Tags

x86-percpu-2024-07-17

x86/percpu changes for v6.11:

 - Enable the named address spaces based percpu accessors optimization
   on all GCC versions that contain this feature, detected through
   a build-time testcase. This effectively enables the feature on
   GCC 6, GCC 7 and GCC 8 versions.

 - Fix operand constraint modifier in __raw_cpu_write()

 - Reorganize the per-CPU headers for better readability

 - Misc cleanups and fixes

Signed-off-by: Ingo Molnar <mingo@kernel.org>

perf-core-2024-07-16

Performance events changes for v6.11:

 - Intel PT support enhancements & fixes
 - Fix leaked SIGTRAP events
 - Improve and fix the Intel uncore driver
 - Add support for Intel HBM and CXL uncore counters
 - Add Intel Lake and Arrow Lake support
 - AMD uncore driver fixes
 - Make SIGTRAP and __perf_pending_irq() work on RT
 - Micro-optimizations
 - Misc cleanups and fixes

Signed-off-by: Ingo Molnar <mingo@kernel.org>

x86_sev_for_v6.11_rc1

 - Add support for running the kernel in a SEV-SNP guest, over a Secure
   VM Service Module (SVSM).

   When running over a SVSM, different services can run at different
   protection levels, apart from the guest OS but still within the
   secure SNP environment.  They can provide services to the guest, like
   a vTPM, for example.

   This series adds the required facilities to interface with such a SVSM
   module.

 - The usual fixlets, refactoring and cleanups

x86_cache_for_v6.11_rc1

 - Enable Sub-NUMA clustering to work with resource control on Intel by
   teaching resctrl to handle scopes due to the clustering which
   partitions the L3 cache into sets. Modify and extend the subsystem to
   handle such scopes properly

sched-core-2024-07-16

Scheduler changes for v6.11:

 - Update Daniel Bristot de Oliveira's entry in MAINTAINERS,
   and credit him in CREDITS.

 - Harmonize the lock-yielding behavior on dynamically selected
   preemption models with static ones.

 - Reorganize the code a bit: split out sched/syscalls.c to reduce
   the size of sched/core.c

 - Micro-optimize psi_group_change()

 - Fix set_load_weight() for SCHED_IDLE tasks

 - Misc cleanups & fixes

Signed-off-by: Ingo Molnar <mingo@kernel.org>

objtool-core-2024-07-16

Objtool changes for v6.11:

 - Fix bug that caused objtool to confuse certain memory ops
   added by KASAN instrumentation as stack accesses

 - Various faddr2line optimizations

 - Improve error messages

Signed-off-by: Ingo Molnar <mingo@kernel.org>

6.6.40-114

x86_cpu_for_v6.11_rc1

- Flip the logic to add feature names to /proc/cpuinfo to having to
  explicitly specify the flag if there's a valid reason to show it in
  /proc/cpuinfo

- Switch a bunch of Intel x86 model checking code to the new CPU model
  defines

- Fixes and cleanups

x86_bugs_for_v6.11_rc1

 - Add a spectre_bhi=vmexit mitigation option aimed at cloud
   environments

 - Remove duplicated Spectre cmdline option documentation

 - Add separate macro definitions for syscall handlers which do not
   return in order to address objtool warnings

x86_vmware_for_v6.11_rc1

 - Add a unified VMware hypercall API layer which should be used by all
   callers instead of them doing homegrown solutions. This will provide for
   adding API support for confidential computing solutions like TDX

x86_misc_for_v6.11_rc1

 - Make error checking of AMD SMN accesses more robust in the callers as
   they're the only ones who can interpret the results properly

 - The usual cleanups and fixes, left and right

irq-core-2024-07-15

Updates for the interrupt subsystem:

  - Core:

    - Provide a new mechanism to create interrupt domains. The existing
      interfaces have already too many parameters and it's a pain to expand
      any of this for new required functionality.

      The new function takes a pointer to a data structure as argument. The
      data structure combines all existing parameters and allows for easy
      extension.

      The first extension for this is to handle the instantiation of
      generic interrupt chips at the core level and to allow drivers to
      provide extra init/exit callbacks.

      This is necessary to do the full interrupt chip initialization before
      the new domain is published, so that concurrent usage sites won't see
      a half initialized interrupt domain. Similar problems exist on
      teardown.

      This has turned out to be a real problem due to the deferred and
      parallel probing which was added in recent years.

      Handling this at the core level allows to remove quite some accrued
      boilerplate code in existing drivers and avoids horrible workarounds
      at the driver level.

    - The usual small improvements all over the place

  - Drivers

    - Add support for LAN966x OIC and RZ/Five SoC

    - Split the STM ExtI driver into a microcontroller and a SMP version to
      allow building the latter as a module for multi-platform kernels.

    - Enable MSI support for Armada 370XP on platforms which do not support
      IPIs.

    - The usual small fixes and enhancements all over the place.

x86_build_for_v6.11_rc1

 - Make sure insn support detection uses the proper compiler flag in bi-arch
   builds

locking-core-2024-07-15

Locking changes for v6.11:

 - Jump label fixes, including a perf events fix that originally
   manifested as jump label failures, but was a serialization bug
   at the usage site.

 - Mark down_write*() helpers as __always_inline, to improve
   WCHAN debuggability.

 - Misc cleanups and fixes.

Signed-off-by: Ingo Molnar <mingo@kernel.org>

x86_core_for_v6.11_rc1

 - Cleanup the 8-byte getuser() asm case

x86_cc_for_v6.11_rc1

- Unrelated x86/cc changes queued here to avoid ugly cross-merges and
  conflicts:

   - Carve out CPU hotplug function declarations into a separate header
     with the goal to be able to use the lockdep assertions in a more
     flexible manner

   - As a result, refactor cacheinfo code after carving out a function
     to return the cache ID associated with a given cache level

   -  Cleanups

- Add support to be able to kexec TDX guests. For that

   - Expand ACPI MADT CPU offlining support

   - Add machinery to prepare CoCo guests memory before kexec-ing into a new
     kernel

   - Cleanup, readjust and massage related code

x86_cleanups_for_v6.11_rc1

 - Remove an unused function and the documentation of an already removed
   cmdline parameter

x86_boot_for_v6.11_rc1

 - Add a check to warn when cmdline parsing happens before the final cmdline
   string has been built and thus arguments can get lost

 - Code cleanups and simplifications

x86_alternatives_for_v6.11_rc1

- This is basically PeterZ's idea to nest the alternative macros to avoid the
  need to "spell out" the number of alternates in an ALTERNATIVE_n() macro and
  thus have an ever-increasing complexity in those definitions.

  For ease of bisection, the old macros are converted to the new, nested
  variants in a step-by-step manner so that in case an issue is encountered
  during testing, one can pinpoint the place where it fails easier. Because
  debugging alternatives is a serious pain.

ras_core_for_v6.11_rc1

 - A cleanup and a correction to the error injection driver to inject
   a MCA_MISC value only when one has actually been supplied by the user