Signed-off-by: Yang Wang <kevinyang.wang@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c | 32 ++++++++++++++++++++++++-
drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 15 ++++++++++++
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 3 +++
4 files changed, 51 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
index c447064fcaec..a460cde20cf2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
@@ -563,12 +563,42 @@ static int add_aca_handle(struct amdgpu_device *adev, struct aca_handle_manager
return 0;
}
+static ssize_t aca_sysfs_read(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct aca_handle *handle = container_of(attr, struct aca_handle, aca_attr);
+
+ /* NOTE: the aca cache will be auto cleared once read,
+ * So the driver should unify the query entry point, forward request to ras query interface directly */
+ return amdgpu_ras_aca_sysfs_read(dev, attr, handle, buf, handle->data);
+}
+
+static int add_aca_sysfs(struct amdgpu_device *adev, struct aca_handle *handle)
+{
+ struct device_attribute *aca_attr = &handle->aca_attr;
+
+ snprintf(handle->attr_name, sizeof(handle->attr_name) - 1, "aca_%s", handle->name);
+ aca_attr->show = aca_sysfs_read;
+ aca_attr->attr.name = handle->attr_name;
+ aca_attr->attr.mode = S_IRUGO;
+ sysfs_attr_init(&aca_attr->attr);
+
+ return sysfs_add_file_to_group(&adev->dev->kobj,
+ &aca_attr->attr,
+ "ras");
+}
+
int amdgpu_aca_add_handle(struct amdgpu_device *adev, struct aca_handle *handle,
const char *name, const struct aca_info *ras_info, void *data)
{
struct amdgpu_aca *aca = &adev->aca;
+ int ret;
+
+ ret = add_aca_handle(adev, &aca->mgr, handle, name, ras_info, data);
+ if (ret)
+ return ret;
- return add_aca_handle(adev, &aca->mgr, handle, name, ras_info, data);
+ return add_aca_sysfs(adev, handle);
}
static void remove_aca(struct aca_handle *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h
index 747150c02609..bb0a3be72cc8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h
@@ -150,6 +150,8 @@ struct aca_handle {
struct aca_handle_manager *mgr;
struct aca_error_cache error_cache;
const struct aca_bank_ops *bank_ops;
+ struct device_attribute aca_attr;
+ char attr_name[64];
const char *name;
u32 mask;
void *data;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index dc11cc98c673..7048bf853cf6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1168,6 +1168,21 @@ static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, s
}
}
+ssize_t amdgpu_ras_aca_sysfs_read(struct device *dev, struct device_attribute *attr,
+ struct aca_handle *handle, char *buf, void *data)
+{
+ struct ras_manager *obj = container_of(handle, struct ras_manager, aca_handle);
+ struct ras_query_if info = {
+ .head = obj->head,
+ };
+
+ if (amdgpu_ras_query_error_status(obj->adev, &info))
+ return -EINVAL;
+
+ return sysfs_emit(buf, "%s: %lu\n%s: %lu\n", "ue", info.ue_count,
+ "ce", info.ce_count);
+}
+
static int amdgpu_ras_query_error_status_helper(struct amdgpu_device *adev,
struct ras_query_if *info,
struct ras_err_data *err_data,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 408e21c3cc88..8c487f3bfbf1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -821,4 +821,7 @@ int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data,
struct amdgpu_smuio_mcm_config_info *mcm_info,
struct ras_err_addr *err_addr, u64 count);
+ssize_t amdgpu_ras_aca_sysfs_read(struct device *dev, struct device_attribute *attr,
+ struct aca_handle *handle, char *buf, void *data);
+
#endif
--
2.34.1
