drm/i915: tune Sandy Bridge DRPS constants

These make us increase our frequency much more readily, and decrease
them only after significant idle time, resulting in a 20% performance
increase for nexuiz.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

3 files changed, 46 insertions, 8 deletions

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index c31e818f8b0..5825a586015 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -862,19 +862,44 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 		u32 rp_state_limits = I915_READ(GEN6_RP_STATE_LIMITS);
 		u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+		u32 rpstat;
+		u32 rpupei, rpcurup, rpprevup;
+		u32 rpdownei, rpcurdown, rpprevdown;
 		int max_freq;
 
 		/* RPSTAT1 is in the GT power well */
 		__gen6_force_wake_get(dev_priv);
 
+		rpstat = I915_READ(GEN6_RPSTAT1);
+		rpupei = I915_READ(GEN6_RP_CUR_UP_EI);
+		rpcurup = I915_READ(GEN6_RP_CUR_UP);
+		rpprevup = I915_READ(GEN6_RP_PREV_UP);
+		rpdownei = I915_READ(GEN6_RP_CUR_DOWN_EI);
+		rpcurdown = I915_READ(GEN6_RP_CUR_DOWN);
+		rpprevdown = I915_READ(GEN6_RP_PREV_DOWN);
+
 		seq_printf(m, "GT_PERF_STATUS: 0x%08x\n", gt_perf_status);
-		seq_printf(m, "RPSTAT1: 0x%08x\n", I915_READ(GEN6_RPSTAT1));
+		seq_printf(m, "RPSTAT1: 0x%08x\n", rpstat);
 		seq_printf(m, "Render p-state ratio: %d\n",
 			   (gt_perf_status & 0xff00) >> 8);
 		seq_printf(m, "Render p-state VID: %d\n",
 			   gt_perf_status & 0xff);
 		seq_printf(m, "Render p-state limit: %d\n",
 			   rp_state_limits & 0xff);
+		seq_printf(m, "CAGF: %dMHz\n", ((rpstat & GEN6_CAGF_MASK) >>
+						GEN6_CAGF_SHIFT) * 100);
+		seq_printf(m, "RP CUR UP EI: %dus\n", rpupei &
+			   GEN6_CURICONT_MASK);
+		seq_printf(m, "RP CUR UP: %dus\n", rpcurup &
+			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "RP PREV UP: %dus\n", rpprevup &
+			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "RP CUR DOWN EI: %dus\n", rpdownei &
+			   GEN6_CURIAVG_MASK);
+		seq_printf(m, "RP CUR DOWN: %dus\n", rpcurdown &
+			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "RP PREV DOWN: %dus\n", rpprevdown &
+			   GEN6_CURBSYTAVG_MASK);
 
 		max_freq = (rp_state_cap & 0xff0000) >> 16;
 		seq_printf(m, "Lowest (RPN) frequency: %dMHz\n",
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index e6b106bb87b..810cfa86641 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3283,15 +3283,28 @@
 #define GEN6_RP_DOWN_TIMEOUT			0xA010
 #define GEN6_RP_INTERRUPT_LIMITS		0xA014
 #define GEN6_RPSTAT1				0xA01C
+#define   GEN6_CAGF_SHIFT			8
+#define   GEN6_CAGF_MASK			(0x7f << GEN6_CAGF_SHIFT)
 #define GEN6_RP_CONTROL				0xA024
 #define   GEN6_RP_MEDIA_TURBO			(1<<11)
 #define   GEN6_RP_USE_NORMAL_FREQ		(1<<9)
 #define   GEN6_RP_MEDIA_IS_GFX			(1<<8)
 #define   GEN6_RP_ENABLE			(1<<7)
-#define   GEN6_RP_UP_BUSY_MAX			(0x2<<3)
-#define   GEN6_RP_DOWN_BUSY_MIN			(0x2<<0)
+#define   GEN6_RP_UP_IDLE_MIN			(0x1<<3)
+#define   GEN6_RP_UP_BUSY_AVG			(0x2<<3)
+#define   GEN6_RP_UP_BUSY_CONT			(0x4<<3)
+#define   GEN6_RP_DOWN_IDLE_CONT		(0x1<<0)
 #define GEN6_RP_UP_THRESHOLD			0xA02C
 #define GEN6_RP_DOWN_THRESHOLD			0xA030
+#define GEN6_RP_CUR_UP_EI			0xA050
+#define   GEN6_CURICONT_MASK			0xffffff
+#define GEN6_RP_CUR_UP				0xA054
+#define   GEN6_CURBSYTAVG_MASK			0xffffff
+#define GEN6_RP_PREV_UP				0xA058
+#define GEN6_RP_CUR_DOWN_EI			0xA05C
+#define   GEN6_CURIAVG_MASK			0xffffff
+#define GEN6_RP_CUR_DOWN			0xA060
+#define GEN6_RP_PREV_DOWN			0xA064
 #define GEN6_RP_UP_EI				0xA068
 #define GEN6_RP_DOWN_EI				0xA06C
 #define GEN6_RP_IDLE_HYSTERSIS			0xA070
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index cb9d547aa42..a90d65dad81 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -6630,18 +6630,18 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN6_RP_INTERRUPT_LIMITS,
 		   18 << 24 |
 		   6 << 16);
-	I915_WRITE(GEN6_RP_UP_THRESHOLD, 90000);
-	I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 100000);
+	I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000);
+	I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000);
 	I915_WRITE(GEN6_RP_UP_EI, 100000);
-	I915_WRITE(GEN6_RP_DOWN_EI, 300000);
+	I915_WRITE(GEN6_RP_DOWN_EI, 5000000);
 	I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10);
 	I915_WRITE(GEN6_RP_CONTROL,
 		   GEN6_RP_MEDIA_TURBO |
 		   GEN6_RP_USE_NORMAL_FREQ |
 		   GEN6_RP_MEDIA_IS_GFX |
 		   GEN6_RP_ENABLE |
-		   GEN6_RP_UP_BUSY_MAX |
-		   GEN6_RP_DOWN_BUSY_MIN);
+		   GEN6_RP_UP_BUSY_AVG |
+		   GEN6_RP_DOWN_IDLE_CONT);
 
 	if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0,
 		     500))