("mount_critical_filesystems local") makes no assumptions about /usr.
Under certain (semi-normal) conditions, /usr is never mounted until
"mount -a" late in rc(8), and This Is Bad.
remove ffs()
add bzero that uses block store insns
save %gsr to fpstate
revert to physical accesses for pmap_zero_page so we don't
have conflicting page mappings
general cleanup.
macro). Add new fwrite_be_offt() function which writes out a big-endian
64bit number regards of the size of off_t on the host machine.
Also fix ``bug'' in previous - even though the off_t pointers were
stored big-endian, the individual card lenghs were in host-order. The
cards.pck files are now identical when make on a big- or little-endian
machine.
Now builds a correct cards.pck file on non-NetBSD machines as well
- tested on Ultrix and Solaris (but Solaris needs -lsocket to get
htonl()).
- replaiced with local die() and dieperror() routines). Use ANSI
prototypes (no dependance on the __P() macro). Add new fwrite_be_offt()
function which writes out a big-endian 64bit number regards of the size
of off_t on the host machine. Remove unused unctrl() function.
Now builds correct fortune .dat files on non-NetBSD machines as well
- tested on Ultrix and Solaris (but Solaris needs -lsocket to get
htonl()).
if you com* at pcmcia?, and com3 and com4 as pcmcia cards, and removed
and reinserted the card that was com3, it would become com5. if you then
removed and reinserted com4, it would become com6. etc.) Now, instead
of incrementing FSTATE_STAR configuration entries for a driver when
a cloning instance is attached, leave it alone, and scan the device softc
array (starting at the first cloning unit number) for units which are
available for use. This wastes a tiny bit of time (can require a linear
scan of the softc table for the device), but device attachment should be
relatively infrequent and the number of units of each type of device
is never particularly large anyway.
windsor