drm/radeon/kms: improve pflip precision on r1xx-r4xx

The update pending bit has a separate enable bit.

Cc: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drivers/gpu/drm/radeon/r100.c

@@ -76,6 +76,8 @@ void r100_pre_page_flip(struct radeon_device *rdev, int crtc)
 	/* make sure flip is at vb rather than hb */
 	tmp = RREG32(RADEON_CRTC_OFFSET_CNTL + radeon_crtc->crtc_offset);
 	tmp &= ~RADEON_CRTC_OFFSET_FLIP_CNTL;
+	/* make sure pending bit is asserted */
+	tmp |= RADEON_CRTC_GUI_TRIG_OFFSET_LEFT_EN;
 	WREG32(RADEON_CRTC_OFFSET_CNTL + radeon_crtc->crtc_offset, tmp);
 
 	/* set pageflip to happen as late as possible in the vblank interval.
@@ -104,9 +106,9 @@ u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* update the scanout addresses */
 	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
 
-	/* Note: We don't wait for update_pending to assert, as this never
-	 * happens for some reason on R1xx - R4xx. Adds a bit of imprecision.
-	 */
+	/* Wait for update_pending to go high. */
+	while (!(RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET));
+	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~RADEON_CRTC_OFFSET__OFFSET_LOCK;

drivers/gpu/drm/radeon/radeon_reg.h

@@ -510,6 +510,8 @@
 #       define RADEON_CRTC_TILE_EN                      (1 << 15)
 #       define RADEON_CRTC_OFFSET_FLIP_CNTL             (1 << 16)
 #       define RADEON_CRTC_STEREO_OFFSET_EN             (1 << 17)
+#       define RADEON_CRTC_GUI_TRIG_OFFSET_LEFT_EN      (1 << 28)
+#       define RADEON_CRTC_GUI_TRIG_OFFSET_RIGHT_EN     (1 << 29)
 
 #define R300_CRTC_TILE_X0_Y0	            0x0350
 #define R300_CRTC2_TILE_X0_Y0	            0x0358