-
Notifications
You must be signed in to change notification settings - Fork 1.7k
Split IO and network crates #1828
Conversation
}).unwrap() | ||
}) | ||
.expect("Error creating worker thread")); | ||
Worker::work_loop(stealer, channel.clone(), wait, wait_mutex.clone(), deleting) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It won't panic? Or is there some other way parity will be closed in case of panic in IO Worker thread?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Set panic = "abort"
in the root crate
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
panic = "abort"
might not be a good idea because every panic will then lead to stuff like rocksdb handles not being closed properly.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That's fine. rocksdb and Parity are designed to be crash resistant. Removing unwinding info speeds up compilation and reduces binary size. Cargo support for this feature seems to be buggy though. Will have to revert to panic handler for now
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
i suppose we will have to keep rocksdb WAL on, though.
5a270a2
to
e26097e
Compare
lgtm, no new logic here except for some error and structs mapping |
No description provided.