix this by changing the name for first generation to
"mtk-iommu-v1".
Fixes: b17336c55d ("iommu/mediatek: add support for mtk iommu generation one HW")
Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Tie in r8a7795 features and update the IOMMU_OF_DECLARE
compat string to include the updated compat string.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Introduce support for two bit SL0 bitfield in IMTTBCR
by using a separate feature flag.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Introduce a feature to allow opt-out of setting up
IMBUSCR. The default case is unchanged.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Write IMCTR both in the root device and the leaf node.
To allow access of IMCTR introduce the following function:
- ipmmu_ctx_write_all()
While at it also rename context functions:
- ipmmu_ctx_read() -> ipmmu_ctx_read_root()
- ipmmu_ctx_write() -> ipmmu_ctx_write_root()
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The r8a7795 IPMMU supports 40-bit bus mastering. Both
the coherent DMA mask and the streaming DMA mask are
set to unlock the 40-bit address space for coherent
allocations and streaming operations.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Hook up IOMMU_OF_DECLARE() support in case CONFIG_IOMMU_DMA
is enabled. The only current supported case for 32-bit ARM
is disabled, however for 64-bit ARM usage of OF is required.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Add support for up to 8 contexts. Each context is mapped to one
domain. One domain is assigned one or more slave devices. Contexts
are allocated dynamically and slave devices are grouped together
based on which IPMMU device they are connected to. This makes slave
devices tied to the same IPMMU device share the same IOVA space.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Add root device handling to the IPMMU driver by allowing certain
DT compat strings to enable has_cache_leaf_nodes that in turn will
support both root devices with interrupts and leaf devices that
face the actual IPMMU consumer devices.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Introduce struct ipmmu_features to track various hardware
and software implementation changes inside the driver for
different kinds of IPMMU hardware. Add use_ns_alias_offset
as a first example of a feature to control if the secure
register bank offset should be used or not.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The remaining difference between the ARM-specific and iommu-dma ops is
in the {add,remove}_device implementations, but even those have some
overlap and duplication. By stubbing out the few arm_iommu_*() calls,
we can get rid of the rest of the inline #ifdeffery to both simplify the
code and improve build coverage.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Now that the IPMMU instance pointer is the only thing remaining in the
private data structure, we no longer need the extra level of indirection
and can simply stash that directlty in the fwspec.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
We go through quite the merry dance in order to find masters behind the
same IPMMU instance, so that we can ensure they are grouped together.
None of which is really necessary, since the master's private data
already points to the particular IPMMU it is associated with, and that
IPMMU instance data is the perfect place to keep track of a per-instance
group directly.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
We have two implementations for ipmmu_ops->alloc depending on
CONFIG_IOMMU_DMA, the difference being whether they accept the
IOMMU_DOMAIN_DMA type or not. However, iommu_dma_get_cookie() is
guaranteed to return an error when !CONFIG_IOMMU_DMA, so if
ipmmu_domain_alloc_dma() was actually checking and handling the return
value correctly, it would behave the same as ipmmu_domain_alloc()
anyway.
Similarly for freeing; iommu_put_dma_cookie() is robust by design.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In case of error, the function iommu_group_get() returns NULL pointer
not ERR_PTR(). The IS_ERR() test in the return value check should be
replaced with NULL test.
Fixes: 3ae4729202 ("iommu/ipmmu-vmsa: Add new IOMMU_DOMAIN_DMA ops")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In intel_svm_unbind_mm(), pasid table entry must be cleared during
svm free. Otherwise, hardware may be set up with a wild pointer.
Suggested-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Currently Page Request Overflow bit in IOMMU Fault Status register
is not cleared. Not clearing this bit would mean that any future
page-request is going to be automatically dropped by IOMMU.
Suggested-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
intel_svm_alloc_pasid_tables() might return an error but never be
checked by the callers. Later when intel_svm_bind_mm() is called,
there are no checks for valid pasid tables before enabling them.
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Liu, Yi L <yi.l.liu@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The extent of pages specified when applying a reserved region should
include up to the last page of the range, but not the page following
the range.
Signed-off-by: Gary R Hook <gary.hook@amd.com>
Fixes: 8d54d6c8b8 ('iommu/amd: Implement apply_dm_region call-back')
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This is quite useful for debugging. Currently, always TERMINATE the
translation when the fault handler returns (since this is all we need
for debugging drivers). But I expect the SVM work should eventually
let us do something more clever.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Variable flush_addr is being assigned but is never read; it
is redundant and can be removed. Cleans up the clang warning:
drivers/iommu/amd_iommu.c:2388:2: warning: Value stored to 'flush_addr'
is never read
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
On an is_allocated() interrupt index, we ALIGN() the current index and
then increment it via the for loop, guaranteeing that it is no longer
aligned for alignments >1. We instead need to align the next index,
to guarantee forward progress, moving the increment-only to the case
where the index was found to be unallocated.
Fixes: 37946d95fc ('iommu/amd: Add align parameter to alloc_irq_index()')
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Many source files in the tree are missing licensing information, which
makes it harder for compliance tools to determine the correct license.
By default all files without license information are under the default
license of the kernel, which is GPL version 2.
Update the files which contain no license information with the 'GPL-2.0'
SPDX license identifier. The SPDX identifier is a legally binding
shorthand, which can be used instead of the full boiler plate text.
This patch is based on work done by Thomas Gleixner and Kate Stewart and
Philippe Ombredanne.
How this work was done:
Patches were generated and checked against linux-4.14-rc6 for a subset of
the use cases:
- file had no licensing information it it.
- file was a */uapi/* one with no licensing information in it,
- file was a */uapi/* one with existing licensing information,
Further patches will be generated in subsequent months to fix up cases
where non-standard license headers were used, and references to license
had to be inferred by heuristics based on keywords.
The analysis to determine which SPDX License Identifier to be applied to
a file was done in a spreadsheet of side by side results from of the
output of two independent scanners (ScanCode & Windriver) producing SPDX
tag:value files created by Philippe Ombredanne. Philippe prepared the
base worksheet, and did an initial spot review of a few 1000 files.
The 4.13 kernel was the starting point of the analysis with 60,537 files
assessed. Kate Stewart did a file by file comparison of the scanner
results in the spreadsheet to determine which SPDX license identifier(s)
to be applied to the file. She confirmed any determination that was not
immediately clear with lawyers working with the Linux Foundation.
Criteria used to select files for SPDX license identifier tagging was:
- Files considered eligible had to be source code files.
- Make and config files were included as candidates if they contained >5
lines of source
- File already had some variant of a license header in it (even if <5
lines).
All documentation files were explicitly excluded.
The following heuristics were used to determine which SPDX license
identifiers to apply.
- when both scanners couldn't find any license traces, file was
considered to have no license information in it, and the top level
COPYING file license applied.
For non */uapi/* files that summary was:
SPDX license identifier # files
---------------------------------------------------|-------
GPL-2.0 11139
and resulted in the first patch in this series.
If that file was a */uapi/* path one, it was "GPL-2.0 WITH
Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was:
SPDX license identifier # files
---------------------------------------------------|-------
GPL-2.0 WITH Linux-syscall-note 930
and resulted in the second patch in this series.
- if a file had some form of licensing information in it, and was one
of the */uapi/* ones, it was denoted with the Linux-syscall-note if
any GPL family license was found in the file or had no licensing in
it (per prior point). Results summary:
SPDX license identifier # files
---------------------------------------------------|------
GPL-2.0 WITH Linux-syscall-note 270
GPL-2.0+ WITH Linux-syscall-note 169
((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21
((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17
LGPL-2.1+ WITH Linux-syscall-note 15
GPL-1.0+ WITH Linux-syscall-note 14
((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5
LGPL-2.0+ WITH Linux-syscall-note 4
LGPL-2.1 WITH Linux-syscall-note 3
((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3
((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1
and that resulted in the third patch in this series.
- when the two scanners agreed on the detected license(s), that became
the concluded license(s).
- when there was disagreement between the two scanners (one detected a
license but the other didn't, or they both detected different
licenses) a manual inspection of the file occurred.
- In most cases a manual inspection of the information in the file
resulted in a clear resolution of the license that should apply (and
which scanner probably needed to revisit its heuristics).
- When it was not immediately clear, the license identifier was
confirmed with lawyers working with the Linux Foundation.
- If there was any question as to the appropriate license identifier,
the file was flagged for further research and to be revisited later
in time.
In total, over 70 hours of logged manual review was done on the
spreadsheet to determine the SPDX license identifiers to apply to the
source files by Kate, Philippe, Thomas and, in some cases, confirmation
by lawyers working with the Linux Foundation.
Kate also obtained a third independent scan of the 4.13 code base from
FOSSology, and compared selected files where the other two scanners
disagreed against that SPDX file, to see if there was new insights. The
Windriver scanner is based on an older version of FOSSology in part, so
they are related.
Thomas did random spot checks in about 500 files from the spreadsheets
for the uapi headers and agreed with SPDX license identifier in the
files he inspected. For the non-uapi files Thomas did random spot checks
in about 15000 files.
In initial set of patches against 4.14-rc6, 3 files were found to have
copy/paste license identifier errors, and have been fixed to reflect the
correct identifier.
Additionally Philippe spent 10 hours this week doing a detailed manual
inspection and review of the 12,461 patched files from the initial patch
version early this week with:
- a full scancode scan run, collecting the matched texts, detected
license ids and scores
- reviewing anything where there was a license detected (about 500+
files) to ensure that the applied SPDX license was correct
- reviewing anything where there was no detection but the patch license
was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied
SPDX license was correct
This produced a worksheet with 20 files needing minor correction. This
worksheet was then exported into 3 different .csv files for the
different types of files to be modified.
These .csv files were then reviewed by Greg. Thomas wrote a script to
parse the csv files and add the proper SPDX tag to the file, in the
format that the file expected. This script was further refined by Greg
based on the output to detect more types of files automatically and to
distinguish between header and source .c files (which need different
comment types.) Finally Greg ran the script using the .csv files to
generate the patches.
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
While CMD_SYNC is unlikely to complete immediately such that we never go
round the polling loop, with a lightly-loaded queue it may still do so
long before the delay period is up. If we have no better completion
notifier, use similar logic as we have for SMMUv2 to spin a number of
times before each backoff, so that we have more chance of catching syncs
which complete relatively quickly and avoid delaying unnecessarily.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
We have separate (identical) timeout values for polling for a queue to
drain and waiting for an MSI to signal CMD_SYNC completion. In reality,
we only wait for the command queue to drain if we're waiting on a sync,
so just merged these two timeouts into a single constant.
Signed-off-by: Will Deacon <will.deacon@arm.com>
arm_smmu_cmdq_issue_sync is a little unwieldy now that it supports both
MSI and event-based polling, so split it into two functions to make things
easier to follow.
Signed-off-by: Will Deacon <will.deacon@arm.com>
As an IRQ, the CMD_SYNC interrupt is not particularly useful, not least
because we often need to wait for sync completion within someone else's
IRQ handler anyway. However, when the SMMU is both coherent and supports
MSIs, we can have a lot more fun by not using it as an interrupt at all.
Following the example suggested in the architecture and using a write
targeting normal memory, we can let callers wait on a status variable
outside the lock instead of having to stall the entire queue or even
touch MMIO registers. Since multiple sync commands are guaranteed to
complete in order, a simple incrementing sequence count is all we need
to unambiguously support any realistic number of overlapping waiters.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
The cmdq-sync interrupt is never going to be particularly useful, since
for stage 1 DMA at least we'll often need to wait for sync completion
within someone else's IRQ handler, thus have to implement polling
anyway. Beyond that, the overhead of taking an interrupt, then still
having to grovel around in the queue to figure out *which* sync command
completed, doesn't seem much more attractive than simple polling either.
Furthermore, if an implementation both has wired interrupts and supports
MSIs, then we don't want to be taking the IRQ unnecessarily if we're
using the MSI write to update memory. Let's just make life simpler by
not even bothering to claim it in the first place.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
CMD_SYNC already has a bit of special treatment here and there, but as
we're about to extend it with more functionality for completing outside
the CMDQ lock, things are going to get rather messy if we keep trying to
cram everything into a single generic command interface. Instead, let's
break out the issuing of CMD_SYNC into its own specific helper where
upcoming changes will have room to breathe.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Slightly confusingly, when reporting a mismatch of the ID register
value, we still refer to the IORT COHACC override flag as the
"dma-coherent property" if we booted with ACPI. Update the message
to be firmware-agnostic in line with SMMUv2.
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reported-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
According to Spec, it is ILLEGAL to set STE.S1STALLD if STALL_MODEL
is not 0b00, which means we should not disable stall mode if stall
or terminate mode is not configuable.
Meanwhile, it is also ILLEGAL when STALL_MODEL==0b10 && CD.S==0 which
means if stall mode is force we should always set CD.S.
As Jean-Philippe's suggestion, this patch introduce a feature bit
ARM_SMMU_FEAT_STALL_FORCE, which means smmu only supports stall force.
Therefore, we can avoid the ILLEGAL setting of STE.S1STALLD.by checking
ARM_SMMU_FEAT_STALL_FORCE.
This patch keeps the ARM_SMMU_FEAT_STALLS as the meaning of stall supported
(force or configuable) to easy to expand the future function, i.e. we can
only use ARM_SMMU_FEAT_STALLS to check whether we should register fault
handle or enable master can_stall, etc to supporte platform SVM.
The feature bit, STE.S1STALLD and CD.S setting will be like:
STALL_MODEL FEATURE S1STALLD CD.S
0b00 ARM_SMMU_FEAT_STALLS 0b1 0b0
0b01 !ARM_SMMU_FEAT_STALLS && !ARM_SMMU_FEAT_STALL_FORCE 0b0 0b0
0b10 ARM_SMMU_FEAT_STALLS && ARM_SMMU_FEAT_STALL_FORCE 0b0 0b1
after apply this patch.
Signed-off-by: Yisheng Xie <xieyisheng1@huawei.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
The ARM SMMU identity mapping performance was poor compared with the
DMA mode. It was found that enable caching would restore the performance
back to normal. The S2CRB_TLBEN bit in the ACR register would allow for
caching of the stream to context register bypass transaction information.
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Feng Kan <fkan@apm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
The SMMUv3 architecture permits caching of data structures deemed to be
"reachable" by the SMU, which includes STEs marked as invalid. When
transitioning an STE to a bypass/fault configuration at init or detach
time, we mistakenly elide the CMDQ_OP_CFGI_STE operation in some cases,
therefore potentially leaving the old STE state cached in the SMMU.
This patch fixes the problem by ensuring that we perform the
CMDQ_OP_CFGI_STE operation irrespective of the validity of the previous
STE.
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Reported-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Now that the kernel headers have synced with the relevant upstream
ACPICA updates, it's time to clean up the temporary local definitions.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
The function only sends the flush command to the IOMMU(s),
but does not wait for its completion when it returns. Fix
that.
Fixes: 601367d76b ('x86/amd-iommu: Remove iommu_flush_domain function')
Cc: stable@vger.kernel.org # >= 2.6.33
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>
Exynos SYSMMU registers standard platform device with sysmmu_of_match
table, what means that this table is accessed every time a new platform
device is registered in a system. This might happen also after the boot,
so the table must not be attributed as initconst to avoid potential kernel
oops caused by access to freed memory.
Fixes: 6b21a5db36 ("iommu/exynos: Support for device tree")
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When SME memory encryption is active it will rely on SWIOTLB to handle
DMA for devices that cannot support the addressing requirements of
having the encryption mask set in the physical address. The IOMMU
currently disables SWIOTLB if it is not running in passthrough mode.
This is not desired as non-PCI devices attempting DMA may fail. Update
the code to check if SME is active and not disable SWIOTLB.
Fixes: 2543a786aa ("iommu/amd: Allow the AMD IOMMU to work with memory encryption")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Make use of the new alignment capability of
alloc_irq_index() to enforce IRQ index alignment
for MSI.
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Fixes: 2b32450634 ('iommu/amd: Add routines to manage irq remapping tables')
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
For multi-MSI IRQ ranges the IRQ index needs to be aligned
to the power-of-two of the requested IRQ count. Extend the
alloc_irq_index() function to allow such an allocation.
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Fixes: 2b32450634 ('iommu/amd: Add routines to manage irq remapping tables')
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Variable did_old is unsigned so checking whether it is
greater or equal to zero is not necessary.
Signed-off-by: Christos Gkekas <chris.gekas@gmail.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The notifier function will take the dmar_global_lock too, so
lockdep complains about inverse locking order when the
notifier is registered under the dmar_global_lock.
Reported-by: Jan Kiszka <jan.kiszka@siemens.com>
Fixes: 59ce0515cd ('iommu/vt-d: Update DRHD/RMRR/ATSR device scope caches when PCI hotplug happens')
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Now that the core API issues its own post-unmap TLB sync call, push that
operation out from the io-pgtable-arm-v7s internals into the users. For
now, we leave the invalidation implicit in the unmap operation, since
none of the current users would benefit much from any change to that.
Note that the conversion of msm_iommu is implicit, since that apparently
has no specific TLB sync operation anyway.
CC: Yong Wu <yong.wu@mediatek.com>
CC: Rob Clark <robdclark@gmail.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Now that the core API issues its own post-unmap TLB sync call, push that
operation out from the io-pgtable-arm internals into the users. For now,
we leave the invalidation implicit in the unmap operation, since none of
the current users would benefit much from any change to that.
CC: Magnus Damm <damm+renesas@opensource.se>
CC: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Anchor nodes are not reserved IOVAs in the way that copy_reserved_iova()
cares about - while the failure from reserve_iova() is benign since the
target domain will already have its own anchor, we still don't want to
be triggering spurious warnings.
Reported-by: kernel test robot <fengguang.wu@intel.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Fixes: bb68b2fbfb ('iommu/iova: Add rbtree anchor node')
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When devices with different DMA masks are using the same domain, or for
PCI devices where we usually try a speculative 32-bit allocation first,
there is a fair possibility that the top PFN of the rcache stack at any
given time may be unsuitable for the lower limit, prompting a fallback
to allocating anew from the rbtree. Consequently, we may end up
artifically increasing pressure on the 32-bit IOVA space as unused IOVAs
accumulate lower down in the rcache stacks, while callers with 32-bit
masks also impose unnecessary rbtree overhead.
In such cases, let's try a bit harder to satisfy the allocation locally
first - scanning the whole stack should still be relatively inexpensive.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When popping a pfn from an rcache, we are currently checking it directly
against limit_pfn for viability. Since this represents iova->pfn_lo, it
is technically possible for the corresponding iova->pfn_hi to be greater
than limit_pfn. Although we generally get away with it in practice since
limit_pfn is typically a power-of-two boundary and the IOVAs are
size-aligned, it's pretty trivial to make the iova_rcache_get() path
take the allocation size into account for complete safety.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
All put_iova_domain() should have to worry about is freeing memory - by
that point the domain must no longer be live, so the act of cleaning up
doesn't need to be concurrency-safe or maintain the rbtree in a
self-consistent state. There's no need to waste time with locking or
emptying the rcache magazines, and we can just use the postorder
traversal helper to clear out the remaining rbtree entries in-place.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The logic of __get_cached_rbnode() is a little obtuse, but then
__get_prev_node_of_cached_rbnode_or_last_node_and_update_limit_pfn()
wouldn't exactly roll off the tongue...
Now that we have the invariant that there is always a valid node to
start searching downwards from, everything gets a bit easier to follow
if we simplify that function to do what it says on the tin and return
the cached node (or anchor node as appropriate) directly. In turn, we
can then deduplicate the rb_prev() and limit_pfn logic into the main
loop itself, further reduce the amount of code under the lock, and
generally make the inner workings a bit less subtle.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Add a permanent dummy IOVA reservation to the rbtree, such that we can
always access the top of the address space instantly. The immediate
benefit is that we remove the overhead of the rb_last() traversal when
not using the cached node, but it also paves the way for further
simplifications.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
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>
The cached node mechanism provides a significant performance benefit for
allocations using a 32-bit DMA mask, but in the case of non-PCI devices
or where the 32-bit space is full, the loss of this benefit can be
significant - on large systems there can be many thousands of entries in
the tree, such that walking all the way down to find free space every
time becomes increasingly awful.
Maintain a similar cached node for the whole IOVA space as a superset of
the 32-bit space so that performance can remain much more consistent.
Inspired by work 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>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The mask for calculating the padding size doesn't change, so there's no
need to recalculate it every loop iteration. Furthermore, Once we've
done that, it becomes clear that we don't actually need to calculate a
padding size at all - by flipping the arithmetic around, we can just
combine the upper limit, size, and mask directly to check against the
lower limit.
For an arm64 build, this alone knocks 20% off the object code size of
the entire alloc_iova() function!
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: simplified more of the arithmetic, rewrote commit message]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Checking the IOVA bounds separately before deciding which direction to
continue the search (if necessary) results in redundantly comparing both
pfns twice each. GCC can already determine that the final comparison op
is redundant and optimise it down to 3 in total, but we can go one
further with a little tweak of the ordering (which makes the intent of
the code that much cleaner as a bonus).
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: rewrote commit message to clarify]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
pr_err() messages should end with a new-line to avoid other messages
being concatenated. So replace '/n' with '\n'.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Fixes: 45a01c4293 ('iommu/amd: Add function copy_dev_tables()')
Signed-off-by: Joerg Roedel <jroedel@suse.de>
