Merge tag 'drm-msm-next-2022-11-28' of https://gitlab.freedesktop.org/drm/msm into drm-next

msm-next for v6.2 (the gpu/gem bits)

- Remove exclusive-fence hack that caused over-synchronization
- Fix speed-bin detection vs. probe-defer
- Enable clamp_to_idle on 7c3
- Improved hangcheck detection

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rob Clark <robdclark@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/CAF6AEGvT1h_S4d=YRgphgR8i7aMaxQaNW8mru7QaoUo9uiUk2A@mail.gmail.com

12 files changed, 150 insertions, 66 deletions

diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
index 2fb32d5552c4..3e09d3a7a0ac 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
@@ -606,8 +606,7 @@ static int a4xx_pm_suspend(struct msm_gpu *gpu) {
 
 static int a4xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 {
-	*value = gpu_read64(gpu, REG_A4XX_RBBM_PERFCTR_CP_0_LO,
-		REG_A4XX_RBBM_PERFCTR_CP_0_HI);
+	*value = gpu_read64(gpu, REG_A4XX_RBBM_PERFCTR_CP_0_LO);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
index 3c537c0016fa..660ba0db8900 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
@@ -605,11 +605,9 @@ static int a5xx_ucode_init(struct msm_gpu *gpu)
 		a5xx_ucode_check_version(a5xx_gpu, a5xx_gpu->pfp_bo);
 	}
 
-	gpu_write64(gpu, REG_A5XX_CP_ME_INSTR_BASE_LO,
-		REG_A5XX_CP_ME_INSTR_BASE_HI, a5xx_gpu->pm4_iova);
+	gpu_write64(gpu, REG_A5XX_CP_ME_INSTR_BASE_LO, a5xx_gpu->pm4_iova);
 
-	gpu_write64(gpu, REG_A5XX_CP_PFP_INSTR_BASE_LO,
-		REG_A5XX_CP_PFP_INSTR_BASE_HI, a5xx_gpu->pfp_iova);
+	gpu_write64(gpu, REG_A5XX_CP_PFP_INSTR_BASE_LO, a5xx_gpu->pfp_iova);
 
 	return 0;
 }
@@ -868,8 +866,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 	 * memory rendering at this point in time and we don't want to block off
 	 * part of the virtual memory space.
 	 */
-	gpu_write64(gpu, REG_A5XX_RBBM_SECVID_TSB_TRUSTED_BASE_LO,
-		REG_A5XX_RBBM_SECVID_TSB_TRUSTED_BASE_HI, 0x00000000);
+	gpu_write64(gpu, REG_A5XX_RBBM_SECVID_TSB_TRUSTED_BASE_LO, 0x00000000);
 	gpu_write(gpu, REG_A5XX_RBBM_SECVID_TSB_TRUSTED_SIZE, 0x00000000);
 
 	/* Put the GPU into 64 bit by default */
@@ -908,8 +905,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 		return ret;
 
 	/* Set the ringbuffer address */
-	gpu_write64(gpu, REG_A5XX_CP_RB_BASE, REG_A5XX_CP_RB_BASE_HI,
-		gpu->rb[0]->iova);
+	gpu_write64(gpu, REG_A5XX_CP_RB_BASE, gpu->rb[0]->iova);
 
 	/*
 	 * If the microcode supports the WHERE_AM_I opcode then we can use that
@@ -936,7 +932,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 		}
 
 		gpu_write64(gpu, REG_A5XX_CP_RB_RPTR_ADDR,
-			REG_A5XX_CP_RB_RPTR_ADDR_HI, shadowptr(a5xx_gpu, gpu->rb[0]));
+			    shadowptr(a5xx_gpu, gpu->rb[0]));
 	} else if (gpu->nr_rings > 1) {
 		/* Disable preemption if WHERE_AM_I isn't available */
 		a5xx_preempt_fini(gpu);
@@ -1239,9 +1235,9 @@ static void a5xx_fault_detect_irq(struct msm_gpu *gpu)
 		gpu_read(gpu, REG_A5XX_RBBM_STATUS),
 		gpu_read(gpu, REG_A5XX_CP_RB_RPTR),
 		gpu_read(gpu, REG_A5XX_CP_RB_WPTR),
-		gpu_read64(gpu, REG_A5XX_CP_IB1_BASE, REG_A5XX_CP_IB1_BASE_HI),
+		gpu_read64(gpu, REG_A5XX_CP_IB1_BASE),
 		gpu_read(gpu, REG_A5XX_CP_IB1_BUFSZ),
-		gpu_read64(gpu, REG_A5XX_CP_IB2_BASE, REG_A5XX_CP_IB2_BASE_HI),
+		gpu_read64(gpu, REG_A5XX_CP_IB2_BASE),
 		gpu_read(gpu, REG_A5XX_CP_IB2_BUFSZ));
 
 	/* Turn off the hangcheck timer to keep it from bothering us */
@@ -1427,8 +1423,7 @@ static int a5xx_pm_suspend(struct msm_gpu *gpu)
 
 static int a5xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 {
-	*value = gpu_read64(gpu, REG_A5XX_RBBM_ALWAYSON_COUNTER_LO,
-		REG_A5XX_RBBM_ALWAYSON_COUNTER_HI);
+	*value = gpu_read64(gpu, REG_A5XX_RBBM_ALWAYSON_COUNTER_LO);
 
 	return 0;
 }
@@ -1465,8 +1460,7 @@ static int a5xx_crashdumper_run(struct msm_gpu *gpu,
 	if (IS_ERR_OR_NULL(dumper->ptr))
 		return -EINVAL;
 
-	gpu_write64(gpu, REG_A5XX_CP_CRASH_SCRIPT_BASE_LO,
-		REG_A5XX_CP_CRASH_SCRIPT_BASE_HI, dumper->iova);
+	gpu_write64(gpu, REG_A5XX_CP_CRASH_SCRIPT_BASE_LO, dumper->iova);
 
 	gpu_write(gpu, REG_A5XX_CP_CRASH_DUMP_CNTL, 1);
 
@@ -1666,8 +1660,7 @@ static u64 a5xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate)
 {
 	u64 busy_cycles;
 
-	busy_cycles = gpu_read64(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_0_LO,
-			REG_A5XX_RBBM_PERFCTR_RBBM_0_HI);
+	busy_cycles = gpu_read64(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_0_LO);
 	*out_sample_rate = clk_get_rate(gpu->core_clk);
 
 	return busy_cycles;
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c
index 8abc9a2b114a..7658e89844b4 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c
@@ -137,7 +137,6 @@ void a5xx_preempt_trigger(struct msm_gpu *gpu)
 
 	/* Set the address of the incoming preemption record */
 	gpu_write64(gpu, REG_A5XX_CP_CONTEXT_SWITCH_RESTORE_ADDR_LO,
-		REG_A5XX_CP_CONTEXT_SWITCH_RESTORE_ADDR_HI,
 		a5xx_gpu->preempt_iova[ring->id]);
 
 	a5xx_gpu->next_ring = ring;
@@ -211,8 +210,7 @@ void a5xx_preempt_hw_init(struct msm_gpu *gpu)
 	}
 
 	/* Write a 0 to signal that we aren't switching pagetables */
-	gpu_write64(gpu, REG_A5XX_CP_CONTEXT_SWITCH_SMMU_INFO_LO,
-		REG_A5XX_CP_CONTEXT_SWITCH_SMMU_INFO_HI, 0);
+	gpu_write64(gpu, REG_A5XX_CP_CONTEXT_SWITCH_SMMU_INFO_LO, 0);
 
 	/* Reset the preemption state */
 	set_preempt_state(a5xx_gpu, PREEMPT_NONE);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index e87196457b9a..36c8fb699b56 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -247,8 +247,7 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 	OUT_RING(ring, submit->seqno);
 
 	trace_msm_gpu_submit_flush(submit,
-		gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER_LO,
-			REG_A6XX_CP_ALWAYS_ON_COUNTER_HI));
+		gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER_LO));
 
 	a6xx_flush(gpu, ring);
 }
@@ -947,8 +946,7 @@ static int a6xx_ucode_init(struct msm_gpu *gpu)
 		}
 	}
 
-	gpu_write64(gpu, REG_A6XX_CP_SQE_INSTR_BASE,
-		REG_A6XX_CP_SQE_INSTR_BASE+1, a6xx_gpu->sqe_iova);
+	gpu_write64(gpu, REG_A6XX_CP_SQE_INSTR_BASE, a6xx_gpu->sqe_iova);
 
 	return 0;
 }
@@ -999,8 +997,7 @@ static int hw_init(struct msm_gpu *gpu)
 	 * memory rendering at this point in time and we don't want to block off
 	 * part of the virtual memory space.
 	 */
-	gpu_write64(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_BASE_LO,
-		REG_A6XX_RBBM_SECVID_TSB_TRUSTED_BASE_HI, 0x00000000);
+	gpu_write64(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_BASE_LO, 0x00000000);
 	gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_SIZE, 0x00000000);
 
 	/* Turn on 64 bit addressing for all blocks */
@@ -1049,11 +1046,9 @@ static int hw_init(struct msm_gpu *gpu)
 
 	if (!adreno_is_a650_family(adreno_gpu)) {
 		/* Set the GMEM VA range [0x100000:0x100000 + gpu->gmem - 1] */
-		gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MIN_LO,
-			REG_A6XX_UCHE_GMEM_RANGE_MIN_HI, 0x00100000);
+		gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MIN_LO, 0x00100000);
 
 		gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MAX_LO,
-			REG_A6XX_UCHE_GMEM_RANGE_MAX_HI,
 			0x00100000 + adreno_gpu->gmem - 1);
 	}
 
@@ -1145,8 +1140,7 @@ static int hw_init(struct msm_gpu *gpu)
 		goto out;
 
 	/* Set the ringbuffer address */
-	gpu_write64(gpu, REG_A6XX_CP_RB_BASE, REG_A6XX_CP_RB_BASE_HI,
-		gpu->rb[0]->iova);
+	gpu_write64(gpu, REG_A6XX_CP_RB_BASE, gpu->rb[0]->iova);
 
 	/* Targets that support extended APRIV can use the RPTR shadow from
 	 * hardware but all the other ones need to disable the feature. Targets
@@ -1178,7 +1172,6 @@ static int hw_init(struct msm_gpu *gpu)
 		}
 
 		gpu_write64(gpu, REG_A6XX_CP_RB_RPTR_ADDR_LO,
-			REG_A6XX_CP_RB_RPTR_ADDR_HI,
 			shadowptr(a6xx_gpu, gpu->rb[0]));
 	}
 
@@ -1499,9 +1492,9 @@ static void a6xx_fault_detect_irq(struct msm_gpu *gpu)
 		gpu_read(gpu, REG_A6XX_RBBM_STATUS),
 		gpu_read(gpu, REG_A6XX_CP_RB_RPTR),
 		gpu_read(gpu, REG_A6XX_CP_RB_WPTR),
-		gpu_read64(gpu, REG_A6XX_CP_IB1_BASE, REG_A6XX_CP_IB1_BASE_HI),
+		gpu_read64(gpu, REG_A6XX_CP_IB1_BASE),
 		gpu_read(gpu, REG_A6XX_CP_IB1_REM_SIZE),
-		gpu_read64(gpu, REG_A6XX_CP_IB2_BASE, REG_A6XX_CP_IB2_BASE_HI),
+		gpu_read64(gpu, REG_A6XX_CP_IB2_BASE),
 		gpu_read(gpu, REG_A6XX_CP_IB2_REM_SIZE));
 
 	/* Turn off the hangcheck timer to keep it from bothering us */
@@ -1712,8 +1705,7 @@ static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 	/* Force the GPU power on so we can read this register */
 	a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_PERFCOUNTER_SET);
 
-	*value = gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER_LO,
-			    REG_A6XX_CP_ALWAYS_ON_COUNTER_HI);
+	*value = gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER_LO);
 
 	a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_PERFCOUNTER_SET);
 
@@ -1824,6 +1816,39 @@ static uint32_t a6xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 	return ring->memptrs->rptr = gpu_read(gpu, REG_A6XX_CP_RB_RPTR);
 }
 
+static bool a6xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	struct msm_cp_state cp_state = {
+		.ib1_base = gpu_read64(gpu, REG_A6XX_CP_IB1_BASE),
+		.ib2_base = gpu_read64(gpu, REG_A6XX_CP_IB2_BASE),
+		.ib1_rem  = gpu_read(gpu, REG_A6XX_CP_IB1_REM_SIZE),
+		.ib2_rem  = gpu_read(gpu, REG_A6XX_CP_IB2_REM_SIZE),
+	};
+	bool progress;
+
+	/*
+	 * Adjust the remaining data to account for what has already been
+	 * fetched from memory, but not yet consumed by the SQE.
+	 *
+	 * This is not *technically* correct, the amount buffered could
+	 * exceed the IB size due to hw prefetching ahead, but:
+	 *
+	 * (1) We aren't trying to find the exact position, just whether
+	 *     progress has been made
+	 * (2) The CP_REG_TO_MEM at the end of a submit should be enough
+	 *     to prevent prefetching into an unrelated submit.  (And
+	 *     either way, at some point the ROQ will be full.)
+	 */
+	cp_state.ib1_rem += gpu_read(gpu, REG_A6XX_CP_CSQ_IB1_STAT) >> 16;
+	cp_state.ib2_rem += gpu_read(gpu, REG_A6XX_CP_CSQ_IB2_STAT) >> 16;
+
+	progress = !!memcmp(&cp_state, &ring->last_cp_state, sizeof(cp_state));
+
+	ring->last_cp_state = cp_state;
+
+	return progress;
+}
+
 static u32 a618_get_speed_bin(u32 fuse)
 {
 	if (fuse == 0)
@@ -1879,7 +1904,7 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse)
 
 	if (val == UINT_MAX) {
 		DRM_DEV_ERROR(dev,
-			"missing support for speed-bin: %u. Some OPPs may not be supported by hardware",
+			"missing support for speed-bin: %u. Some OPPs may not be supported by hardware\n",
 			fuse);
 		return UINT_MAX;
 	}
@@ -1889,7 +1914,7 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse)
 
 static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev)
 {
-	u32 supp_hw = UINT_MAX;
+	u32 supp_hw;
 	u32 speedbin;
 	int ret;
 
@@ -1901,15 +1926,13 @@ static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev)
 	if (ret == -ENOENT) {
 		return 0;
 	} else if (ret) {
-		DRM_DEV_ERROR(dev,
-			      "failed to read speed-bin (%d). Some OPPs may not be supported by hardware",
-			      ret);
-		goto done;
+		dev_err_probe(dev, ret,
+			      "failed to read speed-bin. Some OPPs may not be supported by hardware\n");
+		return ret;
 	}
 
 	supp_hw = fuse_to_supp_hw(dev, rev, speedbin);
 
-done:
 	ret = devm_pm_opp_set_supported_hw(dev, &supp_hw, 1);
 	if (ret)
 		return ret;
@@ -1942,6 +1965,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.create_address_space = a6xx_create_address_space,
 		.create_private_address_space = a6xx_create_private_address_space,
 		.get_rptr = a6xx_get_rptr,
+		.progress = a6xx_progress,
 	},
 	.get_timestamp = a6xx_get_timestamp,
 };
@@ -1978,13 +2002,6 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 			adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), info->rev)))
 		adreno_gpu->base.hw_apriv = true;
 
-	/*
-	 * For now only clamp to idle freq for devices where this is known not
-	 * to cause power supply issues:
-	 */
-	if (info && (info->revn == 618))
-		gpu->clamp_to_idle = true;
-
 	a6xx_llc_slices_init(pdev, a6xx_gpu);
 
 	ret = a6xx_set_supported_hw(&pdev->dev, config->rev);
@@ -1999,6 +2016,13 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 		return ERR_PTR(ret);
 	}
 
+	/*
+	 * For now only clamp to idle freq for devices where this is known not
+	 * to cause power supply issues:
+	 */
+	if (adreno_is_a618(adreno_gpu) || adreno_is_7c3(adreno_gpu))
+		gpu->clamp_to_idle = true;
+
 	/* Check if there is a GMU phandle and set it up */
 	node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
index a5c3d1ed255a..a023d5f962dc 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
@@ -147,8 +147,7 @@ static int a6xx_crashdumper_run(struct msm_gpu *gpu,
 	/* Make sure all pending memory writes are posted */
 	wmb();
 
-	gpu_write64(gpu, REG_A6XX_CP_CRASH_SCRIPT_BASE_LO,
-		REG_A6XX_CP_CRASH_SCRIPT_BASE_HI, dumper->iova);
+	gpu_write64(gpu, REG_A6XX_CP_CRASH_SCRIPT_BASE_LO, dumper->iova);
 
 	gpu_write(gpu, REG_A6XX_CP_CRASH_DUMP_CNTL, 1);
 
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index cffba9ca7ec2..8b0b0ac74a6f 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -418,7 +418,6 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv)
 	priv->dev = ddev;
 
 	priv->wq = alloc_ordered_workqueue("msm", 0);
-	priv->hangcheck_period = DRM_MSM_HANGCHECK_DEFAULT_PERIOD;
 
 	INIT_LIST_HEAD(&priv->objects);
 	mutex_init(&priv->obj_lock);
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index b2ea262296a4..d4e0ef608950 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -224,7 +224,13 @@ struct msm_drm_private {
 
 	struct drm_atomic_state *pm_state;
 
-	/* For hang detection, in ms */
+	/**
+	 * hangcheck_period: For hang detection, in ms
+	 *
+	 * Note that in practice, a submit/job will get at least two hangcheck
+	 * periods, due to checking for progress being implemented as simply
+	 * "have the CP position registers changed since last time?"
+	 */
 	unsigned int hangcheck_period;
 
 	/**
diff --git a/drivers/gpu/drm/msm/msm_gem_shrinker.c b/drivers/gpu/drm/msm/msm_gem_shrinker.c
index 1de14e67f96b..051bdbc093cf 100644
--- a/drivers/gpu/drm/msm/msm_gem_shrinker.c
+++ b/drivers/gpu/drm/msm/msm_gem_shrinker.c
@@ -15,7 +15,7 @@
 /* Default disabled for now until it has some more testing on the different
  * iommu combinations that can be paired with the driver:
  */
-static bool enable_eviction = false;
+static bool enable_eviction = true;
 MODULE_PARM_DESC(enable_eviction, "Enable swappable GEM buffers");
 module_param(enable_eviction, bool, 0600);
 
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 45a3e5cadc7d..73a2ca122c57 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -334,8 +334,7 @@ static int submit_fence_sync(struct msm_gem_submit *submit, bool no_implicit)
 		if (ret)
 			return ret;
 
-		/* exclusive fences must be ordered */
-		if (no_implicit && !write)
+		if (no_implicit)
 			continue;
 
 		ret = drm_sched_job_add_implicit_dependencies(&submit->base,
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 021f4e29b613..30ed45af76ad 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -492,6 +492,21 @@ static void hangcheck_timer_reset(struct msm_gpu *gpu)
 			round_jiffies_up(jiffies + msecs_to_jiffies(priv->hangcheck_period)));
 }
 
+static bool made_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	if (ring->hangcheck_progress_retries >= DRM_MSM_HANGCHECK_PROGRESS_RETRIES)
+		return false;
+
+	if (!gpu->funcs->progress)
+		return false;
+
+	if (!gpu->funcs->progress(gpu, ring))
+		return false;
+
+	ring->hangcheck_progress_retries++;
+	return true;
+}
+
 static void hangcheck_handler(struct timer_list *t)
 {
 	struct msm_gpu *gpu = from_timer(gpu, t, hangcheck_timer);
@@ -502,9 +517,12 @@ static void hangcheck_handler(struct timer_list *t)
 	if (fence != ring->hangcheck_fence) {
 		/* some progress has been made.. ya! */
 		ring->hangcheck_fence = fence;
-	} else if (fence_before(fence, ring->fctx->last_fence)) {
+		ring->hangcheck_progress_retries = 0;
+	} else if (fence_before(fence, ring->fctx->last_fence) &&
+			!made_progress(gpu, ring)) {
 		/* no progress and not done.. hung! */
 		ring->hangcheck_fence = fence;
+		ring->hangcheck_progress_retries = 0;
 		DRM_DEV_ERROR(dev->dev, "%s: hangcheck detected gpu lockup rb %d!\n",
 				gpu->name, ring->id);
 		DRM_DEV_ERROR(dev->dev, "%s:     completed fence: %u\n",
@@ -830,6 +848,7 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 		struct msm_gpu *gpu, const struct msm_gpu_funcs *funcs,
 		const char *name, struct msm_gpu_config *config)
 {
+	struct msm_drm_private *priv = drm->dev_private;
 	int i, ret, nr_rings = config->nr_rings;
 	void *memptrs;
 	uint64_t memptrs_iova;
@@ -857,6 +876,16 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	kthread_init_work(&gpu->recover_work, recover_worker);
 	kthread_init_work(&gpu->fault_work, fault_worker);
 
+	priv->hangcheck_period = DRM_MSM_HANGCHECK_DEFAULT_PERIOD;
+
+	/*
+	 * If progress detection is supported, halve the hangcheck timer
+	 * duration, as it takes two iterations of the hangcheck handler
+	 * to detect a hang.
+	 */
+	if (funcs->progress)
+		priv->hangcheck_period /= 2;
+
 	timer_setup(&gpu->hangcheck_timer, hangcheck_handler, 0);
 
 	spin_lock_init(&gpu->perf_lock);
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index 58a72e6b1400..651786bc55e5 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -78,6 +78,15 @@ struct msm_gpu_funcs {
 	struct msm_gem_address_space *(*create_private_address_space)
 		(struct msm_gpu *gpu);
 	uint32_t (*get_rptr)(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
+
+	/**
+	 * progress: Has the GPU made progress?
+	 *
+	 * Return true if GPU position in cmdstream has advanced (or changed)
+	 * since the last call.  To avoid false negatives, this should account
+	 * for cmdstream that is buffered in this FIFO upstream of the CP fw.
+	 */
+	bool (*progress)(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
 };
 
 /* Additional state for iommu faults: */
@@ -237,6 +246,7 @@ struct msm_gpu {
 #define DRM_MSM_INACTIVE_PERIOD   66 /* in ms (roughly four frames) */
 
 #define DRM_MSM_HANGCHECK_DEFAULT_PERIOD 500 /* in ms */
+#define DRM_MSM_HANGCHECK_PROGRESS_RETRIES 3
 	struct timer_list hangcheck_timer;
 
 	/* Fault info for most recent iova fault: */
@@ -540,7 +550,7 @@ static inline void gpu_rmw(struct msm_gpu *gpu, u32 reg, u32 mask, u32 or)
 	msm_rmw(gpu->mmio + (reg << 2), mask, or);
 }
 
-static inline u64 gpu_read64(struct msm_gpu *gpu, u32 lo, u32 hi)
+static inline u64 gpu_read64(struct msm_gpu *gpu, u32 reg)
 {
 	u64 val;
 
@@ -558,17 +568,17 @@ static inline u64 gpu_read64(struct msm_gpu *gpu, u32 lo, u32 hi)
 	 * when the lo is read, so make sure to read the lo first to trigger
 	 * that
 	 */
-	val = (u64) msm_readl(gpu->mmio + (lo << 2));
-	val |= ((u64) msm_readl(gpu->mmio + (hi << 2)) << 32);
+	val = (u64) msm_readl(gpu->mmio + (reg << 2));
+	val |= ((u64) msm_readl(gpu->mmio + ((reg + 1) << 2)) << 32);
 
 	return val;
 }
 
-static inline void gpu_write64(struct msm_gpu *gpu, u32 lo, u32 hi, u64 val)
+static inline void gpu_write64(struct msm_gpu *gpu, u32 reg, u64 val)
 {
 	/* Why not a writeq here? Read the screed above */
-	msm_writel(lower_32_bits(val), gpu->mmio + (lo << 2));
-	msm_writel(upper_32_bits(val), gpu->mmio + (hi << 2));
+	msm_writel(lower_32_bits(val), gpu->mmio + (reg << 2));
+	msm_writel(upper_32_bits(val), gpu->mmio + ((reg + 1) << 2));
 }
 
 int msm_gpu_pm_suspend(struct msm_gpu *gpu);
diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.h b/drivers/gpu/drm/msm/msm_ringbuffer.h
index 2a5045abe46e..698b333abccd 100644
--- a/drivers/gpu/drm/msm/msm_ringbuffer.h
+++ b/drivers/gpu/drm/msm/msm_ringbuffer.h
@@ -35,6 +35,11 @@ struct msm_rbmemptrs {
 	volatile u64 ttbr0;
 };
 
+struct msm_cp_state {
+	uint64_t ib1_base, ib2_base;
+	uint32_t ib1_rem, ib2_rem;
+};
+
 struct msm_ringbuffer {
 	struct msm_gpu *gpu;
 	int id;
@@ -64,6 +69,29 @@ struct msm_ringbuffer {
 	uint64_t memptrs_iova;
 	struct msm_fence_context *fctx;
 
+	/**
+	 * hangcheck_progress_retries:
+	 *
+	 * The number of extra hangcheck duration cycles that we have given
+	 * due to it appearing that the GPU is making forward progress.
+	 *
+	 * For GPU generations which support progress detection (see.
+	 * msm_gpu_funcs::progress()), if the GPU appears to be making progress
+	 * (ie. the CP has advanced in the command stream, we'll allow up to
+	 * DRM_MSM_HANGCHECK_PROGRESS_RETRIES expirations of the hangcheck timer
+	 * before killing the job.  But to detect progress we need two sample
+	 * points, so the duration of the hangcheck timer is halved.  In other
+	 * words we'll let the submit run for up to:
+	 *
+	 * (DRM_MSM_HANGCHECK_DEFAULT_PERIOD / 2) * (DRM_MSM_HANGCHECK_PROGRESS_RETRIES + 1)
+	 */
+	int hangcheck_progress_retries;
+
+	/**
+	 * last_cp_state: The state of the CP at the last call to gpu->progress()
+	 */
+	struct msm_cp_state last_cp_state;
+
 	/*
 	 * preempt_lock protects preemption and serializes wptr updates against
 	 * preemption.  Can be aquired from irq context.