Pull iommu fixes from Joerg Roedel:

 - Intel VT-d fixes from Lu Baolu:
     - Boot kdump kernels with VT-d scalable mode on
     - Calculate the right page table levels
     - Fix two recursive locking issues
     - Fix a lockdep splat issue

 - AMD IOMMU fixes:
     - Fix for completion-wait command to use full 64 bits of data
     - Fix PASID related issue where GPU sound devices failed to
       initialize

 - Fix for Virtio-IOMMU to report correct caching behavior, needed for
   use with VFIO

* tag 'iommu-fixes-v6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu: Fix false ownership failure on AMD systems with PASID activated
  iommu/vt-d: Fix possible recursive locking in intel_iommu_init()
  iommu/virtio: Fix interaction with VFIO
  iommu/vt-d: Fix lockdep splat due to klist iteration in atomic context
  iommu/vt-d: Fix recursive lock issue in iommu_flush_dev_iotlb()
  iommu/vt-d: Correctly calculate sagaw value of IOMMU
  iommu/vt-d: Fix kdump kernels boot failure with scalable mode
  iommu/amd: use full 64-bit value in build_completion_wait()

The global rwsem dmar_global_lock was introduced by commit 3a5670e8ac932
("iommu/vt-d: Introduce a rwsem to protect global data structures"). It
is used to protect DMAR related global data from DMAR hotplug operations.

The dmar_global_lock used in the intel_iommu_init() might cause recursive
locking issue, for example, intel_iommu_get_resv_regions() is taking the
dmar_global_lock from within a section where intel_iommu_init() already
holds it via probe_acpi_namespace_devices().

Using dmar_global_lock in intel_iommu_init() could be relaxed since it is
unlikely that any IO board must be hot added before the IOMMU subsystem is
initialized. This eliminates the possible recursive locking issue by moving
down DMAR hotplug support after the IOMMU is initialized and removing the
uses of dmar_global_lock in intel_iommu_init().

Fixes: d5692d4af08cd ("iommu/vt-d: Fix suspicious RCU usage in probe_acpi_namespace_devices()")
Reported-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/894db0ccae854b35c73814485569b634237b5538.1657034828.git.robin.murphy@arm.com
Link: https://lore.kernel.org/r/20220718235325.3952426-1-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Pull dma-mapping fixes from Christoph Hellwig:

 - revert a panic on swiotlb initialization failure (Yu Zhao)

 - fix the lookup for partial syncs in dma-debug (Robin Murphy)

 - fix a shift overflow in swiotlb (Chao Gao)

 - fix a comment typo in swiotlb (Chao Gao)

 - mark a function static now that all abusers are gone (Christoph
   Hellwig)

* tag 'dma-mapping-6.0-2022-09-10' of git://git.infradead.org/users/hch/dma-mapping:
  dma-mapping: mark dma_supported static
  swiotlb: fix a typo
  swiotlb: avoid potential left shift overflow
  dma-debug: improve search for partial syncs
  Revert "swiotlb: panic if nslabs is too small"

Pull driver core fixes from Greg KH:
 "Here are some small driver core and debugfs fixes for 6.0-rc5.

  Included in here are:

   - multiple attempts to get the arch_topology code to work properly on
     non-cluster SMT systems. First attempt caused build breakages in
     linux-next and 0-day, second try worked.

   - debugfs fixes for a long-suffering memory leak. The pattern of
     debugfs_remove(debugfs_lookup(...)) turns out to leak dentries, so
     add debugfs_lookup_and_remove() to fix this problem. Also fix up
     the scheduler debug code that highlighted this problem. Fixes for
     other subsystems will be trickling in over the next few months for
     this same issue once the debugfs function is merged.

  All of these have been in linux-next since Wednesday with no reported
  problems"

* tag 'driver-core-6.0-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  arch_topology: Make cluster topology span at least SMT CPUs
  sched/debug: fix dentry leak in update_sched_domain_debugfs
  debugfs: add debugfs_lookup_and_remove()
  driver core: fix driver_set_override() issue with empty strings
  Revert "arch_topology: Make cluster topology span at least SMT CPUs"
  arch_topology: Make cluster topology span at least SMT CPUs

Pull rdma fixes from Jason Gunthorpe:
 "Many bug fixes in several drivers:

   - Fix misuse of the DMA API in rtrs

   - Several irdma issues: hung task due to SQ flushing, incorrect
     capability reporting to userspace, improper error handling for MW
     corners, touching an uninitialized SGL for during invalidation.

   - hns was using the wrong page size limits for the HW, an incorrect
     calculation of wqe_shift causing WQE corruption, and mis computed a
     timer id.

   - Fix a crash in SRP triggered by blktests

   - Fix compiler errors by calling virt_to_page() with the proper type
     in siw

   - Userspace triggerable deadlock in ODP

   - mlx5 could use the wrong profile due to some driver loading races,
     counters were not working in some device configurations, and a
     crash on error unwind"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
  RDMA/irdma: Report RNR NAK generation in device caps
  RDMA/irdma: Use s/g array in post send only when its valid
  RDMA/irdma: Return correct WC error for bind operation failure
  RDMA/irdma: Return error on MR deregister CQP failure
  RDMA/irdma: Report the correct max cqes from query device
  MAINTAINERS: Update maintainers of HiSilicon RoCE
  RDMA/mlx5: Fix UMR cleanup on error flow of driver init
  RDMA/mlx5: Set local port to one when accessing counters
  RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile
  IB/core: Fix a nested dead lock as part of ODP flow
  RDMA/siw: Pass a pointer to virt_to_page()
  RDMA/srp: Set scmnd->result only when scmnd is not NULL
  RDMA/hns: Remove the num_qpc_timer variable
  RDMA/hns: Fix wrong fixed value of qp->rq.wqe_shift
  RDMA/hns: Fix supported page size
  RDMA/cma: Fix arguments order in net device validation
  RDMA/irdma: Fix drain SQ hang with no completion
  RDMA/rtrs-srv: Pass the correct number of entries for dma mapped SGL
  RDMA/rtrs-clt: Use the right sg_cnt after ib_dma_map_sg

Pull spi fixes from Mark Brown:
 "Several fixes that came in since the merge window, the major one being
  a fix for the spi-mux driver which was broken by the performance
  optimisations due to it peering inside the core's data structures more
  than it should"

* tag 'spi-fix-v6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
  spi: spi: Fix queue hang if previous transfer failed
  spi: mux: Fix mux interaction with fast path optimisations
  spi: cadence-quadspi: Disable irqs during indirect reads
  spi: bitbang: Fix lsb-first Rx

Pull networking fixes from Paolo Abeni:
 "Including fixes from rxrpc, netfilter, wireless and bluetooth
  subtrees.

  Current release - regressions:

   - skb: export skb drop reaons to user by TRACE_DEFINE_ENUM

   - bluetooth: fix regression preventing ACL packet transmission

  Current release - new code bugs:

   - dsa: microchip: fix kernel oops on ksz8 switches

   - dsa: qca8k: fix NULL pointer dereference for
     of_device_get_match_data

  Previous releases - regressions:

   - netfilter: clean up hook list when offload flags check fails

   - wifi: mt76: fix crash in chip reset fail

   - rxrpc: fix ICMP/ICMP6 error handling

   - ice: fix DMA mappings leak

   - i40e: fix kernel crash during module removal

  Previous releases - always broken:

   - ipv6: sr: fix out-of-bounds read when setting HMAC data.

   - tcp: TX zerocopy should not sense pfmemalloc status

   - sch_sfb: don't assume the skb is still around after
     enqueueing to child

   - netfilter: drop dst references before setting

   - wifi: wilc1000: fix DMA on stack objects

   - rxrpc: fix an insufficiently large sglist in
     rxkad_verify_packet_2()

   - fec: use a spinlock to guard `fep->ptp_clk_on`

  Misc:

   - usb: qmi_wwan: add Quectel RM520N"

* tag 'net-6.0-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (50 commits)
  sch_sfb: Also store skb len before calling child enqueue
  net: phy: lan87xx: change interrupt src of link_up to comm_ready
  net/smc: Fix possible access to freed memory in link clear
  net: ethernet: mtk_eth_soc: check max allowed hash in mtk_ppe_check_skb
  net: skb: export skb drop reaons to user by TRACE_DEFINE_ENUM
  net: ethernet: mtk_eth_soc: fix typo in __mtk_foe_entry_clear
  net: dsa: felix: access QSYS_TAG_CONFIG under tas_lock in vsc9959_sched_speed_set
  net: dsa: felix: disable cut-through forwarding for frames oversized for tc-taprio
  net: dsa: felix: tc-taprio intervals smaller than MTU should send at least one packet
  net: usb: qmi_wwan: add Quectel RM520N
  net: dsa: qca8k: fix NULL pointer dereference for of_device_get_match_data
  tcp: fix early ETIMEDOUT after spurious non-SACK RTO
  stmmac: intel: Simplify intel_eth_pci_remove()
  net: mvpp2: debugfs: fix memory leak when using debugfs_lookup()
  ipv6: sr: fix out-of-bounds read when setting HMAC data.
  bonding: accept unsolicited NA message
  bonding: add all node mcast address when slave up
  bonding: use unspecified address if no available link local address
  wifi: use struct_group to copy addresses
  wifi: mac80211_hwsim: check length for virtio packets
  ...

Commit d4252071b97d ("add barriers to buffer_uptodate and
set_buffer_uptodate") added proper memory barriers to the buffer head
BH_Uptodate bit, so that anybody who tests a buffer for being up-to-date
will be guaranteed to actually see initialized state.

However, that commit didn't _just_ add the memory barrier, it also ended
up dropping the "was it already set" logic that the BUFFER_FNS() macro
had.

That's conceptually the right thing for a generic "this is a memory
barrier" operation, but in the case of the buffer contents, we really
only care about the memory barrier for the _first_ time we set the bit,
in that the only memory ordering protection we need is to avoid anybody
seeing uninitialized memory contents.

Any other access ordering wouldn't be about the BH_Uptodate bit anyway,
and would require some other proper lock (typically BH_Lock or the folio
lock).  A reader that races with somebody invalidating the buffer head
isn't an issue wrt the memory ordering, it's a serialization issue.

Now, you'd think that the buffer head operations don't matter in this
day and age (and I certainly thought so), but apparently some loads
still end up being heavy users of buffer heads.  In particular, the
kernel test robot reported that not having this bit access optimization
in place caused a noticeable direct IO performance regression on ext4:

  fxmark.ssd_ext4_no_jnl_DWTL_54_directio.works/sec -26.5% regression

although you presumably need a fast disk and a lot of cores to actually
notice.

Link: https://lore.kernel.org/all/Yw8L7HTZ%2FdE2%2Fo9C@xsang-OptiPlex-9020/
Reported-by: kernel test robot <oliver.sang@intel.com>
Tested-by: Fengwei Yin <fengwei.yin@intel.com>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Now that the remaining users in drivers are gone, this function can be
marked static.

Signed-off-by: Christoph Hellwig <hch@lst.de>

Pull ARM fix from Russell King:
 "Just one fix for now for the AMBA bus code from Isaac Manjarres"

* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
  ARM: 9229/1: amba: Fix use-after-free in amba_read_periphid()