recognize it any more. This makes the following significant changes:
- The main body of the program doesn't know a thing about URIs,
HTTP, or FTP. This makes it possible to easily plug in other
protocols. (The next revision will probably be able to dynamically
add new recognizers.)
- There are no longer arbitrary timeouts for the protocols. If you want
to set one for yourself, use the environment variables.
- FTP proxies are now supported (if I implemented it right).
- The HTTP implementation is much more complete, and can now do restarts,
preserve modtimes, and mrun in mirror mode. It's not yet up to 1.1,
but it's getting there.
- Transaction TCP is now used for sending HTTP requests. The HTTP/1.1 syntax
for requesting that the connection be closed after one request is
implemented.
In all of this, I have doubtless broken somebody. Please test it and tell me
about the bugs.
key "print scrn".
It used to stop at the first non-open vty, now it skips the non-open
ones and thereby enable one to cycle around all open vty by pressing
"print scrn".
when parsing a printf-like arg list. Looking for someone to blame,
I noticed that the man page has a bad example. It clearly says at
the top that types following the last known argument are passed after
their default type conversions, and then later the example uses
va_arg (..., char);
so I fixed it.
1. Pass argc and argv to getarg and process them with getopt().
2. Instead of using an array to save arg characters, use array of
pointers and call backgammon/teachgammon with execv, instead of execl.
This should fix problems with calling teachgammon.
2.2 candidate.
I have code to calibrate the overhead fairly accurately, but there
is little point in using it since it is most accurate on machines
where an estimate of 0 works well. On slow machines, the accuracy
of DELAY() has a large variance since it is limited by the resolution
of getit() even if the initial delay is calibrated perfectly.
Use fixed point and long longs to speed up scaling in DELAY().
The old method slowed down a lot when the frequency became variable.
Assume the default frequency for short delays so that the fixed
point calculation can be exact.
Fast scaling is only important for small delays. Scaling is done
after looking at the counter and outside the loop, so it doesn't
decrease accuracy or resolution provided it completes before the
delay is up. The comment in the code is still confused about this.
- don't uselessly initialize the fifo "DMA" bit at attach time.
- initialize the fifo "DMA" bit at open time. Without this, the device
interrupts for every character received, reducing input performance
to that of an 8250.
- don't uselessly initialize the fifo trigger level to 8 (scaled to
256) at attach time.
- don't scale the fifo trigger level to 512 bytes. The driver's pseudo-
dma buffer has size 256, so it can't handle bursts of size 512 or 256.
It should be able to handle the second lowest ftl (2 scaled to 64).
- don't reset the fifos in siostop(). Reset triggers a hardware bug
involving wedging of the output interrupt bit This workaround
unfortunately requires ESP support to be configured.
slices in sd_open() after a media change when the previous sd_open()
discards the previous slices and then fails. sd_open() just handles
media changes poorly and fails too often.
