>>>>> "Jim" == Cathey, Jim <[email protected]> writes:
Hi, >> This thread started by Michal, who's trying to use a very slow >> serial-over-jtag setup on microblaze, and had issues with characters >> getting lost when getty started up (because of the tcflush), so we >> wanted to add a tcdrain() instead. Jim> A design problem? Why even issue the tcflush(stdout) (in essence) Jim> if you _don't_ want to discard pending output? The tcflush() is only to workaround the issue with SW flow control like noted earlier. If tcdrain() works correctly (E.G. finishes before the alarm fires), then the tcflush is a noop. If we can ignore that error situation or only need to flush input, then sure - Let's just do that. Jim> If the flush problem is from stuff sent by a process that Jim> preceded getty, rather than getty's own early output, then Jim> the lack of the flush is _its_ problem, not getty's, given Jim> that getty's charter is to program the serial port (if only Jim> partially) in the first place! Perhaps the parent shouldn't Jim> even be calling getty, but rather login instead? True, but this is typically not done. The parent of getty is most likely init, and the unsent output is from other init scripts. Our very own examples/inittab calls getty as well. -- Bye, Peter Korsgaard _______________________________________________ busybox mailing list [email protected] http://lists.busybox.net/mailman/listinfo/busybox
