romain@pyrnj.UUCP (11/03/85)
<frivolous on>
I need a vacation, but don't have time for it right now. So to
compensate, I started running infinite loops of Rog-O-Matic (from
Carnegie-Mellon University) on my officemate's terminal. However, it
took so long to start up I noticed it even when I was working at other
things on my own terminal. I got curious and gcore'd the player
process to find it trying to lock
"\027F$=/games/lib/rgm/rlog/GeneLock521".
A little more detective work showed that the lockfile name was getting
created correctly, but getting trashed. The fix:
Change this line in rand.c
if (seed == 0) seed = time();
to
if (seed == 0) seed = time(0);
since 4.2 time() stuffs the time into *(parameter register 0) if it's
non-null. In this case, pr0 always pointed at genelock[] because of
the Pyramid register window mechanism.
There remain other things in Rog-O-Matic that need to be patched up.
It looks as if some strings or pointers passed through match() are
sometimes getting corrupted. I've been seeing a lot of loops where rgm
doesn't recognize strings of the form "you can't*". (How's that for
optimistic programming?)
Moral of the story: Even good programmers can write non-portable code
on a bad day. All those little warnings from lint may eventually catch
up with you...
Romain Kang, Pyramid Technology Corporation
US Mail: 900 Route 9, Woodbridge, NJ 07095
Ma Bell: (201) 750-2626
UUCPnet: {allegra,cmcl2,pyramid,topaz}!pyrnj!romain
"!!!x09 dimaryP a fo edisni deppart m'I !pleH"
"oNhwre eenraa sab dsab iegnt arppdei n aAV X117/05!!!"