tetragon: Add buffer between perf reader and events processing code #593
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
8c44927 to
cd655e6
Compare
f5e3f16 to
af73cf1
Compare
|
this change makes big difference in being able to get and process more events from kernel |
af73cf1 to
1b2845c
Compare
kkourt
reviewed
Aug 22, 2023
1b2845c to
e13eea1
Compare
kkourt
approved these changes
Aug 23, 2023
Adding extra layer between perf ring buffer reader and event processing by sensor code. Basically it's new channel (default size 65k) that is written by perf ringbuffer reader and read by new go routine that processes these events. This way we keep the perf reader fast and we won't get lost events if sensor processing code have a hiccup and takes longer. When the eventsQueue channel is full we drop the event. It's configurable by new --rb-queue-size option, the default is 65k. Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Adding following metric counters for events buffer queue: ringbufqueuemetrics.Received - number of records received from perf event reader go routine ringbufqueuemetrics.Lost - number of records lost because the RB Queue channel was full Signed-off-by: Jiri Olsa <jolsa@kernel.org>
e13eea1 to
dcd1faa
Compare
|
OK lets try it. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adding extra layer between perf ring buffer reader and event
processing by sensor code.
This way we keep the perf reader fast and we won't get lost events
if sensor processing code have a hiccup and takes longer.