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.

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

Fix hang caused by OUT_ALLOC and remove macro #523

Merged
merged 2 commits into from
Apr 17, 2020

Conversation

tianjunwork
Copy link
Contributor

@tianjunwork tianjunwork commented Apr 15, 2020

  1. The reason of hang caused by OUT_ALLOC is lack of available output stream buffer.
    The fix is to seperate the outputBufferFifoInitCount from inputOutputBufferFifoInitCount.
    And increase its size.
  2. Remove the OUT_ALLOC macro.

Signed-off-by: Jun Tian jun.tian@intel.com

closes #514, #497

1. The reason of hang caused by OUT_ALLOC is lack of available output stream buffer.
   The fix is to seperate the outputBufferFifoInitCount from inputOutputBufferFifoInitCount.
   And increase its size.
2. Clean up the OUT_ALLOC macro.
3. Fix typo in PK.

Signed-off-by: Jun Tian <jun.tian@intel.com>
@tianjunwork tianjunwork marked this pull request as ready for review April 15, 2020 03:07
@tianjunwork tianjunwork added the bug Something isn't working label Apr 15, 2020
@tianjunwork
Copy link
Contributor Author

tianjunwork commented Apr 15, 2020

Memory allocated at init.

1080p 4k 8k
with this PR 3.91GB 13.45GB 47.21GB
master 4.24GB 13.9GB 48.11GB

Speed gain for resolution below 1080p. No gain for resolution >= 1080p(4k, 8k almost the same).
image
image

Copy link
Contributor

@intelmark intelmark left a comment

Choose a reason for hiding this comment

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

Seems like the PR is removing the OUT_ALLOC define and always outputting bitstream allocations at run time. "Fix and remove OUT_ALLOC" would be more appropriate

Signed-off-by: Jun Tian <jun.tian@intel.com>
@tianjunwork tianjunwork changed the title Fix and clean up OUT_ALLOC Fix hang caused by OUT_ALLOC and remove macro Apr 15, 2020
@tianjunwork
Copy link
Contributor Author

tianjunwork commented Apr 17, 2020

Finished test with -n 1000. No hang. No new failed cases other than the existing issues.

---------------------------------------------------------
Test Begin... 
Total Number of Tests: 2685
Total Passed: 2671
Percentage Passed: 99.47858473%
Time Elapsed: 1 day(s), 23 hour(s), 56 minute(s), 39 second(s)
---------------------------------------------------------
asm_test.txt:------------Failed-------------
asm_test.txt:------------Failed-------------
asm_test.txt:------------Failed-------------
asm_test.txt:------------Failed-------------
buffered_test.txt:------------Failed-------------
buffered_test.txt:------------Failed-------------
buffered_test.txt:------------Failed-------------
buffered_test.txt:------------Failed-------------
defield_test.txt:------------Failed-------------
defield_test.txt:------------Failed-------------
defield_test.txt:------------Failed-------------
defield_test.txt:------------Failed-------------
multi_channel_test.txt:------------Failed-------------
multi_channel_test.txt:------------Failed-------------

@tianjunwork
Copy link
Contributor Author

Hi @intelmark @Austin-Hu , good to go?

@intelmark
Copy link
Contributor

looks good! (aside from those test failures which we'll tackle separately)

@tianjunwork tianjunwork merged commit cb1d6a9 into OpenVisualCloud:master Apr 17, 2020
@tianjunwork tianjunwork deleted the out_alloc branch April 17, 2020 16:35
@tianjunwork tianjunwork mentioned this pull request Apr 17, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Encoding hangs with OUT_ALLOC enabled
3 participants