[PATCH] drm/amdgpu: use GPU_HDP_FLUSH for sriov

[Victor Zhao <Victor.Zhao@amd.com>]

Currently SRIOV runtime will use kiq to write HDP_MEM_FLUSH_CNTL for
hdp flush. This register need to be write from CPU for nbif to aware,
otherwise it will not work.
Add kiq ring callback to emit GPU_HDP_FLUSH, and use as generic hdp
flush option.

Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 61 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c | 12 ++---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c  |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c  |  5 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c  |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c   |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c   |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c |  1 +
 drivers/gpu/drm/amd/amdgpu/vi.c         |  8 ++--
 10 files changed, 80 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 7f02e36ccc1e..9cb13424ab8d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1194,6 +1194,67 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint3
 	dev_err(adev->dev, "failed to write reg:%x\n", reg);
 }
 
+void amdgpu_kiq_hdp_flush(struct amdgpu_device *adev, uint32_t xcc_id)
+{
+	signed long r, cnt = 0;
+	unsigned long flags;
+	uint32_t seq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
+	struct amdgpu_ring *ring = &kiq->ring;
+
+	BUG_ON(!ring->funcs->emit_hdp_flush);
+
+	if (amdgpu_device_skip_hw_access(adev))
+		return;
+
+	spin_lock_irqsave(&kiq->ring_lock, flags);
+	r = amdgpu_ring_alloc(ring, 32);
+	if (r)
+		goto failed_unlock;
+
+	amdgpu_ring_emit_hdp_flush(ring);
+	r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
+	if (r)
+		goto failed_undo;
+
+	amdgpu_ring_commit(ring);
+	spin_unlock_irqrestore(&kiq->ring_lock, flags);
+
+	r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
+
+	/* don't wait anymore for gpu reset case because this way may
+	 * block gpu_recover() routine forever, e.g. this virt_kiq_rreg
+	 * is triggered in TTM and ttm_bo_lock_delayed_workqueue() will
+	 * never return if we keep waiting in virt_kiq_rreg, which cause
+	 * gpu_recover() hang there.
+	 *
+	 * also don't wait anymore for IRQ context
+	 * */
+	if (r < 1 && (amdgpu_in_reset(adev) || in_interrupt()))
+		goto failed_kiq_hdp_flush;
+
+	might_sleep();
+	while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
+		if (amdgpu_in_reset(adev))
+			goto failed_kiq_hdp_flush;
+
+		msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
+		r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
+	}
+
+	if (cnt > MAX_KIQ_REG_TRY)
+		goto failed_kiq_hdp_flush;
+
+	return;
+
+failed_undo:
+	amdgpu_ring_undo(ring);
+failed_unlock:
+	spin_unlock_irqrestore(&kiq->ring_lock, flags);
+failed_kiq_hdp_flush:
+	dev_err(adev->dev, "failed to flush HDP via KIQ\n");
+}
+
 int amdgpu_gfx_get_num_kcq(struct amdgpu_device *adev)
 {
 	if (amdgpu_num_kcq == -1) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index fb5f7a0ee029..5bccd2cc9518 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -615,6 +615,7 @@ int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev,
 				  struct amdgpu_iv_entry *entry);
 uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t xcc_id);
 void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t xcc_id);
+void amdgpu_kiq_hdp_flush(struct amdgpu_device *adev, uint32_t xcc_id);
 int amdgpu_gfx_get_num_kcq(struct amdgpu_device *adev);
 void amdgpu_gfx_cp_init_microcode(struct amdgpu_device *adev, uint32_t ucode_id);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c
index 6e02fb9ac2f6..803f65dc4563 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c
@@ -51,18 +51,16 @@ int amdgpu_hdp_ras_sw_init(struct amdgpu_device *adev)
 void amdgpu_hdp_generic_flush(struct amdgpu_device *adev,
 			      struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg) {
+	if (ring && ring->funcs->emit_hdp_flush) {
+		amdgpu_ring_emit_hdp_flush(ring);
+	} else if (amdgpu_sriov_runtime(adev)) {
+		amdgpu_kiq_hdp_flush(adev, 0);
+	} else {
 		WREG32((adev->rmmio_remap.reg_offset +
 			KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >>
 			       2,
 		       0);
 		if (adev->nbio.funcs->get_memsize)
 			adev->nbio.funcs->get_memsize(adev);
-	} else {
-		amdgpu_ring_emit_wreg(ring,
-				      (adev->rmmio_remap.reg_offset +
-				       KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >>
-					      2,
-				      0);
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 8841d7213de4..751732f3e883 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -9951,6 +9951,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_kiq = {
 	.emit_wreg = gfx_v10_0_ring_emit_wreg,
 	.emit_reg_wait = gfx_v10_0_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait,
+	.emit_hdp_flush = gfx_v10_0_ring_emit_hdp_flush,
 };
 
 static void gfx_v10_0_set_ring_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 66c47c466532..10d2219866f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -2438,7 +2438,7 @@ static int gfx_v11_0_rlc_load_microcode(struct amdgpu_device *adev)
 			if (version_minor == 3)
 				gfx_v11_0_load_rlcp_rlcv_microcode(adev);
 		}
-		
+
 		return 0;
 	}
 
@@ -3886,7 +3886,7 @@ static int gfx_v11_0_cp_compute_load_microcode(struct amdgpu_device *adev)
 	}
 
 	memcpy(fw, fw_data, fw_size);
-	
+
 	amdgpu_bo_kunmap(adev->gfx.mec.mec_fw_obj);
 	amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_obj);
 
@@ -7320,6 +7320,7 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_kiq = {
 	.emit_wreg = gfx_v11_0_ring_emit_wreg,
 	.emit_reg_wait = gfx_v11_0_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v11_0_ring_emit_reg_write_reg_wait,
+	.emit_hdp_flush = gfx_v11_0_ring_emit_hdp_flush,
 };
 
 static void gfx_v11_0_set_ring_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index 710ec9c34e43..e2bb8668150d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -5597,6 +5597,7 @@ static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_kiq = {
 	.emit_wreg = gfx_v12_0_ring_emit_wreg,
 	.emit_reg_wait = gfx_v12_0_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v12_0_ring_emit_reg_write_reg_wait,
+	.emit_hdp_flush = gfx_v12_0_ring_emit_hdp_flush,
 };
 
 static void gfx_v12_0_set_ring_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 0856ff65288c..d3d0a4b0380c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -6939,6 +6939,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_kiq = {
 	.pad_ib = amdgpu_ring_generic_pad_ib,
 	.emit_rreg = gfx_v8_0_ring_emit_rreg,
 	.emit_wreg = gfx_v8_0_ring_emit_wreg,
+	.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
 };
 
 static void gfx_v8_0_set_ring_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index dd19a97436db..f1a2efc2a8d0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -7586,6 +7586,7 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_kiq = {
 	.emit_wreg = gfx_v9_0_ring_emit_wreg,
 	.emit_reg_wait = gfx_v9_0_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait,
+	.emit_hdp_flush = gfx_v9_0_ring_emit_hdp_flush,
 };
 
 static void gfx_v9_0_set_ring_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 77f9d5b9a556..b1fa4036befb 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -4798,6 +4798,7 @@ static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_kiq = {
 	.emit_wreg = gfx_v9_4_3_ring_emit_wreg,
 	.emit_reg_wait = gfx_v9_4_3_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v9_4_3_ring_emit_reg_write_reg_wait,
+	.emit_hdp_flush = gfx_v9_4_3_ring_emit_hdp_flush,
 };
 
 static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index a611a7345125..2f549f1a9018 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -1309,11 +1309,13 @@ static uint32_t vi_get_rev_id(struct amdgpu_device *adev)
 
 static void vi_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring)
 {
-	if (!ring || !ring->funcs->emit_wreg) {
+	if (ring && ring->funcs->emit_hdp_flush) {
+		amdgpu_ring_emit_hdp_flush(ring);
+	} else if (amdgpu_sriov_runtime(adev)) {
+		amdgpu_kiq_hdp_flush(adev, 0);
+	} else {
 		WREG32(mmHDP_MEM_COHERENCY_FLUSH_CNTL, 1);
 		RREG32(mmHDP_MEM_COHERENCY_FLUSH_CNTL);
-	} else {
-		amdgpu_ring_emit_wreg(ring, mmHDP_MEM_COHERENCY_FLUSH_CNTL, 1);
 	}
 }
 
-- 
2.25.1