Closed GoogleCodeExporter closed 9 years ago
Issue 2 not r2.
Original comment by bri...@gmail.com
on 5 Dec 2008 at 10:03
[deleted comment]
Original comment by bri...@gmail.com
on 5 Dec 2008 at 11:30
Move and resize implemented for window objects.
Original comment by bri...@gmail.com
on 24 May 2009 at 9:48
Segfaults still occur during terminal resizing. When World is first starting
up a
resize will casue a segfault. Segfaults can be forced also when resizing while
windows, out of bounds of the new terminal size, are being printed to.
Original comment by bri...@gmail.com
on 25 May 2009 at 5:26
R17 slows down the eventual race condition that still causes random crashes.
Original comment by bri...@gmail.com
on 29 May 2009 at 10:00
Some terminals will send multiple SIGWINCH signals as the window is resized.
This
causes a crash. Putty and XTERM send only one which is sent after the desired
window
size has been set.
Original comment by bri...@gmail.com
on 9 Jun 2009 at 10:24
Catching SIGWINCH with a handler that busy waits for a few seconds will crash
if repeated signals are sent.
Original comment by bri...@gmail.com
on 29 Jun 2010 at 5:13
The code that dispatched a thread when an interrupt occurred was not correct
and was free to mutate the wait thread queue during a virtual machine context.
The signal handler now keeps track of a signal allowing the routing scheduler
to dispatch the new thread interrupt handler. The scheduler is only called
after the virtual machine normalizes its active registers such as the
instruction pointer which must also happen before a garbage collection.
Original comment by bri...@gmail.com
on 2 Jul 2010 at 1:06
Original issue reported on code.google.com by
bri...@gmail.com
on 5 Dec 2008 at 10:02