Re: [PATCH] x86/cstate: Replace vendor check with X86_FEATURE_MWAIT in ffh_cstate_init

From: Tony W Wang-oc
Date: Mon Jul 18 2022 - 23:51:37 EST


On 2022/7/6 上午2:45, Rafael J. Wysocki wrote:
On Fri, Jul 1, 2022 at 9:05 AM Tony W Wang-oc <TonyWWang-oc@xxxxxxxxxxx> wrote:

The original commit 991528d73486 ("ACPI: Processor native C-states using
MWAIT") has a vendor check for Intel in the function of ffh_cstate_init().

Commit 5209654a46ee ("x86/ACPI/cstate: Allow ACPI C1 FFH MWAIT use on AMD
systems") and commit 280b68a3b3b9 ("x86/cstate: Allow ACPI C1 FFH MWAIT
use on Hygon systems") add vendor check for AMD and HYGON in the function
of ffh_cstate_init().

Recent Zhaoxin and Centaur CPUs support MONITOR/MWAIT instructions that
can be used for ACPI Cx state in the same way as Intel. So expected to
add the support of these CPUs in the function of ffh_cstate_init() too.

The CPU feature X86_FEATURE_MWAIT indicates processor supports MONITOR/
MWAIT instructions. So the check for many CPU vendors in ffh_cstate_init()
is unnecessary, use X86_FEATURE_MWAIT to replace the CPU vendor check.

Signed-off-by: Tony W Wang-oc <TonyWWang-oc@xxxxxxxxxxx>

Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>


Thanks a lot.
Add some original commit authors this patch related to.

or please let me know if I'm expected to pick up this one. Thanks!


Yes for me. Thanks a lot.


---
arch/x86/kernel/acpi/cstate.c | 6 +-----
1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c
index 7945eae..a64c38f 100644
--- a/arch/x86/kernel/acpi/cstate.c
+++ b/arch/x86/kernel/acpi/cstate.c
@@ -209,11 +209,7 @@ EXPORT_SYMBOL_GPL(acpi_processor_ffh_cstate_enter);

static int __init ffh_cstate_init(void)
{
- struct cpuinfo_x86 *c = &boot_cpu_data;
-
- if (c->x86_vendor != X86_VENDOR_INTEL &&
- c->x86_vendor != X86_VENDOR_AMD &&
- c->x86_vendor != X86_VENDOR_HYGON)
+ if (!boot_cpu_has(X86_FEATURE_MWAIT))
return -1;

cpu_cstate_entry = alloc_percpu(struct cstate_entry);
--
2.7.4

.