[PATCH 4/4] drm/amdkfd: use queue reset for SDMA RAS poison consumption

[Tao Zhou]

Replace mode2 reset with queue reset to improve system performance.

v2: fallback to mode1 reset if queue reset fails.
    move the queue reset handling from KFD to KGD.

Signed-off-by: Tao Zhou <tao.zhou1 at amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index 0018d04f4b33..b9429774a006 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -197,6 +197,9 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 	case SOC15_IH_CLIENTID_SDMA2:
 	case SOC15_IH_CLIENTID_SDMA3:
 	case SOC15_IH_CLIENTID_SDMA4:
+	case SOC15_IH_CLIENTID_SDMA5:
+	case SOC15_IH_CLIENTID_SDMA6:
+	case SOC15_IH_CLIENTID_SDMA7:
 		block = AMDGPU_RAS_BLOCK__SDMA;
 		if (amdgpu_ip_version(dev->adev, SDMA0_HWIP, 0) == IP_VERSION(4, 4, 2)) {
 			/* driver mode-2 for gfx poison is only supported by
@@ -204,7 +207,7 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 			if (dev->adev->pm.fw_version < 0x00557300)
 				reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
 			else
-				reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+				reset = 0;
 		} else if (amdgpu_ip_version(dev->adev, SDMA0_HWIP, 0) == IP_VERSION(4, 4, 5)) {
 			/* driver mode-2 for gfx poison is only supported by
 			 * pmfw 0x05550C00 and onwards */
@@ -223,6 +226,14 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 		return;
 	}
 
+	if (!reset) {
+		ret = amdgpu_amdkfd_ras_poison_queue_reset(dev->adev, block,
+				 client_id, vmid, node_id);
+		/* fallback to mode1 reset */
+		if (ret)
+			reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+	}
+
 	ret = amdgpu_ras_mark_ras_event(dev->adev, type);
 	if (ret)
 		return;
-- 
2.34.1