Merge tag 'drm-fixes-2024-12-06' of https://gitlab.freedesktop.org/drm/kernel

Pull more drm fixes from Simona Vetter:
 "Due to mailing list unreliability we missed the amdgpu pull, hence
  part two with that now included:

   - amdgu: mostly display fixes + jpeg vcn 1.0, sriov, dcn4.0 resume
     fixes

   - amdkfd fixes"

* tag 'drm-fixes-2024-12-06' of https://gitlab.freedesktop.org/drm/kernel:
  drm/amdgpu: rework resume handling for display (v2)
  drm/amd/pm: fix and simplify workload handling
  Revert "drm/amd/pm: correct the workload setting"
  drm/amdgpu: fix sriov reinit late orders
  drm/amdgpu: Fix ISP hw init issue
  drm/amd/display: Add hblank borrowing support
  drm/amd/display: Limit VTotal range to max hw cap minus fp
  drm/amd/display: Correct prefetch calculation
  drm/amd/display: Add option to retrieve detile buffer size
  drm/amd/display: Add a left edge pixel if in YCbCr422 or YCbCr420 and odm
  drm/amdkfd: hard-code cacheline for gc943,gc944
  drm/amdkfd: add MEC version that supports no PCIe atomics for GFX12
  drm/amd/display: Fix programming backlight on OLED panels
  drm/amd: Sanity check the ACPI EDID
  drm/amdgpu/hdp7.0: do a posting read when flushing HDP
  drm/amdgpu/hdp6.0: do a posting read when flushing HDP
  drm/amdgpu/hdp5.2: do a posting read when flushing HDP
  drm/amdgpu/hdp5.0: do a posting read when flushing HDP
  drm/amdgpu/hdp4.0: do a posting read when flushing HDP
  drm/amdgpu/jpeg1.0: fix idle work handler

7 files changed, 90 insertions, 39 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 9095c05e0269..96316111300a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -145,7 +145,7 @@ const char *amdgpu_asic_name[] = {
 	"LAST",
 };
 
-#define AMDGPU_IP_BLK_MASK_ALL GENMASK(AMDGPU_MAX_IP_NUM - 1, 0)
+#define AMDGPU_IP_BLK_MASK_ALL GENMASK(AMDGPU_MAX_IP_NUM, 0)
 /*
  * Default init level where all blocks are expected to be initialized. This is
  * the level of initialization expected by default and also after a full reset
@@ -3670,9 +3670,11 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
 				continue;
 
 			r = block->version->funcs->hw_init(&adev->ip_blocks[i]);
-			DRM_INFO("RE-INIT-early: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
-			if (r)
+			if (r) {
+				dev_err(adev->dev, "RE-INIT-early: %s failed\n",
+					 block->version->funcs->name);
 				return r;
+			}
 			block->status.hw = true;
 		}
 	}
@@ -3682,7 +3684,8 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
 
 static int amdgpu_device_ip_reinit_late_sriov(struct amdgpu_device *adev)
 {
-	int i, r;
+	struct amdgpu_ip_block *block;
+	int i, r = 0;
 
 	static enum amd_ip_block_type ip_order[] = {
 		AMD_IP_BLOCK_TYPE_SMC,
@@ -3697,34 +3700,28 @@ static int amdgpu_device_ip_reinit_late_sriov(struct amdgpu_device *adev)
 	};
 
 	for (i = 0; i < ARRAY_SIZE(ip_order); i++) {
-		int j;
-		struct amdgpu_ip_block *block;
-
-		for (j = 0; j < adev->num_ip_blocks; j++) {
-			block = &adev->ip_blocks[j];
+		block = amdgpu_device_ip_get_ip_block(adev, ip_order[i]);
 
-			if (block->version->type != ip_order[i] ||
-				!block->status.valid ||
-				block->status.hw)
-				continue;
+		if (!block)
+			continue;
 
+		if (block->status.valid && !block->status.hw) {
 			if (block->version->type == AMD_IP_BLOCK_TYPE_SMC) {
-				r = amdgpu_ip_block_resume(&adev->ip_blocks[i]);
-				if (r)
-					return r;
+				r = amdgpu_ip_block_resume(block);
 			} else {
-				r = block->version->funcs->hw_init(&adev->ip_blocks[i]);
-				if (r) {
-					DRM_ERROR("hw_init of IP block <%s> failed %d\n",
-						  adev->ip_blocks[i].version->funcs->name, r);
-					return r;
-				}
-				block->status.hw = true;
+				r = block->version->funcs->hw_init(block);
 			}
+
+			if (r) {
+				dev_err(adev->dev, "RE-INIT-late: %s failed\n",
+					 block->version->funcs->name);
+				break;
+			}
+			block->status.hw = true;
 		}
 	}
 
-	return 0;
+	return r;
 }
 
 /**
@@ -3765,7 +3762,7 @@ static int amdgpu_device_ip_resume_phase1(struct amdgpu_device *adev)
  *
  * @adev: amdgpu_device pointer
  *
- * First resume function for hardware IPs.  The list of all the hardware
+ * Second resume function for hardware IPs.  The list of all the hardware
  * IPs that make up the asic is walked and the resume callbacks are run for
  * all blocks except COMMON, GMC, and IH.  resume puts the hardware into a
  * functional state after a suspend and updates the software state as
@@ -3783,6 +3780,7 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev)
 		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON ||
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC ||
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH ||
+		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_DCE ||
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP)
 			continue;
 		r = amdgpu_ip_block_resume(&adev->ip_blocks[i]);
@@ -3794,6 +3792,36 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev)
 }
 
 /**
+ * amdgpu_device_ip_resume_phase3 - run resume for hardware IPs
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Third resume function for hardware IPs.  The list of all the hardware
+ * IPs that make up the asic is walked and the resume callbacks are run for
+ * all DCE.  resume puts the hardware into a functional state after a suspend
+ * and updates the software state as necessary.  This function is also used
+ * for restoring the GPU after a GPU reset.
+ *
+ * Returns 0 on success, negative error code on failure.
+ */
+static int amdgpu_device_ip_resume_phase3(struct amdgpu_device *adev)
+{
+	int i, r;
+
+	for (i = 0; i < adev->num_ip_blocks; i++) {
+		if (!adev->ip_blocks[i].status.valid || adev->ip_blocks[i].status.hw)
+			continue;
+		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_DCE) {
+			r = amdgpu_ip_block_resume(&adev->ip_blocks[i]);
+			if (r)
+				return r;
+		}
+	}
+
+	return 0;
+}
+
+/**
  * amdgpu_device_ip_resume - run resume for hardware IPs
  *
  * @adev: amdgpu_device pointer
@@ -3822,6 +3850,13 @@ static int amdgpu_device_ip_resume(struct amdgpu_device *adev)
 	if (adev->mman.buffer_funcs_ring->sched.ready)
 		amdgpu_ttm_set_buffer_funcs_status(adev, true);
 
+	if (r)
+		return r;
+
+	amdgpu_fence_driver_hw_init(adev);
+
+	r = amdgpu_device_ip_resume_phase3(adev);
+
 	return r;
 }
 
@@ -4902,7 +4937,6 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
 		dev_err(adev->dev, "amdgpu_device_ip_resume failed (%d).\n", r);
 		goto exit;
 	}
-	amdgpu_fence_driver_hw_init(adev);
 
 	if (!adev->in_s0ix) {
 		r = amdgpu_amdkfd_resume(adev, adev->in_runpm);
@@ -5487,6 +5521,10 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context)
 				if (tmp_adev->mman.buffer_funcs_ring->sched.ready)
 					amdgpu_ttm_set_buffer_funcs_status(tmp_adev, true);
 
+				r = amdgpu_device_ip_resume_phase3(tmp_adev);
+				if (r)
+					goto out;
+
 				if (vram_lost)
 					amdgpu_device_fill_reset_magic(tmp_adev);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
index e019249883fb..194026e9be33 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
@@ -40,10 +40,12 @@
 static void hdp_v4_0_flush_hdp(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg)
+	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	else
+		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+	} else {
 		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
+	}
 }
 
 static void hdp_v4_0_invalidate_hdp(struct amdgpu_device *adev,
@@ -54,11 +56,13 @@ static void hdp_v4_0_invalidate_hdp(struct amdgpu_device *adev,
 	    amdgpu_ip_version(adev, HDP_HWIP, 0) == IP_VERSION(4, 4, 5))
 		return;
 
-	if (!ring || !ring->funcs->emit_wreg)
+	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE, 1);
-	else
+		RREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE);
+	} else {
 		amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET(
 			HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1);
+	}
 }
 
 static void hdp_v4_0_query_ras_error_count(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c
index ed7facacf2fe..d3962d469088 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c
@@ -31,10 +31,12 @@
 static void hdp_v5_0_flush_hdp(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg)
+	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	else
+		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+	} else {
 		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
+	}
 }
 
 static void hdp_v5_0_invalidate_hdp(struct amdgpu_device *adev,
@@ -42,6 +44,7 @@ static void hdp_v5_0_invalidate_hdp(struct amdgpu_device *adev,
 {
 	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE, 1);
+		RREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE);
 	} else {
 		amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET(
 					HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1);
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
index 29c3484ae1f1..f52552c5fa27 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
@@ -31,13 +31,15 @@
 static void hdp_v5_2_flush_hdp(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg)
+	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2,
 			0);
-	else
+		RREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+	} else {
 		amdgpu_ring_emit_wreg(ring,
 			(adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2,
 			0);
+	}
 }
 
 static void hdp_v5_2_update_mem_power_gating(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c
index 33736d361dd0..6948fe9956ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c
@@ -34,10 +34,12 @@
 static void hdp_v6_0_flush_hdp(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg)
+	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	else
+		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+	} else {
 		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
+	}
 }
 
 static void hdp_v6_0_update_clock_gating(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c
index 1c99bb09e2a1..63820329f67e 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c
@@ -31,10 +31,12 @@
 static void hdp_v7_0_flush_hdp(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg)
+	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	else
+		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+	} else {
 		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
+	}
 }
 
 static void hdp_v7_0_update_clock_gating(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
index 7319299f25ae..03b8b7cd5229 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
@@ -604,7 +604,7 @@ static void jpeg_v1_0_set_irq_funcs(struct amdgpu_device *adev)
 static void jpeg_v1_0_ring_begin_use(struct amdgpu_ring *ring)
 {
 	struct	amdgpu_device *adev = ring->adev;
-	bool	set_clocks = !cancel_delayed_work_sync(&adev->jpeg.idle_work);
+	bool	set_clocks = !cancel_delayed_work_sync(&adev->vcn.idle_work);
 	int		cnt = 0;
 
 	mutex_lock(&adev->vcn.vcn1_jpeg1_workaround);