linux_dsm_epyc7002/Documentation
Alexey Kardashevskiy 121f80ba68 KVM: PPC: VFIO: Add in-kernel acceleration for VFIO
This allows the host kernel to handle H_PUT_TCE, H_PUT_TCE_INDIRECT
and H_STUFF_TCE requests targeted an IOMMU TCE table used for VFIO
without passing them to user space which saves time on switching
to user space and back.

This adds H_PUT_TCE/H_PUT_TCE_INDIRECT/H_STUFF_TCE handlers to KVM.
KVM tries to handle a TCE request in the real mode, if failed
it passes the request to the virtual mode to complete the operation.
If it a virtual mode handler fails, the request is passed to
the user space; this is not expected to happen though.

To avoid dealing with page use counters (which is tricky in real mode),
this only accelerates SPAPR TCE IOMMU v2 clients which are required
to pre-register the userspace memory. The very first TCE request will
be handled in the VFIO SPAPR TCE driver anyway as the userspace view
of the TCE table (iommu_table::it_userspace) is not allocated till
the very first mapping happens and we cannot call vmalloc in real mode.

If we fail to update a hardware IOMMU table unexpected reason, we just
clear it and move on as there is nothing really we can do about it -
for example, if we hot plug a VFIO device to a guest, existing TCE tables
will be mirrored automatically to the hardware and there is no interface
to report to the guest about possible failures.

This adds new attribute - KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE - to
the VFIO KVM device. It takes a VFIO group fd and SPAPR TCE table fd
and associates a physical IOMMU table with the SPAPR TCE table (which
is a guest view of the hardware IOMMU table). The iommu_table object
is cached and referenced so we do not have to look up for it in real mode.

This does not implement the UNSET counterpart as there is no use for it -
once the acceleration is enabled, the existing userspace won't
disable it unless a VFIO container is destroyed; this adds necessary
cleanup to the KVM_DEV_VFIO_GROUP_DEL handler.

This advertises the new KVM_CAP_SPAPR_TCE_VFIO capability to the user
space.

This adds real mode version of WARN_ON_ONCE() as the generic version
causes problems with rcu_sched. Since we testing what vmalloc_to_phys()
returns in the code, this also adds a check for already existing
vmalloc_to_phys() call in kvmppc_rm_h_put_tce_indirect().

This finally makes use of vfio_external_user_iommu_id() which was
introduced quite some time ago and was considered for removal.

Tests show that this patch increases transmission speed from 220MB/s
to 750..1020MB/s on 10Gb network (Chelsea CXGB3 10Gb ethernet card).

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
2017-04-20 11:39:26 +10:00
..
ABI RTC for 4.11 2017-02-27 19:59:21 -08:00
accounting tools: move accounting tool from Documentation 2016-09-23 13:07:15 -06:00
acpi scripts/spelling.txt: add "followings" pattern and fix typo instances 2017-02-27 18:43:47 -08:00
admin-guide Power management updates for v4.11-rc2 2017-03-09 16:30:37 -08:00
aoe
arm arm: sunxi: add support for V3s SoC 2017-01-20 21:31:34 +01:00
arm64 irqchip/gicv3-its: Add workaround for QDF2400 ITS erratum 0065 2017-03-07 14:34:27 +00:00
auxdisplay samples: move auxdisplay example code from Documentation 2016-09-23 11:52:32 -06:00
backlight
blackfin samples: move blackfin gptimers-example from Documentation 2016-10-10 07:12:02 -06:00
block A slightly quieter cycle for documentation this time around. 2017-02-22 18:51:29 -08:00
blockdev scripts/spelling.txt: add "followings" pattern and fix typo instances 2017-02-27 18:43:47 -08:00
bus-devices
cdrom cdrom: Make device operations read-only 2017-02-14 08:29:56 -07:00
cgroup-v1 Merge branch 'for-4.11' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup 2017-02-27 21:41:08 -08:00
cma
connector samples: connector: from Documentation to samples directory 2016-04-28 07:47:35 -06:00
console
core-api Documentation: Update CPU hotplug and move it to core-api 2017-01-13 10:32:32 -07:00
cpu-freq A slightly quieter cycle for documentation this time around. 2017-02-22 18:51:29 -08:00
cpuidle
cris
crypto crypto: doc - fix typo 2017-02-15 13:23:49 +08:00
dev-tools scripts/spelling.txt: add "disble(d)" pattern and fix typo instances 2017-03-09 17:01:09 -08:00
device-mapper scripts/spelling.txt: add "explictely" pattern and fix typo instances 2017-02-27 18:43:47 -08:00
devicetree Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-03-14 21:31:23 -07:00
dmaengine dmaengine: Documentation: Fix typo in pxa_dma.txt 2016-11-14 08:14:24 +05:30
doc-guide docs-rst: parse-headers.pl: cleanup the documentation 2016-11-30 17:08:09 -07:00
DocBook A few fixes for the docs tree, including one for a 4.11 build regression. 2017-03-04 11:32:18 -08:00
driver-api Less anger inducing pull request for 4.11 2017-02-23 18:58:18 -08:00
driver-model irqdesc: Add a resource managed version of irq_alloc_descs() 2017-02-10 14:39:20 +01:00
early-userspace
EDID
extcon extcon: int3496: Add Intel INT3496 ACPI device extcon driver 2017-01-09 10:04:11 +09:00
fault-injection
fb Documentation: fb: fix spelling mistakes 2016-05-10 12:05:27 +03:00
features 2nd round of ARC udpates for 4.10rc1 2016-12-23 10:22:47 -08:00
filesystems statx: Add a system call to make enhanced file info available 2017-03-02 20:51:15 -05:00
firmware_class firmware: revamp firmware documentation 2017-01-11 09:42:59 +01:00
fmc
fpga fpga: Add scatterlist based programming 2017-02-10 15:20:44 +01:00
frv docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
gpio gpio: random documentation update 2017-01-31 15:43:05 +01:00
gpu Less anger inducing pull request for 4.11 2017-02-23 18:58:18 -08:00
hid Documentation: HID: Intel ISH HID document 2016-08-17 11:13:07 +02:00
hwmon A slightly quieter cycle for documentation this time around. 2017-02-22 18:51:29 -08:00
i2c i2c: i801: Add support for Intel Gemini Lake 2017-02-09 17:39:16 +01:00
ia64 selftests: move ia64 tests from Documentation/ia64 2016-09-20 09:58:12 -06:00
ide
iio iio: Documentation: Correct the path used to create triggers. 2016-10-01 00:49:58 -06:00
infiniband IB/hfi1: Document new sysfs entries for hfi1 driver 2016-10-02 08:42:19 -04:00
input Documentation: input: fix path to input code definitions 2017-02-12 15:19:00 -07:00
ioctl rpmsg updates for v4.11 2017-02-23 09:41:03 -08:00
isdn docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
kbuild Kconfig: Introduce the "imply" keyword 2016-11-16 09:26:33 +01:00
kdump Documentation: kdump: Add description of enable multi-cpus support 2016-09-20 18:02:54 -06:00
laptops platform/x86: thinkpad_acpi: Add support for X1 Yoga (2016) Tablet Mode 2016-12-13 09:29:06 -08:00
leds leds: class: Add new optional brightness_hw_changed attribute 2017-01-29 19:59:42 +01:00
livepatch A slightly quieter cycle for documentation this time around. 2017-02-22 18:51:29 -08:00
locking locking/ww_mutex/Documentation: Update the design document 2017-01-14 11:14:55 +01:00
m68k docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
md MD: add doc for raid5-cache 2017-02-13 09:17:54 -08:00
media A few fixes for the docs tree, including one for a 4.11 build regression. 2017-03-04 11:32:18 -08:00
memory-devices
metag
mic samples: move mic/mpssd example code from Documentation 2016-09-20 12:38:48 -06:00
mips
misc-devices samples: move misc-devices/mei example code from Documentation 2016-09-23 11:51:43 -06:00
mmc mmc: core: Extend sysfs with DSR register 2016-07-25 10:34:51 +02:00
mn10300
mtd spi-nor: Add support for Intel SPI serial flash controller 2017-01-03 17:33:36 +00:00
namespaces
netlabel
networking Make IP 'forwarding' doc more precise 2017-03-12 23:28:43 -07:00
nfc
nios2
nvdimm libnvdimm, btt: update the usage section in Documentation 2016-06-17 16:23:23 -07:00
nvmem
parisc
PCI A few fixes for the docs tree, including one for a 4.11 build regression. 2017-03-04 11:32:18 -08:00
pcmcia tools: move pcmcia crc32hash tool from Documentation 2016-09-23 13:07:27 -06:00
perf perf: add qcom l2 cache perf events driver 2017-02-08 19:32:24 +00:00
phy
platform
power More power management updates for v4.11-rc1 2017-03-02 17:33:52 -08:00
powerpc powerpc updates for 4.9 2016-10-07 20:19:31 -07:00
pps Doc: clarify source of jitter in USB1.1, and USB2.0 2017-01-04 14:40:52 -07:00
prctl selftests: move prctl tests from Documentation/prctl 2016-09-20 09:09:09 -06:00
process Doc: Correct typo, "Introdution" => "Introduction" 2016-12-01 10:44:08 -07:00
pti
ptp selftests: move ptp tests from Documentation/ptp 2016-09-20 09:54:38 -06:00
rapidio rapidio/documentation/mport_cdev: add missing parameter description 2016-09-01 17:52:02 -07:00
RCU Merge branches 'doc.2017.01.15b', 'dyntick.2017.01.23a', 'fixes.2017.01.23a', 'srcu.2017.01.25a' and 'torture.2017.01.15b' into HEAD 2017-01-25 12:56:05 -08:00
s390 Documentation: Update path to sysrq.txt 2017-03-03 15:48:38 -07:00
scheduler sched/Documentation/sched-rt-group: Fix incorrect example 2017-01-22 10:34:17 +01:00
scsi scripts/spelling.txt: add "varible" pattern and fix typo instances 2017-02-27 18:43:47 -08:00
security KEYS: Differentiate uses of rcu_dereference_key() and user_key_payload() 2017-03-02 10:09:00 +11:00
serial Documentation: rs485: Do not define manually the ioctl 2016-08-18 11:08:33 -06:00
sh
sound scripts/spelling.txt: add "followings" pattern and fix typo instances 2017-02-27 18:43:47 -08:00
sparc Documentation/sparc: Steps for sending break on sunhv console 2017-02-23 08:27:25 -08:00
sphinx docs: sphinx-extensions: make rstFlatTable work with docutils 0.13 2016-12-18 13:30:29 -07:00
sphinx-static This is the documentation update pull for the 4.9 merge window. 2016-10-04 13:54:07 -07:00
spi spi: spi-ep93xx: simplify GPIO chip selects 2017-02-16 20:10:26 +00:00
sysctl A few fixes for the docs tree, including one for a 4.11 build regression. 2017-03-04 11:32:18 -08:00
target target: make close_session optional 2016-05-10 01:19:26 -07:00
thermal Documentation: fix spelling mistakes of "Celcius" -- > "Celsius" 2017-01-04 14:36:17 -07:00
timers time: Remove CONFIG_TIMER_STATS 2017-02-10 11:15:08 +01:00
trace perf/core: Rename CONFIG_[UK]PROBE_EVENT to CONFIG_[UK]PROBE_EVENTS 2017-03-01 10:26:39 +01:00
translations doc/ko_KR/memory-barriers: Update control-dependencies section 2017-03-03 15:54:55 -07:00
usb A slightly quieter cycle for documentation this time around. 2017-02-22 18:51:29 -08:00
virtual KVM: PPC: VFIO: Add in-kernel acceleration for VFIO 2017-04-20 11:39:26 +10:00
vm userfaultfd: non-cooperative: rollback userfaultfd_exit 2017-03-09 17:01:09 -08:00
w1 w1: add ability to set (SRAM) and store (EEPROM) configuration for temp sensors like DS18B20 2016-05-01 14:37:49 -07:00
watchdog watchdog: Introduce watchdog_stop_on_unregister helper 2017-02-24 14:00:23 -08:00
wimax
x86 Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2017-02-28 11:46:00 -08:00
xtensa xtensa: cleanup MMU setup and kernel layout macros 2016-07-24 06:33:58 +03:00
.gitignore Add .pyc files to .gitignore 2016-06-30 13:07:33 -06:00
00-INDEX Documentation: move MD related doc into a separate dir 2017-02-13 09:17:53 -08:00
bcache.txt bcache: documentation formatting, edited for clarity, stripe alignment notes 2016-06-23 07:58:38 -06:00
bt8xxgpio.txt
btmrvl.txt
bus-virt-phys-mapping.txt
cachetlb.txt
cgroup-v2.txt cgroup: Fix indenting in PID controller documentation 2017-03-06 14:46:27 -05:00
Changes docs: add back 'Documentation/Changes' file (as symlink) 2016-12-14 16:30:12 -08:00
circular-buffers.txt Documentation: circular-buffers: use READ_ONCE() 2016-11-16 16:17:45 -07:00
clk.txt Documentation: clk: update file names containing referenced structures 2016-08-14 12:12:36 -06:00
CodingStyle doc: re-add CodingStyle and SubmittingPatches 2016-10-24 08:12:35 -02:00
conf.py Documentation/sphinx: fix primary_domain configuration 2017-03-03 16:12:30 -07:00
cpu-load.txt
cputopology.txt topology/sysfs: provide drawer id and siblings attributes 2016-06-13 15:58:27 +02:00
crc32.txt
dcdbas.txt
debugging-modules.txt
debugging-via-ohci1394.txt
dell_rbu.txt
digsig.txt
DMA-API-HOWTO.txt Documentation: DMA-API-HOWTO: Fix a typo 2016-09-20 17:58:46 -06:00
DMA-API.txt dma-mapping: add dma_{map,unmap}_resource 2016-09-26 22:16:41 +05:30
DMA-attributes.txt common: DMA-mapping: add DMA_ATTR_PRIVILEGED attribute 2017-01-19 15:56:19 +00:00
DMA-ISA-LPC.txt Documentation: DMA-ISA-LPC.txt 2017-02-12 15:20:07 -07:00
docutils.conf doc-rst: add docutils config file 2016-08-14 11:52:40 -06:00
dontdiff Documentation: dontdiff: Update with additional entries 2017-01-26 15:08:06 -07:00
efi-stub.txt
eisa.txt
flexible-arrays.txt
futex-requeue-pi.txt
gcc-plugins.txt GCC plugin infrastructure 2016-06-07 22:57:10 +02:00
highuid.txt
hw_random.txt
hwspinlock.txt
index.rst docs/zh_CN: Add coding-style into docs build system 2017-01-26 15:30:34 -07:00
intel_txt.txt
Intel-IOMMU.txt
io_ordering.txt
io-mapping.txt
iostats.txt
IPMI.txt Documentation: Fix a typo in IPMI.txt. 2017-01-05 15:01:54 -06:00
IRQ-affinity.txt
IRQ-domain.txt Documentation/IRQ-domain.txt: Document irq_domain_create_{linear, tree} 2016-03-31 00:32:59 -06:00
IRQ.txt
irqflags-tracing.txt
isa.txt Documentation: Add ISA bus driver documentation 2016-05-02 09:32:04 -07:00
isapnp.txt
kernel-doc-nano-HOWTO.txt docs-rst: doc-guide: split the kernel-documentation.rst contents 2016-11-19 10:22:04 -07:00
kernel-per-CPU-kthreads.txt docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
kobject.txt
kprobes.txt Documentation: kprobes: Document jprobes stack copying limitations 2016-08-15 10:19:11 -06:00
kref.txt
kselftest.txt scripts/spelling.txt: add "an user" pattern and fix typo instances 2017-02-27 18:43:46 -08:00
ldm.txt
lockup-watchdogs.txt docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
logo.gif
logo.txt
lzo.txt Documentation: lzo: fix spelling mistakes 2016-04-28 07:23:11 -06:00
mailbox.txt
Makefile samples: move blackfin gptimers-example from Documentation 2016-10-10 07:12:02 -06:00
Makefile.sphinx Add a target to check broken external links in the Documentation 2017-02-15 15:22:47 -07:00
memory-barriers.txt doc: Update control-dependencies section of memory-barriers.txt 2017-01-14 21:29:15 -08:00
memory-hotplug.txt scripts/spelling.txt: add "followings" pattern and fix typo instances 2017-02-27 18:43:47 -08:00
men-chameleon-bus.txt
nommu-mmap.txt
ntb.txt
numastat.txt
padata.txt
parport-lowlevel.txt
percpu-rw-semaphore.txt
phy.txt phy: core: Allow children node to be overridden 2016-04-29 16:39:39 +02:00
pi-futex.txt
pinctrl.txt pinctrl: core: Fix regression caused by delayed work for hogs 2017-01-13 16:25:17 +01:00
pnp.txt
preempt-locking.txt
printk-formats.txt
pwm.txt pwm: Update documentation 2016-05-17 14:48:04 +02:00
rbtree.txt
remoteproc.txt remoteproc: Split driver and consumer dereferencing 2016-10-02 22:50:21 -07:00
rfkill.txt docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
robust-futex-ABI.txt
robust-futexes.txt Documentation: robust-futexes: fix spelling mistakes 2016-04-28 07:26:41 -06:00
rpmsg.txt rpmsg: use module_rpmsg_driver in existing drivers and examples 2016-05-06 11:09:01 -07:00
rtc.txt
SAK.txt
sgi-ioc4.txt
siphash.txt siphash: implement HalfSipHash1-3 for hash tables 2017-01-09 13:58:57 -05:00
SM501.txt
smsc_ece1099.txt
static-keys.txt jump_label: Reduce the size of struct static_key 2017-02-15 09:02:26 -05:00
SubmittingPatches doc: re-add CodingStyle and SubmittingPatches 2016-10-24 08:12:35 -02:00
svga.txt
sync_file.txt dma-buf: Rename struct fence to dma_fence 2016-10-25 14:40:39 +02:00
this_cpu_ops.txt
unaligned-memory-access.txt Documentation/unaligned-memory-access.txt: fix incorrect comparison operator 2016-12-27 13:08:42 -07:00
unshare.txt
vfio-mediated-device.txt vfio-mdev: Make mdev_parent private 2016-12-30 08:13:41 -07:00
vfio.txt
video-output.txt
xillybus.txt Documentation: xillybus: fix spelling mistake 2016-04-28 07:44:54 -06:00
xz.txt
zorro.txt