Pull PCI fixes from Bjorn Helgaas:

 - Add PCI_DYNAMIC_OF_NODES dependency on OF_IRQ to fix sparc64 build
   error (Lizhi Hou)

 - After coalescing host bridge resources, free any released resources
   to avoid a leak (Ross Lagerwall)

 - Revert a quirk that prevented NVIDIA T4 GPUs from using Secondary Bus
   Reset. The quirk worked around an issue that we now think is related
   to the Root Port, not the GPU (Bjorn Helgaas)

* tag 'pci-v6.6-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
  Revert "PCI: Mark NVIDIA T4 GPUs to avoid bus reset"
  PCI: Free released resource after coalescing
  PCI: Fix CONFIG_PCI_DYNAMIC_OF_NODES kconfig dependencies

This reverts commit d5af729dc2071273f14cbb94abbc60608142fd83.

d5af729dc207 ("PCI: Mark NVIDIA T4 GPUs to avoid bus reset") avoided
Secondary Bus Reset on the T4 because the reset seemed to not work when the
T4 was directly attached to a Root Port.

But NVIDIA thinks the issue is probably related to some issue with the Root
Port, not with the T4.  The T4 provides neither PM nor FLR reset, so
masking bus reset compromises this device for assignment scenarios.

Revert d5af729dc207 as requested by Wu Zongyong.  This will leave SBR
broken in the specific configuration Wu tested, as it was in v6.5, so Wu
will debug that further.

Link: https://lore.kernel.org/r/ZPqMCDWvITlOLHgJ@wuzongyong-alibaba
Link: https://lore.kernel.org/r/20230908201104.GA305023@bhelgaas
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>

release_resource() doesn't actually free the resource or resource list
entry so free the resource list entry to avoid a leak.

Closes: https://lore.kernel.org/r/878r9sga1t.fsf@kernel.org/
Fixes: e54223275ba1 ("PCI: Release resource invalidated by coalescing")
Link: https://lore.kernel.org/r/20230906110846.225369-1-ross.lagerwall@citrix.com
Reported-by: Kalle Valo <kvalo@kernel.org>
Tested-by: Kalle Valo <kvalo@kernel.org>
Signed-off-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org      # v5.16+

Generating interrupt-map property depends on of_irq_parse_raw() which
is enabled by CONFIG_OF_IRQ. Change CONFIG_PCI_DYNAMIC_OF_NODES
dependency from CONFIG_OF to CONFIG_OF_IRQ.

Closes: https://lore.kernel.org/linux-devicetree/2187619d-55bc-41bb-bbb4-6059399db997@roeck-us.net/
Fixes: 407d1a51921e ("PCI: Create device tree node for bridge")
Link: https://lore.kernel.org/r/1693505947-29786-1-git-send-email-lizhi.hou@amd.com
Reported-by: Guenter Roeck <groeck7@gmail.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>

Pull powerpc updates from Michael Ellerman:

 - Add HOTPLUG_SMT support (/sys/devices/system/cpu/smt) and honour the
   configured SMT state when hotplugging CPUs into the system

 - Combine final TLB flush and lazy TLB mm shootdown IPIs when using the
   Radix MMU to avoid a broadcast TLBIE flush on exit

 - Drop the exclusion between ptrace/perf watchpoints, and drop the now
   unused associated arch hooks

 - Add support for the "nohlt" command line option to disable CPU idle

 - Add support for -fpatchable-function-entry for ftrace, with GCC >=
   13.1

 - Rework memory block size determination, and support 256MB size on
   systems with GPUs that have hotpluggable memory

 - Various other small features and fixes

Thanks to Andrew Donnellan, Aneesh Kumar K.V, Arnd Bergmann, Athira
Rajeev, Benjamin Gray, Christophe Leroy, Frederic Barrat, Gautam
Menghani, Geoff Levand, Hari Bathini, Immad Mir, Jialin Zhang, Joel
Stanley, Jordan Niethe, Justin Stitt, Kajol Jain, Kees Cook, Krzysztof
Kozlowski, Laurent Dufour, Liang He, Linus Walleij, Mahesh Salgaonkar,
Masahiro Yamada, Michal Suchanek, Nageswara R Sastry, Nathan Chancellor,
Nathan Lynch, Naveen N Rao, Nicholas Piggin, Nick Desaulniers, Omar
Sandoval, Randy Dunlap, Reza Arbab, Rob Herring, Russell Currey, Sourabh
Jain, Thomas Gleixner, Trevor Woerner, Uwe Kleine-König, Vaibhav Jain,
Xiongfeng Wang, Yuan Tan, Zhang Rui, and Zheng Zengkai.

* tag 'powerpc-6.6-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (135 commits)
  macintosh/ams: linux/platform_device.h is needed
  powerpc/xmon: Reapply "Relax frame size for clang"
  powerpc/mm/book3s64: Use 256M as the upper limit with coherent device memory attached
  powerpc/mm/book3s64: Fix build error with SPARSEMEM disabled
  powerpc/iommu: Fix notifiers being shared by PCI and VIO buses
  powerpc/mpc5xxx: Add missing fwnode_handle_put()
  powerpc/config: Disable SLAB_DEBUG_ON in skiroot
  powerpc/pseries: Remove unused hcall tracing instruction
  powerpc/pseries: Fix hcall tracepoints with JUMP_LABEL=n
  powerpc: dts: add missing space before {
  powerpc/eeh: Use pci_dev_id() to simplify the code
  powerpc/64s: Move CPU -mtune options into Kconfig
  powerpc/powermac: Fix unused function warning
  powerpc/pseries: Rework lppaca_shared_proc() to avoid DEBUG_PREEMPT
  powerpc: Don't include lppaca.h in paca.h
  powerpc/pseries: Move hcall_vphn() prototype into vphn.h
  powerpc/pseries: Move VPHN constants into vphn.h
  cxl: Drop unused detach_spa()
  powerpc: Drop zalloc_maybe_bootmem()
  powerpc/powernv: Use struct opal_prd_msg in more places
  ...

Pull PCI updates from Bjorn Helgaas:
 "Enumeration:
   - Add locking to read/modify/write PCIe Capability Register accessors
     for Link Control and Root Control
   - Use pci_dev_id() when possible instead of manually composing ID
     from dev->bus->number and dev->devfn

  Resource management:
   - Move prototypes for __weak sysfs resource files to linux/pci.h to
     fix 'no previous prototype' warnings
   - Make more I/O port accesses depend on HAS_IOPORT
   - Use devm_platform_get_and_ioremap_resource() instead of open-coding
     platform_get_resource() followed by devm_ioremap_resource()

  Power management:
   - Ensure devices are powered up while accessing VPD
   - If device is powered-up, keep it that way while polling for PME
   - Only read PCI_PM_CTRL register when available, to avoid reading the
     wrong register and corrupting dev->current_state

  Virtualization:
   - Avoid Secondary Bus Reset on NVIDIA T4 GPUs

  Error handling:
   - Remove unused pci_disable_pcie_error_reporting()
   - Unexport pci_enable_pcie_error_reporting(), used only by aer.c
   - Unexport pcie_port_bus_type, used only by PCI core

  VGA:
   - Simplify and clean up typos in VGA arbiter

  Apple PCIe controller driver:
   - Initialize pcie->nvecs (number of available MSIs) before use

  Broadcom iProc PCIe controller driver:
   - Use of_property_read_bool() instead of low-level accessors for
     boolean properties

  Broadcom STB PCIe controller driver:
   - Assert PERST# when probing BCM2711 because some bootloaders don't
     do it

  Freescale i.MX6 PCIe controller driver:
   - Add .host_deinit() callback so we can clean up things like
     regulators on probe failure or driver unload

  Freescale Layerscape PCIe controller driver:
   - Add support for link-down notification so the endpoint driver can
     process LINK_DOWN events
   - Add suspend/resume support, including manual
     PME_Turn_off/PME_TO_Ack handshake
   - Save Link Capabilities during probe so they can be restored when
     handling a link-up event, since the controller loses the Link Width
     and Link Speed values during reset

  Intel VMD host bridge driver:
   - Fix disable of bridge windows during domain reset; previously we
     cleared the base/limit registers, which actually left the windows
     enabled

  Marvell MVEBU PCIe controller driver:
   - Remove unused busn member

  Microchip PolarFlare PCIe controller driver:
   - Fix interrupt bit definitions so the SEC and DED interrupt handlers
     work correctly
   - Make driver buildable as a module
   - Read FPGA MSI configuration parameters from hardware instead of
     hard-coding them

  Microsoft Hyper-V host bridge driver:
   - To avoid a NULL pointer dereference, skip MSI restore after
     hibernate if MSI/MSI-X hasn't been enabled

  NVIDIA Tegra194 PCIe controller driver:
   - Revert 'PCI: tegra194: Enable support for 256 Byte payload' because
     Linux doesn't know how to reduce MPS from to 256 to 128 bytes for
     endpoints below a switch (because other devices below the switch
     might already be operating), which leads to 'Malformed TLP' errors

  Qualcomm PCIe controller driver:
   - Add DT and driver support for interconnect bandwidth voting for
     'pcie-mem' and 'cpu-pcie' interconnects
   - Fix broken SDX65 'compatible' DT property
   - Configure controller so MHI bus master clock will be switched off
     while in ASPM L1.x states
   - Use alignment restriction from EPF core in EPF MHI driver
   - Add Endpoint eDMA support
   - Add MHI eDMA support
   - Add Snapdragon SM8450 support to the EPF MHI driversupport
   - Add MHI eDMA support
   - Add Snapdragon SM8450 support to the EPF MHI driversupport
   - Add MHI eDMA support
   - Add Snapdragon SM8450 support to the EPF MHI driversupport
   - Add MHI eDMA support
   - Add Snapdragon SM8450 support to the EPF MHI driver
   - Use iATU for EPF MHI transfers smaller than 4K to avoid eDMA setup
     latency
   - Add sa8775p DT binding and driver support

  Rockchip PCIe controller driver:
   - Use 64-bit mask on MSI 64-bit PCI address to avoid zeroing out the
     upper 32 bits

  SiFive FU740 PCIe controller driver:
   - Set the supported number of MSI vectors so we can use all available
     MSI interrupts

  Synopsys DesignWare PCIe controller driver:
   - Add generic dwc suspend/resume APIs (dw_pcie_suspend_noirq() and
     dw_pcie_resume_noirq()) to be called by controller driver
     suspend/resume ops, and a controller callback to send PME_Turn_Off

  MicroSemi Switchtec management driver:
   - Add support for PCIe Gen5 devices

  Miscellaneous:
   - Reorder and compress to reduce size of struct pci_dev
   - Fix race in DOE destroy_work_on_stack()
   - Add stubs to avoid casts between incompatible function types
   - Explicitly include correct DT includes to untangle headers"

* tag 'pci-v6.6-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci: (96 commits)
  PCI: qcom-ep: Add ICC bandwidth voting support
  dt-bindings: PCI: qcom: ep: Add interconnects path
  PCI: qcom-ep: Treat unknown IRQ events as an error
  dt-bindings: PCI: qcom: Fix SDX65 compatible
  PCI: endpoint: Add kernel-doc for pci_epc_mem_init() API
  PCI: epf-mhi: Use iATU for small transfers
  PCI: epf-mhi: Add support for SM8450
  PCI: epf-mhi: Add eDMA support
  PCI: qcom-ep: Add eDMA support
  PCI: epf-mhi: Make use of the alignment restriction from EPF core
  PCI/PM: Only read PCI_PM_CTRL register when available
  PCI: qcom: Add support for sa8775p SoC
  dt-bindings: PCI: qcom: Add sa8775p compatible
  PCI: qcom-ep: Pass alignment restriction to the EPF core
  PCI: Simplify pcie_capability_clear_and_set_word() control flow
  PCI: Tidy config space save/restore messages
  PCI: Fix code formatting inconsistencies
  PCI: Fix typos in docs and comments
  PCI: Fix pci_bus_resetable(), pci_slot_resetable() name typos
  PCI: Simplify pci_dev_driver()
  ...

Pull devicetree updates from Rob Herring:
 "DT core:

   - Add support for generating DT nodes for PCI devices. This is the
     groundwork for applying overlays to PCI devices containing
     non-discoverable downstream devices.

   - DT unittest additions to check reverted changesets, to test for
     refcount issues, and to test unresolved symbols. Also, various
     clean-ups of the unittest along the way.

   - Refactor node and property manipulation functions to better share
     code with old API and changeset API

   - Refactor changeset print functions to a common implementation

   - Move some platform_device specific functions into of_platform.c

  Bindings:

   - Treewide fixing of typos

   - Treewide clean-up of SPDX tags to use 'OR' consistently

   - Last chunk of dropping unnecessary quotes. With that, the check for
     unnecessary quotes is enabled in yamllint.

   - Convert ftgmac100, zynqmp-genpd, pps-gpio, syna,rmi4, and qcom,ssbi
     bindings to DT schema format

   - Add Allwinner V3s xHCI USB, Saef SF-TC154B display, QCom SM8450
     Inline Crypto Engine, QCom SM6115 UFS, QCom SDM670 PDC interrupt
     controller, Arm 2022 Cortex cores, and QCom IPQ9574 Crypto bindings

   - Fixes for Rockchip DWC PCI binding

   - Ensure all properties are evaluated on USB connector schema

   - Fix dt-check-compatible script to find of_device_id instances with
     compiler annotations"

* tag 'devicetree-for-6.6' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux: (64 commits)
  dt-bindings: usb: Add V3s compatible string for OHCI
  dt-bindings: usb: Add V3s compatible string for EHCI
  dt-bindings: display: panel: mipi-dbi-spi: add Saef SF-TC154B
  dt-bindings: vendor-prefixes: document Saef Technology
  dt-bindings: thermal: lmh: update maintainer address
  of: unittest: Fix of_unittest_pci_node() kconfig dependencies
  dt-bindings: crypto: ice: Document sm8450 inline crypto engine
  dt-bindings: ufs: qcom: Add ICE to sm8450 example
  dt-bindings: ufs: qcom: Add sm6115 binding
  dt-bindings: ufs: qcom: Add reg-names property for ICE
  dt-bindings: yamllint: Enable quoted string check
  dt-bindings: Drop remaining unneeded quotes
  of: unittest-data: Fix whitespace - angular brackets
  of: unittest-data: Fix whitespace - indentation
  of: unittest-data: Fix whitespace - blank lines
  of: unittest-data: Convert remaining overlay DTS files to sugar syntax
  of: overlay: unittest: Add test for unresolved symbol
  of: unittest: Add separators to of_unittest_overlay_high_level()
  of: unittest: Cleanup partially-applied overlays
  of: unittest: Merge of_unittest_apply{,_revert}_overlay_check()
  ...

Pull dma-maping updates from Christoph Hellwig:

 - allow dynamic sizing of the swiotlb buffer, to cater for secure
   virtualization workloads that require all I/O to be bounce buffered
   (Petr Tesarik)

 - move a declaration to a header (Arnd Bergmann)

 - check for memory region overlap in dma-contiguous (Binglei Wang)

 - remove the somewhat dangerous runtime swiotlb-xen enablement and
   unexport is_swiotlb_active (Christoph Hellwig, Juergen Gross)

 - per-node CMA improvements (Yajun Deng)

* tag 'dma-mapping-6.6-2023-08-29' of git://git.infradead.org/users/hch/dma-mapping:
  swiotlb: optimize get_max_slots()
  swiotlb: move slot allocation explanation comment where it belongs
  swiotlb: search the software IO TLB only if the device makes use of it
  swiotlb: allocate a new memory pool when existing pools are full
  swiotlb: determine potential physical address limit
  swiotlb: if swiotlb is full, fall back to a transient memory pool
  swiotlb: add a flag whether SWIOTLB is allowed to grow
  swiotlb: separate memory pool data from other allocator data
  swiotlb: add documentation and rename swiotlb_do_find_slots()
  swiotlb: make io_tlb_default_mem local to swiotlb.c
  swiotlb: bail out of swiotlb_init_late() if swiotlb is already allocated
  dma-contiguous: check for memory region overlap
  dma-contiguous: support numa CMA for specified node
  dma-contiguous: support per-numa CMA for all architectures
  dma-mapping: move arch_dma_set_mask() declaration to header
  swiotlb: unexport is_swiotlb_active
  x86: always initialize xen-swiotlb when xen-pcifront is enabling
  xen/pci: add flag for PCI passthrough being possible

- Reorder struct pci_dev to avoid holes and reduce size (Christophe
  JAILLET)

- Change pdev->rom_attr_enabled to single bit since it's only a boolean
  value (Christophe JAILLET)

- Use struct_size() in pirq_convert_irt_table() instead of hand-writing it
  (Christophe JAILLET)

- Explicitly include correct DT includes to untangle headers (Rob Herring)

- Fix a DOE race between destroy_work_on_stack() and the stack-allocated
  task->work struct going out of scope in pci_doe() (Ira Weiny)

- Use pci_dev_id() when possible instead of manually composing ID from
  dev->bus->number and dev->devfn (Xiongfeng Wang, Zheng Zengkai)

- Move pci_create_resource_files() declarations to linux/pci.h for alpha
  build warnings (Arnd Bergmann)

- Remove unused hotplug function declarations (Yue Haibing)

- Remove unused mvebu struct mvebu_pcie.busn (Pali Rohár)

- Unexport pcie_port_bus_type (Bjorn Helgaas)

- Remove unnecessary sysfs ID local variable initialization (Bjorn Helgaas)

- Fix BAR value printk formatting to accommodate 32-bit values (Bjorn
  Helgaas)

- Use consistent pointer types for config access syscall get_user() and
  put_user() uses (Bjorn Helgaas)

- Simplify AER_RECOVER_RING_SIZE definition (Bjorn Helgaas)

- Simplify pci_pio_to_address() (Bjorn Helgaas)

- Simplify pci_dev_driver() (Bjorn Helgaas)

- Fix pci_bus_resetable(), pci_slot_resetable() name typos (Bjorn Helgaas)

- Fix code and doc typos and code formatting (Bjorn Helgaas)

- Tidy config space save/restore messages (Bjorn Helgaas)

* pci/misc:
  PCI: Tidy config space save/restore messages
  PCI: Fix code formatting inconsistencies
  PCI: Fix typos in docs and comments
  PCI: Fix pci_bus_resetable(), pci_slot_resetable() name typos
  PCI: Simplify pci_dev_driver()
  PCI: Simplify pci_pio_to_address()
  PCI/AER: Simplify AER_RECOVER_RING_SIZE definition
  PCI: Use consistent put_user() pointer types
  PCI: Fix printk field formatting
  PCI: Remove unnecessary initializations
  PCI: Unexport pcie_port_bus_type
  PCI: mvebu: Remove unused busn member
  PCI: Remove unused function declarations
  PCI/sysfs: Move declarations to linux/pci.h
  PCI/P2PDMA: Use pci_dev_id() to simplify the code
  PCI/IOV: Use pci_dev_id() to simplify the code
  PCI/AER: Use pci_dev_id() to simplify the code
  PCI: apple: Use pci_dev_id() to simplify the code
  PCI/DOE: Fix destroy_work_on_stack() race
  PCI: Explicitly include correct DT includes
  x86/PCI: Use struct_size() in pirq_convert_irt_table()
  PCI: Change pdev->rom_attr_enabled to single bit
  PCI: Reorder pci_dev fields to reduce holes

- Add support for Switechtec PCIe Gen5 devices (Kelvin Cao)

* pci/controller/switchtec:
  PCI: switchtec: Add support for PCIe Gen5 devices
  PCI: switchtec: Use normal comment style