From fe732d6a5aaffc328676c075f8a3abf2b3b032a4 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Tue, 15 Mar 2022 17:48:18 +0800
Subject: drm/amdgpu: add UTCL2 RAS poison query for Aldebaran (v2)
Git-commit: 6475ae2b742876aa9b2a0aff7ba60f5c81917614
Patch-mainline: v5.19-rc1
References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225
Add help functions to query and reset RAS UTCL2 poison status.
v2: implement it on amdgpu side and kfd only calls it.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Patrik Jakobsson <pjakobsson@suse.de>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 8 ++++++++
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 1 +
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c | 14 ++++++++++++++
4 files changed, 24 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 6ca1db3c243f..c18c4be1e4ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -724,3 +724,11 @@ void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bo
else if (reset)
amdgpu_amdkfd_gpu_reset(adev);
}
+
+bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev)
+{
+ if (adev->gfx.ras->query_utcl2_poison_status)
+ return adev->gfx.ras->query_utcl2_poison_status(adev);
+ else
+ return false;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 4cb14c2fe53f..0838926a8ef0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -301,6 +301,7 @@ void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev,
bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem);
void amdgpu_amdkfd_block_mmu_notifications(void *p);
int amdgpu_amdkfd_criu_resume(void *p);
+bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev);
#if IS_ENABLED(CONFIG_HSA_AMD)
void amdgpu_amdkfd_gpuvm_init_mem_limits(void);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index dcb3c7871c73..5ed9b8a4c571 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -202,6 +202,7 @@ struct amdgpu_cu_info {
struct amdgpu_gfx_ras {
struct amdgpu_ras_block_object ras_block;
void (*enable_watchdog_timer)(struct amdgpu_device *adev);
+ bool (*query_utcl2_poison_status)(struct amdgpu_device *adev);
};
struct amdgpu_gfx_funcs {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index 7653ebd0e67b..3a797424579c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -1930,6 +1930,19 @@ static void gfx_v9_4_2_reset_sq_timeout_status(struct amdgpu_device *adev)
mutex_unlock(&adev->grbm_idx_mutex);
}
+static bool gfx_v9_4_2_query_uctl2_poison_status(struct amdgpu_device *adev)
+{
+ u32 status = 0;
+ struct amdgpu_vmhub *hub;
+
+ hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+ status = RREG32(hub->vm_l2_pro_fault_status);
+ /* reset page fault status */
+ WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
+
+ return REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, FED);
+}
+
struct amdgpu_ras_block_hw_ops gfx_v9_4_2_ras_ops = {
.ras_error_inject = &gfx_v9_4_2_ras_error_inject,
.query_ras_error_count = &gfx_v9_4_2_query_ras_error_count,
@@ -1943,4 +1956,5 @@ struct amdgpu_gfx_ras gfx_v9_4_2_ras = {
.hw_ops = &gfx_v9_4_2_ras_ops,
},
.enable_watchdog_timer = &gfx_v9_4_2_enable_watchdog_timer,
+ .query_utcl2_poison_status = gfx_v9_4_2_query_uctl2_poison_status,
};
--
2.38.1