With the current IOMMU-API the hardware TLBs have to be
flushed in every iommu_ops->unmap() call-back.

For unmapping large amounts of address space, like it
happens when a KVM domain with assigned devices is
destroyed, this causes thousands of unnecessary TLB flushes
in the IOMMU hardware because the unmap call-back runs for
every unmapped physical page.

With the TLB Flush Interface and the new iommu_unmap_fast()
function introduced here the need to clean the hardware TLBs
is removed from the unmapping code-path. Users of
iommu_unmap_fast() have to explicitly call the TLB-Flush
functions to sync the page-table changes to the hardware.

Three functions for TLB-Flushes are introduced:

	* iommu_flush_tlb_all() - Flushes all TLB entries
	                          associated with that
				  domain. TLBs entries are
				  flushed when this function
				  returns.

	* iommu_tlb_range_add() - This will add a given
				  range to the flush queue
				  for this domain.

	* iommu_tlb_sync() - Flushes all queued ranges from
			     the hardware TLBs. Returns when
			     the flush is finished.

The semantic of this interface is intentionally similar to
the iommu_gather_ops from the io-pgtable code.

Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Rename a few iommu cache-flush functions that start with
iommu_ to start with amd_iommu now. This is to prevent name
collisions with generic iommu code later on.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

The recently-removed FIXME in iommu_get_domain_for_dev() turns out to
have been a little misleading, since that check is still worthwhile even
when groups *are* universal. We have a few IOMMU-aware drivers which
only care whether their device is already attached to an existing domain
or not, for which the previous behaviour of iommu_get_domain_for_dev()
was ideal, and who now crash if their device does not have an IOMMU.

With IOMMU groups now serving as a reliable indicator of whether a
device has an IOMMU or not (barring false-positives from VFIO no-IOMMU
mode), drivers could arguably do this:

	group = iommu_group_get(dev);
	if (group) {
		domain = iommu_get_domain_for_dev(dev);
		iommu_group_put(group);
	}

However, rather than duplicate that code across multiple callsites,
particularly when it's still only the domain they care about, let's skip
straight to the next step and factor out the check into the common place
it applies - in iommu_get_domain_for_dev() itself. Sure, it ends up
looking rather familiar, but now it's backed by the reasoning of having
a robust API able to do the expected thing for all devices regardless.

Fixes: 05f80300 ("iommu: Finish making iommu_group support mandatory")
Reported-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Remove the deferred flushing implementation in the Intel
VT-d driver and use the one from the common iova code
instead.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

The shift qi_flush_dev_iotlb() is done on an int, which
limits the mask to 32 bits. Make the mask 64 bits wide so
that more than 4GB of address range can be flushed at once.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Rip out the implementation in the AMD IOMMU driver and use
the one in the common iova code instead.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Add a timer to flush entries from the Flush-Queues every
10ms. This makes sure that no stale TLB entries remain for
too long after an IOVA has been unmapped.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

The lock is taken from the same CPU most of the time. But
having it allows to flush the queue also from another CPU if
necessary.

This will be used by a timer to regularily flush any pending
IOVAs from the Flush-Queues.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

There are two counters:

	* fq_flush_start_cnt  - Increased when a TLB flush
	                        is started.

	* fq_flush_finish_cnt - Increased when a TLB flush
				is finished.

The fq_flush_start_cnt is assigned to every Flush-Queue
entry on its creation. When freeing entries from the
Flush-Queue, the value in the entry is compared to the
fq_flush_finish_cnt. The entry can only be freed when its
value is less than the value of fq_flush_finish_cnt.

The reason for these counters it to take advantage of IOMMU
TLB flushes that happened on other CPUs. These already
flushed the TLB for Flush-Queue entries on other CPUs so
that they can already be freed without flushing the TLB
again.

This makes it less likely that the Flush-Queue is full and
saves IOMMU TLB flushes.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Add a function to add entries to the Flush-Queue ring
buffer. If the buffer is full, call the flush-callback and
free the entries.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

This patch adds the basic data-structures to implement
flush-queues in the generic IOVA code. It also adds the
initialization and destroy routines for these data
structures.

The initialization routine is designed so that the use of
this feature is optional for the users of IOVA code.
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Sudeep reports that the logic got slightly broken when a PCI iommu-map
entry targets an IOMMU marked as disabled in DT, since of_pci_map_rid()
succeeds in following a phandle, and of_iommu_xlate() doesn't return an
error value, but we miss checking whether ops was actually non-NULL.
Whilst this could be solved with a point fix in of_pci_iommu_init(), it
suggests that all the juggling of ERR_PTR values through the ops pointer
is proving rather too complicated for its own good, so let's instead
simplify the whole flow (with a side-effect of eliminating the cause of
the bug).

The fact that we now rely on iommu_fwspec means that we no longer need
to pass around an iommu_ops pointer at all - we can simply propagate a
regular int return value until we know whether we have a viable IOMMU,
then retrieve the ops from the fwspec if and when we actually need them.
This makes everything a bit more uniform and certainly easier to follow.

Fixes: d87beb74 ("iommu/of: Handle PCI aliases properly")
Reported-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Now that all the drivers properly implementing the IOMMU API support
groups (I'm ignoring the etnaviv GPU MMUs which seemingly only do just
enough to convince the ARM DMA mapping ops), we can remove the FIXME
workarounds from the core code. In the process, it also seems logical to
make the .device_group callback non-optional for drivers calling
iommu_group_get_for_dev() - the current callers all implement it anyway,
and it doesn't make sense for any future callers not to either.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

As the last step to making groups mandatory, clean up the remaining
drivers by adding basic support. Whilst it may not perfectly reflect the
isolation capabilities of the hardware, using generic_device_group()
should at least maintain existing behaviour with respect to the API.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

As the last step to making groups mandatory, clean up the remaining
drivers by adding basic support. Whilst it may not perfectly reflect
the isolation capabilities of the hardware (tegra_smmu_swgroup sounds
suspiciously like something that might warrant representing at the
iommu_group level), using generic_device_group() should at least
maintain existing behaviour with respect to the API.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

As the last step to making groups mandatory, clean up the remaining
drivers by adding basic support. Whilst it may not perfectly reflect the
isolation capabilities of the hardware, using generic_device_group()
should at least maintain existing behaviour with respect to the API.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Now that we have a custom printf format specifier, convert users of
full_name to use %pOF instead. This is preparation to remove storing
of the full path string for each node.
Signed-off-by: Rob Herring <robh@kernel.org>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: iommu@lists.linux-foundation.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-rockchip@lists.infradead.org
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Commit 7d3002cc ("iommu/core: split mapping to page sizes as supported
by the hardware") replaced 'int gfp_order' with a 'size_t size' of
iommu_map / iommu_unmap function arguments, but missed the function
prototypes for the disabled CONFIG_IOMMU_API case, let's correct them
for consistency.
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

When a PCI device has DMA quirks, we need to ensure that an upstream
IOMMU knows about all possible aliases, since the presence of a DMA
quirk does not preclude the device still also emitting transactions
(e.g. MSIs) on its 'real' RID. Similarly, the rules for bridge aliasing
are relatively complex, and some bridges may only take ownership of
transactions under particular transient circumstances, leading again to
multiple RIDs potentially being seen at the IOMMU for the given device.

Take all this into account in the OF code by translating every RID
produced by the alias walk, not just whichever one comes out last.
Happily, this also makes things tidy enough that we can reduce the
number of both total lines of code, and confusing levels of indirection,
by pulling the "iommus"/"iommu-map" parsing helpers back in-line again.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

This adds a perl script to actually parse the MAINTAINERS file, clean up
some whitespace in it, warn about errors in it, and then properly sort
the end result.

My perl-fu is atrocious, so the script has basically been created by
randomly putting various characters in a pile, mixing them around, and
then looking it the end result does anything interesting when used as a
perl script.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Prepping for scripting the MAINTAINERS file cleanup (and possible split)
showed a couple of cases where the headers for a couple of entries were
bogus.

There's a few different kinds of bogosities:

 - the X-GENE SOC EDAC case was confused and split over two lines

 - there were four entries for "GREYBUS PROTOCOLS DRIVERS" that were all
   different things.

 - the NOKIA N900 CAMERA SUPPORT" was duplicated

all of which were more obvious when you started doing associative arrays
in perl to track these things by the header (so that we can alphabetize
this thing properly, and so that we might split it up by the data too).

Cc: Joe Perches <joe@perches.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull xen fixes from Juergen Gross:
 "Some fixes and cleanups for running under Xen"

* tag 'for-linus-4.13b-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/balloon: don't online new memory initially
  xen/x86: fix cpu hotplug
  xen/grant-table: log the lack of grants
  xen/x86: Don't BUG on CPU0 offlining

When setting up the Xenstore watch for the memory target size the new
watch will fire at once. Don't try to reach the configured target size
by onlining new memory in this case, as the current memory size will
be smaller in almost all cases due to e.g. BIOS reserved pages.

Onlining new memory will lead to more problems e.g. undesired conflicts
with NVMe devices meant to be operated as block devices.

Instead remember the difference between target size and current size
when the watch fires for the first time and apply it to any further
size changes, too.

In order to avoid races between balloon.c and xen-balloon.c init calls
do the xen-balloon.c initialization from balloon.c.
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Juergen Gross <jgross@suse.com>

Commit dc6416f1 ("xen/x86: Call
cpu_startup_entry(CPUHP_AP_ONLINE_IDLE) from xen_play_dead()")
introduced an error leading to a stack overflow of the idle task when
a cpu was brought offline/online many times: by calling
cpu_startup_entry() instead of returning at the end of xen_play_dead()
do_idle() would be entered again and again.

Don't use cpu_startup_entry(), but cpuhp_online_idle() instead allowing
to return from xen_play_dead().

Cc: <stable@vger.kernel.org> # 4.12
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Juergen Gross <jgross@suse.com>

log a message when we enter this situation:
1) we already allocated the max number of available grants from hypervisor
and
2) we still need more (but the request fails because of 1)).

Sometimes the lack of grants causes IO hangs in xen_blkfront devices.
Adding this log would help debuging.
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Juergen Gross <jgross@suse.com>

CONFIG_BOOTPARAM_HOTPLUG_CPU0 allows to offline CPU0 but Xen HVM guests
BUG() in xen_teardown_timer(). Remove the BUG_ON(), this is probably a
leftover from ancient times when CPU0 hotplug was impossible, it works
just fine for HVM.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Acked-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Juergen Gross <jgross@suse.com>

Merge tag 'hwmon-for-linus-v4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging

Pull hwmon fix from Guenter Roeck:
 "Avoid buffer overruns in applesmc driver"

* tag 'hwmon-for-linus-v4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: (applesmc) Avoid buffer overruns

Pull tty/serial fixes from Greg KH:
 "Here are some small tty and serial driver fixes for 4.13-rc2. Nothing
  huge at all, a revert of a patch that turned out to break things, a
  fix up for a new tty ioctl we added in 4.13-rc1 to get the uapi
  definition correct, and a few minor serial driver fixes for reported
  issues.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'tty-4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  tty: Fix TIOCGPTPEER ioctl definition
  tty: hide unused pty_get_peer function
  tty: serial: lpuart: Fix the logic for detecting the 32-bit type UART
  serial: imx: Prevent TX buffer PIO write when a DMA has been started
  Revert "serial: imx-serial - move DMA buffer configuration to DT"
  serial: sh-sci: Uninitialized variables in sysfs files
  serial: st-asc: Potential error pointer dereference

Pull char/misc driver fixes from Greg KH:
 "Here are some small char and misc driver fixes for 4.13-rc2. All fix
  reported problems with 4.13-rc1 or older kernels (like the binder
  fixes). Full details in the shortlog.

  All have been in linux-next with no reported issues"

* tag 'char-misc-4.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  w1: omap-hdq: fix error return code in omap_hdq_probe()
  regmap: regmap-w1: Fix build troubles
  w1: Fix slave count on 1-Wire bus (resend)
  mux: mux-core: unregister mux_class in mux_exit()
  mux: remove the Kconfig question for the subsystem
  nvmem: rockchip-efuse: amend compatible rk322x-efuse to rk3228-efuse
  drivers/fsi: fix fsi_slave_mode prototype
  fsi: core: register with postcore_initcall
  thunderbolt: Correct access permissions for active NVM contents
  vmbus: re-enable channel tasklet
  spmi: pmic-arb: Always allocate ppid_to_apid table
  MAINTAINERS: Add entry for SPMI subsystem
  spmi: Include OF based modalias in device uevent
  binder: Use wake up hint for synchronous transactions.
  binder: use group leader instead of open thread
  Revert "android: binder: Sanity check at binder ioctl"

Pull USB fixes from Greg KH:
 "Here are some small USB fixes for 4.13-rc2.

  The usual batch, gadget fixes for reported issues, as well as xhci
  fixes, and a small random collection of other fixes for reported
  issues.

  All have been in linux-next with no reported issues"

* tag 'usb-4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (25 commits)
  xhci: fix memleak in xhci_run()
  usb: xhci: fix spinlock recursion for USB2 test mode
  xhci: fix 20000ms port resume timeout
  usb: xhci: Issue stop EP command only when the EP state is running
  xhci: Bad Ethernet performance plugged in ASM1042A host
  xhci: Fix NULL pointer dereference when cleaning up streams for removed host
  usb: renesas_usbhs: gadget: disable all eps when the driver stops
  usb: renesas_usbhs: fix usbhsc_resume() for !USBHSF_RUNTIME_PWCTRL
  usb: gadget: udc: renesas_usb3: protect usb3_ep->started in usb3_start_pipen()
  usb: gadget: udc: renesas_usb3: fix zlp transfer by the dmac
  usb: gadget: udc: renesas_usb3: fix free size in renesas_usb3_dma_free_prd()
  usb: gadget: f_uac2: endianness fixes.
  usb: gadget: f_uac1: endianness fixes.
  include: usb: audio: specify exact endiannes of descriptors
  usb: gadget: udc: start_udc() can be static
  usb: dwc2: gadget: On USB RESET reset device address to zero
  usb: storage: return on error to avoid a null pointer dereference
  usb: typec: include linux/device.h in ucsi.h
  USB: cdc-acm: add device-id for quirky printer
  usb: dwc3: gadget: only unmap requests from DMA if mapped
  ...

Pull staging driver fixes from Greg KH:
 "Here are some small staging driver fixes for reported issues for
  4.13-rc2.

  Also in here is a new driver, the virtualbox DRM driver. It's
  stand-alone and got acks from the DRM developers to go in through this
  tree. It's a new thing, but it should be fine for this point in the rc
  cycle due to it being independent.

  All of this has been in linux-next for a while with no reported
  issues"

* tag 'staging-4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: rtl8188eu: add TL-WN722N v2 support
  staging: speakup: safely register and unregister ldisc
  staging: speakup: add functions to register and unregister ldisc
  staging: speakup: safely close tty
  staging: sm750fb: avoid conflicting vesafb
  staging: lustre: ko2iblnd: check copy_from_iter/copy_to_iter return code
  staging: vboxvideo: Add vboxvideo to drivers/staging
  staging: sm750fb: fixed a assignment typo
  staging: rtl8188eu: memory leak in rtw_free_cmd_obj()
  staging: vchiq_arm: fix error codes in probe
  staging: comedi: ni_mio_common: fix AO timer off-by-one regression

Fix major alphabetic errors.  No attempt to fix items that all begin
with the same word (like ARM, BROADCOM, DRM, EDAC, FREESCALE, INTEL,
OMAP, PCI, SAMSUNG, TI, USB, etc.).

(diffstat +/- is different by one line because TI KEYSTONE MULTICORE
had 2 blank lines after it.)
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull NFS client bugfixes from Anna Schumaker:
 "Stable bugfix:
   - Fix error reporting regression

  Bugfixes:
   - Fix setting filelayout ds address race
   - Fix subtle access bug when using ACLs
   - Fix setting mnt3_counts array size
   - Fix a couple of pNFS commit races"

* tag 'nfs-for-4.13-2' of git://git.linux-nfs.org/projects/anna/linux-nfs:
  NFS/filelayout: Fix racy setting of fl->dsaddr in filelayout_check_deviceid()
  NFS: Be more careful about mapping file permissions
  NFS: Store the raw NFS access mask in the inode's access cache
  NFSv3: Convert nfs3_proc_access() to use nfs_access_set_mask()
  NFS: Refactor NFS access to kernel access mask calculation
  net/sunrpc/xprt_sock: fix regression in connection error reporting.
  nfs: count correct array for mnt3_counts array size
  Revert commit 722f0b89 ("pNFS: Don't send COMMITs to the DSes if...")
  pNFS/flexfiles: Handle expired layout segments in ff_layout_initiate_commit()
  NFS: Fix another COMMIT race in pNFS
  NFS: Fix a COMMIT race in pNFS
  mount: copy the port field into the cloned nfs_server structure.
  NFS: Don't run wake_up_bit() when nobody is waiting...
  nfs: add export operations

Pull overlayfs fixes from Miklos Szeredi:
 "This fixes a crash with SELinux and several other old and new bugs"

* 'overlayfs-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
  ovl: check for bad and whiteout index on lookup
  ovl: do not cleanup directory and whiteout index entries
  ovl: fix xattr get and set with selinux
  ovl: remove unneeded check for IS_ERR()
  ovl: fix origin verification of index dir
  ovl: mark parent impure on ovl_link()
  ovl: fix random return value on mount

Pull block fixes from Jens Axboe:
 "A small set of fixes for -rc2 - two fixes for BFQ, documentation and
  code, and a removal of an unused variable in nbd. Outside of that, a
  small collection of fixes from the usual crew on the nvme side"

* 'for-linus' of git://git.kernel.dk/linux-block:
  nvmet: don't report 0-bytes in serial number
  nvmet: preserve controller serial number between reboots
  nvmet: Move serial number from controller to subsystem
  nvmet: prefix version configfs file with attr
  nvme-pci: Fix an error handling path in 'nvme_probe()'
  nvme-pci: Remove nvme_setup_prps BUG_ON
  nvme-pci: add another device ID with stripe quirk
  nvmet-fc: fix byte swapping in nvmet_fc_ls_create_association
  nvme: fix byte swapping in the streams code
  nbd: kill unused ret in recv_work
  bfq: dispatch request to prevent queue stalling after the request completion
  bfq: fix typos in comments about B-WF2Q+ algorithm

Pull more rdma fixes from Doug Ledford:
 "As per my previous pull request, there were two drivers that each had
  a rather large number of legitimate fixes still to be sent.

  As it turned out, I also missed a reasonably large set of fixes from
  one person across the stack that are all important fixes. All in all,
  the bnxt_re, i40iw, and Dan Carpenter are 3/4 to 2/3rds of this pull
  request.

  There were some other random fixes that I didn't send in the last pull
  request that I added to this one. This catches the rdma stack up to
  the fixes from up to about the beginning of this week. Any more fixes
  I'll wait and batch up later in the -rc cycle. This will give us a
  good base to start with for basing a for-next branch on -rc2.

  Summary:

   - i40iw fixes

   - bnxt_re fixes

   - Dan Carpenter bugfixes across stack

   - ten more random fixes, no more than two from any one person"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (37 commits)
  RDMA/core: Initialize port_num in qp_attr
  RDMA/uverbs: Fix the check for port number
  IB/cma: Fix reference count leak when no ipv4 addresses are set
  RDMA/iser: don't send an rkey if all data is written as immadiate-data
  rxe: fix broken receive queue draining
  RDMA/qedr: Prevent memory overrun in verbs' user responses
  iw_cxgb4: don't use WR keys/addrs for 0 byte reads
  IB/mlx4: Fix CM REQ retries in paravirt mode
  IB/rdmavt: Setting of QP timeout can overflow jiffies computation
  IB/core: Fix sparse warnings
  RDMA/bnxt_re: Fix the value reported for local ack delay
  RDMA/bnxt_re: Report MISSED_EVENTS in req_notify_cq
  RDMA/bnxt_re: Fix return value of poll routine
  RDMA/bnxt_re: Enable atomics only if host bios supports
  RDMA/bnxt_re: Specify RDMA component when allocating stats context
  RDMA/bnxt_re: Fixed the max_rd_atomic support for initiator and destination QP
  RDMA/bnxt_re: Report supported value to IB stack in query_device
  RDMA/bnxt_re: Do not free the ctx_tbl entry if delete GID fails
  RDMA/bnxt_re: Fix WQE Size posted to HW to prevent it from throwing error
  RDMA/bnxt_re: Free doorbell page index (DPI) during dealloc ucontext
  ...

Pull drm fixes from Dave Airlie:
 "A bunch of fixes for rc2: two imx regressions, vc4 fix, dma-buf fix,
  some displayport mst fixes, and an amdkfd fix.

  Nothing too crazy, I assume we just haven't see much rc1 testing yet"

* tag 'drm-fixes-for-v4.13-rc2' of git://people.freedesktop.org/~airlied/linux:
  drm/mst: Avoid processing partially received up/down message transactions
  drm/mst: Avoid dereferencing a NULL mstb in drm_dp_mst_handle_up_req()
  drm/mst: Fix error handling during MST sideband message reception
  drm/imx: parallel-display: Accept drm_of_find_panel_or_bridge failure
  drm/imx: fix typo in ipu_plane_formats[]
  drm/vc4: Fix VBLANK handling in crtc->enable() path
  dma-buf/fence: Avoid use of uninitialised timestamp
  drm/amdgpu: Remove unused field kgd2kfd_shared_resources.num_mec
  drm/radeon: Remove initialization of shared_resources.num_mec
  drm/amdkfd: Remove unused references to shared_resources.num_mec
  drm/amdgpu: Fix KFD oversubscription by tracking queues correctly

Pull tracing fixes from Steven Rostedt:
 "Three minor updates

   - Use the new GFP_RETRY_MAYFAIL to be more aggressive in allocating
     memory for the ring buffer without causing OOMs

   - Fix a memory leak in adding and removing instances

   - Add __rcu annotation to be able to debug RCU usage of function
     tracing a bit better"

* tag 'trace-v4.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  trace: fix the errors caused by incompatible type of RCU variables
  tracing: Fix kmemleak in instance_rmdir
  tracing/ring_buffer: Try harder to allocate

Pull KVM fixes from Radim Krčmář:
 "A bunch of small fixes for x86"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  kvm: x86: hyperv: avoid livelock in oneshot SynIC timers
  KVM: VMX: Fix invalid guest state detection after task-switch emulation
  x86: add MULTIUSER dependency for KVM
  KVM: nVMX: Disallow VM-entry in MOV-SS shadow
  KVM: nVMX: track NMI blocking state separately for each VMCS
  KVM: x86: masking out upper bits