Commit Graph

118223 Commits

Author SHA1 Message Date
wiz
6181ed442a Bump date for previous; grammar improvements. 2003-10-02 07:06:36 +00:00
itojun
2c96e1a283 use netmask of /32. as discusssed on tech-userlevel 2003-10-02 06:25:28 +00:00
itojun
ea3d98658c plug mbuf leak due to manual mbuf handling. PR kern/13807.
(martti confirmed that it stabilizes the situation described in kren/13807)
2003-10-02 06:01:51 +00:00
itojun
b3cd345741 more fixes from 0.9.7c, from openbsd 2003-10-02 02:26:17 +00:00
itojun
ae91503b5d from openbsd:
Correct some off-by-ones.  They currently don't matter, but this
is for future safety and consistency.
OK krw@, markus@
2003-10-02 02:25:05 +00:00
itojun
269a200e21 minor KNF 2003-10-02 01:27:20 +00:00
itojun
30b8d01871 suggest setting netmask to 32 for security reasons; see tech-userlevel 2003-10-02 01:21:16 +00:00
itojun
d3e55680aa correct typo. PR misc/22944 2003-10-02 00:09:06 +00:00
enami
ae9b5cba84 Rewrite uvm_map_findspace() to improve readability and to fix a bug that
it may return space already in use as free space under some condition.
The symptom of the bug is that exec fails if stack is unlimited on
topdown VM kernel.
2003-10-02 00:02:10 +00:00
itojun
899b67c09a correct ip_fragment() wrt ip->ip_off handling.
do not send out incomplete fragment due to ENOBUFS (behavior change from 4.4BSD)
2003-10-01 23:54:40 +00:00
matt
3aafdaac96 Make the kernel makefiles pay attention to NOGCCERROR 2003-10-01 23:11:29 +00:00
enami
0ca733e759 Some whitespace fixes. 2003-10-01 23:08:32 +00:00
enami
aa87bee0c5 ansi'fy. 2003-10-01 22:50:15 +00:00
uwe
c313f02140 Whitespace nit. 2003-10-01 21:51:15 +00:00
itojun
d104a1a89a make debug_sysctl() sysctl MIB check more strict. from smak.
attack similar to NetBSD-SA2003-014 can be mounted due to this flaw.
2003-10-01 21:45:14 +00:00
jmmv
457f6dbe79 Add several memory related terms. 2003-10-01 20:11:11 +00:00
fvdl
7baac02537 Reenable negotiation by default for discovery commands, until I figure out
what the issue is.
2003-10-01 18:01:06 +00:00
simonb
457cdc6624 Quieten down lint a little with a /* LONGLONG */ comment. 2003-10-01 15:47:27 +00:00
cl
1453f2ca48 make sa_vp_donate update p_nrlwps when it retires an LWP 2003-10-01 14:50:06 +00:00
kleink
b4a47d2e02 Query _POSIX_SEMAPHORES before the functional tests are run. 2003-10-01 12:18:53 +00:00
seb
e969ce2f8d Adjust pathnames to {md{4,5},sha1}.h after their removal from
src/include.

Approved by mrg@
2003-10-01 11:59:06 +00:00
tron
cc5dbcf926 Add dynamically generated sources to "CLEANFILES". 2003-10-01 09:39:50 +00:00
itojun
c6eceb7a58 avoid out-of-bounds memory access (array size out of sync). found by
openbsd 'bound' attribute.
2003-10-01 08:07:39 +00:00
itojun
f8641ea112 avoid out-of-bound memory access. found by openbsd "bound" C attribute. 2003-10-01 08:03:48 +00:00
itojun
83ddaf1cbf print prefixlen on -show 2003-10-01 06:24:19 +00:00
mycroft
de16a50404 Enabling caching of the video memory, and preset the modified bits. 2003-10-01 06:14:57 +00:00
jhawk
74f5f0da45 Redirect stderr as well as stdout in run_calendar (calendar -a) 2003-10-01 05:34:50 +00:00
jhawk
6a61a211cf Suppress output when running security.local if it produces no output.
/etc/security should produce no output (and thus suppress the report)
when nothing is wrong.

While we're here, use printf instead of two echos, like the rest of
the script.
2003-10-01 04:29:03 +00:00
itojun
ae7fc0a7d4 fix out-of-bounds access to ifindex2ifnet[]. found by iij seil team. 2003-10-01 04:22:33 +00:00
petrov
0545d82fe3 remove unused fastvec. 2003-10-01 02:01:20 +00:00
christos
ad9b12ae7e make symlinks for sha1.h md4.h and md5.h and don't install kernel.h 2003-10-01 01:37:13 +00:00
christos
46a53ee354 make some files symlinks from sys. 2003-10-01 01:35:39 +00:00
christos
d91f070b69 remove kernel.h 2003-10-01 01:31:20 +00:00
mycroft
e0933f24da Some small cleanup -- make the base,limit values vaddr_t, not char*.
Also add some code that attempts to deal with C16_VIDEO interrupts, though it
does not seem to make my color slabs work again.
2003-10-01 01:25:06 +00:00
mycroft
5f4d907d3f Add some debugging code to log intrstat history so I can track down stray
interrupts.
2003-10-01 00:07:48 +00:00
christos
3e10887300 Pass -D_LIBC 2003-09-30 22:30:16 +00:00
matt
86b06b2e50 Rearrange and with COMPAT_* properly 2003-09-30 21:45:09 +00:00
martin
1483742f8a Fix typo (DP83020 -> DP83820) from HITOSHI Osada in PR kern/23023. 2003-09-30 21:21:34 +00:00
christos
b6f1a886f4 Turn on strict warnings and add a main() 2003-09-30 21:05:04 +00:00
manu
2522a3f784 Build PowerPC again with _HAS_SIGINFO. Remove the ifdef on _HAS_SIGINFO
in the header file since all platforms supported by COMPAT_DARWIN now have it.
2003-09-30 21:04:54 +00:00
christos
2be3afdc37 don't check for vaddr_t yet, because all ports define it. 2003-09-30 21:01:59 +00:00
christos
2d148569b5 Make sure that our headers are clean enough by including all of them, and
checking for symbols that we leak in userland that we should not. Currently
only vaddr_t checked.
2003-09-30 20:47:23 +00:00
christos
a3fa6a15ef - Protect clockctl.h consistently with the correct ifdef.
- Move the ifdef before we include the world.
2003-09-30 20:38:38 +00:00
christos
f088ebb248 Don't make declarations visible from userland, unless _LIBC is defined,
because we don't define semid_t.
XXX: Should we be defining __semid_t instead?
2003-09-30 20:37:36 +00:00
christos
7bd9c302eb Regen 2003-09-30 20:36:40 +00:00
christos
4420def0dc #ifdef _LIBC the semid_t syscalls since we don't define semid_t in userland
except in the libc case.
XXX: Should we define __semid_t instead and make those visible from userland
again? Only ktrace and syscall() get affected.
2003-09-30 20:36:10 +00:00
christos
a474dc2106 Keep track of which prototypes we declared already so that we don't
declare them again.
2003-09-30 20:34:28 +00:00
wiz
8d0f21a650 Fix regexo in previous. 2003-09-30 20:11:26 +00:00
wiz
084f5623b9 Prepend doc- to have the same names as the groff distribution
(stripped out by strip.sed during installation).
2003-09-30 20:10:14 +00:00
manu
45fa851e14 Report keyup and keydown events
Qhen the event queue is full, sleep so that the process will be able to
consume entries
2003-09-30 19:56:54 +00:00