drm/i915/gt: Move vlv GT workarounds from init_clock_gating to workarounds

Rescue the GT workarounds from being buried inside init_clock_gating so
that we remember to apply them after a GT reset, and that they are
included in our verification that the workarounds are applied.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: stable@vger.kernel.org
Link: https://patchwork.freedesktop.org/patch/msgid/20200611080140.30228-3-chris@chris-wilson.co.uk
(cherry picked from commit 7331c356)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

drivers/gpu/drm/i915/gt/intel_workarounds.c

@@ -752,6 +752,63 @@ ivb_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 	       GEN6_WIZ_HASHING_16x4);
 }
 
+static void
+vlv_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
+{
+	/* WaDisableEarlyCull:vlv */
+	wa_masked_en(wal, _3D_CHICKEN3, _3D_CHICKEN_SF_DISABLE_OBJEND_CULL);
+
+	/* WaPsdDispatchEnable:vlv */
+	/* WaDisablePSDDualDispatchEnable:vlv */
+	wa_masked_en(wal,
+		     GEN7_HALF_SLICE_CHICKEN1,
+		     GEN7_MAX_PS_THREAD_DEP |
+		     GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE);
+
+	/* WaDisable_RenderCache_OperationalFlush:vlv */
+	wa_masked_dis(wal, CACHE_MODE_0_GEN7, RC_OP_FLUSH_ENABLE);
+
+	/* WaForceL3Serialization:vlv */
+	wa_write_clr(wal, GEN7_L3SQCREG4, L3SQ_URB_READ_CAM_MATCH_DISABLE);
+
+	/*
+	 * WaVSThreadDispatchOverride:ivb,vlv
+	 *
+	 * This actually overrides the dispatch
+	 * mode for all thread types.
+	 */
+	wa_write_masked_or(wal,
+			   GEN7_FF_THREAD_MODE,
+			   GEN7_FF_SCHED_MASK,
+			   GEN7_FF_TS_SCHED_HW |
+			   GEN7_FF_VS_SCHED_HW |
+			   GEN7_FF_DS_SCHED_HW);
+
+	/*
+	 * BSpec says this must be set, even though
+	 * WaDisable4x2SubspanOptimization isn't listed for VLV.
+	 */
+	wa_masked_en(wal, CACHE_MODE_1, PIXEL_SUBSPAN_COLLECT_OPT_DISABLE);
+
+	/*
+	 * BSpec recommends 8x4 when MSAA is used,
+	 * however in practice 16x4 seems fastest.
+	 *
+	 * Note that PS/WM thread counts depend on the WIZ hashing
+	 * disable bit, which we don't touch here, but it's good
+	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
+	 */
+	wa_add(wal, GEN7_GT_MODE, 0,
+	       _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4),
+	       GEN6_WIZ_HASHING_16x4);
+
+	/*
+	 * WaIncreaseL3CreditsForVLVB0:vlv
+	 * This is the hardware default actually.
+	 */
+	wa_write(wal, GEN7_L3SQCREG1, VLV_B0_WA_L3SQCREG1_VALUE);
+}
+
 static void
 hsw_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 {
@@ -1071,6 +1128,8 @@ gt_init_workarounds(struct drm_i915_private *i915, struct i915_wa_list *wal)
 		skl_gt_workarounds_init(i915, wal);
 	else if (IS_HASWELL(i915))
 		hsw_gt_workarounds_init(i915, wal);
+	else if (IS_VALLEYVIEW(i915))
+		vlv_gt_workarounds_init(i915, wal);
 	else if (IS_IVYBRIDGE(i915))
 		ivb_gt_workarounds_init(i915, wal);
 	else if (INTEL_GEN(i915) <= 8)

drivers/gpu/drm/i915/intel_pm.c

@@ -6986,24 +6986,6 @@ static void gen6_init_clock_gating(struct drm_i915_private *dev_priv)
 	gen6_check_mch_setup(dev_priv);
 }
 
-static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv)
-{
-	u32 reg = I915_READ(GEN7_FF_THREAD_MODE);
-
-	/*
-	 * WaVSThreadDispatchOverride:ivb,vlv
-	 *
-	 * This actually overrides the dispatch
-	 * mode for all thread types.
-	 */
-	reg &= ~GEN7_FF_SCHED_MASK;
-	reg |= GEN7_FF_TS_SCHED_HW;
-	reg |= GEN7_FF_VS_SCHED_HW;
-	reg |= GEN7_FF_DS_SCHED_HW;
-
-	I915_WRITE(GEN7_FF_THREAD_MODE, reg);
-}
-
 static void lpt_init_clock_gating(struct drm_i915_private *dev_priv)
 {
 	/*
@@ -7290,28 +7272,11 @@ static void ivb_init_clock_gating(struct drm_i915_private *dev_priv)
 
 static void vlv_init_clock_gating(struct drm_i915_private *dev_priv)
 {
-	/* WaDisableEarlyCull:vlv */
-	I915_WRITE(_3D_CHICKEN3,
-		   _MASKED_BIT_ENABLE(_3D_CHICKEN_SF_DISABLE_OBJEND_CULL));
-
 	/* WaDisableBackToBackFlipFix:vlv */
 	I915_WRITE(IVB_CHICKEN3,
 		   CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE |
 		   CHICKEN3_DGMG_DONE_FIX_DISABLE);
 
-	/* WaPsdDispatchEnable:vlv */
-	/* WaDisablePSDDualDispatchEnable:vlv */
-	I915_WRITE(GEN7_HALF_SLICE_CHICKEN1,
-		   _MASKED_BIT_ENABLE(GEN7_MAX_PS_THREAD_DEP |
-				      GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE));
-
-	/* WaDisable_RenderCache_OperationalFlush:vlv */
-	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
-
-	/* WaForceL3Serialization:vlv */
-	I915_WRITE(GEN7_L3SQCREG4, I915_READ(GEN7_L3SQCREG4) &
-		   ~L3SQ_URB_READ_CAM_MATCH_DISABLE);
-
 	/* WaDisableDopClockGating:vlv */
 	I915_WRITE(GEN7_ROW_CHICKEN2,
 		   _MASKED_BIT_ENABLE(DOP_CLOCK_GATING_DISABLE));
@@ -7321,8 +7286,6 @@ static void vlv_init_clock_gating(struct drm_i915_private *dev_priv)
 		   I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) |
 		   GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB);
 
-	gen7_setup_fixed_func_scheduler(dev_priv);
-
 	/*
 	 * According to the spec, bit 13 (RCZUNIT) must be set on IVB.
 	 * This implements the WaDisableRCZUnitClockGating:vlv workaround.
@@ -7336,30 +7299,6 @@ static void vlv_init_clock_gating(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN7_UCGCTL4,
 		   I915_READ(GEN7_UCGCTL4) | GEN7_L3BANK2X_CLOCK_GATE_DISABLE);
 
-	/*
-	 * BSpec says this must be set, even though
-	 * WaDisable4x2SubspanOptimization isn't listed for VLV.
-	 */
-	I915_WRITE(CACHE_MODE_1,
-		   _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE));
-
-	/*
-	 * BSpec recommends 8x4 when MSAA is used,
-	 * however in practice 16x4 seems fastest.
-	 *
-	 * Note that PS/WM thread counts depend on the WIZ hashing
-	 * disable bit, which we don't touch here, but it's good
-	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
-	 */
-	I915_WRITE(GEN7_GT_MODE,
-		   _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4));
-
-	/*
-	 * WaIncreaseL3CreditsForVLVB0:vlv
-	 * This is the hardware default actually.
-	 */
-	I915_WRITE(GEN7_L3SQCREG1, VLV_B0_WA_L3SQCREG1_VALUE);
-
 	/*
 	 * WaDisableVLVClockGating_VBIIssue:vlv
 	 * Disable clock gating on th GCFG unit to prevent a delay