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

Supporting aggregator elements #111

Open
djns99 opened this issue Jun 21, 2022 · 2 comments
Open

Supporting aggregator elements #111

djns99 opened this issue Jun 21, 2022 · 2 comments
Assignees

Comments

@djns99
Copy link

djns99 commented Jun 21, 2022

I have some custom classes that inherit the base class GstAggregator that combines buffers from two source elements into one buffer and I wanted to get interlatency from each source element. It appears that the default behaviour is to report the interlatency for the last latency event received. There is a couple of issues with this approach a) it is non-deterministic since the latency events are each produced by separate threads and b) there is no way to get the latency from other source elements.

Locally I have implemented a workaround in the interlatency tracer (commit 8ef0e7b) that replaces the stored latency_probe_id qdata with a list of GstEvents instead of the event directly. Patch with my changes

This appears to work as I need and I can plot the latency if I include a filter that matches a specific source element to gstshark-plot. Though in general more work would be needed to handle plotting multiple source elements with gstshark-plot. Currently it treats the duplicate entries as part of the same series which results in erratic plotting if the latencies from each source are not similar. My initial thought would be to produce a separate plot for each source element, though I have not verified if this is practical.

@michaelgruner
Copy link
Collaborator

Hey @djns99 sorry for the slow response. Thanks for your patch, do you mind opening a PR with the changes so we start from there? Thanks!

@djns99
Copy link
Author

djns99 commented Jan 26, 2023

Sure thing I opened PR #120

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants