causes an end-less loop in ci(1) which uses "break" inside one of the
macros.
"/etc/security" will now no longer fill up "/" (or "/var" depending on
file-system layout).
magic number or versioning, relied on random(3) never changing to a
different implementation, and were also saving pointers to disk and
reading them back again. It *looks* as if the pointers thus loaded
were reset before being used, but it's not particularly clear as the
main loop of this thing is goto-based FORTRAN translated lightly to C.
I've changed the logic to null these pointers instead of saving and
loading them, and things seem to still work.
The new save files have a header, support versioning, write only sized
types in network byte order, and for the toy encryption to discourage
cheating do something self-contained instead of using random(3) as a
stream cipher.
Because between the original import from 4.4 until earlier today
trying to save would result in SIGSEGV on most platforms, it's
unlikely anyone has a save file, but just in case (since the pointer
issue appears to be nonlethal) I've kept compat code for old save
files.
chars by default (i.e., almost all machines). Makes it possible to save
the game. This has been broken since 4.4 and probably ever since the
FORTRAN -> C translation.
Crash reported by Petri Laakso in private mail.
First, there are legitimate cases where we want to return -1 and errno == 0,
second, all paths go through time1() and/or time2sub() which explicitly
take care to set errno in cases of real overflows.
Should fix atf test failure on various ports in
lib/libc/time/t_mktime:timegm_epoch.
- pass the expected errno to it, not "errno"
- make sure to have errno set already before invoking the macro, i.e.
do not use it to test errno changes as side effect of the asserted
expression
Spotted by mlelstv, makes the epoch tests correctly fail on amd64 as
well.
- fix the handling of subdir additional LIBDPLIBS so they're present
when <bsd.lib.mk> is loaded. fixes missing shlib links relative to
what the normal package does. (noted by wiz.)
jmcneill