summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2002-01-07use flags instead of options to choose the tea5757 vs tea5759.Michael Shalayeff
from Vladimir Popov <jumbo@narod.ru>
2002-01-07Next issue:Mike Pechkin
af is always u_int8_t, not int; dhartmei@ ok
2002-01-07remove 3rd argument from ipmask(), not used.Mike Pechkin
dhartmei@ ok
2002-01-07While I'm here (next issue):Mike Pechkin
o) __progname; o) remove #define LOG_DAEMON; o) use openlog() before syslog(); o) write $version to stderr; millert@ ok
2002-01-07If the handler for SIGCHLD is set to SIG_IGN, act as if theTodd C. Miller
SA_NOCLDWAIT (don't create zombies) flag has been specified. This is consistent with most other operating systems and is what XPG4.2 specifies.
2002-01-07Revert local changes that should not have been committedTodd C. Miller
2002-01-07o) int -> {uid,gid}_t; %d -> %u;Mike Pechkin
o) int -> pid_t; deraadt@ ok
2002-01-07syncTodd T. Fries
2002-01-07generate i386 (regress diff = more devices than before, ugen, etc..)Todd T. Fries
2002-01-07add i386 'source'Todd T. Fries
2002-01-07o more devices for 'all' targetTodd T. Fries
o ttyCcfg is 600 not 666 o add radio o joy is 666 not 600 (current i386 MAKEDEV does this)
2002-01-07add macppc 'source'Todd T. Fries
2002-01-07syncTodd T. Fries
2002-01-07add /dev/pciTodd T. Fries
s/tzs/czs/ typo
2002-01-07sprinklings of 'ifstep()'Todd T. Fries
'the big disk case' no longer sets vars and calls dodisk, would take more space to additionally do the 'stepping' argument (see cd\* in generated MAKEDEV)
2002-01-07syncTodd T. Fries
2002-01-07DISK_STEPPING is not used, and in any event, is not global to all disksTodd T. Fries
prepare dodisk* script functions for a 7th argument, what I call disk stepping, the number to add from rd0a to rd1a minor numbers, defaulted to 16; some archs *cough*macppc*cough* still have 8 for cdroms and such
2002-01-07syncTodd T. Fries
2002-01-07add more disks to the 'all' target, rd shows up on sparc64 now ;-)Todd T. Fries
s/tty/tth/
2002-01-07comment tweakingTodd T. Fries
2002-01-07Paranoia on my part, do not let the condition exist where kvm space couldDale Rahn
be claimed by the pre kvm init stealing process and kvm.
2002-01-07On cache flushing, if start is not cacheline aligned, add to the lenghtDale Rahn
to make sure the whole region is flushed. from conversation with pefo.
2002-01-07To disable a SIGCHLD signal handler use SIG_DFL, *not* SIG_IGNTodd C. Miller
which now has a different meaning.
2002-01-07To disable a SIGCHLD signal handler use SIG_DFL, *not* SIG_IGNTodd C. Miller
which now has a different meaning.
2002-01-07Do not enter ddb twice in a row on console break.Miod Vallat
2002-01-07- increment uvmexp.intrs on interrupts (oops)Miod Vallat
- change more fatal situations into panics - reorganize the execution flow to be shorter if no device interrupt handler are registered - minor style/KNF changes as well
2002-01-07better.Niels Provos
2002-01-07add kqueueNiels Provos
2002-01-07simple kqueue test for pipesNiels Provos
2002-01-06remove bogus diagnostic check.Eric Jackson
pr#2298
2002-01-06add ums and wsmuxJason Wright
2002-01-06Do not allow root to login via an insecure tty even if the auth methodTodd C. Miller
does not use plaintext passwords due to the possibility of connection hijacking. This restores the pre-3.0 behavior and closes PR 2299.
2002-01-06-x needs read-write access for DIOCSETDEBUG ioctlDaniel Hartmeier
2002-01-06Don't let root change it's password via login_l?chpass(8).Todd C. Miller
At some point in the future, passwd(1) may call these auth methods at which time we will need a flag to allow root to do a password change and also to allow account w/o a password to be given one.
2002-01-06That was no fix, that broke things. If the pte entry is currently foundDale Rahn
in the po lists, it will NOT have the PTE_VALID bit set. Thus valid mappings could be ignored if enough mappings existed for that PTEG pair. This explains the bus_dma panics.
2002-01-06pte_spill() is executed on a special stack in real mode (vm not enabled).Dale Rahn
It is not valid to call pool_put() from that context. If called from that context, put the freed item on one of two lists (race safe), poalloc() will attempt to fetch from there, and pofree() will clean up if called from a normal context.
2002-01-06Claim to support WSSCREEN_WSCOLORS (drivers support it, but rcons doesn'tJason Wright
appear to).
2002-01-05Sync with reality: make mj2 a trusted user and add 'C' (nocanonify)Todd C. Miller
flag to MSA for fast message injection.
2002-01-05some missing and misplaced periodsKevin Steves
2002-01-05Documentation update. Adapted from NetBSD.Thierry Deval
2002-01-05mark signal racesTheo de Raadt
2002-01-05fix hanging x11 channels for rejected cookies (e.g. XAUTHORITY=/dev/null xbiff)Markus Friedl
bug #36, based on patch from djast@cs.toronto.edu
2002-01-05support detach, from netbsdMichael Shalayeff
2002-01-05Avoid using an uninitialized variable. dhartmei@ okThomas Nordin
2002-01-04Document some.Artur Grabowski
2002-01-04Use [eid]cache-size to determine cache sizes and fall back to (?cache-nlines ↵Jason Wright
* ?cache-linesize) if it fails (from NetBSD).
2002-01-04Fix byte order: 0x55AA -> 0xAA55Kjell Wooding
2002-01-04system calls are made thru _thread_sys_* with weak symbols of the real nameDale Rahn
pointing to the _thread_sys_ version. ok pval, fgs
2002-01-04sync to reality.Kjell Wooding
Note that the NT serial number is now whacked with an update, or -u Fix signature in example. Explicitly state the MBR signature (0xAA55)
2002-01-04add 600x read for 744 rev < 2 - from netbsdEric Jackson
cleanup dmesg output
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-04 17:04:41 +0000