gpu: nvgpu: Add correct nomenclature for NVS ioctls

Its preferable to use the following naming convention
NVGPU_<group>_IOCTL_<function>.

The IOCTL interfaces are updated accordingly.

Also, all KMD based defines as part of the UAPI need
to be prefixed by NVGPU.

Jira NVGPU-8619

Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Change-Id: I2210336536cbcc0415885f3f92a2f7fa982fa39c
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2814484
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Debarshi Dutta
2022-11-23 10:42:07 +05:30
committed by mobile promotions
parent eb60e7f1f1
commit 8e60795b9c
3 changed files with 59 additions and 71 deletions

View File

@@ -285,18 +285,6 @@ enum nvgpu_nvs_ctrl_queue_direction {
*/
#define NVGPU_NVS_CTRL_FIFO_QUEUE_CLIENT_EVENTS_READ 4U
/*
* Direction of the requested queue is from CLIENT(producer)
* to SCHEDULER(consumer).
*/
#define NVGPU_NVS_CTRL_FIFO_QUEUE_DIRECTION_CLIENT_TO_SCHEDULER 0
/*
* Direction of the requested queue is from SCHEDULER(producer)
* to CLIENT(consumer).
*/
#define NVGPU_NVS_CTRL_FIFO_QUEUE_DIRECTION_SCHEDULER_TO_CLIENT 1
/* Structure to hold control_queues. This can be then passed to GSP or Rm based subscheduler. */
struct nvgpu_nvs_ctrl_queue {
struct nvgpu_mem mem;