gpu: nvgpu: rename redundant runlist names in HAL

Drop the "runlist_" part in the runlist section of the HAL ops. For
example:

- old: g->ops.runlist.runlist_wait_pending
- new: g->ops.runlist.wait_pending

At the same time, drop the "fifo_" part from the function names. For
example:

- old: gk20a_fifo_runlist_wait_pending
- new: gk20a_runlist_wait_pending

Also rename eng_runlist_base_size to count_max. The size of the
eng_runlist_base register array depicts the maximum possible number of
runlists in the chip for which count_max is more descriptive.

Jira NVGPU-1309

Change-Id: Ie9e94b9f65cd10d3e682d19954f240adb6e311be
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2017403
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Konsta Holtta
2019-02-12 15:44:36 +02:00
committed by mobile promotions
parent ac24a33a76
commit 93e15f9c43
25 changed files with 125 additions and 129 deletions

View File

@@ -1366,7 +1366,7 @@ long gk20a_channel_ioctl(struct file *filp,
err = -EPERM;
break;
}
if (!ch->g->ops.runlist.reschedule_runlist) {
if (!ch->g->ops.runlist.reschedule) {
err = -ENOSYS;
break;
}
@@ -1377,7 +1377,7 @@ long gk20a_channel_ioctl(struct file *filp,
__func__, cmd);
break;
}
err = ch->g->ops.runlist.reschedule_runlist(ch,
err = ch->g->ops.runlist.reschedule(ch,
NVGPU_RESCHEDULE_RUNLIST_PREEMPT_NEXT &
((struct nvgpu_reschedule_runlist_args *)buf)->flags);
gk20a_idle(ch->g);