PCI/MSI: Prevent MSI hardware interrupt number truncation

[Brad Figg]

    While calculating the hardware interrupt number for a MSI interrupt, the
    higher bits (i.e. from bit-5 onwards a.k.a domain_nr >= 32) of the PCI
    domain number gets truncated because of the shifted value casting to return
    type of pci_domain_nr() which is 'int'. This for example is resulting in
    same hardware interrupt number for devices 0019:00:00.0 and 0039:00:00.0.

    To address this cast the PCI domain number to 'irq_hw_number_t' before left
    shifting it to calculate the hardware interrupt number.

    Please note that this fixes the issue only on 64-bit systems and doesn't
    change the behavior for 32-bit systems i.e. the 32-bit systems continue to
    have the issue. Since the issue surfaces only if there are too many PCIe
    controllers in the system which usually is the case in modern server
    systems and they don't tend to run 32-bit kernels.

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This bug is awaiting verification that the linux-nvidia-6.5/6.5.0-1021.22 kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-jammy-linux-nvidia-6.5' to 'verification-done-jammy-linux-nvidia-6.5'. If the problem still exists, change the tag 'verification-needed-jammy-linux-nvidia-6.5' to 'verification-failed-jammy-linux-nvidia-6.5'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux-nvidia-6.5 - 6.5.0-1021.22

---------------
linux-nvidia-6.5 (6.5.0-1021.22) jammy; urgency=medium

  * jammy/linux-nvidia-6.5: 6.5.0-1021.21 -proposed tracker (LP: #2065892)

  * Packaging resync (LP: #1786013)
    - [Packaging] debian.nvidia-6.5/dkms-versions -- update from kernel-versions
      (main/2024.04.29)

  * Address out-of-bounds issue when using TPM SPI interface (LP: #2067429)
    - tpm_tis_spi: Account for SPI header when allocating TPM SPI xfer buffer

  * Perf-stat tool does not support ipc and ipc_rate monitoring on NVIDIA Grace
    system (LP: #2063461)
    - perf vendor events arm64: Update stall_slot workaround for N2 r0p3
    - perf vendor events arm64: Update N2 and V2 metrics and events using Arm
      telemetry repo
    - perf jevents: Add a new expression builtin strcmp_cpuid_str()
    - perf jevents metric: Fix type of strcmp_cpuid_str

  * PCI/MSI: Prevent MSI hardware interrupt number truncation (LP: #2065721)
    - PCI/MSI: Prevent MSI hardware interrupt number truncation

  * Apply patch to set CONFIG_EFI_CAPSULE_LOADER=y for arm64 (LP: #2067111)
    - NVIDIA: [Config] EFI: set CAPSULE_LOADER=y for arm64

  * pull-request: Fixes: b2b56a163230 ("gpio: tegra186: Check GPIO pin
    permission before access.") (LP: #2064549)
    - Revert "NVIDIA: SAUCE: Revert "gpio: tegra186: Check GPIO pin permission
      before access.""
    - gpio: tegra186: Fix tegra186_gpio_is_accessible() check

  [ Ubuntu: 6.5.0-41.41 ]

  * mantic/linux: 6.5.0-41.41 -proposed tracker (LP: #2065893)
  * CVE-2024-21823
    - VFIO: Add the SPR_DSA and SPR_IAX devices to the denylist
    - dmaengine: idxd: add a new security check to deal with a hardware erratum
    - dmaengine: idxd: add a write() method for applications to submit work

  [ Ubuntu: 6.5.0-40.40 ]

  * mantic/linux: 6.5.0-40.40 -proposed tracker (LP: #2063709)
  * [Mantic] Compile broken on armhf (cc1 out of memory) (LP: #2060446)
    - Revert "minmax: relax check to allow comparison between unsigned arguments
      and signed constants"
    - Revert "minmax: allow comparisons of 'int' against 'unsigned char/short'"
    - Revert "minmax: allow min()/max()/clamp() if the arguments have the same
      signedness."
    - Revert "minmax: add umin(a, b) and umax(a, b)"
  * Drop fips-checks script from trees (LP: #2055083)
    - [Packaging] Remove fips-checks script
  * alsa/realtek: adjust max output valume for headphone on 2 LG machines
    (LP: #2058573)
    - ALSA: hda/realtek: fix the hp playback volume issue for LG machines
  * Mantic update: upstream stable patchset 2024-03-27 (LP: #2059284)
    - asm-generic: make sparse happy with odd-sized put_unaligned_*()
    - powerpc/mm: Fix null-pointer dereference in pgtable_cache_add
    - arm64: irq: set the correct node for VMAP stack
    - drivers/perf: pmuv3: don't expose SW_INCR event in sysfs
    - powerpc: Fix build error due to is_valid_bugaddr()
    - powerpc/mm: Fix build failures due to arch_reserved_kernel_pages()
    - powerpc/64s: Fix CONFIG_NUMA=n build due to create_section_mapping()
    - x86/boot: Ignore NMIs during very early boot
    - powerpc: pmd_move_must_withdraw() is on...