Commit Graph

158227 Commits

Author SHA1 Message Date
pooka
fa71b0706d Make it possible to interrupt waiters for fs operation completion
again.  This is useful until locking is further developed and basically
any deadlocks can be solved by killing appropriate processes.

Thanks especially to Tommi Kyntola and Antti Louko for sitting down
with me and discussing resource ownership and locking strategies
in implementing this.
2007-04-04 16:13:51 +00:00
pooka
ad018126a3 s/ppark/park/ to make all the variable names consistent - park is
always a pointer now.  no functional change
2007-04-04 15:50:56 +00:00
mishka
2f5bd28732 Revert previous commit until further investigation because it is
known as problematic.  Some problems obscured are dead Del key (I
can reproduce this with wscons scroll support) and panic (seen by
<elad> only).
2007-04-04 14:50:21 +00:00
vanhu
ace683e685 checks proto_id in ipsecdoi_chkcmpids() 2007-04-04 13:09:36 +00:00
vanhu
f31c3aee8e dumps peer's ID and peer's certificate subject /subjectaltname if they don't match 2007-04-04 13:07:31 +00:00
tnn
02ade499f6 pci_find_ino(): Use interrupt pin from pci_attach_args when mapping interrupts.
Use *ihp as a pure output. Fix case where interrupt pin #D was incorrectly
mapped. Some cosmetic changes while here. Ok'd by martin.
2007-04-04 11:04:33 +00:00
rmind
330166a2eb Fix tbl, which was misstyped as tlb. 2007-04-04 11:01:10 +00:00
rmind
0a747ea89c Unfortunately, missed procfs_proc_unlock() in previous.
Pointed out by pooka@
2007-04-04 10:50:42 +00:00
rmind
f49599932b k8_powernow_init_main: Initialize k8pnow_current_state. 2007-04-04 02:56:25 +00:00
rmind
9464b9ab22 Initialize k7pnow_current_state in k7_powernow_init(). Free the memory
in LKM's case. Found via CID: 4149
2007-04-04 02:50:02 +00:00
jmcneill
b25953f4b1 Note that i2s controllers lacking a TAS3004 are also supported. 2007-04-04 02:37:16 +00:00
wrstuden
db2a757990 Add a KASSERT to address CID: 4335.
Make sure that if we pull a buffer off of the read queue that it really
is a read request. Lower in this routine we base which queue we
dequeue the request from on its read/write state. Thus if a write
op ever ended up on the read queue, we'd explode (dereference NULL).
2007-04-04 02:30:43 +00:00
jmcneill
f972216e48 Simplify sw volume control, attach to AOAK2 as well, make mono work as
well as stereo, and print a friendlier message when we don't find a
hardware mixer.
2007-04-04 02:14:57 +00:00
rmind
523c1fc3e1 filecore_mountfs: Remove the dead code - fcmp cannot be non-zero here.
CID: 714
2007-04-04 02:00:58 +00:00
rmind
0dbab2d2b2 clockmod_sysctl_helper: For the sake of clarity - avoid magic number.
No functional change.
2007-04-04 01:50:15 +00:00
rmind
4e3ec8a076 vald_acpi_libright_get_bus: Plug a possible memory leak.
CID: 4358
2007-04-04 01:34:44 +00:00
rmind
199691e947 procfs_readlink: Handle a possible fail of fd_getfile(), also, we
do not need to check for error again.
CID: 4436
2007-04-04 01:27:32 +00:00
rmind
084180cb58 drm_do_addbufs_pci: Fix the mess of malloc()s and free()s, which
do not accept NULL arguments.
Found via CID: 4439
2007-04-04 00:30:58 +00:00
dyoung
5770bb4c31 Add extio, gpio. Remove admflash. Makes the compile again. 2007-04-03 23:53:34 +00:00
rmind
1db98c3723 socreate: l cannot be NULL.
CID: 4314
2007-04-03 23:44:53 +00:00
rmind
497adf4cc1 drvctlioctl: Plug a possible memory leak.
CID: 4363
2007-04-03 23:02:39 +00:00
rmind
5174ef0ece fileassoc_file_delete: Handle an error case of fileassoc_table_lookup().
CID: 4431
2007-04-03 22:29:23 +00:00
dyoung
eabf98a2cc Add extio and gpio so that this compiles again. 2007-04-03 21:54:01 +00:00
dyoung
aa13eb6f2f Add extio and gpio at mainbus so that this builds. 2007-04-03 18:06:31 +00:00
dyoung
29bf5964b4 Auto-detect RAM size. 2007-04-03 18:05:57 +00:00
hannken
0adf7298aa Remove calls to now obsolete vn_start_write() and vn_finished_write(). 2007-04-03 16:11:31 +00:00
pooka
9cc76110cd In sync, skip vnodes which have a null v_data pointer. This is a
workaround for the problem analyzed more deeply in kern/30831.  In
short, the problem is keeping the vnode on the mount point vnode
list during reclaim.  If reclaim happens to sleep (as is a possibility
with smbfs due to calling vrele() and therefore possibly VOP_INACTIVE),
code going through the entire mountpoint vnode list will hit
half-reclaimed vnodes.
2007-04-03 14:18:19 +00:00
jmcneill
71195759ef In the AOAKeylargo case, there is no hardware volume control. Instead,
implement volume control in software via audio filter framework.

XXX: audio filter framework doesn't let me pass parameters from the hw
     softc to the filter, so the volume variables are globals for now.
2007-04-03 13:11:21 +00:00
yamt
ef47d5b67e - don't use veriexec_mountspecific_key uninitialized.
- fix weird "Veriexec: IDS mode, allowing unmount" messages.
2007-04-03 09:17:31 +00:00
xtraeme
02e5ad5584 Initialize msr_read explicitly to avoid a hang, from FUKUMUTO Atsushi. 2007-04-03 07:56:08 +00:00
lukem
b234edd83d * Support reading multiple keys with the same name if '-D' is given,
using db->seq() and match comparison rather than db->get().
* Deleting multiple keys of the same name appears to already work as-is
  with a single db->del() call; I had code to implement this if '-D'
  is given but it appears to be unnecessary.
* Cosmetic/KNF tweaks.
2007-04-03 04:52:32 +00:00
macallan
f5a25c2db5 regen 2007-04-03 03:33:54 +00:00
macallan
8ec5fb10e2 add 1680x1050 @ 60.00Hz 2007-04-03 03:33:41 +00:00
jmcneill
c38a78ca65 Match AOAKeylargo, and don't make finding deq a requirement. Sound now
comes out of my last gen G4 Mac Mini, minus mixer controls. ok macallan@
2007-04-03 03:26:52 +00:00
macallan
d0ebad929a adapt to pmu changes 2007-04-03 03:15:17 +00:00
macallan
06802d0b24 cleanup a bit, fix some stupid pasto
thanks to jmcneill for finding the problem
2007-04-03 03:09:25 +00:00
matt
086f0329dc Fix constraints on ffs. 2007-04-02 16:45:39 +00:00
christos
7074d009d5 Don't panic if we don't find a symbol or string table. Perhaps the kernel
has been stripped?
2007-04-02 16:44:44 +00:00
matt
20feb5b064 Fix cut-and-pasto. 2007-04-02 16:44:31 +00:00
matt
831f4686c1 Get rid of old unused constraints. 2007-04-02 16:44:17 +00:00
njoly
5c37954c94 Fix select() behaviour with NULL timeout, to blocks indefinitely
instead of returning immediately.
Reported by Edgar Fuß on port-amd64@.
2007-04-02 16:29:55 +00:00
christos
e3845a8924 Merge the multibyte routines into one file, and add stub implementations
for one more that is needed, and commented out partial implementations for
the test of the functions. This should stop the re-defined symbol errors
for now.
2007-04-02 15:53:25 +00:00
mishka
833cf25850 Provide yet more functionality to wscons.conf's "setvar" (see
examples in the wscons.conf). Now, you have to specify wscons
control device and put variable name and parameter on separate
fields.

NB: changes ARE NOT backward compatible and config file must be
checked (but default config doesn't utilizing setvar).
2007-04-02 12:42:42 +00:00
apb
2347a3fcd8 The "${progname} started: ${build_start}" message will appear in the
"Summary of results", so don't print it again just before that.
2007-04-02 11:00:10 +00:00
apb
daaa947e26 * Report the NetBSD version being built.
* Report the host platform.
* Print more of the initial messages in the same part of the code.
2007-04-02 10:57:36 +00:00
rmind
086bf77aab yamt-idlelwp: Note few WIPs, people are working on that. 2007-04-02 10:18:52 +00:00
mishka
933b97ecac Implement locking switcher for alternative layots (many languages
are based on completely different alphabet, and typesetting with
one of modifiers pressed all the time is just a PITA). See
http://mail-index.netbsd.org/tech-kern/2007/03/29/0002.html for
usage details.

While here, perform the following small cleanups:
- KS_Cmd_Scroll* execution should be explicitly interrupted with 'break'.
- Some 'human' cleanups in ksym extraction algorithm (no func. changes).
- Add a note: implemented ksym translation isn't sufficient for l10n.
2007-04-02 10:01:31 +00:00
tnn
a99b3971af Add myself (tnn@) to the list of developers. 2007-04-02 08:21:46 +00:00
christos
28712ce84b bring in FreeBSD's vfscanf() to gain multi-byte/collation support.
Unfortunately it is too difficult to make vfwscanf and this share
the same code like I did with printf, because for string parsing
the code is too different.
2007-04-01 19:23:55 +00:00
tnozaki
fd2dd8ec0d add new encoding support to iconv(3):
- RISCOS-LATIN1
	- DEC-MCS
	- DEC-HANYU(libDECHanyu)
2007-04-01 18:52:28 +00:00