Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "Okay this is the big one, I was stalled on the fbdev pull req as I
  stupidly let fbdev guys merge a patch I required to fix a warning with
  some patches I had, they ended up merging the patch from the wrong
  place, but the warning should be fixed.  In future I'll just take the
  patch myself!

  Outside drm:

  There are some snd changes for the HDMI audio interactions on haswell,
  they've been acked for inclusion via my tree.  This relies on the
  wound/wait tree from Ingo which is already merged.

  Major changes:

  AMD finally released the dynamic power management code for all their
  GPUs from r600->present day, this is great, off by default for now but
  also a huge amount of code, in fact it is most of this pull request.

  Since it landed there has been a lot of community testing and Alex has
  sent a lot of fixes for any bugs found so far.  I suspect radeon might
  now be the biggest kernel driver ever :-P p.s.  radeon.dpm=1 to enable
  dynamic powermanagement for anyone.

  New drivers:

  Renesas r-car display unit.

  Other highlights:

   - core: GEM CMA prime support, use new w/w mutexs for TTM
     reservations, cursor hotspot, doc updates
   - dvo chips: chrontel 7010B support
   - i915: Haswell (fbc, ips, vecs, watermarks, audio powerwell),
     Valleyview (enabled by default, rc6), lots of pll reworking, 30bpp
     support (this time for sure)
   - nouveau: async buffer object deletion, context/register init
     updates, kernel vp2 engine support, GF117 support, GK110 accel
     support (with external nvidia ucode), context cleanups.
   - exynos: memory leak fixes, Add S3C64XX SoC series support, device
     tree updates, common clock framework support,
   - qxl: cursor hotspot support, multi-monitor support, suspend/resume
     support
   - mgag200: hw cursor support, g200 mode limiting
   - shmobile: prime support
   - tegra: fixes mostly

  I've been banging on this quite a lot due to the size of it, and it
  seems to okay on everything I've tested it on."

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (811 commits)
  drm/radeon/dpm: implement vblank_too_short callback for si
  drm/radeon/dpm: implement vblank_too_short callback for cayman
  drm/radeon/dpm: implement vblank_too_short callback for btc
  drm/radeon/dpm: implement vblank_too_short callback for evergreen
  drm/radeon/dpm: implement vblank_too_short callback for 7xx
  drm/radeon/dpm: add checks against vblank time
  drm/radeon/dpm: add helper to calculate vblank time
  drm/radeon: remove stray line in old pm code
  drm/radeon/dpm: fix display_gap programming on rv7xx
  drm/nvc0/gr: fix gpc firmware regression
  drm/nouveau: fix minor thinko causing bo moves to not be async on kepler
  drm/radeon/dpm: implement force performance level for TN
  drm/radeon/dpm: implement force performance level for ON/LN
  drm/radeon/dpm: implement force performance level for SI
  drm/radeon/dpm: implement force performance level for cayman
  drm/radeon/dpm: implement force performance levels for 7xx/eg/btc
  drm/radeon/dpm: add infrastructure to force performance levels
  drm/radeon: fix surface setup on r1xx
  drm/radeon: add support for 3d perf states on older asics
  drm/radeon: set default clocks for SI when DPM is disabled
  ...

1 files changed, 132 insertions, 92 deletions

diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 21c7d7b26e55..8ea1573ae820 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -937,13 +937,16 @@ struct rv515_watermark {
 };
 
 static void rv515_crtc_bandwidth_compute(struct radeon_device *rdev,
-				  struct radeon_crtc *crtc,
-				  struct rv515_watermark *wm)
+					 struct radeon_crtc *crtc,
+					 struct rv515_watermark *wm,
+					 bool low)
 {
 	struct drm_display_mode *mode = &crtc->base.mode;
 	fixed20_12 a, b, c;
 	fixed20_12 pclk, request_fifo_depth, tolerable_latency, estimated_width;
 	fixed20_12 consumption_time, line_time, chunk_time, read_delay_latency;
+	fixed20_12 sclk;
+	u32 selected_sclk;
 
 	if (!crtc->base.enabled) {
 		/* FIXME: wouldn't it better to set priority mark to maximum */
@@ -951,6 +954,18 @@ static void rv515_crtc_bandwidth_compute(struct radeon_device *rdev,
 		return;
 	}
 
+	/* rv6xx, rv7xx */
+	if ((rdev->family >= CHIP_RV610) &&
+	    (rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled)
+		selected_sclk = radeon_dpm_get_sclk(rdev, low);
+	else
+		selected_sclk = rdev->pm.current_sclk;
+
+	/* sclk in Mhz */
+	a.full = dfixed_const(100);
+	sclk.full = dfixed_const(selected_sclk);
+	sclk.full = dfixed_div(sclk, a);
+
 	if (crtc->vsc.full > dfixed_const(2))
 		wm->num_line_pair.full = dfixed_const(2);
 	else
@@ -1016,7 +1031,7 @@ static void rv515_crtc_bandwidth_compute(struct radeon_device *rdev,
 	 * sclk = system clock(Mhz)
 	 */
 	a.full = dfixed_const(600 * 1000);
-	chunk_time.full = dfixed_div(a, rdev->pm.sclk);
+	chunk_time.full = dfixed_div(a, sclk);
 	read_delay_latency.full = dfixed_const(1000);
 
 	/* Determine the worst case latency
@@ -1077,152 +1092,177 @@ static void rv515_crtc_bandwidth_compute(struct radeon_device *rdev,
 	}
 }
 
-void rv515_bandwidth_avivo_update(struct radeon_device *rdev)
+static void rv515_compute_mode_priority(struct radeon_device *rdev,
+					struct rv515_watermark *wm0,
+					struct rv515_watermark *wm1,
+					struct drm_display_mode *mode0,
+					struct drm_display_mode *mode1,
+					u32 *d1mode_priority_a_cnt,
+					u32 *d2mode_priority_a_cnt)
 {
-	struct drm_display_mode *mode0 = NULL;
-	struct drm_display_mode *mode1 = NULL;
-	struct rv515_watermark wm0;
-	struct rv515_watermark wm1;
-	u32 tmp;
-	u32 d1mode_priority_a_cnt = MODE_PRIORITY_OFF;
-	u32 d2mode_priority_a_cnt = MODE_PRIORITY_OFF;
 	fixed20_12 priority_mark02, priority_mark12, fill_rate;
 	fixed20_12 a, b;
 
-	if (rdev->mode_info.crtcs[0]->base.enabled)
-		mode0 = &rdev->mode_info.crtcs[0]->base.mode;
-	if (rdev->mode_info.crtcs[1]->base.enabled)
-		mode1 = &rdev->mode_info.crtcs[1]->base.mode;
-	rs690_line_buffer_adjust(rdev, mode0, mode1);
-
-	rv515_crtc_bandwidth_compute(rdev, rdev->mode_info.crtcs[0], &wm0);
-	rv515_crtc_bandwidth_compute(rdev, rdev->mode_info.crtcs[1], &wm1);
-
-	tmp = wm0.lb_request_fifo_depth;
-	tmp |= wm1.lb_request_fifo_depth << 16;
-	WREG32(LB_MAX_REQ_OUTSTANDING, tmp);
+	*d1mode_priority_a_cnt = MODE_PRIORITY_OFF;
+	*d2mode_priority_a_cnt = MODE_PRIORITY_OFF;
 
 	if (mode0 && mode1) {
-		if (dfixed_trunc(wm0.dbpp) > 64)
-			a.full = dfixed_div(wm0.dbpp, wm0.num_line_pair);
+		if (dfixed_trunc(wm0->dbpp) > 64)
+			a.full = dfixed_div(wm0->dbpp, wm0->num_line_pair);
 		else
-			a.full = wm0.num_line_pair.full;
-		if (dfixed_trunc(wm1.dbpp) > 64)
-			b.full = dfixed_div(wm1.dbpp, wm1.num_line_pair);
+			a.full = wm0->num_line_pair.full;
+		if (dfixed_trunc(wm1->dbpp) > 64)
+			b.full = dfixed_div(wm1->dbpp, wm1->num_line_pair);
 		else
-			b.full = wm1.num_line_pair.full;
+			b.full = wm1->num_line_pair.full;
 		a.full += b.full;
-		fill_rate.full = dfixed_div(wm0.sclk, a);
-		if (wm0.consumption_rate.full > fill_rate.full) {
-			b.full = wm0.consumption_rate.full - fill_rate.full;
-			b.full = dfixed_mul(b, wm0.active_time);
+		fill_rate.full = dfixed_div(wm0->sclk, a);
+		if (wm0->consumption_rate.full > fill_rate.full) {
+			b.full = wm0->consumption_rate.full - fill_rate.full;
+			b.full = dfixed_mul(b, wm0->active_time);
 			a.full = dfixed_const(16);
 			b.full = dfixed_div(b, a);
-			a.full = dfixed_mul(wm0.worst_case_latency,
-						wm0.consumption_rate);
+			a.full = dfixed_mul(wm0->worst_case_latency,
+						wm0->consumption_rate);
 			priority_mark02.full = a.full + b.full;
 		} else {
-			a.full = dfixed_mul(wm0.worst_case_latency,
-						wm0.consumption_rate);
+			a.full = dfixed_mul(wm0->worst_case_latency,
+						wm0->consumption_rate);
 			b.full = dfixed_const(16 * 1000);
 			priority_mark02.full = dfixed_div(a, b);
 		}
-		if (wm1.consumption_rate.full > fill_rate.full) {
-			b.full = wm1.consumption_rate.full - fill_rate.full;
-			b.full = dfixed_mul(b, wm1.active_time);
+		if (wm1->consumption_rate.full > fill_rate.full) {
+			b.full = wm1->consumption_rate.full - fill_rate.full;
+			b.full = dfixed_mul(b, wm1->active_time);
 			a.full = dfixed_const(16);
 			b.full = dfixed_div(b, a);
-			a.full = dfixed_mul(wm1.worst_case_latency,
-						wm1.consumption_rate);
+			a.full = dfixed_mul(wm1->worst_case_latency,
+						wm1->consumption_rate);
 			priority_mark12.full = a.full + b.full;
 		} else {
-			a.full = dfixed_mul(wm1.worst_case_latency,
-						wm1.consumption_rate);
+			a.full = dfixed_mul(wm1->worst_case_latency,
+						wm1->consumption_rate);
 			b.full = dfixed_const(16 * 1000);
 			priority_mark12.full = dfixed_div(a, b);
 		}
-		if (wm0.priority_mark.full > priority_mark02.full)
-			priority_mark02.full = wm0.priority_mark.full;
+		if (wm0->priority_mark.full > priority_mark02.full)
+			priority_mark02.full = wm0->priority_mark.full;
 		if (dfixed_trunc(priority_mark02) < 0)
 			priority_mark02.full = 0;
-		if (wm0.priority_mark_max.full > priority_mark02.full)
-			priority_mark02.full = wm0.priority_mark_max.full;
-		if (wm1.priority_mark.full > priority_mark12.full)
-			priority_mark12.full = wm1.priority_mark.full;
+		if (wm0->priority_mark_max.full > priority_mark02.full)
+			priority_mark02.full = wm0->priority_mark_max.full;
+		if (wm1->priority_mark.full > priority_mark12.full)
+			priority_mark12.full = wm1->priority_mark.full;
 		if (dfixed_trunc(priority_mark12) < 0)
 			priority_mark12.full = 0;
-		if (wm1.priority_mark_max.full > priority_mark12.full)
-			priority_mark12.full = wm1.priority_mark_max.full;
-		d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
-		d2mode_priority_a_cnt = dfixed_trunc(priority_mark12);
+		if (wm1->priority_mark_max.full > priority_mark12.full)
+			priority_mark12.full = wm1->priority_mark_max.full;
+		*d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
+		*d2mode_priority_a_cnt = dfixed_trunc(priority_mark12);
 		if (rdev->disp_priority == 2) {
-			d1mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
-			d2mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
+			*d1mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
+			*d2mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
 		}
 	} else if (mode0) {
-		if (dfixed_trunc(wm0.dbpp) > 64)
-			a.full = dfixed_div(wm0.dbpp, wm0.num_line_pair);
+		if (dfixed_trunc(wm0->dbpp) > 64)
+			a.full = dfixed_div(wm0->dbpp, wm0->num_line_pair);
 		else
-			a.full = wm0.num_line_pair.full;
-		fill_rate.full = dfixed_div(wm0.sclk, a);
-		if (wm0.consumption_rate.full > fill_rate.full) {
-			b.full = wm0.consumption_rate.full - fill_rate.full;
-			b.full = dfixed_mul(b, wm0.active_time);
+			a.full = wm0->num_line_pair.full;
+		fill_rate.full = dfixed_div(wm0->sclk, a);
+		if (wm0->consumption_rate.full > fill_rate.full) {
+			b.full = wm0->consumption_rate.full - fill_rate.full;
+			b.full = dfixed_mul(b, wm0->active_time);
 			a.full = dfixed_const(16);
 			b.full = dfixed_div(b, a);
-			a.full = dfixed_mul(wm0.worst_case_latency,
-						wm0.consumption_rate);
+			a.full = dfixed_mul(wm0->worst_case_latency,
+						wm0->consumption_rate);
 			priority_mark02.full = a.full + b.full;
 		} else {
-			a.full = dfixed_mul(wm0.worst_case_latency,
-						wm0.consumption_rate);
+			a.full = dfixed_mul(wm0->worst_case_latency,
+						wm0->consumption_rate);
 			b.full = dfixed_const(16);
 			priority_mark02.full = dfixed_div(a, b);
 		}
-		if (wm0.priority_mark.full > priority_mark02.full)
-			priority_mark02.full = wm0.priority_mark.full;
+		if (wm0->priority_mark.full > priority_mark02.full)
+			priority_mark02.full = wm0->priority_mark.full;
 		if (dfixed_trunc(priority_mark02) < 0)
 			priority_mark02.full = 0;
-		if (wm0.priority_mark_max.full > priority_mark02.full)
-			priority_mark02.full = wm0.priority_mark_max.full;
-		d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
+		if (wm0->priority_mark_max.full > priority_mark02.full)
+			priority_mark02.full = wm0->priority_mark_max.full;
+		*d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
 		if (rdev->disp_priority == 2)
-			d1mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
+			*d1mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
 	} else if (mode1) {
-		if (dfixed_trunc(wm1.dbpp) > 64)
-			a.full = dfixed_div(wm1.dbpp, wm1.num_line_pair);
+		if (dfixed_trunc(wm1->dbpp) > 64)
+			a.full = dfixed_div(wm1->dbpp, wm1->num_line_pair);
 		else
-			a.full = wm1.num_line_pair.full;
-		fill_rate.full = dfixed_div(wm1.sclk, a);
-		if (wm1.consumption_rate.full > fill_rate.full) {
-			b.full = wm1.consumption_rate.full - fill_rate.full;
-			b.full = dfixed_mul(b, wm1.active_time);
+			a.full = wm1->num_line_pair.full;
+		fill_rate.full = dfixed_div(wm1->sclk, a);
+		if (wm1->consumption_rate.full > fill_rate.full) {
+			b.full = wm1->consumption_rate.full - fill_rate.full;
+			b.full = dfixed_mul(b, wm1->active_time);
 			a.full = dfixed_const(16);
 			b.full = dfixed_div(b, a);
-			a.full = dfixed_mul(wm1.worst_case_latency,
-						wm1.consumption_rate);
+			a.full = dfixed_mul(wm1->worst_case_latency,
+						wm1->consumption_rate);
 			priority_mark12.full = a.full + b.full;
 		} else {
-			a.full = dfixed_mul(wm1.worst_case_latency,
-						wm1.consumption_rate);
+			a.full = dfixed_mul(wm1->worst_case_latency,
+						wm1->consumption_rate);
 			b.full = dfixed_const(16 * 1000);
 			priority_mark12.full = dfixed_div(a, b);
 		}
-		if (wm1.priority_mark.full > priority_mark12.full)
-			priority_mark12.full = wm1.priority_mark.full;
+		if (wm1->priority_mark.full > priority_mark12.full)
+			priority_mark12.full = wm1->priority_mark.full;
 		if (dfixed_trunc(priority_mark12) < 0)
 			priority_mark12.full = 0;
-		if (wm1.priority_mark_max.full > priority_mark12.full)
-			priority_mark12.full = wm1.priority_mark_max.full;
-		d2mode_priority_a_cnt = dfixed_trunc(priority_mark12);
+		if (wm1->priority_mark_max.full > priority_mark12.full)
+			priority_mark12.full = wm1->priority_mark_max.full;
+		*d2mode_priority_a_cnt = dfixed_trunc(priority_mark12);
 		if (rdev->disp_priority == 2)
-			d2mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
+			*d2mode_priority_a_cnt |= MODE_PRIORITY_ALWAYS_ON;
 	}
+}
+
+void rv515_bandwidth_avivo_update(struct radeon_device *rdev)
+{
+	struct drm_display_mode *mode0 = NULL;
+	struct drm_display_mode *mode1 = NULL;
+	struct rv515_watermark wm0_high, wm0_low;
+	struct rv515_watermark wm1_high, wm1_low;
+	u32 tmp;
+	u32 d1mode_priority_a_cnt, d1mode_priority_b_cnt;
+	u32 d2mode_priority_a_cnt, d2mode_priority_b_cnt;
+
+	if (rdev->mode_info.crtcs[0]->base.enabled)
+		mode0 = &rdev->mode_info.crtcs[0]->base.mode;
+	if (rdev->mode_info.crtcs[1]->base.enabled)
+		mode1 = &rdev->mode_info.crtcs[1]->base.mode;
+	rs690_line_buffer_adjust(rdev, mode0, mode1);
+
+	rv515_crtc_bandwidth_compute(rdev, rdev->mode_info.crtcs[0], &wm0_high, false);
+	rv515_crtc_bandwidth_compute(rdev, rdev->mode_info.crtcs[1], &wm1_high, false);
+
+	rv515_crtc_bandwidth_compute(rdev, rdev->mode_info.crtcs[0], &wm0_low, false);
+	rv515_crtc_bandwidth_compute(rdev, rdev->mode_info.crtcs[1], &wm1_low, false);
+
+	tmp = wm0_high.lb_request_fifo_depth;
+	tmp |= wm1_high.lb_request_fifo_depth << 16;
+	WREG32(LB_MAX_REQ_OUTSTANDING, tmp);
+
+	rv515_compute_mode_priority(rdev,
+				    &wm0_high, &wm1_high,
+				    mode0, mode1,
+				    &d1mode_priority_a_cnt, &d2mode_priority_a_cnt);
+	rv515_compute_mode_priority(rdev,
+				    &wm0_low, &wm1_low,
+				    mode0, mode1,
+				    &d1mode_priority_b_cnt, &d2mode_priority_b_cnt);
 
 	WREG32(D1MODE_PRIORITY_A_CNT, d1mode_priority_a_cnt);
-	WREG32(D1MODE_PRIORITY_B_CNT, d1mode_priority_a_cnt);
+	WREG32(D1MODE_PRIORITY_B_CNT, d1mode_priority_b_cnt);
 	WREG32(D2MODE_PRIORITY_A_CNT, d2mode_priority_a_cnt);
-	WREG32(D2MODE_PRIORITY_B_CNT, d2mode_priority_a_cnt);
+	WREG32(D2MODE_PRIORITY_B_CNT, d2mode_priority_b_cnt);
 }
 
 void rv515_bandwidth_update(struct radeon_device *rdev)