Merge tag 'drm-for-v4.13' into drm-intel-next-queued

Resync with the main drm-next pull request for 4.13. What we really
need is to fully resync with pending drm-misc, but that's not yet
possible due to the still ongoing merge window.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

9 files changed, 81 insertions, 78 deletions

diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index 8bba38fa19b8..700050556242 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -769,8 +769,26 @@ static void init_vgpu_execlist(struct intel_vgpu *vgpu, int ring_id)
 	vgpu_vreg(vgpu, ctx_status_ptr_reg) = ctx_status_ptr.dw;
 }
 
+static void clean_workloads(struct intel_vgpu *vgpu, unsigned long engine_mask)
+{
+	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	struct intel_engine_cs *engine;
+	struct intel_vgpu_workload *pos, *n;
+	unsigned int tmp;
+
+	/* free the unsubmited workloads in the queues. */
+	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
+		list_for_each_entry_safe(pos, n,
+			&vgpu->workload_q_head[engine->id], list) {
+			list_del_init(&pos->list);
+			free_workload(pos);
+		}
+	}
+}
+
 void intel_vgpu_clean_execlist(struct intel_vgpu *vgpu)
 {
+	clean_workloads(vgpu, ALL_ENGINES);
 	kmem_cache_destroy(vgpu->workloads);
 }
 
@@ -801,17 +819,9 @@ void intel_vgpu_reset_execlist(struct intel_vgpu *vgpu,
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine;
-	struct intel_vgpu_workload *pos, *n;
 	unsigned int tmp;
 
-	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
-		/* free the unsubmited workload in the queue */
-		list_for_each_entry_safe(pos, n,
-			&vgpu->workload_q_head[engine->id], list) {
-			list_del_init(&pos->list);
-			free_workload(pos);
-		}
-
+	clean_workloads(vgpu, engine_mask);
+	for_each_engine_masked(engine, dev_priv, engine_mask, tmp)
 		init_vgpu_execlist(vgpu, engine->id);
-	}
 }
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 35a9bff6b0cf..1414d7e6148d 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1372,18 +1372,28 @@ static int skl_misc_ctl_write(struct intel_vgpu *vgpu, unsigned int offset,
 		void *p_data, unsigned int bytes)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
-	i915_reg_t reg = {.reg = offset};
+	u32 v = *(u32 *)p_data;
+
+	if (!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv))
+		return intel_vgpu_default_mmio_write(vgpu,
+				offset, p_data, bytes);
 
 	switch (offset) {
 	case 0x4ddc:
-		vgpu_vreg(vgpu, offset) = 0x8000003c;
-		/* WaCompressedResourceSamplerPbeMediaNewHashMode:skl */
-		I915_WRITE(reg, vgpu_vreg(vgpu, offset));
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 31);
 		break;
 	case 0x42080:
-		vgpu_vreg(vgpu, offset) = 0x8000;
-		/* WaCompressedResourceDisplayNewHashMode:skl */
-		I915_WRITE(reg, vgpu_vreg(vgpu, offset));
+		/* bypass WaCompressedResourceDisplayNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 15);
+		break;
+	case 0xe194:
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 8);
+		break;
+	case 0x7014:
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 13);
 		break;
 	default:
 		return -EINVAL;
@@ -1658,7 +1668,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(GAM_ECOCHK, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(GEN7_COMMON_SLICE_CHICKEN1, D_ALL, F_MODE_MASK | F_CMD_ACCESS,
 		NULL, NULL);
-	MMIO_DFH(COMMON_SLICE_CHICKEN2, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(COMMON_SLICE_CHICKEN2, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL,
+		 skl_misc_ctl_write);
 	MMIO_DFH(0x9030, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x20a0, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x2420, D_ALL, F_CMD_ACCESS, NULL, NULL);
@@ -2583,7 +2594,8 @@ static int init_broadwell_mmio_info(struct intel_gvt *gvt)
 	MMIO_D(0x6e570, D_BDW_PLUS);
 	MMIO_D(0x65f10, D_BDW_PLUS);
 
-	MMIO_DFH(0xe194, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(0xe194, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL,
+		 skl_misc_ctl_write);
 	MMIO_DFH(0xe188, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(HALF_SLICE_CHICKEN2, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x2580, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index a7db1ab60884..5e70f5711fc8 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -610,7 +610,8 @@ struct intel_link_m_n {
 
 void intel_link_compute_m_n(int bpp, int nlanes,
 			    int pixel_clock, int link_clock,
-			    struct intel_link_m_n *m_n);
+			    struct intel_link_m_n *m_n,
+			    bool reduce_m_n);
 
 /* Interface history:
  *
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 9a3919b19413..bb9c9c3c391f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -6143,7 +6143,7 @@ retry:
 	pipe_config->fdi_lanes = lane;
 
 	intel_link_compute_m_n(pipe_config->pipe_bpp, lane, fdi_dotclock,
-			       link_bw, &pipe_config->fdi_m_n);
+			       link_bw, &pipe_config->fdi_m_n, false);
 
 	ret = ironlake_check_fdi_lanes(dev, intel_crtc->pipe, pipe_config);
 	if (ret == -EINVAL && pipe_config->pipe_bpp > 6*3) {
@@ -6319,7 +6319,8 @@ intel_reduce_m_n_ratio(uint32_t *num, uint32_t *den)
 }
 
 static void compute_m_n(unsigned int m, unsigned int n,
-			uint32_t *ret_m, uint32_t *ret_n)
+			uint32_t *ret_m, uint32_t *ret_n,
+			bool reduce_m_n)
 {
 	/*
 	 * Reduce M/N as much as possible without loss in precision. Several DP
@@ -6327,9 +6328,11 @@ static void compute_m_n(unsigned int m, unsigned int n,
 	 * values. The passed in values are more likely to have the least
 	 * significant bits zero than M after rounding below, so do this first.
 	 */
-	while ((m & 1) == 0 && (n & 1) == 0) {
-		m >>= 1;
-		n >>= 1;
+	if (reduce_m_n) {
+		while ((m & 1) == 0 && (n & 1) == 0) {
+			m >>= 1;
+			n >>= 1;
+		}
 	}
 
 	*ret_n = min_t(unsigned int, roundup_pow_of_two(n), DATA_LINK_N_MAX);
@@ -6340,16 +6343,19 @@ static void compute_m_n(unsigned int m, unsigned int n,
 void
 intel_link_compute_m_n(int bits_per_pixel, int nlanes,
 		       int pixel_clock, int link_clock,
-		       struct intel_link_m_n *m_n)
+		       struct intel_link_m_n *m_n,
+		       bool reduce_m_n)
 {
 	m_n->tu = 64;
 
 	compute_m_n(bits_per_pixel * pixel_clock,
 		    link_clock * nlanes * 8,
-		    &m_n->gmch_m, &m_n->gmch_n);
+		    &m_n->gmch_m, &m_n->gmch_n,
+		    reduce_m_n);
 
 	compute_m_n(pixel_clock, link_clock,
-		    &m_n->link_m, &m_n->link_n);
+		    &m_n->link_m, &m_n->link_n,
+		    reduce_m_n);
 }
 
 static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
@@ -11448,6 +11454,7 @@ static bool check_digital_port_conflicts(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
 	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
 	unsigned int used_ports = 0;
 	unsigned int used_mst_ports = 0;
 
@@ -11456,7 +11463,8 @@ static bool check_digital_port_conflicts(struct drm_atomic_state *state)
 	 * list to detect the problem on ddi platforms
 	 * where there's just one encoder per digital port.
 	 */
-	drm_for_each_connector(connector, dev) {
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
 		struct drm_connector_state *connector_state;
 		struct intel_encoder *encoder;
 
@@ -11495,6 +11503,7 @@ static bool check_digital_port_conflicts(struct drm_atomic_state *state)
 			break;
 		}
 	}
+	drm_connector_list_iter_end(&conn_iter);
 
 	/* can't mix MST and SST/HDMI on the same port */
 	if (used_ports & used_mst_ports)
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index a9022d13e959..2d42d09428c9 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1548,37 +1548,6 @@ static void intel_dp_print_rates(struct intel_dp *intel_dp)
 	DRM_DEBUG_KMS("common rates: %s\n", str);
 }
 
-bool
-__intel_dp_read_desc(struct intel_dp *intel_dp, struct intel_dp_desc *desc)
-{
-	u32 base = drm_dp_is_branch(intel_dp->dpcd) ? DP_BRANCH_OUI :
-						      DP_SINK_OUI;
-
-	return drm_dp_dpcd_read(&intel_dp->aux, base, desc, sizeof(*desc)) ==
-	       sizeof(*desc);
-}
-
-bool intel_dp_read_desc(struct intel_dp *intel_dp)
-{
-	struct intel_dp_desc *desc = &intel_dp->desc;
-	bool oui_sup = intel_dp->dpcd[DP_DOWN_STREAM_PORT_COUNT] &
-		       DP_OUI_SUPPORT;
-	int dev_id_len;
-
-	if (!__intel_dp_read_desc(intel_dp, desc))
-		return false;
-
-	dev_id_len = strnlen(desc->device_id, sizeof(desc->device_id));
-	DRM_DEBUG_KMS("DP %s: OUI %*phD%s dev-ID %*pE HW-rev %d.%d SW-rev %d.%d\n",
-		      drm_dp_is_branch(intel_dp->dpcd) ? "branch" : "sink",
-		      (int)sizeof(desc->oui), desc->oui, oui_sup ? "" : "(NS)",
-		      dev_id_len, desc->device_id,
-		      desc->hw_rev >> 4, desc->hw_rev & 0xf,
-		      desc->sw_major_rev, desc->sw_minor_rev);
-
-	return true;
-}
-
 int
 intel_dp_max_link_rate(struct intel_dp *intel_dp)
 {
@@ -1660,6 +1629,8 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	int link_avail, link_clock;
 	int common_len;
 	uint8_t link_bw, rate_select;
+	bool reduce_m_n = drm_dp_has_quirk(&intel_dp->desc,
+					   DP_DPCD_QUIRK_LIMITED_M_N);
 
 	common_len = intel_dp_common_len_rate_limit(intel_dp,
 						    intel_dp->max_link_rate);
@@ -1797,7 +1768,8 @@ found:
 	intel_link_compute_m_n(bpp, lane_count,
 			       adjusted_mode->crtc_clock,
 			       pipe_config->port_clock,
-			       &pipe_config->dp_m_n);
+			       &pipe_config->dp_m_n,
+			       reduce_m_n);
 
 	if (intel_connector->panel.downclock_mode != NULL &&
 		dev_priv->drrs.type == SEAMLESS_DRRS_SUPPORT) {
@@ -1805,7 +1777,8 @@ found:
 			intel_link_compute_m_n(bpp, lane_count,
 				intel_connector->panel.downclock_mode->clock,
 				pipe_config->port_clock,
-				&pipe_config->dp_m2_n2);
+				&pipe_config->dp_m2_n2,
+				reduce_m_n);
 	}
 
 	/*
@@ -3672,7 +3645,8 @@ intel_edp_init_dpcd(struct intel_dp *intel_dp)
 	if (!intel_dp_read_dpcd(intel_dp))
 		return false;
 
-	intel_dp_read_desc(intel_dp);
+	drm_dp_read_desc(&intel_dp->aux, &intel_dp->desc,
+			 drm_dp_is_branch(intel_dp->dpcd));
 
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
 		dev_priv->no_aux_handshake = intel_dp->dpcd[DP_MAX_DOWNSPREAD] &
@@ -4741,7 +4715,8 @@ intel_dp_long_pulse(struct intel_connector *intel_connector)
 
 	intel_dp_print_rates(intel_dp);
 
-	intel_dp_read_desc(intel_dp);
+	drm_dp_read_desc(&intel_dp->aux, &intel_dp->desc,
+			 drm_dp_is_branch(intel_dp->dpcd));
 
 	intel_dp_configure_mst(intel_dp);
 
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 3715386e4272..2cf046beae0f 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -44,6 +44,8 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	int lane_count, slots;
 	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	int mst_pbn;
+	bool reduce_m_n = drm_dp_has_quirk(&intel_dp->desc,
+					   DP_DPCD_QUIRK_LIMITED_M_N);
 
 	pipe_config->has_pch_encoder = false;
 	bpp = 24;
@@ -80,7 +82,8 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	intel_link_compute_m_n(bpp, lane_count,
 			       adjusted_mode->crtc_clock,
 			       pipe_config->port_clock,
-			       &pipe_config->dp_m_n);
+			       &pipe_config->dp_m_n,
+			       reduce_m_n);
 
 	pipe_config->dp_m_n.tu = slots;
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index f73474a22b6b..d17a32437f07 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -944,14 +944,6 @@ enum link_m_n_set {
 	M2_N2
 };
 
-struct intel_dp_desc {
-	u8 oui[3];
-	u8 device_id[6];
-	u8 hw_rev;
-	u8 sw_major_rev;
-	u8 sw_minor_rev;
-} __packed;
-
 struct intel_dp_compliance_data {
 	unsigned long edid;
 	uint8_t video_pattern;
@@ -999,7 +991,7 @@ struct intel_dp {
 	/* Max rate for the current link */
 	int max_link_rate;
 	/* sink or branch descriptor */
-	struct intel_dp_desc desc;
+	struct drm_dp_desc desc;
 	struct drm_dp_aux aux;
 	enum intel_display_power_domain aux_power_domain;
 	uint8_t train_set[4];
@@ -1578,9 +1570,6 @@ static inline unsigned int intel_dp_unused_lane_mask(int lane_count)
 }
 
 bool intel_dp_read_dpcd(struct intel_dp *intel_dp);
-bool __intel_dp_read_desc(struct intel_dp *intel_dp,
-			  struct intel_dp_desc *desc);
-bool intel_dp_read_desc(struct intel_dp *intel_dp);
 int intel_dp_link_required(int pixel_clock, int bpp);
 int intel_dp_max_data_rate(int max_link_clock, int max_lanes);
 bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c
index 71cbe9c08932..5abef482eacf 100644
--- a/drivers/gpu/drm/i915/intel_lspcon.c
+++ b/drivers/gpu/drm/i915/intel_lspcon.c
@@ -240,7 +240,7 @@ bool lspcon_init(struct intel_digital_port *intel_dig_port)
 		return false;
 	}
 
-	intel_dp_read_desc(dp);
+	drm_dp_read_desc(&dp->aux, &dp->desc, drm_dp_is_branch(dp->dpcd));
 
 	DRM_DEBUG_KMS("Success: LSPCON init\n");
 	return true;
diff --git a/drivers/gpu/drm/i915/intel_uc.h b/drivers/gpu/drm/i915/intel_uc.h
index 69daf4c01cd0..22ae52b17b0f 100644
--- a/drivers/gpu/drm/i915/intel_uc.h
+++ b/drivers/gpu/drm/i915/intel_uc.h
@@ -59,6 +59,10 @@ struct drm_i915_gem_request;
  *                available in the work queue (note, the queue is shared,
  *                not per-engine). It is OK for this to be nonzero, but
  *                it should not be huge!
+ *   b_fail: failed to ring the doorbell. This should never happen, unless
+ *           somehow the hardware misbehaves, or maybe if the GuC firmware
+ *           crashes? We probably need to reset the GPU to recover.
+ *   retcode: errno from last guc_submit()
  */
 struct i915_guc_client {
 	struct i915_vma *vma;