Re: [PATCH v2 12/17] cpufreq/amd-pstate: Cache CPPC request in shared mem case too
From: Gautham R. Shenoy
Date: Mon Feb 17 2025 - 08:54:43 EST
On Fri, Feb 14, 2025 at 06:52:39PM -0600, Mario Limonciello wrote:
> From: Mario Limonciello <mario.limonciello@xxxxxxx>
>
> In order to prevent a potential write for shmem_update_perf()
> cache the request into the cppc_req_cached variable normally only
> used for the MSR case.
>
> This adds symmetry into the code and potentially avoids extra writes.
>
> Reviewed-by: Dhananjay Ugwekar <dhananjay.ugwekar@xxxxxxx>
> Signed-off-by: Mario Limonciello <mario.limonciello@xxxxxxx>
> ---
> drivers/cpufreq/amd-pstate.c | 22 +++++++++++++++++++++-
> 1 file changed, 21 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
> index 0a7e69fd32dbf..9517da9b7e692 100644
> --- a/drivers/cpufreq/amd-pstate.c
> +++ b/drivers/cpufreq/amd-pstate.c
> @@ -496,6 +496,8 @@ static int shmem_update_perf(struct amd_cpudata *cpudata, u8 min_perf,
> u8 des_perf, u8 max_perf, u8 epp, bool fast_switch)
> {
> struct cppc_perf_ctrls perf_ctrls;
> + u64 value, prev;
> + int ret;
>
> if (cppc_state == AMD_PSTATE_ACTIVE) {
> int ret = shmem_set_epp(cpudata, epp);
> @@ -504,11 +506,29 @@ static int shmem_update_perf(struct amd_cpudata *cpudata, u8 min_perf,
> return ret;
> }
>
> + value = prev = READ_ONCE(cpudata->cppc_req_cached);
> +
> + value &= ~(AMD_CPPC_MAX_PERF_MASK | AMD_CPPC_MIN_PERF_MASK |
> + AMD_CPPC_DES_PERF_MASK | AMD_CPPC_EPP_PERF_MASK);
> + value |= FIELD_PREP(AMD_CPPC_MAX_PERF_MASK, max_perf);
> + value |= FIELD_PREP(AMD_CPPC_DES_PERF_MASK, des_perf);
> + value |= FIELD_PREP(AMD_CPPC_MIN_PERF_MASK, min_perf);
> + value |= FIELD_PREP(AMD_CPPC_EPP_PERF_MASK, epp);
Perhaps this can be further simplified in the future using a new macro
FIELD_SET(var, mask, val) where
#define FIELD_SET(var, mask, val) var = ((var) & ~(mask)) | FIELD_PREP(mask, val))
For this patch,
Reviewed-by: Gautham R. Shenoy <gautham.shenoy@xxxxxxx>
--
Thanks and Regards
gautham.