[RPC] android process isolation/watchdog #1387
Merged
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
|
CC @yzhliu |
|
@PariksheetPinjari909 @srkreddy1238 can you also help review this? |
tqchen
approved these changes
Jul 8, 2018
tqchen
pushed a commit
to tqchen/tvm
that referenced
this pull request
Aug 4, 2018
sergei-mironov
pushed a commit
to sergei-mironov/tvm
that referenced
this pull request
Aug 8, 2018
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Outdated suggestions cannot be applied.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.
Summary: add more robustness to android RPC to facilitate automated schedule optimization via autotvm.
Previously the android RPC app would crash if passed a bad schedule as the RPC server ran in the same process as the app (different thread). This PR moves the RPC service to an android service that runs in a separate process to provide process isolation.
There is also now a watchdog which periodically polls the RPC service to check if it is alive. The watchdog starts a new RPC service if a bad schedule crashed the service.
The watchdog also instructs the RPC service to check if it is currently in a session that has timed out (currently a hard timeout of 30 seconds is enforced). The RPC service kills itself in this case; the app will wait for a respawn from the watchdog. This step is necessary as some bad schedules have been found to leave OpenCL in a bad state where the entire process must be completely restarted while locking up the RPC service.