Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

schedule: config to build LL without task rescheduling #9816

Merged
merged 2 commits into from
Feb 10, 2025
Merged

schedule: config to build LL without task rescheduling #9816

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
68701fe
schedule: config to build LL without task rescheduling
johnylin76 Feb 6, 2025
35eff4c
mt8196: set config SCHEDULE_LL_NO_RESCHEDULE_TASK=y
johnylin76 Feb 6, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions src/schedule/Kconfig
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,3 +41,13 @@ config SCHEDULE_LL_STATS_LOG_WINDOW_SIZE
Size of the statistics window as a power of two. The window size
setting also impacts the rate of reporting. With 1ms scheduler tick,
default of 10 results in 1024msec window size.

config SCHEDULE_LL_NO_RESCHEDULE_TASK
bool "Low-latency scheduler skips task rescheduling"
default n
help
Select this to instantiate the low-latency scheduler without task
rescheduling, given that the operation is optional. Under such cases,
scheduler_ops::reschedule_task will set to NULL instead, tasks with
the attempt to reschedule (e.g. DMA trace works) will be relinguished
directly and return no error.
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess this is ok. One option would be to track the dependency to features (like DMA trace) also in Kconfig, but given this is opt-out, this might be too klungy.

6 changes: 6 additions & 0 deletions src/schedule/ll_schedule.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -684,6 +684,11 @@ static int schedule_ll_task_cancel(void *data, struct task *task)
return 0;
}

#if CONFIG_SCHEDULE_LL_NO_RESCHEDULE_TASK
/* As a null function pointer */
#define reschedule_ll_task ((void*)0)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why not just NULL?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Both ways are probably fine. The reason I chose ((void*)0) is because NULL in c is also a macro, and I'm not sure if there is any risk when a macro represents as another macro.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'd be very surprised if this didn't work. We certainly have macros using macros like across a dozen or more layers. And this is in a C file, so it won't be included in other files to potentially be used in further macro definitions.


#else
static int reschedule_ll_task(void *data, struct task *task, uint64_t start)
{
struct ll_schedule_data *sch = data;
Expand Down Expand Up @@ -717,6 +722,7 @@ static int reschedule_ll_task(void *data, struct task *task, uint64_t start)

return 0;
}
#endif

static void scheduler_free_ll(void *data, uint32_t flags)
{
Expand Down