Consider this situation:
1. Process A locks a file, using fcntl(..., F_SETLKW, ... ).
2. Process B tries to lock the file the same way, and so goes to
sleep waiting for the lock.
3. Process A releases the lock, but then immediately tries to
reaquire it.
On Linux (2.2.9 is the only one I tested), process A gets the lock back in
step #3. Anyone else think that step #3 should put A to sleep and let B
get the lock? An application that wants that behavior can get it by throwing
in a sleep(0) between A's releasing the lock and attempting to reaquire it,
but is it supposed to have to do that?
--Tim Smith
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/