Commit Graph

163635 Commits

Author SHA1 Message Date
yamt
f3ea2e999d 4.99.37: kmutex_t and krwlock_t changes. 2007-11-21 10:20:55 +00:00
yamt
38d5e34116 make kmutex_t and krwlock_t smaller by killing lock id.
ok'ed by Andrew Doran.
2007-11-21 10:19:06 +00:00
rillig
55bf992f36 The e* functions don't call the error handler in every case, just in
case of an error.

Mentioned the error conditions for the various functions.
2007-11-21 09:17:24 +00:00
sborrill
3a63ee360d While boot.cfg support is currently i386 only, it is hoped that its use
will be extended to other appropriate ports in future. Note as such in the
bugs section.

This is an MI man page, so should not be flagged as i386 in the header.
2007-11-21 08:14:10 +00:00
dyoung
3b205d80f9 Detect cardbus_setpowerstate() errors and aprint_debug() a warning.
Leave alone the Latency Timer set by the Cardbus bridge.

With version 1.152 of dev/pci/pccbb.c, we can power down the card
when it is not in-use, so do that.
2007-11-21 02:10:45 +00:00
dyoung
a5a3f28dce Sometimes a change of CSTSCHG# accompanies the first interrupt from
an Atheros WLAN.  That generates a CB_SOCKET_EVENT_CSTS event on
the bridge.  The event isn't interesting to pccbb(4), so we used
to ignore the interrupt.  Now, let the child devices try to handle
the interrupt, instead.  The Atheros NIC produces interrupts more
reliably, now: used to be that it would only interrupt if the driver
avoided powering down the NIC's cardslot, and then the NIC would
only work after it was reset a second time.
2007-11-21 02:07:09 +00:00
dogcow
37ec50ec33 silence, might be used uninitialized warning! 2007-11-21 01:31:34 +00:00
dogcow
f5f98b502d add <dev/pud> (hi, pooka!) 2007-11-21 01:28:44 +00:00
dsl
aa21c2800b When copying system call arguments into kernel space, either copy 4
arguments (enough for the majority of calls) or the maximum of 10 (8 plus
2 for sys__syscall).
Avoids all the unlikely-to-be-predicted-correctly in the unrolled loop.
This might be a problem if a thread's stack doesn't have the extra args
mapped - but that is extremely unlikely.
2007-11-21 00:15:08 +00:00
drochner
ad37bd5da3 fix gcc bug #34130 (wrong multiplication with negative constants
inside abs()) which got some attention in the linux kernel mailing
list recently
2007-11-20 23:27:56 +00:00
xtraeme
f0927a591f Put back again softintr_init() that was removed in rev 1.67, this fixes
the panics that people reported on current-users.

Tested and verified by Greg Oster.
2007-11-20 23:10:49 +00:00
joerg
08598b7d45 Merge from jmcneill-pm: XXX allow pci_find_mem to map a 64-bit BAR. 2007-11-20 22:48:49 +00:00
xtraeme
da3a7579a9 Use BUILDSYMLINKS to avoid a compilation error where the source file
has the same name than the LKM.
2007-11-20 21:47:34 +00:00
plunky
166b860e06 allow listening to channel RFCOMM_CHANNEL_ANY, which means that
we should allocate the first available channel at the given
device address.
2007-11-20 20:25:57 +00:00
plunky
fb5c91f9fc when listening, don't attach to the 'any' address unless
we wanted to listen to the 'any' address.
2007-11-20 20:19:24 +00:00
plunky
cd286130a6 don't explicitly list unused and NULL entries in the domain declaration 2007-11-20 20:18:00 +00:00
pooka
b305014b90 add /dev/pud 2007-11-20 18:58:46 +00:00
pooka
378e2f0d3e Add a very simple intro-level example on how to use pud. 2007-11-20 18:58:17 +00:00
pooka
b7737261b0 pud (userspace char/block drivers) build goo 2007-11-20 18:54:31 +00:00
pooka
001b8e986e Add initial support for userspace block/char device drivers. This
is a very rough cut supporting only open/close/read/write for char
devices for the time being.  It also takes some extra liberties in
error value processing wrt. to the request-response protocol, but
that will be fixed later.  On the bright side, I haven't managed
to crash my test kernel in several hours with this anymore, so it
can be considered extra-stable ...  All in all this is remarkably
similar to puffs from multiple angles (well, actually, big surprise
there) and these two will share more and more code in the future.

The hardest part, coming up with the name, was done by agc.  I'm
not exactly sure what "pud" is supposed to stand for, but if my
British is anything close to par, I understand that "pud" over
there can stand for just about anything, so it's a safe bet.
2007-11-20 18:47:05 +00:00
pooka
b6cf4ea195 Allow to register configuration routines to putter. Once a
configuration routine has been attached to the putter device minor
number, the configuration routine will be called when the device
is opened.  This is required for subsystems which enter the kernel
only through the file descriptor and cannot configure themselves
via another route (like puffs does via the mount() syscall).  Keep
the minor number 0 for "wildcard" uses where we do not require a
hardwired config routine (e.g. the puffs case).
2007-11-20 18:35:22 +00:00
xtraeme
58406f0ac9 Update a comment. 2007-11-20 17:27:35 +00:00
xtraeme
714201def5 After comments from Andrew Doran:
- The mutex for the callout handler must run at IPL_SOFTCLOCK.
- Just stop the callout in sysmon_envsys_unregister() and don't wait
  for the callout to finish.
2007-11-20 17:24:32 +00:00
ad
10564377d6 Use the softint API. 2007-11-20 17:13:04 +00:00
macallan
bc546cac16 don't bother looking for console-specific properties on devices that aren't
PCI_CLASS_DISPLAY
2007-11-20 17:06:18 +00:00
tsutsui
ddd9ad0a5e Fix e2fs_itpg calculation. 2007-11-20 16:34:37 +00:00
sborrill
e14a5d3249 Fix some typos and note that bootxx_cd9660 can load from a path name longer
than 8 chars.
2007-11-20 16:12:50 +00:00
sborrill
980d2ea0f0 Support for boot.cfg has been added to the i386 bootloader 2007-11-20 15:49:14 +00:00
sborrill
d849e6b356 Add support for /boot.cfg configuration file in x86 boot loader.
This allows easy configuration of banner text, console device and timeout
as well as allowing menus of commands to be displayed. If /boot.cfg
is not present, then the existing behaviour does not change.

The sections in the boot loader source are surrounded by #ifdef SMALL
allowing this functionality to be removed if space is at a premium.
2007-11-20 15:37:36 +00:00
pooka
e0122367ec Don't pass devname to {b,c}devsw_attach(), it's not used. 2007-11-20 14:47:14 +00:00
ad
b7b14106bd Don't rely on others being kind enough to pull in sys/device.h 2007-11-20 14:33:58 +00:00
yamt
3a40ff100a put back IPL_IPI as it's still used. 2007-11-20 14:31:46 +00:00
yamt
c77d808475 build rb.c. 2007-11-20 13:49:01 +00:00
yamt
c64de7a608 sync the prototype of cpu_switchto with sys/cpu.h. 2007-11-20 13:37:37 +00:00
yamt
3acf687ede rename
RB_LEFT
	RB_RIGHT
	RB_OTHER
to
	RB_DIR_LEFT
	RB_DIR_RIGHT
	RB_DIR_OTHER
so that it can coexist with sys/tree.h.
2007-11-20 12:32:55 +00:00
sborrill
3cdec1aaba Add IYSWIM 2007-11-20 12:26:12 +00:00
pooka
7c0ef2ee17 Retire M_PUFFS, use kmem(9) instead. 2007-11-20 11:51:01 +00:00
sborrill
12e455f905 Some manufacturers use the same vendor and product id for different devices.
We need to sanity check the DeviceClass in this case.

Currently known guilty products:
0x050d/0x0121 Belkin Bluetooth and USB2LAN

If this turns out to be more common, we could use a quirk table.
2007-11-20 10:43:01 +00:00
xtraeme
53cbf2f68b Improve description for ENVSYS_FMONNOTSUPP. 2007-11-20 10:08:47 +00:00
yamt
0ae55456d7 dzstart: remove a variable which is no longer used after the previous change. 2007-11-20 09:16:53 +00:00
xtraeme
21ba58354e Indentation fix for the units list. 2007-11-20 09:16:36 +00:00
yamt
636f153f4d fix compilation in the case of _KERNEL && DIAGNOSTIC && !RBDEBUG. 2007-11-20 08:30:05 +00:00
ad
a72bb4a2f0 For the last time, compile! 2007-11-20 01:53:50 +00:00
uebayasi
dfd4cf0aaf Indent. 2007-11-20 01:33:32 +00:00
uebayasi
146861b4fa Sort variable definitions. No functional change expected. 2007-11-20 01:30:07 +00:00
ad
bf2e3dcbb6 Call ttstart() with tty_lock held. 2007-11-20 00:45:23 +00:00
ad
ce2527b4f4 IPL_IPI -> IPL_HIGH 2007-11-19 23:07:48 +00:00
ad
772ae8c190 G/C the PDP cache generation stuff. When allocating, just check for a valid
entry where nkpde currently points.
2007-11-19 21:46:51 +00:00
mjf
0b2e8c0fd9 Document mjf-devfs branch. Thanks to xtraeme for reminding me to do this. 2007-11-19 20:45:41 +00:00
ad
e27aeee1b7 Work around another funny until I figure out what is going wrong: somehow,
tp->t_rawq.c_cv.cv_waiters is dropping to zero while there are still LWPs
waiting on the queue. dtrace would be really handy here :-/
2007-11-19 19:53:48 +00:00