-
Notifications
You must be signed in to change notification settings - Fork 243
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
allow filtering for --exportTo
#524
Comments
While possible, this would require special code as the current filtering is applied to the analyzed hotspot specific data, not the raw perfparser data. Meaning this isn't that trivial to implement. |
More details: the following entries of
The use case is the option to have one big perf recording (for example from user-wide recording or from a recoding that has many sub-processes) and limiting it to the interesting parts and/or to split that into several smaller perfparser files that hotspot will handle quick later on. |
Is your feature request related to a problem? Please describe.
After running system-wide recordings (actually limited to a service user already) I often get huge amount of warning messages like "Invalid memory read requested by dwf" and "failed to attach state Couldn't find architecture of any ELF", "No DWARF information found 2", "No DWARF information found 5" during open/export.
Then I'm in Hotspot seeing a lot of processes I'm not interested in at all (bash, grep, ls, and others), so I need to always filter when inspecting the recording.
Describe the solution you'd like
Provide an option to filter when converting to perfparser file like it is possible in the timeline view ([list of] name/pid).
It may be possible to skip several things then, that not match the filter during import, possibly also reducing the amount of warning messages (for example adding a filter "only processes that contain myApp") - but in any case the .perfparser output file would be much smaller, could be opened faster in Hotspot, would not need the filter to be post-applied each time it is opened.
The text was updated successfully, but these errors were encountered: