[PATCH v2 0/2] x86/fpu: Make AMX state ready for CPU idle

[Chang S. Bae]

AMX state is a large state (at least 8KB or more). Entering CPU idle with
this non-initialized large state may result in shallow states while a
deeper low-power state is available.

We can confirm this behavior is implementation-specific. Section 3.3 in [1]
will be updated to clarify this. Sorry for the delay in this conclusion
and the update here.

Changes from v1 [2]:
* Drop the code to call TILERELEASE from arch_cpu_idle_enter(). Instead,
  do it from the intel-idle driver, as the behavior is *not* architectural.
* Simplify the driver code change (Rui).
* Improve the dynamic state flag check in the helper function.
* Drop the new C-state table as it is done by another patchset [3].
* Drop the opcode map changes as tip's x86/misc [4] has merged it already.

With the last two changes, the series depends on these:
* The C-state table update for Sapphire Rapids in linux-pm [3].
* The update of the x86 opcode map in the tip's branch [4].

The series on top of the above is available here:
  git://github.com/intel/amx-linux.git tilerelease

Thanks,
Chang

[1]: Intel Architecture Instruction Set Extension Programming Reference
     May 2021, https://software.intel.com/content/dam/develop/external/us/en/documents-tps/architecture-instruction-set-extensions-programming-reference.pdf
[2]: https://lore.kernel.org/lkml/20211104225226.5031-1-chang.seok.bae@xxxxxxxxx/
[3]: https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git/commit/?h=linux-next&id=9edf3c0ffef0ec1bed8300315852b5c6a0997130
[4]: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=x86/misc&id=9dd94df75b30eca03ed2151dd5bbc152a6f19abf

Chang S. Bae (2):
  x86/fpu: Add a helper to prepare AMX state for low-power CPU idle
  intel_idle: Add a new flag to initialize the AMX state

 arch/x86/include/asm/fpu/api.h       |  2 ++
 arch/x86/include/asm/special_insns.h |  9 +++++++++
 arch/x86/kernel/fpu/core.c           | 15 +++++++++++++++
 drivers/idle/intel_idle.c            | 18 ++++++++++++++++--
 4 files changed, 42 insertions(+), 2 deletions(-)

-- 
2.17.1