| Message ID | 1490366919-34715-1-git-send-email-daniele.ceraolospurio@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com> On 3/24/2017 8:18 PM, Daniele Ceraolo Spurio wrote: > The forcewake_get call in the guc_send_mmio function was added to > avoid getting and releasing forcewake on each register access. > While this makes sense, all GuC registers are in the blitter range > so no need to wake all the wells. > > Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> > --- > drivers/gpu/drm/i915/intel_uc.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/i915/intel_uc.c b/drivers/gpu/drm/i915/intel_uc.c > index 4a872cd..e016227 100644 > --- a/drivers/gpu/drm/i915/intel_uc.c > +++ b/drivers/gpu/drm/i915/intel_uc.c > @@ -245,7 +245,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len) > return -EINVAL; > > mutex_lock(&guc->send_mutex); > - intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL); > + intel_uncore_forcewake_get(dev_priv, FORCEWAKE_BLITTER); > > dev_priv->guc.action_count += 1; > dev_priv->guc.action_cmd = action[0]; > @@ -283,7 +283,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len) > } > dev_priv->guc.action_status = status; > > - intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); > + intel_uncore_forcewake_put(dev_priv, FORCEWAKE_BLITTER); > mutex_unlock(&guc->send_mutex); > > return ret;
On Fri, Mar 24, 2017 at 09:12:13PM +0530, Kamble, Sagar A wrote:
> Reviewed-by: Sagar Arun Kamble [1]<sagar.a.kamble@intel.com>
and pushed.
Sagar, you might want to turn off sending html email for intel-gfx@.
-Chris
diff --git a/drivers/gpu/drm/i915/intel_uc.c b/drivers/gpu/drm/i915/intel_uc.c index 4a872cd..e016227 100644 --- a/drivers/gpu/drm/i915/intel_uc.c +++ b/drivers/gpu/drm/i915/intel_uc.c @@ -245,7 +245,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len) return -EINVAL; mutex_lock(&guc->send_mutex); - intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL); + intel_uncore_forcewake_get(dev_priv, FORCEWAKE_BLITTER); dev_priv->guc.action_count += 1; dev_priv->guc.action_cmd = action[0]; @@ -283,7 +283,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len) } dev_priv->guc.action_status = status; - intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); + intel_uncore_forcewake_put(dev_priv, FORCEWAKE_BLITTER); mutex_unlock(&guc->send_mutex); return ret;
The forcewake_get call in the guc_send_mmio function was added to avoid getting and releasing forcewake on each register access. While this makes sense, all GuC registers are in the blitter range so no need to wake all the wells. Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> --- drivers/gpu/drm/i915/intel_uc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)