Merge tag 'drm-fixes-2023-04-13' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Daniel Vetter:

 - two fbcon regressions

 - amdgpu: dp mst, smu13

 - i915: dual link dsi for tgl+

 - armada, nouveau, drm/sched, fbmem

* tag 'drm-fixes-2023-04-13' of git://anongit.freedesktop.org/drm/drm:
  fbcon: set_con2fb_map needs to set con2fb_map!
  fbcon: Fix error paths in set_con2fb_map
  drm/amd/pm: correct the pcie link state check for SMU13
  drm/amd/pm: correct SMU13.0.7 max shader clock reporting
  drm/amd/pm: correct SMU13.0.7 pstate profiling clock settings
  drm/amd/display: Pass the right info to drm_dp_remove_payload
  drm/armada: Fix a potential double free in an error handling path
  fbmem: Reject FB_ACTIVATE_KD_TEXT from userspace
  drm/nouveau/fb: add missing sysmen flush callbacks
  drm/i915/dsi: fix DSS CTL register offsets for TGL+
  drm/scheduler: Fix UAF race in drm_sched_entity_push_job()

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c

@@ -177,6 +177,40 @@ void dm_helpers_dp_update_branch_info(
 	const struct dc_link *link)
 {}
 
+static void dm_helpers_construct_old_payload(
+			struct dc_link *link,
+			int pbn_per_slot,
+			struct drm_dp_mst_atomic_payload *new_payload,
+			struct drm_dp_mst_atomic_payload *old_payload)
+{
+	struct link_mst_stream_allocation_table current_link_table =
+									link->mst_stream_alloc_table;
+	struct link_mst_stream_allocation *dc_alloc;
+	int i;
+
+	*old_payload = *new_payload;
+
+	/* Set correct time_slots/PBN of old payload.
+	 * other fields (delete & dsc_enabled) in
+	 * struct drm_dp_mst_atomic_payload are don't care fields
+	 * while calling drm_dp_remove_payload()
+	 */
+	for (i = 0; i < current_link_table.stream_count; i++) {
+		dc_alloc =
+			&current_link_table.stream_allocations[i];
+
+		if (dc_alloc->vcp_id == new_payload->vcpi) {
+			old_payload->time_slots = dc_alloc->slot_count;
+			old_payload->pbn = dc_alloc->slot_count * pbn_per_slot;
+			break;
+		}
+	}
+
+	/* make sure there is an old payload*/
+	ASSERT(i != current_link_table.stream_count);
+
+}
+
 /*
  * Writes payload allocation table in immediate downstream device.
  */
@@ -188,7 +222,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 {
 	struct amdgpu_dm_connector *aconnector;
 	struct drm_dp_mst_topology_state *mst_state;
-	struct drm_dp_mst_atomic_payload *payload;
+	struct drm_dp_mst_atomic_payload *target_payload, *new_payload, old_payload;
 	struct drm_dp_mst_topology_mgr *mst_mgr;
 
 	aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context;
@@ -204,17 +238,26 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 	mst_state = to_drm_dp_mst_topology_state(mst_mgr->base.state);
 
 	/* It's OK for this to fail */
-	payload = drm_atomic_get_mst_payload_state(mst_state, aconnector->mst_output_port);
-	if (enable)
-		drm_dp_add_payload_part1(mst_mgr, mst_state, payload);
-	else
-		drm_dp_remove_payload(mst_mgr, mst_state, payload, payload);
+	new_payload = drm_atomic_get_mst_payload_state(mst_state, aconnector->mst_output_port);
+
+	if (enable) {
+		target_payload = new_payload;
+
+		drm_dp_add_payload_part1(mst_mgr, mst_state, new_payload);
+	} else {
+		/* construct old payload by VCPI*/
+		dm_helpers_construct_old_payload(stream->link, mst_state->pbn_div,
+						new_payload, &old_payload);
+		target_payload = &old_payload;
+
+		drm_dp_remove_payload(mst_mgr, mst_state, &old_payload, new_payload);
+	}
 
 	/* mst_mgr->->payloads are VC payload notify MST branch using DPCD or
 	 * AUX message. The sequence is slot 1-63 allocated sequence for each
 	 * stream. AMD ASIC stream slot allocation should follow the same
 	 * sequence. copy DRM MST allocation to dc */
-	fill_dc_mst_payload_table_from_drm(stream->link, enable, payload, proposed_table);
+	fill_dc_mst_payload_table_from_drm(stream->link, enable, target_payload, proposed_table);
 
 	return true;
 }

drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h

@@ -61,6 +61,12 @@
 #define CTF_OFFSET_HOTSPOT		5
 #define CTF_OFFSET_MEM			5
 
+static const int pmfw_decoded_link_speed[5] = {1, 2, 3, 4, 5};
+static const int pmfw_decoded_link_width[7] = {0, 1, 2, 4, 8, 12, 16};
+
+#define DECODE_GEN_SPEED(gen_speed_idx)		(pmfw_decoded_link_speed[gen_speed_idx])
+#define DECODE_LANE_WIDTH(lane_width_idx)	(pmfw_decoded_link_width[lane_width_idx])
+
 struct smu_13_0_max_sustainable_clocks {
 	uint32_t display_clock;
 	uint32_t phy_clock;

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c

@@ -1144,8 +1144,8 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu,
 					(pcie_table->pcie_lane[i] == 5) ? "x12" :
 					(pcie_table->pcie_lane[i] == 6) ? "x16" : "",
 					pcie_table->clk_freq[i],
-					((gen_speed - 1) == pcie_table->pcie_gen[i]) &&
-					(lane_width == link_width[pcie_table->pcie_lane[i]]) ?
+					(gen_speed == DECODE_GEN_SPEED(pcie_table->pcie_gen[i])) &&
+					(lane_width == DECODE_LANE_WIDTH(link_width[pcie_table->pcie_lane[i]])) ?
 					"*" : "");
 		break;
 

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c

@@ -575,6 +575,14 @@ static int smu_v13_0_7_set_default_dpm_table(struct smu_context *smu)
 						     dpm_table);
 		if (ret)
 			return ret;
+
+		if (skutable->DriverReportedClocks.GameClockAc &&
+			(dpm_table->dpm_levels[dpm_table->count - 1].value >
+			skutable->DriverReportedClocks.GameClockAc)) {
+			dpm_table->dpm_levels[dpm_table->count - 1].value =
+				skutable->DriverReportedClocks.GameClockAc;
+			dpm_table->max = skutable->DriverReportedClocks.GameClockAc;
+		}
 	} else {
 		dpm_table->count = 1;
 		dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.gfxclk / 100;
@@ -828,6 +836,57 @@ static int smu_v13_0_7_get_smu_metrics_data(struct smu_context *smu,
 	return ret;
 }
 
+static int smu_v13_0_7_get_dpm_ultimate_freq(struct smu_context *smu,
+					     enum smu_clk_type clk_type,
+					     uint32_t *min,
+					     uint32_t *max)
+{
+	struct smu_13_0_dpm_context *dpm_context =
+		smu->smu_dpm.dpm_context;
+	struct smu_13_0_dpm_table *dpm_table;
+
+	switch (clk_type) {
+	case SMU_MCLK:
+	case SMU_UCLK:
+		/* uclk dpm table */
+		dpm_table = &dpm_context->dpm_tables.uclk_table;
+		break;
+	case SMU_GFXCLK:
+	case SMU_SCLK:
+		/* gfxclk dpm table */
+		dpm_table = &dpm_context->dpm_tables.gfx_table;
+		break;
+	case SMU_SOCCLK:
+		/* socclk dpm table */
+		dpm_table = &dpm_context->dpm_tables.soc_table;
+		break;
+	case SMU_FCLK:
+		/* fclk dpm table */
+		dpm_table = &dpm_context->dpm_tables.fclk_table;
+		break;
+	case SMU_VCLK:
+	case SMU_VCLK1:
+		/* vclk dpm table */
+		dpm_table = &dpm_context->dpm_tables.vclk_table;
+		break;
+	case SMU_DCLK:
+	case SMU_DCLK1:
+		/* dclk dpm table */
+		dpm_table = &dpm_context->dpm_tables.dclk_table;
+		break;
+	default:
+		dev_err(smu->adev->dev, "Unsupported clock type!\n");
+		return -EINVAL;
+	}
+
+	if (min)
+		*min = dpm_table->min;
+	if (max)
+		*max = dpm_table->max;
+
+	return 0;
+}
+
 static int smu_v13_0_7_read_sensor(struct smu_context *smu,
 				   enum amd_pp_sensors sensor,
 				   void *data,
@@ -1074,8 +1133,8 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu,
 					(pcie_table->pcie_lane[i] == 5) ? "x12" :
 					(pcie_table->pcie_lane[i] == 6) ? "x16" : "",
 					pcie_table->clk_freq[i],
-					(gen_speed == pcie_table->pcie_gen[i]) &&
-					(lane_width == pcie_table->pcie_lane[i]) ?
+					(gen_speed == DECODE_GEN_SPEED(pcie_table->pcie_gen[i])) &&
+					(lane_width == DECODE_LANE_WIDTH(pcie_table->pcie_lane[i])) ?
 					"*" : "");
 		break;
 
@@ -1329,9 +1388,17 @@ static int smu_v13_0_7_populate_umd_state_clk(struct smu_context *smu)
 				&dpm_context->dpm_tables.fclk_table;
 	struct smu_umd_pstate_table *pstate_table =
 				&smu->pstate_table;
+	struct smu_table_context *table_context = &smu->smu_table;
+	PPTable_t *pptable = table_context->driver_pptable;
+	DriverReportedClocks_t driver_clocks =
+		pptable->SkuTable.DriverReportedClocks;
 
 	pstate_table->gfxclk_pstate.min = gfx_table->min;
-	pstate_table->gfxclk_pstate.peak = gfx_table->max;
+	if (driver_clocks.GameClockAc &&
+		(driver_clocks.GameClockAc < gfx_table->max))
+		pstate_table->gfxclk_pstate.peak = driver_clocks.GameClockAc;
+	else
+		pstate_table->gfxclk_pstate.peak = gfx_table->max;
 
 	pstate_table->uclk_pstate.min = mem_table->min;
 	pstate_table->uclk_pstate.peak = mem_table->max;
@@ -1348,12 +1415,12 @@ static int smu_v13_0_7_populate_umd_state_clk(struct smu_context *smu)
 	pstate_table->fclk_pstate.min = fclk_table->min;
 	pstate_table->fclk_pstate.peak = fclk_table->max;
 
-	/*
-	 * For now, just use the mininum clock frequency.
-	 * TODO: update them when the real pstate settings available
-	 */
-	pstate_table->gfxclk_pstate.standard = gfx_table->min;
-	pstate_table->uclk_pstate.standard = mem_table->min;
+	if (driver_clocks.BaseClockAc &&
+		driver_clocks.BaseClockAc < gfx_table->max)
+		pstate_table->gfxclk_pstate.standard = driver_clocks.BaseClockAc;
+	else
+		pstate_table->gfxclk_pstate.standard = gfx_table->max;
+	pstate_table->uclk_pstate.standard = mem_table->max;
 	pstate_table->socclk_pstate.standard = soc_table->min;
 	pstate_table->vclk_pstate.standard = vclk_table->min;
 	pstate_table->dclk_pstate.standard = dclk_table->min;
@@ -1676,7 +1743,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.dpm_set_jpeg_enable = smu_v13_0_set_jpeg_enable,
 	.init_pptable_microcode = smu_v13_0_init_pptable_microcode,
 	.populate_umd_state_clk = smu_v13_0_7_populate_umd_state_clk,
-	.get_dpm_ultimate_freq = smu_v13_0_get_dpm_ultimate_freq,
+	.get_dpm_ultimate_freq = smu_v13_0_7_get_dpm_ultimate_freq,
 	.get_vbios_bootup_values = smu_v13_0_get_vbios_bootup_values,
 	.read_sensor = smu_v13_0_7_read_sensor,
 	.feature_is_enabled = smu_cmn_feature_is_enabled,

drivers/gpu/drm/armada/armada_drv.c

@@ -99,7 +99,6 @@ static int armada_drm_bind(struct device *dev)
 	if (ret) {
 		dev_err(dev, "[" DRM_NAME ":%s] can't kick out simple-fb: %d\n",
 			__func__, ret);
-		kfree(priv);
 		return ret;
 	}
 

drivers/gpu/drm/i915/display/icl_dsi.c

@@ -300,9 +300,21 @@ static void configure_dual_link_mode(struct intel_encoder *encoder,
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
+	i915_reg_t dss_ctl1_reg, dss_ctl2_reg;
 	u32 dss_ctl1;
 
-	dss_ctl1 = intel_de_read(dev_priv, DSS_CTL1);
+	/* FIXME: Move all DSS handling to intel_vdsc.c */
+	if (DISPLAY_VER(dev_priv) >= 12) {
+		struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
+
+		dss_ctl1_reg = ICL_PIPE_DSS_CTL1(crtc->pipe);
+		dss_ctl2_reg = ICL_PIPE_DSS_CTL2(crtc->pipe);
+	} else {
+		dss_ctl1_reg = DSS_CTL1;
+		dss_ctl2_reg = DSS_CTL2;
+	}
+
+	dss_ctl1 = intel_de_read(dev_priv, dss_ctl1_reg);
 	dss_ctl1 |= SPLITTER_ENABLE;
 	dss_ctl1 &= ~OVERLAP_PIXELS_MASK;
 	dss_ctl1 |= OVERLAP_PIXELS(intel_dsi->pixel_overlap);
@@ -323,16 +335,16 @@ static void configure_dual_link_mode(struct intel_encoder *encoder,
 
 		dss_ctl1 &= ~LEFT_DL_BUF_TARGET_DEPTH_MASK;
 		dss_ctl1 |= LEFT_DL_BUF_TARGET_DEPTH(dl_buffer_depth);
-		dss_ctl2 = intel_de_read(dev_priv, DSS_CTL2);
+		dss_ctl2 = intel_de_read(dev_priv, dss_ctl2_reg);
 		dss_ctl2 &= ~RIGHT_DL_BUF_TARGET_DEPTH_MASK;
 		dss_ctl2 |= RIGHT_DL_BUF_TARGET_DEPTH(dl_buffer_depth);
-		intel_de_write(dev_priv, DSS_CTL2, dss_ctl2);
+		intel_de_write(dev_priv, dss_ctl2_reg, dss_ctl2);
 	} else {
 		/* Interleave */
 		dss_ctl1 |= DUAL_LINK_MODE_INTERLEAVE;
 	}
 
-	intel_de_write(dev_priv, DSS_CTL1, dss_ctl1);
+	intel_de_write(dev_priv, dss_ctl1_reg, dss_ctl1);
 }
 
 /* aka DSI 8X clock */

drivers/gpu/drm/nouveau/nvkm/subdev/fb/gf108.c

@@ -31,6 +31,7 @@ gf108_fb = {
 	.init = gf100_fb_init,
 	.init_page = gf100_fb_init_page,
 	.intr = gf100_fb_intr,
+	.sysmem.flush_page_init = gf100_fb_sysmem_flush_page_init,
 	.ram_new = gf108_ram_new,
 	.default_bigpage = 17,
 };

drivers/gpu/drm/nouveau/nvkm/subdev/fb/gk104.c

@@ -77,6 +77,7 @@ gk104_fb = {
 	.init = gf100_fb_init,
 	.init_page = gf100_fb_init_page,
 	.intr = gf100_fb_intr,
+	.sysmem.flush_page_init = gf100_fb_sysmem_flush_page_init,
 	.ram_new = gk104_ram_new,
 	.default_bigpage = 17,
 	.clkgate_pack = gk104_fb_clkgate_pack,

drivers/gpu/drm/nouveau/nvkm/subdev/fb/gk110.c

@@ -59,6 +59,7 @@ gk110_fb = {
 	.init = gf100_fb_init,
 	.init_page = gf100_fb_init_page,
 	.intr = gf100_fb_intr,
+	.sysmem.flush_page_init = gf100_fb_sysmem_flush_page_init,
 	.ram_new = gk104_ram_new,
 	.default_bigpage = 17,
 	.clkgate_pack = gk110_fb_clkgate_pack,

drivers/gpu/drm/nouveau/nvkm/subdev/fb/gm107.c

@@ -31,6 +31,7 @@ gm107_fb = {
 	.init = gf100_fb_init,
 	.init_page = gf100_fb_init_page,
 	.intr = gf100_fb_intr,
+	.sysmem.flush_page_init = gf100_fb_sysmem_flush_page_init,
 	.ram_new = gm107_ram_new,
 	.default_bigpage = 17,
 };

drivers/gpu/drm/scheduler/sched_entity.c

@@ -507,12 +507,19 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job)
 {
 	struct drm_sched_entity *entity = sched_job->entity;
 	bool first;
+	ktime_t submit_ts;
 
 	trace_drm_sched_job(sched_job, entity);
 	atomic_inc(entity->rq->sched->score);
 	WRITE_ONCE(entity->last_user, current->group_leader);
+
+	/*
+	 * After the sched_job is pushed into the entity queue, it may be
+	 * completed and freed up at any time. We can no longer access it.
+	 * Make sure to set the submit_ts first, to avoid a race.
+	 */
+	sched_job->submit_ts = submit_ts = ktime_get();
 	first = spsc_queue_push(&entity->job_queue, &sched_job->queue_node);
-	sched_job->submit_ts = ktime_get();
 
 	/* first job wakes up scheduler */
 	if (first) {
@@ -529,7 +536,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job)
 		spin_unlock(&entity->rq_lock);
 
 		if (drm_sched_policy == DRM_SCHED_POLICY_FIFO)
-			drm_sched_rq_update_fifo(entity, sched_job->submit_ts);
+			drm_sched_rq_update_fifo(entity, submit_ts);
 
 		drm_sched_wakeup(entity->rq->sched);
 	}

drivers/video/fbdev/core/fbcon.c

@@ -823,7 +823,7 @@ static int set_con2fb_map(int unit, int newidx, int user)
 	int oldidx = con2fb_map[unit];
 	struct fb_info *info = fbcon_registered_fb[newidx];
 	struct fb_info *oldinfo = NULL;
-	int found, err = 0, show_logo;
+	int err = 0, show_logo;
 
 	WARN_CONSOLE_UNLOCKED();
 
@@ -841,26 +841,26 @@ static int set_con2fb_map(int unit, int newidx, int user)
 	if (oldidx != -1)
 		oldinfo = fbcon_registered_fb[oldidx];
 
-	found = search_fb_in_map(newidx);
-
-	if (!err && !found) {
+	if (!search_fb_in_map(newidx)) {
 		err = con2fb_acquire_newinfo(vc, info, unit);
-		if (!err)
-			con2fb_map[unit] = newidx;
+		if (err)
+			return err;
+
+		fbcon_add_cursor_work(info);
 	}
 
+	con2fb_map[unit] = newidx;
+
 	/*
 	 * If old fb is not mapped to any of the consoles,
 	 * fbcon should release it.
 	 */
-	if (!err && oldinfo && !search_fb_in_map(oldidx))
+	if (oldinfo && !search_fb_in_map(oldidx))
 		con2fb_release_oldinfo(vc, oldinfo, info);
 
 	show_logo = (fg_console == 0 && !user &&
 			 logo_shown != FBCON_LOGO_DONTSHOW);
 
-	if (!found)
-		fbcon_add_cursor_work(info);
 	con2fb_map_boot[unit] = newidx;
 	con2fb_init_display(vc, info, unit, show_logo);
 

drivers/video/fbdev/core/fbmem.c

@@ -1116,6 +1116,8 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 	case FBIOPUT_VSCREENINFO:
 		if (copy_from_user(&var, argp, sizeof(var)))
 			return -EFAULT;
+		/* only for kernel-internal use */
+		var.activate &= ~FB_ACTIVATE_KD_TEXT;
 		console_lock();
 		lock_fb_info(info);
 		ret = fbcon_modechange_possible(info, &var);