drm/i915: Wait for pipe to start on i830 as well
authorVille Syrjälä <ville.syrjala@linux.intel.com>
Wed, 29 Nov 2017 15:37:31 +0000 (17:37 +0200)
committerVille Syrjälä <ville.syrjala@linux.intel.com>
Fri, 1 Dec 2017 15:05:16 +0000 (17:05 +0200)
We should make sure the pipe has fully started when we enable it from
the i830 "power well". Otherwise theoretically i830 could also hit
problems with vblank timestamps jumping around (since we skip the
wait during modeset on i830). Additionally moving planes between the
pipes etc. might not work correctly until both pipes are actually up and
running.

v2: Less pointless duplication in the code (Chris)

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171129153732.3612-2-ville.syrjala@linux.intel.com
drivers/gpu/drm/i915/intel_display.c

index f712418..7678caa 100644 (file)
@@ -14718,6 +14718,7 @@ int intel_modeset_init(struct drm_device *dev)
 
 void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)
 {
+       struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
        /* 640x480@60Hz, ~25175 kHz */
        struct dpll clock = {
                .m1 = 18,
@@ -14781,6 +14782,8 @@ void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)
 
        I915_WRITE(PIPECONF(pipe), PIPECONF_ENABLE | PIPECONF_PROGRESSIVE);
        POSTING_READ(PIPECONF(pipe));
+
+       intel_wait_for_pipe_scanline_moving(crtc);
 }
 
 void i830_disable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)