On Tue, Aug 5, 2025 at 10:38 AM Asad Kamal <asad.ka...@amd.com> wrote:
>
> Increase cache interval time to 50 ms while fetching system
> metrics table for smu_v13_0_12
>
Can we move the caching up to the device independent level? It would
be nice to have this for other GPUs.
Alex
> Signed-off-by: Asad Kamal <asad.ka...@amd.com>
> ---
> drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
> b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
> index fca50f6a8ef6..2ee61c7e8bc8 100644
> --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
> +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
> @@ -372,7 +372,7 @@ static int smu_v13_0_12_get_system_metrics_table(struct
> smu_context *smu, void *
> if (bypass_cache ||
> !smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time ||
> time_after(jiffies,
> smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time
> +
> - msecs_to_jiffies(1))) {
> + msecs_to_jiffies(50))) {
> ret = smu_cmn_send_smc_msg(smu,
> SMU_MSG_GetSystemMetricsTable, NULL);
> if (ret) {
> dev_info(smu->adev->dev,
> --
> 2.46.0
>
