Pull ACPI fix from Rafael Wysocki:
 "One more urgent ACPI support fix for 6.18

  There is one more commit that needs to be reverted after reverting
  problematic commit 7a8c994cbb2d ("ACPI: processor: idle: Optimize ACPI
  idle driver registration"), so revert it"

* tag 'acpi-6.18-rc8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  Revert "ACPI: processor: Update cpuidle driver check in __acpi_processor_start()"

Revert commit 8a1b5d412cb4 ("ACPI: processor: Update cpuidle driver
check in __acpi_processor_start()") which depends on commit
7a8c994cbb2d ("ACPI: processor: idle: Optimize ACPI idle driver
registration") that got reverted.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Pull ACPI fix from Rafael Wysocki:
 "Revert a commit that attempted to make the code in the ACPI processor
  driver more straightforward, but it turned out to cause the kernel to
  crash on at least one system, along with some further cleanups on top
  of it"

* tag 'acpi-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  Revert "ACPI: processor: idle: Optimize ACPI idle driver registration"
  Revert "ACPI: processor: Remove unused empty stubs of some functions"
  Revert "ACPI: processor: idle: Rearrange declarations in header file"
  Revert "ACPI: processor: idle: Redefine two functions as void"
  Revert "ACPI: processor: Do not expose global variable acpi_idle_driver"

Pull arm64 fixes from Will Deacon:
 "We've got a revert due to one of the recent CCA commits breaking ACPI
  firmware-based error reporting, a fix for a hard-lockup introduced by
  a prior fix affecting non-default (CONFIG_EXPERT) configurations and
  another ACPI fix for systems using MMIO-based timers.

  Other than that, we're looking pretty good.

   - Avoid hardlockup when CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY=n

   - Fix regression in APEI/GHES error handling

   - Fix MMIO timers when probed via ACPI"

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: proton-pack: Fix hard lockup when !MITIGATE_SPECTRE_BRANCH_HISTORY
  ACPI: GTDT: Correctly number platform devices for MMIO timers
  Revert "arm64: acpi: Enable ACPI CCEL support"

Revert commit 7a8c994cbb2d ("ACPI: processor: idle: Optimize ACPI idle
driver registration") because it is reported to introduce a cpuidle
regression leading to a kernel crash on a platform using the ACPI idle
driver.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reported-by: Borislav Petkov <bp@alien8.de>
Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
Closes: https://lore.kernel.org/lkml/20251124200019.GIaSS5U9HhsWBotrQZ@fat_crate.local/

Revert commit fbd401e95e56 ("ACPI: processor: idle: Redefine two
functions as void") because it depends on a problematic one.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Revert commit 559f2eacc8a2 ACPI: processor: Do not expose global variable
acpi_idle_driver" because it depends on a problematic one.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Use the actual timer counter instead of the watchdog counter.

Fixes: 5669d92f3efa ("ACPI: GTDT: Generate platform devices for MMIO timers")
Reported-by: Pavan Kondeti <pavan.kondeti@oss.qualcomm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Cc: Hanjun Guo <guohanjun@huawei.com>
Cc: Sudeep Holla <sudeep.holla@arm.com>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Acked-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Will Deacon <will@kernel.org>

ACPI 6.6 specification for EINJV2 appends an extra structure to
the end of the existing struct set_error_type_with_address.

Several issues showed up in testing.

 1) Initialization was broken by an earlier fix [1] since is_v2 is only
    set while performing an injection, not during initialization.

 2) A buggy BIOS provided invalid "revision" and "length" for the
    extension structure. Add several sanity checks.

 3) When injecting legacy error types on an EINJV2 capable system,
    don't copy the component arrays.

Fixes: 6c7058514991 ("ACPI: APEI: EINJ: Check if user asked for EINJV2 injection") # [1]
Fixes: b47610296d17 ("ACPI: APEI: EINJ: Enable EINJv2 error injections")
Signed-off-by: Tony Luck <tony.luck@intel.com>
[ rjw: Changelog edits ]
Cc: 6.17+ <stable@vger.kernel.org> # 6.17+
Link: https://patch.msgid.link/20251119012712.178715-1-tony.luck@intel.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Pull cxl fixes from Dave Jiang:

 - Fix incorrect device handle check for Generic Initiator

 - Fix offset calculation for extended linear cache poison injection

 - Fix lockdep warning for hmem_register_resource()

* tag 'cxl-fixes-6.18-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
  acpi/hmat: Fix lockdep warning for hmem_register_resource()
  cxl: Adjust offset calculation for poison injection
  acpi,srat: Fix incorrect device handle check for Generic Initiator