Commit Graph

168734 Commits

Author SHA1 Message Date
he d7ce7c850c Make racoonctl build again, sockmisc.c needs -DNOUSE_PRIVSEP in order
to not pull in all the rest of the racoon/ipsec code.
2008-03-28 14:37:08 +00:00
yamt 24840272de whitespace. 2008-03-28 14:29:57 +00:00
ad 4bd84ff96a Prevent overlapping calls to bind() and/or connect() on a Unix socket. 2008-03-28 12:14:22 +00:00
ad c2f3592995 Prevent listen() on a socket that is already connected - we already prevent
connect() on a listening socket.
2008-03-28 12:12:20 +00:00
kiyohara c65c7603a1 The status buffer must synchronize. If your CPU is cache writeback,
synchronous processing is necessary.  However, the size of a status
buffer will overwrite the memory while synchronously processing it
because it is very small.  We clear the flag SK_Y2_STOPC_OWN at more
late.
2008-03-28 08:46:01 +00:00
kiyohara 2b2915bee1 Fix ugly format in aprint_normal{,_dev}. 2008-03-28 08:35:44 +00:00
dholland f47dfee9a0 Correct documentation of default value of FILESMODE.
PR bin/38059 from Aleksey Cheusov.
2008-03-28 06:04:40 +00:00
lukem c63660b4d2 Fix fat-finger in previous.
Use "e.g.," instead of "E.g.,".
Noted by Geoff Wing.
2008-03-28 05:47:27 +00:00
dholland 000424deb2 Assorted updates and clarifications, grammar fixes and wording improvements,
and so on. Significant highlights:
   - mention src/tests and reference atf;
   - provide examples of MACHINE and MACHINE_ARCH;
   - mention that MKOBJ=no is not recommended;
   - correct the description of the default setting of USETOOLS;
   - document the interactions of build.sh -[uo] with various things;
   - document the interactions of build.sh tools and kernel=FOO;
   - use a lot fewer parentheses.
Discussed on netbsd-docs.
2008-03-28 05:23:44 +00:00
dholland d868f7242f Yet another rename workaround - this time check for . and .. early because
relookup() objects to being asked to handle them.
2008-03-28 05:02:08 +00:00
dholland fc5b5e6ac9 The phase of the build that -u prevents is "make cleandir", not merely
"make clean"; adjust build.sh's usage message to reflect this.
2008-03-28 04:24:18 +00:00
manu 1d223a6207 From Cyrus Rahman: Allow interface reconfiguration when running in privilege separation mode, document privilege separation 2008-03-28 04:18:52 +00:00
manu 182dbe8881 From Cyrus Rahman <crahman@gmail.com>
Allow interface reconfiguration when running in privilege separation mode,
document privilege separation
2008-03-28 04:18:51 +00:00
kiyohara b5ce3aaf20 Split device_t/softc. 2008-03-28 02:40:19 +00:00
uwe 5f35af71bf aprint_naive("\n"); 2008-03-28 01:37:25 +00:00
lukem c4851a8c8e Do not mark "old" major and major.minor shared libraries as
"obsolete"; just remove the entry, as third-party applications
may be linked against the old major shared library, and
that is a symlink to the old major.minor shared library.
E.g., "lib<name>.so.<N>" and "lib<name>.so.<N>.<M>"
Exceptions to this rule may include shared libraries that
are dlopen()ed at run-time, such as extra locales, etc.
2008-03-28 01:22:44 +00:00
uwe fd8017cbe4 Use aprint_*. 2008-03-28 01:19:12 +00:00
he c2dbd6e374 Make this build again. 2008-03-28 00:56:54 +00:00
he 4f00010385 Remove <m68k/saframe.h> here as well, patterned after hp300. 2008-03-28 00:35:04 +00:00
uwe 43e743115c Split device_t and softc. 2008-03-28 00:13:15 +00:00
uwe 9694bbc3bd Convert to use CFATTACH_DECL_NEW (no softc). 2008-03-27 23:57:58 +00:00
uwe e6030fb8a9 Split device_t and softc. 2008-03-27 23:57:21 +00:00
uwe 99111ebac3 Convert to use CFATTACH_DECL_NEW (no softc). 2008-03-27 23:42:34 +00:00
lukem 978f5e87de Remove all entries for obsolete major shlibs ("lib<name>.so.<N>");
third-party applications may be linked against them.
2008-03-27 23:16:33 +00:00
mlelstv 2d66ab6e51 kerberos includes are installed in krb5 subdirectory 2008-03-27 22:01:56 +00:00
christos 23b2e3cd80 Fix another integer overflow issue discovered by Maksymilian Arciemowicz.
On top of this, limit the range of getnumber to 0x00ffffff to make sure
that adding two of them does not cause an integer overflow.
2008-03-27 21:50:30 +00:00
mlelstv 930dfcfc6c The previous versions of kdc got different levels of patches
to make them work like a standard detaching daemon. Now keep
the original and just supply the "work as a daemon" command line
option as a default.
2008-03-27 21:42:22 +00:00
xtraeme db64271c36 Add back the bus_dmamap_sync() calls that were in the old
mfi_despatch_cmd() func after merging changes in rev 1.12.

Merge rev 1.69 and 1.64 from openbsd, fixing two bugs:

Fix subtle x ? y : z bug when calling mfi_create_sgl. This worked because
during runtime the bug condition does not happen.

When determining the size of the volumes and disks we want how many
bytes each of them has, not the number of kilobytes. this solves the
"why am i dividing by 2" question. the answer is "you dont, you multiply
by 512".

The first item fixes a panic with Dell Perc 5/i reported by oster@.
Changes tested on Dell Perc 5/i by oster@.
2008-03-27 21:15:29 +00:00
gdt 53ede206ae Following the change to remove GENERIC.local from the repository and
cinclude it, remove INSTALL.local and adjust kernels that referenced
it.
2008-03-27 21:13:29 +00:00
hauke e8d5382c3a Add explanation for the sbc flags to the kernel configurations that
use the sbc 53c80 driver. Until now, this information has lived in the
source code, only; neither sbc nor ncrscsi have a man page.

Triggered by PR port-mac68k/2921
2008-03-27 20:00:24 +00:00
ad bb61e73cd5 Add code for dynamically allocated mutexes, as posted on tech-kern. 2008-03-27 19:11:05 +00:00
ad be04ac4896 Make rusage collection per-LWP and collate in the appropriate places.
cloned threads need a little bit more work but the locking needs to
be fixed first.
2008-03-27 19:06:51 +00:00
ad f280ebae4b ALIGNBYTES + 1, not ALIGNBYTES in previous. 2008-03-27 18:44:21 +00:00
tnn b2e98d2a76 Revert for DOM0 as well at request of Christoph Egger.
I guess this bootstatic stuff is supposed to play nicely with dhcp,
but it still doesn't for me. I suspect it's due to misdetecting the bootdev.
2008-03-27 18:42:16 +00:00
ad feb4783fdf Replace use of CACHE_LINE_SIZE in some obvious places. 2008-03-27 18:30:15 +00:00
ad 0f33676c6c Introduce COHERENCY_UNIT/coherency_unit as proposed on tech-kern. 2008-03-27 18:28:20 +00:00
phx 14f850861b SmartFirmware only reports the first 256MB as available.
Also use the remaining memory by looking at the "reg" property.
2008-03-27 18:01:08 +00:00
phx 61737214f3 Restrict RTAS instantiation to the first 256MB. Other memory regions may not
be mapped by OpenFirmware.
2008-03-27 17:54:46 +00:00
ad b19c5369ab Make it compile again. 2008-03-27 17:34:54 +00:00
ad 2ffc44f47b Regen. 2008-03-27 17:14:21 +00:00
ad a19a515177 Put kqueue/kevent back as MPSAFE. 2008-03-27 17:13:25 +00:00
tnn 7b0c772876 Revert previous for DOMU kernels. They use a custom xennet bootstatic
callback for cmdline configuration. Keep for DOM0 kernels.
2008-03-27 16:48:35 +00:00
tnn 65052bdca1 Comment out NFS_BOOT_BOOTSTATIC. There no use having it in the default
kernel as it needs manual configuration anyway. Also, if it's enabled
but no static options given, then the kernel will happily configure the
interface as INADDY_ANY and not even try to use DHCP.
2008-03-27 16:33:32 +00:00
tsutsui 5155960c3f Note cobalt Z85C30 serial support. 2008-03-27 15:42:07 +00:00
tsutsui ec8e6f5b4b Add entries for zstty(4) serial on Qube 2700.
Also add tty0[1-3] entries to ttys(5) for com(4) at puc(4).
2008-03-27 15:26:39 +00:00
tsutsui dde8b66a5c Add zsc(4) and zstty(4) for Qube 2700. 2008-03-27 15:24:35 +00:00
tsutsui fc21b00873 Add kernel support for optional Z85C30 serial console on Cobalt Qube 2700.
There is some interrupt related issue (in MI z8530tty driver?) on heavy load,
but useful enough for debugging.
2008-03-27 15:21:46 +00:00
martin 5405d9ca93 Restrict the "mach prom" command to the primary cpu - it hangs the machine
hard on every other try on some machines.
2008-03-27 15:21:35 +00:00
martin afa2677a01 Fix a panic message and bump the number of retries untill we manage to
send an IPI a bit up again.
2008-03-27 15:20:47 +00:00
martin 1f9fec2e30 Get rid of delay() while waiting for IPI results; limit the waiting
time via the cycle counter. Suggested by macallan.
2008-03-27 14:51:02 +00:00