| Message ID | 1430408363-20905-8-git-send-email-damien.lespiau@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, Apr 30, 2015 at 04:39:22PM +0100, Damien Lespiau wrote: > Signed-off-by: Damien Lespiau <damien.lespiau@intel.com> Why? I.e. way too terse commit message. -Daniel > --- > drivers/gpu/drm/i915/intel_display.c | 8 ++++++-- > 1 file changed, 6 insertions(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c > index 9c8338f..2cbae9a 100644 > --- a/drivers/gpu/drm/i915/intel_display.c > +++ b/drivers/gpu/drm/i915/intel_display.c > @@ -5525,6 +5525,7 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq) > { > int ret; > u32 val, freq_select, freq_decimal, pcu_ack; > + bool do_pcu_ack = true; > > DRM_DEBUG_DRIVER("Changing CDCLK to %dKHz\n", freq); > > @@ -5534,8 +5535,8 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq) > ret = sandybridge_pcode_read(dev_priv, SKL_PCODE_CDCLK_CONTROL, &val); > mutex_unlock(&dev_priv->rps.hw_lock); > if (ret || !(val & SKL_CDCLK_READY_FOR_CHANGE)) { > - DRM_ERROR("failed to inform PCU about cdclk change\n"); > - return; > + DRM_DEBUG_KMS("failed to inform PCU about cdclk change\n"); > + do_pcu_ack = false; > } > > /* set CDCLK_CTL */ > @@ -5567,6 +5568,9 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq) > I915_WRITE(CDCLK_CTL, freq_select | freq_decimal); > POSTING_READ(CDCLK_CTL); > > + if (!do_pcu_ack) > + return; > + > /* inform PCU of the change */ > mutex_lock(&dev_priv->rps.hw_lock); > sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, pcu_ack); > -- > 2.1.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
On Wed, May 06, 2015 at 12:53:14PM +0200, Daniel Vetter wrote: > On Thu, Apr 30, 2015 at 04:39:22PM +0100, Damien Lespiau wrote: > > Signed-off-by: Damien Lespiau <damien.lespiau@intel.com> > > Why? I.e. way too terse commit message. Because I couldn't get the PCU to answer to the handcheck we should be doing. Ville noticed (and, to be fair, I knew this as well but didn't have time to investigate before an early week-end) that I'm not exactly doing what's documented and we need to wait in a busy loop for the PCU ack. Will try that when coming back at it.
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 9c8338f..2cbae9a 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -5525,6 +5525,7 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq) { int ret; u32 val, freq_select, freq_decimal, pcu_ack; + bool do_pcu_ack = true; DRM_DEBUG_DRIVER("Changing CDCLK to %dKHz\n", freq); @@ -5534,8 +5535,8 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq) ret = sandybridge_pcode_read(dev_priv, SKL_PCODE_CDCLK_CONTROL, &val); mutex_unlock(&dev_priv->rps.hw_lock); if (ret || !(val & SKL_CDCLK_READY_FOR_CHANGE)) { - DRM_ERROR("failed to inform PCU about cdclk change\n"); - return; + DRM_DEBUG_KMS("failed to inform PCU about cdclk change\n"); + do_pcu_ack = false; } /* set CDCLK_CTL */ @@ -5567,6 +5568,9 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq) I915_WRITE(CDCLK_CTL, freq_select | freq_decimal); POSTING_READ(CDCLK_CTL); + if (!do_pcu_ack) + return; + /* inform PCU of the change */ mutex_lock(&dev_priv->rps.hw_lock); sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, pcu_ack);
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com> --- drivers/gpu/drm/i915/intel_display.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)