See comment inline. Other than that, the series looks good to me.
On 2019-12-16 2:02, Huang Rui wrote:
Thunk driver would like to know the num_cp_queues data, however this data relied
on different asic specific. So it's better to get it from kfd driver.
Signed-off-by: Huang Rui <ray.huang@xxxxxxx>
---
drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 3 +++
drivers/gpu/drm/amd/amdkfd/kfd_topology.h | 3 ++-
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index cc01ccd..203c823 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -488,6 +488,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
dev->node_props.num_sdma_xgmi_engines);
sysfs_show_32bit_prop(buffer, "num_sdma_queues_per_engine",
dev->node_props.num_sdma_queues_per_engine);
+ sysfs_show_32bit_prop(buffer, "num_cp_queues",
+ dev->node_props.num_cp_queues);
if (dev->gpu) {
log_max_watch_addr =
@@ -1316,6 +1318,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
dev->node_props.num_gws = (hws_gws_support &&
dev->gpu->dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) ?
amdgpu_amdkfd_get_num_gws(dev->gpu->kgd) : 0;
+ dev->node_props.num_cp_queues = get_queues_num(dev->gpu->dqm);
kfd_fill_mem_clk_max_info(dev);
kfd_fill_iolink_non_crat_info(dev);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
index 9346cc1..e447901 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
@@ -27,7 +27,7 @@
#include <linux/list.h>
#include "kfd_crat.h"
-#define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 28
+#define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 24
I don't see why you need to change the name size here. I'm not aware of
any requirement that the structure size cannot change. This comment
applies to patch 1 as well.
Regards,
Felix
#define HSA_CAP_HOT_PLUGGABLE 0x00000001
#define HSA_CAP_ATS_PRESENT 0x00000002
@@ -82,6 +82,7 @@ struct kfd_node_properties {
uint32_t num_sdma_engines;
uint32_t num_sdma_xgmi_engines;
uint32_t num_sdma_queues_per_engine;
+ uint32_t num_cp_queues;
char name[KFD_TOPOLOGY_PUBLIC_NAME_SIZE];
};
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
