Skip to content

schedule: ll: refinement low latency queue - #1149

Merged
lgirdwood merged 1 commit into
thesofproject:masterfrom
bkokoszx:low-latency-one-queue
Mar 25, 2019
Merged

lgirdwood merged 1 commit into
thesofproject:masterfrom
bkokoszx:low-latency-one-queue

Conversation

@bkokoszx

Copy link
Copy Markdown
Collaborator

I've removed queue per priority in low latency scheduler. Now there is only one queue for all priorities. Scheduler adds tasks to queue in descending order based on their priority.

Comment thread test/cmocka/src/list/list_add_item_after_ref_item.c Outdated
@bkokoszx
bkokoszx force-pushed the low-latency-one-queue branch from df25e11 to 4730c6e Compare March 20, 2019 07:19
Comment thread src/lib/ll_schedule.c Outdated
Comment thread src/lib/ll_schedule.c Outdated
Comment thread src/include/sof/list.h Outdated
@bkokoszx
bkokoszx force-pushed the low-latency-one-queue branch from 4730c6e to f2cd433 Compare March 20, 2019 11:42
@jajanusz
jajanusz requested a review from tlauda March 20, 2019 12:18
@bkokoszx
bkokoszx force-pushed the low-latency-one-queue branch 2 times, most recently from 11c066f to 94f2eec Compare March 21, 2019 09:36
Comment thread src/include/sof/list.h Outdated

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Can we not use the existing list_item_prepend() and list_item_append() here ?

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

list_item_prepend() adds to the beginning of the list and list_item_append() at the end. These new functions allow to insert into specific location in the middle of the list.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

yes, but by changing the list head passed to list_item_prepend() should do the same.

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

@lgirdwood
Yes, you are right. If we replace list head, list_item_prepend() do the same as list_item_add_after() and list_item_append() do the same list_item_add_before(). I can use "old" ones, but it can enter some misunderstanding by function names.

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

@lgirdwood
I've removed list_item_add_after() and list_item_add_before() definitions. I'm using list_item_append() instead of list_item_add_before() now.

Comment thread src/lib/ll_schedule.c Outdated
@bkokoszx
bkokoszx force-pushed the low-latency-one-queue branch from 94f2eec to ad63982 Compare March 22, 2019 11:09

@lgirdwood lgirdwood left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Blocking on CI.

@zrombel

zrombel commented Mar 22, 2019

Copy link
Copy Markdown

We are seeing regression with DMIC tests on CI, @bkokoszx please check it and confirm.

@bkokoszx

Copy link
Copy Markdown
Collaborator Author

@zrombel @lgirdwood
I will check. Please do not merge.

I've removed queue per priority in low latency scheduler.
Now there is only one queue for all priorities. Scheduler adds
tasks to queue in descending order based on ther priority.

Signed-off-by: Bartosz Kokoszko <bartoszx.kokoszko@linux.intel.com>
@bkokoszx
bkokoszx force-pushed the low-latency-one-queue branch from ad63982 to a32be94 Compare March 25, 2019 08:26
@bkokoszx

Copy link
Copy Markdown
Collaborator Author

I have modified insert_task_to_queue() function. At the end of the function if task has not been added (means that it has the lowest priority), it should be added at the end of the list.
@zrombel
Could you take a look whether DMIC regression still exists?

@zrombel

zrombel commented Mar 25, 2019

Copy link
Copy Markdown

With those changes I don't see any regression in python tests. Looks ready for merging.

@lgirdwood
lgirdwood merged commit 5ea5da8 into thesofproject:master Mar 25, 2019
@bkokoszx
bkokoszx deleted the low-latency-one-queue branch April 10, 2019 07:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants