Register our DRHD IOMMUs, cross link devices, and provide a base set
of attributes for the IOMMU. Note that IRQ remapping support parses
the DMAR table very early in boot, well before the iommu_class can
reasonably be setup, so our registration is split between
intel_iommu_init(), which occurs later, and alloc_iommu(), which
typically occurs much earlier, but may happen at any time later
with IOMMU hot-add support.
On a typical desktop system, this provides the following (pruned):
$ find /sys | grep dmar
/sys/devices/virtual/iommu/dmar0
/sys/devices/virtual/iommu/dmar0/devices
/sys/devices/virtual/iommu/dmar0/devices/0000:00:02.0
/sys/devices/virtual/iommu/dmar0/intel-iommu
/sys/devices/virtual/iommu/dmar0/intel-iommu/cap
/sys/devices/virtual/iommu/dmar0/intel-iommu/ecap
/sys/devices/virtual/iommu/dmar0/intel-iommu/address
/sys/devices/virtual/iommu/dmar0/intel-iommu/version
/sys/devices/virtual/iommu/dmar1
/sys/devices/virtual/iommu/dmar1/devices
/sys/devices/virtual/iommu/dmar1/devices/0000:00:00.0
/sys/devices/virtual/iommu/dmar1/devices/0000:00:01.0
/sys/devices/virtual/iommu/dmar1/devices/0000:00:16.0
/sys/devices/virtual/iommu/dmar1/devices/0000:00:1a.0
/sys/devices/virtual/iommu/dmar1/devices/0000:00:1b.0
/sys/devices/virtual/iommu/dmar1/devices/0000:00:1c.0
...
/sys/devices/virtual/iommu/dmar1/intel-iommu
/sys/devices/virtual/iommu/dmar1/intel-iommu/cap
/sys/devices/virtual/iommu/dmar1/intel-iommu/ecap
/sys/devices/virtual/iommu/dmar1/intel-iommu/address
/sys/devices/virtual/iommu/dmar1/intel-iommu/version
/sys/class/iommu/dmar0
/sys/class/iommu/dmar1
(devices also link back to the dmar units)
This makes address, version, capabilities, and extended capabilities
available, just like printed on boot. I've tried not to duplicate
data that can be found in the DMAR table, with the exception of the
address, which provides an easy way to associate the sysfs device with
a DRHD entry in the DMAR. It's tempting to add scopes and RMRR data
here, but the full DMAR table is already exposed under /sys/firmware/
and therefore already provides a way for userspace to learn such
details.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
IOMMUs currently have no common representation to userspace, most
seem to have no representation at all aside from a few printks
on bootup. There are however features of IOMMUs that are useful
to know about. For instance the IOMMU might support superpages,
making use of processor large/huge pages more important in a device
assignment scenario. It's also useful to create cross links between
devices and IOMMU hardware units, so that users might be able to
load balance their devices to avoid thrashing a single hardware unit.
This patch adds a device create and destroy interface as well as
device linking, making it very lightweight for an IOMMU driver to add
basic support. IOMMU drivers can provide additional attributes
automatically by using an attribute_group.
The attributes exposed are expected to be relatively device specific,
the means to retrieve them certainly are, so there are currently no
common attributes for the new class created here.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The single helper here no longer has any users.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Drop custom code and use IOMMU provided grouping support for PCI.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Varun Sethi <varun.sethi@freescale.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
VT-d code currently makes use of pci_find_upstream_pcie_bridge() in
order to find the topology based alias of a device. This function has
a few problems. First, it doesn't check the entire alias path of the
device to the root bus, therefore if a PCIe device is masked upstream,
the wrong result is produced. Also, it's known to get confused and
give up when it crosses a bridge from a conventional PCI bus to a PCIe
bus that lacks a PCIe capability. The PCI-core provided DMA alias
support solves both of these problems and additionally adds support
for DMA function quirks allowing VT-d to work with devices like
Marvell and Ricoh with known broken requester IDs.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The IOMMU code now provides a common interface for finding or
creating an IOMMU group for a device on PCI buses. Make use of it
and remove piles of code.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: David Woodhouse <dwmw2@infradead.org>
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>
Currently each IOMMU driver that supports IOMMU groups has its own
code for discovering the base device used in grouping. This code
is generally not specific to the IOMMU hardware, but to the bus of
the devices managed by the IOMMU. We can therefore create a common
interface for supporting devices on different buses.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Commit e79df31 introduced mmu_notifer_count to protect
against parallel mmu_notifier_invalidate_range_start/end
calls. The patch left a small race condition when
invalidate_range_end() races with a new
invalidate_range_start() the empty page-table may be
reverted leading to stale TLB entries in the IOMMU and the
device. Use a spin_lock instead of just an atomic variable
to eliminate the race.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Function dmar_iommu_notify_scope_dev() makes a wrong assumption that
there's one RMRR for each PCI device at most, which causes DMA failure
on some HP platforms. So enhance dmar_iommu_notify_scope_dev() to
handle multiple RMRRs for the same PCI device.
Fixbug: https://bugzilla.novell.com/show_bug.cgi?id=879482
Cc: <stable@vger.kernel.org> # 3.15
Reported-by: Tom Mingarelli <thomas.mingarelli@hp.com>
Tested-by: Linda Knippers <linda.knippers@hp.com>
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The changes include:
* A new IOMMU driver for ARM Renesas SOCs
* Updates and fixes for the ARM Exynos driver to bring it closer
to a usable state again
* Convert the AMD IOMMUv2 driver to use the
mmu_notifier->release call-back instead of the task_exit
notifier
* Random other fixes and minor improvements to a number of other
IOMMU drivers
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=1DJL
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v3.16' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu into next
Pull IOMMU updates from Joerg Roedel:
"The changes include:
- a new IOMMU driver for ARM Renesas SOCs
- updates and fixes for the ARM Exynos driver to bring it closer to a
usable state again
- convert the AMD IOMMUv2 driver to use the mmu_notifier->release
call-back instead of the task_exit notifier
- random other fixes and minor improvements to a number of other
IOMMU drivers"
* tag 'iommu-updates-v3.16' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (54 commits)
iommu/msm: Use devm_ioremap_resource to simplify code
iommu/amd: Fix recently introduced compile warnings
arm/ipmmu-vmsa: Fix compile error
iommu/exynos: Fix checkpatch warning
iommu/exynos: Fix trivial typo
iommu/exynos: Remove invalid symbol dependency
iommu: fsl_pamu.c: Fix for possible null pointer dereference
iommu/amd: Remove duplicate checking code
iommu/amd: Handle parallel invalidate_range_start/end calls correctly
iommu/amd: Remove IOMMUv2 pasid_state_list
iommu/amd: Implement mmu_notifier_release call-back
iommu/amd: Convert IOMMUv2 state_table into state_list
iommu/amd: Don't access IOMMUv2 state_table directly
iommu/ipmmu-vmsa: Support clearing mappings
iommu/ipmmu-vmsa: Remove stage 2 PTE bits definitions
iommu/ipmmu-vmsa: Support 2MB mappings
iommu/ipmmu-vmsa: Rewrite page table management
iommu/ipmmu-vmsa: PMD is never folded, PUD always is
iommu/ipmmu-vmsa: Set the PTE contiguous hint bit when possible
iommu/ipmmu-vmsa: Define driver-specific page directory sizes
...
Merge misc updates from Andrew Morton:
- a few fixes for 3.16. Cc'ed to stable so they'll get there somehow.
- various misc fixes and cleanups
- most of the ocfs2 queue. Review is slow...
- most of MM. The MM queue is pretty huge this time, but not much in
the way of feature work.
- some tweaks under kernel/
- printk maintenance work
- updates to lib/
- checkpatch updates
- tweaks to init/
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (276 commits)
fs/autofs4/dev-ioctl.c: add __init to autofs_dev_ioctl_init
fs/ncpfs/getopt.c: replace simple_strtoul by kstrtoul
init/main.c: remove an ifdef
kthreads: kill CLONE_KERNEL, change kernel_thread(kernel_init) to avoid CLONE_SIGHAND
init/main.c: add initcall_blacklist kernel parameter
init/main.c: don't use pr_debug()
fs/binfmt_flat.c: make old_reloc() static
fs/binfmt_elf.c: fix bool assignements
fs/efs: convert printk(KERN_DEBUG to pr_debug
fs/efs: add pr_fmt / use __func__
fs/efs: convert printk to pr_foo()
scripts/checkpatch.pl: device_initcall is not the only __initcall substitute
checkpatch: check stable email address
checkpatch: warn on unnecessary void function return statements
checkpatch: prefer kstrto<foo> to sscanf(buf, "%<lhuidx>", &bar);
checkpatch: add warning for kmalloc/kzalloc with multiply
checkpatch: warn on #defines ending in semicolon
checkpatch: make --strict a default for files in drivers/net and net/
checkpatch: always warn on missing blank line after variable declaration block
checkpatch: fix wildcard DT compatible string checking
...
This adds support for the DMA Contiguous Memory Allocator for
intel-iommu. This change enables dma_alloc_coherent() to allocate big
contiguous memory.
It is achieved in the same way as nommu_dma_ops currently does, i.e.
trying to allocate memory by dma_alloc_from_contiguous() and
alloc_pages() is used as a fallback.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Don Dutile <ddutile@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull core irq updates from Thomas Gleixner:
"The irq department delivers:
- Another tree wide update to get rid of the horrible create_irq
interface along with its even more horrible variants. That also
gets rid of the last leftovers of the initial sparse irq hackery.
arch/driver specific changes have been either acked or ignored.
- A fix for the spurious interrupt detection logic with threaded
interrupts.
- A new ARM SoC interrupt controller
- The usual pile of fixes and improvements all over the place"
* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (40 commits)
Documentation: brcmstb-l2: Add Broadcom STB Level-2 interrupt controller binding
irqchip: brcmstb-l2: Add Broadcom Set Top Box Level-2 interrupt controller
genirq: Improve documentation to match current implementation
ARM: iop13xx: fix msi support with sparse IRQ
genirq: Provide !SMP stub for irq_set_affinity_notifier()
irqchip: armada-370-xp: Move the devicetree binding documentation
irqchip: gic: Use mask field in GICC_IAR
genirq: Remove dynamic_irq mess
ia64: Use irq_init_desc
genirq: Replace dynamic_irq_init/cleanup
genirq: Remove irq_reserve_irq[s]
genirq: Replace reserve_irqs in core code
s390: Avoid call to irq_reserve_irqs()
s390: Remove pointless arch_show_interrupts()
s390: pci: Check return value of alloc_irq_desc() proper
sh: intc: Remove pointless irq_reserve_irqs() invocation
x86, irq: Remove pointless irq_reserve_irqs() call
genirq: Make create/destroy_irq() ia64 private
tile: Use SPARSE_IRQ
tile: pci: Use irq_alloc/free_hwirq()
...
Here is the "big" pull request for 3.16-rc1.
Not a lot of changes here, some kernfs work, a revert of a very old
driver core change that ended up cauing some memory leaks on driver
probe error paths, and other minor things.
As was pointed out earlier today, one commit here,
26fc9cd200 (kernfs: move the last
knowledge of sysfs out from kernfs) is also needed in your 3.15-final
branch as well. If you could cherry-pick it there, it would be most
appreciated by Andy Lutomirski to prevent a regression there.
All of these have been in linux-next for a while.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iEYEABECAAYFAlONV9YACgkQMUfUDdst+yn0sQCfWWYg1oVXyu6f0uJjYbVBFkpD
UHgAoJxxfwTZJq/xYrnk6+RqUowIsUlh
=ojAS
-----END PGP SIGNATURE-----
Merge tag 'driver-core-3.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core into next
Pull driver core / kernfs changes from Greg KH:
"Here is the "big" pull request for 3.16-rc1.
Not a lot of changes here, some kernfs work, a revert of a very old
driver core change that ended up cauing some memory leaks on driver
probe error paths, and other minor things.
As was pointed out earlier today, one commit here, 26fc9cd200
("kernfs: move the last knowledge of sysfs out from kernfs") is also
needed in your 3.15-final branch as well. If you could cherry-pick it
there, it would be most appreciated by Andy Lutomirski to prevent a
regression there.
All of these have been in linux-next for a while"
* tag 'driver-core-3.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
crypto/nx/nx-842: dev_set_drvdata can no longer fail
kernfs: move the last knowledge of sysfs out from kernfs
sysfs: fix attribute_group bin file path on removal
sysfs.h: don't return a void-valued expression in sysfs_remove_file
init.h: Update initcall_sync variants to fix build errors
driver core: Inline dev_set/get_drvdata
driver core: dev_get_drvdata: Don't check for NULL dev
driver core: dev_set_drvdata returns void
driver core: dev_set_drvdata can no longer fail
driver core: Move driver_data back to struct device
lib/devres.c: fix checkpatch warnings
lib/devres.c: use dev in devm_request_and_ioremap
kobject: Make support for uevent_helper optional.
kernfs: make kernfs_notify() trigger inotify events too
kernfs: implement kernfs_root->supers list
Use devm_ioremap_resource() to make the code simpler, drop unused variable,
redundant return value check, and error-handing code.
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
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>
So there is no point in checking its return value, which will soon
disappear.
Signed-off-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function arm_iommu_create_mapping lost the order
parameter. Remove it from this IOMMU driver too to make it
build.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Silences the following type of warnings:
WARNING: Missing a blank line after declarations
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
EXYNOS_DEV_SYSMMU symbol is not defined anywhere and prevents building
the Exynos driver. Remove it.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
There is otherwise a risk of a possible null pointer dereference.
Was largely found by using a static code analysis program called cppcheck.
Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Reviewed-by: Bharat Bhushan <bharat.bhushan@freescale.com>
Acked-by: Varun Sethi <Varun.Sethi@freescale.com>
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>
Add a counter to the pasid_state so that we do not restore
the original page-table before all invalidate_range_start
to invalidate_range_end sections have finished.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This list was only used for the task_exit notifier function.
Now that it is gone we can remove it.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Tested-by: Jay Cornwall <Jay.Cornwall@amd.com>
Since mmu_notifier call-backs can sleep (because they use
SRCU now) we can use them to tear down PASID mappings. This
allows us to finally remove the hack to use the task_exit
notifier from oprofile to get notified when a process dies.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Tested-by: Jay Cornwall <Jay.Cornwall@amd.com>
The state_table consumes 512kb of memory and is only sparsly
populated. Convert it into a list to save memory. There
should be no measurable performance impact.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Tested-by: Jay Cornwall <Jay.Cornwall@amd.com>
This is a preparation for converting the state_table into a
state_list.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Tested-by: Jay Cornwall <Jay.Cornwall@amd.com>
Add support for 2MB block mappings at the PMD level.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The IOMMU core will only call us with page sizes advertized as supported
by the driver. We can thus simplify the code by removing loops over PGD
and PMD entries.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The driver only supports the 3-level long descriptor format that has no
PUD and always has a PMD.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The contiguous hint bit signals to the IOMMU that a range of 16 PTEs
refer to physically contiguous memory. It improves performances by
dividing the number of TLB lookups by 16, effectively implementing 64kB
page sizes.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The PTRS_PER_(PUD|PGD|PMD|PTE) macros evaluate to different values
depending on whether LPAE is enabled. The IPMMU driver uses a long
descriptor format regardless of LPAE, making those macros mismatch the
IPMMU configuration on non-LPAE systems.
Replace the macros by driver-specific versions that always evaluate to
the right value.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Cache the micro-TLB number in archdata allocated in the .add_device
handler instead of looking it up when the deviced is attached and
detached. This simplifies the .attach_dev and .detach_dev operations and
prepares for DT support.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
ia64 and x86 share this driver. x86 is moving to a different irq
allocation and ia64 keeps its private irq_create/destroy stuff.
Use macros to redirect to one or the other. Yes, macros to avoid
include hell.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Grant Likely <grant.likely@linaro.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Acked-by: Joerg Roedel <joro@8bytes.org>
Cc: x86@kernel.org
Cc: linux-ia64@vger.kernel.org
Cc: iommu@lists.linux-foundation.org
Link: http://lkml.kernel.org/r/20140507154336.372289825@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The create_irq variants are going away. Use the new interface. The
core and arch code already excludes the gsi interrupts from the
allocation, so no functional change.
This does not replace the requirement to move x86 to irq domains, but
it limits the mess to some degree.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Grant Likely <grant.likely@linaro.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: Joerg Roedel <joro@8bytes.org>
Cc: x86@kernel.org
Cc: iommu@lists.linux-foundation.org
Link: http://lkml.kernel.org/r/20140507154334.741805075@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch contains 2 workaround for the System MMU v3.x.
System MMU v3.2 and v3.3 has FLPD cache that caches first level page
table entries to reduce page table walking latency. However, the
FLPD cache is filled with a first level page table entry even though
it is not accessed by a master H/W because System MMU v3.3
speculatively prefetches page table entries that may be accessed
in the near future by the master H/W.
The prefetched FLPD cache entries are not invalidated by iommu_unmap()
because iommu_unmap() only unmaps and invalidates the page table
entries that is mapped.
Because exynos-iommu driver discards a second level page table when
it needs to be replaced with another second level page table or
a first level page table entry with 1MB mapping, It is required to
invalidate FLPD cache that may contain the first level page table
entry that points to the second level page table.
Another workaround of System MMU v3.3 is initializing the first level
page table entries with the second level page table which is filled
with all zeros. This prevents System MMU prefetches 'fault' first
level page table entry which may lead page fault on access to 16MiB
wide.
System MMU 3.x fetches consecutive page table entries by a page
table walking to maximize bus utilization and to minimize TLB miss
panelty.
Unfortunately, functional problem is raised with the fetching behavior
because it fetches 'fault' page table entries that specifies no
translation information and that a valid translation information will
be written to in the near future. The logic in the System MMU generates
page fault with the cached fault entries that is no longer coherent
with the page table which is updated.
There is another workaround that must be implemented by I/O virtual
memory manager: any two consecutive I/O virtual memory area must have
a hole between the two that is larger than or equal to 128KiB.
Also, next I/O virtual memory area must be started from the next
128KiB boundary.
0 128K 256K 384K 512K
|-------------|---------------|-----------------|----------------|
|area1---------------->|.........hole...........|<--- area2 -----
The constraint is depicted above.
The size is selected by the calculation followed:
- System MMU can fetch consecutive 64 page table entries at once
64 * 4KiB = 256KiB. This is the size between 128K ~ 384K of the
above picture. This style of fetching is 'block fetch'. It fetches
the page table entries predefined consecutive page table entries
including the entry that is the reason of the page table walking.
- System MMU can prefetch upto consecutive 32 page table entries.
This is the size between 256K ~ 384K.
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
Signed-off-by: Shaik Ameer Basha <shaik.ameer@samsung.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This turns on FLPD_CACHE, ACGEN and SYSSEL.
FLPD_CACHE is a cache of 1st level page table entries that contains
the address of a 2nd level page table to reduce latency of page table
walking.
ACGEN is architectural clock gating that gates clocks by System MMU
itself if it is not active. Note that ACGEN is different from clock
gating by the CPU. ACGEN just gates clocks to the internal logic of
System MMU while clock gating by the CPU gates clocks to the System
MMU.
SYSSEL selects System MMU version in some Exynos SoCs. Some Exynos
SoCs have an option to select System MMU versions exclusively because
the SoCs adopts new System MMU version experimentally.
This also always selects LRU as TLB replacement policy. Selecting TLB
replacement policy is deprecated from System MMU 3.2. TLB in System
MMU 3.3 has single TLB replacement policy, LRU. The bit of MMU_CFG
selecting TLB replacement policy is remained as reserved.
QoS value of page table walking is set to 15 (highst value). System
MMU 3.3 can inherit QoS value of page table walking from its master
H/W's transaction. This new feature is enabled by default and QoS
value written to MMU_CFG is ignored.
This patch also adds simplifies the sysmmu version checking by
introducing some macros.
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
Signed-off-by: Shaik Ameer Basha <shaik.ameer@samsung.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This commit adds device tree support for System MMU.
Also, system mmu handling is improved. Previously, an IOMMU domain is
bound to a System MMU which is not correct. This patch binds an IOMMU
domain with the master device of a System MMU.
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
Signed-off-by: Shaik Ameer Basha <shaik.ameer@samsung.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Alex Williamson