[PATCH 2/2] drm/amd/powerplay: add registry key to disable ACG
Huang Rui
ray.huang at amd.com
Thu Apr 19 07:47:55 UTC 2018
On Thu, Apr 19, 2018 at 03:11:36PM +0800, Kenneth Feng wrote:
> For the dummy ACG fuses,need to disable ACG, otherwise
> corruption will be caused.
>
> Change-Id: I03db372bb6e934e9230f0745aa19c683adfdeccb
> Signed-off-by: Kenneth Feng <kenneth.feng at amd.com>
> ---
> drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c | 5 +++++
> drivers/gpu/drm/amd/powerplay/inc/hwmgr.h | 1 +
> 2 files changed, 6 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c
> index 7fa1ba8..888ddca 100644
> --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c
> +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c
> @@ -224,6 +224,11 @@ static int append_vbios_pptable(struct pp_hwmgr *hwmgr, PPTable_t *ppsmc_pptable
> ppsmc_pptable->AcgGfxclkSpreadPercent = smc_dpm_table.acggfxclkspreadpercent;
> ppsmc_pptable->AcgGfxclkSpreadFreq = smc_dpm_table.acggfxclkspreadfreq;
>
> + /* 0xFFFF will disable the ACG feature */
> + if (!(hwmgr->feature_mask & PP_ACG_MASK)) {
> + ppsmc_pptable->AcgThresholdFreqHigh = 0xFFFF;
> + ppsmc_pptable->AcgThresholdFreqLow = 0xFFFF;
> + }
>
> return 0;
> }
> diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
> index 9b3dd7d..9afa5fe 100644
> --- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
> +++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
> @@ -82,6 +82,7 @@ enum PP_FEATURE_MASK {
> PP_SOCCLK_DPM_MASK = 0x1000,
> PP_DCEFCLK_DPM_MASK = 0x2000,
> PP_OVERDRIVE_MASK = 0x4000,
> + PP_ACG_MASK = 0x10000000,
Actually, we use bit map to indicate different power features, and use bit
one by one on that feature_mask.
PP_DCEFCLK_DPM_MASK = 1 << 13
PP_OVERDRIVE_MASK = 1 << 14
And in another patch, I add gfxoff feature mask below:
PP_GFXOFF_MASK = 0x8000 = 1 << 15
So I suggest you can set
PP_ACG_MASK = 0x10000 = 1 << 16 here.
Thanks,
Ray
More information about the amd-gfx
mailing list