[Intel-gfx] [PATCH v2 3/8] drm/i915: Use cdclk_state->voltage on VLV/CHV
Ville Syrjälä
ville.syrjala at linux.intel.com
Fri Oct 20 16:20:20 UTC 2017
On Thu, Oct 19, 2017 at 04:42:52PM -0700, Rodrigo Vivi wrote:
> On Thu, Oct 19, 2017 at 05:43:29PM +0000, Ville Syrjala wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> >
> > Store the punit DSPFREQUAR value into cdclk_state->voltage on
> > VLV/CHV. Since we can actually read that out from the hardware
> > this can give us a bit more cross checking between the hardware
> > and software state.
> >
> > v2: Don't break waiting for cdclk change on VLV/CHV
> >
> > Cc: Mika Kahola <mika.kahola at intel.com>
> > Cc: Manasi Navare <manasi.d.navare at intel.com>
> > Cc: Rodrigo Vivi <rodrigo.vivi at intel.com>
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > ---
> > drivers/gpu/drm/i915/intel_cdclk.c | 66 +++++++++++++++++++++++++++++---------
> > 1 file changed, 50 insertions(+), 16 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/intel_cdclk.c b/drivers/gpu/drm/i915/intel_cdclk.c
> > index 522222f8bb50..9cc4374797eb 100644
> > --- a/drivers/gpu/drm/i915/intel_cdclk.c
> > +++ b/drivers/gpu/drm/i915/intel_cdclk.c
> > @@ -437,13 +437,45 @@ static int vlv_calc_cdclk(struct drm_i915_private *dev_priv, int min_cdclk)
> > return 200000;
> > }
> >
> > +static u8 vlv_calc_voltage(struct drm_i915_private *dev_priv, int cdclk)
> > +{
> > + if (IS_VALLEYVIEW(dev_priv)) {
> > + if (cdclk >= 320000) /* jump to highest voltage for 400MHz too */
> > + return 2;
> > + else if (cdclk >= 266667)
> > + return 1;
> > + else
> > + return 0;
> > + } else {
> > + /*
> > + * Specs are full of misinformation, but testing on actual
> > + * hardware has shown that we just need to write the desired
> > + * CCK divider into the Punit register.
> > + */
> > + return DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, cdclk) - 1;
> > + }
> > +}
> > +
> > static void vlv_get_cdclk(struct drm_i915_private *dev_priv,
> > struct intel_cdclk_state *cdclk_state)
> > {
> > + u32 val;
> > +
> > cdclk_state->vco = vlv_get_hpll_vco(dev_priv);
> > cdclk_state->cdclk = vlv_get_cck_clock(dev_priv, "cdclk",
> > CCK_DISPLAY_CLOCK_CONTROL,
> > cdclk_state->vco);
> > +
> > + mutex_lock(&dev_priv->pcu_lock);
> > + val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
> > + mutex_unlock(&dev_priv->pcu_lock);
> > +
> > + if (IS_VALLEYVIEW(dev_priv))
> > + cdclk_state->voltage = (val & DSPFREQGUAR_MASK) >>
> > + DSPFREQGUAR_SHIFT;
> > + else
> > + cdclk_state->voltage = (val & DSPFREQGUAR_MASK_CHV) >>
> > + DSPFREQGUAR_SHIFT_CHV;
>
> Oh interresting that vlv is the only one where we can actually read it. :(
>
> I was going to ask pointers to spec, but just by {vlv,chv}_set_cdclk it was
> possible to verify this is right.
Yeah, the iosf sideband mailbox is much nicer because you have actual
read vs. write opcodes in addition to the port+register offset
information which just specify what you want to access.
>
> > }
> >
> > static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv)
> > @@ -486,7 +518,19 @@ static void vlv_set_cdclk(struct drm_i915_private *dev_priv,
> > const struct intel_cdclk_state *cdclk_state)
> > {
> > int cdclk = cdclk_state->cdclk;
> > - u32 val, cmd;
> > + u32 val, cmd = cdclk_state->voltage;
> > +
> > + switch (cdclk) {
> > + case 400000:
> > + case 333333:
> > + case 320000:
> > + case 266667:
> > + case 200000:
> > + break;
> > + default:
> > + MISSING_CASE(cdclk);
> > + return;
> > + }
>
> I believe this should be in a separated patch, no?!
> Seems something we were already missing anyways.
I was going back and forth between removing this from the
chv_set_cdclk(), and in the end I decided to keep it. And looks like I
then accidentally added it into vlv_set_cdclk() as well thinking both
had it originally. I guess I'm still leaning towards keeping it, so
I'll go ahead and split this out to a separate patch.
>
> But anyways, if you decide to split or not
>
> Reviewed-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
>
> >
> > /* There are cases where we can end up here with power domains
> > * off and a CDCLK frequency other than the minimum, like when
> > @@ -496,13 +540,6 @@ static void vlv_set_cdclk(struct drm_i915_private *dev_priv,
> > */
> > intel_display_power_get(dev_priv, POWER_DOMAIN_PIPE_A);
> >
> > - if (cdclk >= 320000) /* jump to highest voltage for 400MHz too */
> > - cmd = 2;
> > - else if (cdclk == 266667)
> > - cmd = 1;
> > - else
> > - cmd = 0;
> > -
> > mutex_lock(&dev_priv->pcu_lock);
> > val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
> > val &= ~DSPFREQGUAR_MASK;
> > @@ -562,7 +599,7 @@ static void chv_set_cdclk(struct drm_i915_private *dev_priv,
> > const struct intel_cdclk_state *cdclk_state)
> > {
> > int cdclk = cdclk_state->cdclk;
> > - u32 val, cmd;
> > + u32 val, cmd = cdclk_state->voltage;
> >
> > switch (cdclk) {
> > case 333333:
> > @@ -583,13 +620,6 @@ static void chv_set_cdclk(struct drm_i915_private *dev_priv,
> > */
> > intel_display_power_get(dev_priv, POWER_DOMAIN_PIPE_A);
> >
> > - /*
> > - * Specs are full of misinformation, but testing on actual
> > - * hardware has shown that we just need to write the desired
> > - * CCK divider into the Punit register.
> > - */
> > - cmd = DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, cdclk) - 1;
> > -
> > mutex_lock(&dev_priv->pcu_lock);
> > val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
> > val &= ~DSPFREQGUAR_MASK_CHV;
> > @@ -1859,11 +1889,15 @@ static int vlv_modeset_calc_cdclk(struct drm_atomic_state *state)
> > cdclk = vlv_calc_cdclk(dev_priv, min_cdclk);
> >
> > intel_state->cdclk.logical.cdclk = cdclk;
> > + intel_state->cdclk.logical.voltage =
> > + vlv_calc_voltage(dev_priv, cdclk);
> >
> > if (!intel_state->active_crtcs) {
> > cdclk = vlv_calc_cdclk(dev_priv, 0);
> >
> > intel_state->cdclk.actual.cdclk = cdclk;
> > + intel_state->cdclk.actual.voltage =
> > + vlv_calc_voltage(dev_priv, cdclk);
> > } else {
> > intel_state->cdclk.actual =
> > intel_state->cdclk.logical;
> > --
> > 2.13.6
> >
--
Ville Syrjälä
Intel OTC
More information about the Intel-gfx
mailing list