Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
161,894 Commits 606 Branches 0 Tags 3.1 GiB
9f0e854e77
Commit Graph

45717 Commits

Author SHA1 Message Date
ad
eb5991fdda x86_ipi: don't wait for the IPI to go, unless DIAGNOSTIC. If it doesn't go, 
the system is going to fail regardless.
 2007-08-07 11:28:26 +00:00
ad
4f5723ab5d Add a couple more calls to pmap_update(). 2007-08-07 11:25:40 +00:00
dyoung
178f1dafa8 Use satocsdl() instead of SDL(). Constify. 2007-08-07 04:42:49 +00:00
macallan
23f9222c11 if we can't determine the console framebuffer's depth assume it's 8bit 
Mainly for /chaos/control which doesn't have any useful properties on its node
 2007-08-07 01:59:23 +00:00
adrianp
475217d719 Remove iso(4) from GENERIC kernels, as discussed on tech-net@ 
Ok'ed by core@
 2007-08-06 21:26:53 +00:00
kiyohara
f668ef5a16 Add support agp_amd64. Imported from FreeBSD. 2007-08-06 06:40:41 +00:00
simonb
3696d5ed67 If an EST frequency table isn't found, fake one up by interpolating 
values from the high/low voltages and frequencies.  Replaces the old
method of using the lower and upper frequencies only.
 2007-08-06 03:38:49 +00:00
ad
c8a5b7d261 Fix traceback through interrupts. This isn't complete but at least it's 
displaying useful output again:

db{0}> bt
cpu_Debugger(c1aa4800,c1aa8b1c,c1aa803c,1,c1aa48ec) at netbsd:cpu_Debugger+0x4
comintr(c1aa4800,cf38fc04,0,0,0) at netbsd:comintr+0x65c
DDB lost frame for netbsd:Xintr_ioapic_edge4+0xa8, trying 0xcf413ff4
Xintr_ioapic_edge4() at netbsd:Xintr_ioapic_edge4+0xa8
--- interrupt ---
--- switch to interrupt stack ---
cpu_idle(cf421e00,cf421e00,0,c0264950,cf421e00) at netbsd:cpu_idle+0x30
idle_loop(cf421e00,0,c01002d2,fbff,c01002d2) at netbsd:idle_loop+0x7a
 2007-08-05 19:27:45 +00:00
ad
474db80a05 Keep alignment small for _STANDALONE. 2007-08-05 14:56:10 +00:00
jmmv
348921e110 Move 'options FONT_VT220L8x16' to a common place for all kernels so that 
the user needn't remember to add this option to get a font that is almost
identical (but fixed!) to the built-in one.
 2007-08-05 13:37:31 +00:00
ad
17be89c2d5 - trace_enter/trace_exit/ktrace don't need kernel_lock. 
- Use a RISC-style copyin optimised for pulling system call arguments.
  In tests I have done on Intel hardware this removes any measurable
  advantage to passing the syscall arguments in registers.
 2007-08-05 10:56:52 +00:00
ad
ebb5b4b828 Don't rely on the ci_self150 hack. 2007-08-05 10:51:03 +00:00
ad
9d841d6e31 - Remove crud. 
- Don't use the ci_self150 hack.
 2007-08-05 10:50:26 +00:00
ad
2818ab2bd1 - Align assembly routines on 32-byte boundaries. 
- Don't bother doing anything special for CPUVAR() when !MULTIPROCESSOR.
 2007-08-05 09:56:48 +00:00
macallan
0591adcde1 fix another typo, now wsfb on ofb should work properly 2007-08-05 04:36:18 +00:00
tsutsui
72e8fd4495 Make clockintr() return 1 since it's always handled so that 
clock interrupt is properly acounted by evcnt(9) in isa_irq.S.
 2007-08-04 20:14:21 +00:00
tsutsui
79f82cdd0a Fix a typo in a debug printf message. 2007-08-04 16:31:21 +00:00
tsutsui
b108f67c05 TAB/space cleanup. 2007-08-04 13:00:45 +00:00
tsutsui
fe3cdffb4a Add ci_cpuid where it's missing. 2007-08-04 12:53:31 +00:00
ad
9c6535a6de Add ci_cpuid where it's missing. 2007-08-04 09:49:51 +00:00
dyoung
d3086ce90d Reduce differences between sgimips/Makefile and evbmips/Makefile, 
making the sgimips/Makefile 'tags' target work.
 2007-08-04 05:32:39 +00:00
jmmv
b77d937244 Only deal with isa dma stuff if this is available/needed. Fixes the build 
of kernels when, e.g. the audio driver is disabled.

Per cube@'s suggestion.
 2007-08-03 16:02:17 +00:00
tsutsui
51b5e07bf9 Misc KNF and cosmetics. 2007-08-03 13:15:56 +00:00
tsutsui
e4daee28e3 Add support for booting off raidframe RAID1 mirrors. 
Should close PR port-cobalt/36639, but with a different way
(which is similar with pmax) from a patch in the PR.
 2007-08-03 12:57:38 +00:00
jmcneill
fa02ff9185 Pull in i82489var.h in the NLAPIC > 0 case, from Geoff Wing. 2007-08-02 11:23:35 +00:00
dogcow
18fbcfc528 remove historical include no longer needed; fixes build breakage. 2007-08-01 23:48:24 +00:00
jmcneill
cedc871a24 Explicitly reinitialize lapic on resume; gets the clock ticking on the 
Lenovo T60 and probably many others.
 2007-08-01 22:41:59 +00:00
macallan
f46b7897c0 fix the fix - do not impose all those extra cache ops on all supersparcs, 
only those without external cache
 2007-08-01 00:06:32 +00:00
jmmv
d291012153 Switch NetBSD/shark's installation kernel and disk image to use wscons 
instead of pccons, just like its GENERIC kernel does.

This obsoletes pccons in this platform, which will probably be removed
later.
 2007-07-31 19:51:58 +00:00
jmmv
f6199f0afb Force vga(4) to override the font built into the VGA ROM by one that 
properly supports all characters expected by wscons.  While doing this
configure a font that resembles the native one as much as possible.

This is a prerequisite for switching INSTALL to use wscons.
 2007-07-31 19:35:37 +00:00
macallan
7add3da5f4 SuperSPARCs without cache controller need some extra cache flushs 
With this a MULTIPROCESSOR kernel works again with a pair of SM50
 2007-07-31 05:21:47 +00:00
macallan
1e54ba9802 add hardware cursor support 2007-07-30 23:28:13 +00:00
macallan
4ba11875bf add wsdisplay support for cg14 
The console will run in 8bit but switch to 32bit in mapped mode so XFree86
with wsfb can run in 24bit colour.
 2007-07-30 18:20:08 +00:00
he
e448d82109 Change one overlooked label reference from "bad" to "done". 2007-07-30 14:32:40 +00:00
jmmv
5575290bc2 Fix build of kernels that use vga or igsfb together with pccons by avoiding 
a warning mentioning an unused static function.
 2007-07-30 13:02:01 +00:00
jmmv
3e495310cf Print some debugging information when DEBUG is set, not when TERSE is 
unset.  The latter is not a "standard" definition, and was used in very
few places anyway.
 2007-07-30 12:25:14 +00:00
jmmv
38b7aff676 Replace usages of bcopy: 
- Use memmove to copy memory blocks that are likely to overlap.  Fixes
  some rendering problems in programs that use, for example, the "insert
  rows" feature.  This was reproducible with sysinst as explained here:
  http://mail-index.netbsd.org/tech-install/2007/07/29/0000.html
- Use memcpy in situations where the blocks cannot overlap.
 2007-07-29 18:01:11 +00:00
jmmv
daea18cac0 Disable the mouse driver during installation. Not needed. 2007-07-29 15:14:08 +00:00
nonaka
1de7f19819 Added support to detect AC adapter and charge battery. 2007-07-29 14:31:23 +00:00
nonaka
9efd71ca75 sync with OpenBSD. 2007-07-29 14:29:38 +00:00
ad
a0d1fd8d0c It's not a good idea for device drivers to modify b_flags, as they don't 
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.
 2007-07-29 13:31:07 +00:00
ad
66fefd117b It's not a good idea for device drivers to modify b_flags, as they don't 
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.
 2007-07-29 12:15:35 +00:00
kiyohara
5fc45c72a5 Add ichsmb(4). 2007-07-28 11:06:23 +00:00
tsutsui
1983a26b2e Check a correct value on a sanity check in jensenio_eisa_intr_map(). 
Fixes yet another bug on Jensen found on tracking PR port-alpha/36628.
 2007-07-27 13:37:07 +00:00
he
40db4f866e Move the include of <machine/intr.h> into the _KERNEL / _LOCORE #ifdefs. 
Fixes build problem with the extent regression test.
 2007-07-26 14:54:09 +00:00
nonaka
a3f23fba88 use WSDISPLAY_TYPE_PXALCD. 2007-07-26 12:17:37 +00:00
macallan
0904724e1a - move register definitions to crmfbreg.h 
- identify as WSDISPLAY_TYPE_CRIME
- always claim we're 32bit
- switch to 32bit colour in WSDISPLAYIO_MODE_MAPPED, switch back to 8bit
  in WSDISPLAYIO_MODE_EMUL
- restore palette, turn off the cursor when switching back to console mode
- add a few delay()s, hopefully that will deal with occasional messed up
  video timings *fingers crossed*
 2007-07-26 02:25:14 +00:00
tsutsui
49773e5f60 Flush all cache data at the beginning of the bootloader. 
Fixes silent hangs after loading 4.0_BETA2 GENERIC kernel
on Qube2 (and RaQ2) with 256MB RAM, reported and tested
by James Hartley on port-cobalt.

Obviously this should be pulled up to netbsd-4.
 2007-07-25 14:11:06 +00:00
xtraeme
bd8aee308c There's no need to check for cpu_vendor before calling est_init(), just 
pass cpu_vendor to it.
 2007-07-25 13:41:53 +00:00
macallan
fbbad8fbc0 add support for ioctl(WSDISPLAYIO_LINEBYTES) so XFree86/wsfb should work now. 
Doh.
 2007-07-24 20:09:57 +00:00
hannken
40135c9d50 Handle IST_LEVEL interrupts better. Disable interrupt while it is 
processed and enable afterwards.  Up to now IST_LEVEL interrupts always
fired again during processing leaving unnecessary pending interrupts.

Tested on EXPLORA451.

Ok: Simon Burge <simonb@netbsd.org>
 2007-07-24 15:22:18 +00:00
hannken
6409dfa660 Use PR_NOWAIT in in pmap_enter_pv() and take care of PMAP_CANFAIL. Built 
after powerpc/oea/pmap.c.  Now a LOCKDEBUG kernel runs on EXPLORA451.

Ok: Simon Burge <simonb@netbsd.org>
 2007-07-24 15:19:09 +00:00
pooka
ac9042f5ce remove TNF advertising clause from file copyrighted to me 2007-07-24 14:41:29 +00:00
he
a4d164aaef Same change as -r1.133 of src/sys/arch/amiga/amiga/pmap.c: 
As per hint from ad@, change uvm.kernel_object to uvm_kernel_object,
  to adapt to the recent uvm changes.
 2007-07-23 23:13:04 +00:00
he
03d59d870c After move to 64-bit paddr_t, cast via vaddr_t before casting to void*. 
OK'ed by macallan@.
 2007-07-23 22:28:23 +00:00
dogcow
0dfe89a6a7 Same change as -r1.133 of src/sys/arch/amiga/amiga/pmap.c : 
As per hint from ad@, change uvm.kernel_object to uvm_kernel_object,
  to adapt to the recent uvm changes.
 2007-07-23 19:21:03 +00:00
tsutsui
c896224c88 Include <sys/proc.h> explicitly. 2007-07-23 17:55:39 +00:00
he
960a35be0a As per hint from ad@, change uvm.kernel_object to uvm_kernel_object, 
to adapt to the recent uvm changes.
 2007-07-23 17:32:25 +00:00
macallan
f6e06b3004 sprinkle (vaddr_t) to deal with sizeof(paddr_t) != sizeof(void *) 
Now the loader should build again.
 2007-07-23 00:45:52 +00:00
mjf
a1e23633a9 Remove newline from format string of aprint_normal. 
Thanks to pooka@ for pointing it out.
 2007-07-22 23:45:50 +00:00
he
ef2c9ff103 Explicitly include <sys/proc.h> for tsleep() prototype. 2007-07-22 20:39:22 +00:00
he
41b06f79ca Explicitly include <sys/proc.h> for tsleep() declaration. 2007-07-22 20:36:13 +00:00
ad
2a051b63b4 +#include <sys/proc.h> 2007-07-22 08:50:26 +00:00
tsutsui
d99606f782 More fixes for Jensen, reported and tested by Christoph Franzen 
in PR port-alpha/36628:
- make jensenio_eisa_maxslots() return 8 (instead of 16) since
  EISA config for slot 8-15 on jensen could return invalid values
- pass eisa_chipset_tag_t to eisa_init() and check eisa_maxslots()
  on probing EISA config space
- pass M_ZERO to malloc(9) and make sure malloc(9) doesn't fail
- fix typo in a debug printf, add more debug printfs, and
  use #ifdef EISA_DEBUG to enable them
- cast uint8_t value to uint32_t before shift more than 8 bits
- check buffer region on reading compressed data from EISA config space
 2007-07-22 02:14:39 +00:00
matt
5aaa0be03a Don't call proc_trampoline_mp when starting a lwp. (XXX why did we do this 
before?)
 2007-07-21 19:26:14 +00:00
matt
ae0a442af3 Make MP kernels build again. Heck, if I'm lucky, maybe these changes will 
even make one boot.
 2007-07-21 19:12:15 +00:00
ad
744a92f0f8 Don't depend on uvm_extern.h pulling in proc.h. 2007-07-21 19:06:20 +00:00
tsutsui
78c4167012 - kern_kern_microtime.c is gone, so replace references of it 
with new kern/kern_cctr.c in TODO comments
- also replace struct cc_microtime_state member in struct cpu_info
  with struct cctr_state defined in <sys/cctr.h>, which is unused yet
  in this port
 2007-07-21 12:15:05 +00:00
tsutsui
13084fc846 Add MI todr(9) support and timercounter(9) support with kern_cctr.c to alpha: 
- use todr(9) API with MI mc146818(4) driver and remove homegrown
  todr stuff from MD alpha/clock.c and alpha/mcclock.c
- also remove obsolete cc_microtime stuff from MD code
- add ci_pcc_freq member in struct cpu_info for cpu_frequency(), and
  calibrate it with mc146818 interval clock in mcclock attachment
- call cc_init() in cpu_initclocks(9) because all alpha cpus have
  a pcc counter

Tested on DEC 3000/300 and AlphaPC 164, but not on any SMP machines yet.
 2007-07-21 11:59:55 +00:00
agc
217f538f17 Add pcn device - useful for VMware guests, as VMware emulates a pcn. 2007-07-21 02:24:03 +00:00
nonaka
c3fdd97b8c Add missing callout_init(). 2007-07-21 01:47:58 +00:00
tsutsui
0681619847 Sync with sun3/locore.s rev 1.88: 
> Pass a frame pointer to trap() rather than the 'entire frame' trick.  Gcc4
> was optimizing away modifications to the frame contents (it's not nice to
> trick gcc).  Pass the pointer as the first argument to reduce the number
> of places that would be changed otherwise.  Fixes the getcwd regression
> test on most m68k ports.
Fixes MMU fault panic in trap() on sun3x.

This should be pulled up to netbsd-4 too.
 2007-07-20 22:26:29 +00:00
rumble
da6e256110 Comment FILECORE and add ADOSFS. 2007-07-20 22:26:12 +00:00
hannken
7111447178 Move intr_init() from cpu_configure() to cpu_startup() so softintrs 
pool_get() works again.
 2007-07-20 13:41:53 +00:00
hannken
89bef33fc3 Bump SYMTAB_SPACE so that it fits again. 2007-07-20 13:39:58 +00:00
mjf
58b2412ea3 s/aprintf_normal/aprint_normal 2007-07-19 14:07:10 +00:00
mjf
ce43347321 Change printf to aprintf_normal and add a newline as requested by Christoph Egger on port-xen. 2007-07-19 13:31:26 +00:00
tsutsui
0d3f4cbf60 Backout changes on lpt_jensenio.c rev 1.2. 
lpt at jensenio doesn't seem to have a specific interrupt vector
but uses a normal EISA interrupt.

Fixes another part of PR port-alpha/36628 and PR port-alpha/20386.
 2007-07-19 12:58:29 +00:00
tsutsui
43c2f4d087 Avoid NULL pointer dereference in MD device_register() function. 
Fixes a part of PR port-alpha/36628.
 2007-07-19 12:46:45 +00:00
tsutsui
9463f365d3 Fix indent in printf message. 2007-07-19 12:42:46 +00:00
tsutsui
1eb143cb4c Use options<space><tab>. 2007-07-18 15:49:09 +00:00
tsutsui
b0e549dc68 - add options COMPAT_40 
- use options<space><tab>
 2007-07-18 15:48:49 +00:00
tsutsui
56738a1953 Just include GENERIC and use <no> keyword of config(9). 2007-07-18 15:42:41 +00:00
tsutsui
64b3321eaf Make KGDB config files just include GENERIC{,3X} and 
use <no> keyword of config(9).
 2007-07-18 15:35:06 +00:00
tsutsui
958c1b32fc Add options COMPAT_40. 2007-07-18 15:19:07 +00:00
tsutsui
a02017d6e7 Don't set PSL_LOWIPL to SR on switching to lwp_trampoline(). 
If interrupts are enabled when kernel is switched to lwp_trampoline(),
we could get lock errors by interrupts before lwp_unlock() in lwp_startup()
is called. The spl is lowered by spl0() in lwp_startup() after lwp_unlock().

Ok'ed by mhitch@ and ad@ on port-mips.
 2007-07-18 13:46:46 +00:00
uwe
0fbcabef5c Implement footbridge_mem_bs_mmap. Needed for mmaping igsfb(4) 
framebuffer on netwinder.
 2007-07-18 09:47:49 +00:00
christos
27e0d2b78b kill MFSNAMELEN 2007-07-17 20:17:52 +00:00
daniel
bcb48b7bb5 Add VIA_PADLOCK. 2007-07-17 16:40:21 +00:00
he
8f6c85ab27 Patterned after sgimips, move include of <machine/intr.h> to only be done 
when _KERNEL is defined and _LOCORE isn't.  Fixes build problem in the
extent regression test.
 2007-07-17 14:36:25 +00:00
he
dc396e3ae7 Adapt to the new signature of kthread_create(), and the removal 
of kthread_create1().
 2007-07-17 11:16:14 +00:00
macallan
02d1fad157 use 64bit paddr_t and bus_addr_t so we can mmap DMA buffers uncached 2007-07-17 04:03:22 +00:00
macallan
57a324e6fa if we have 64bit paddr_t add a flag which indicates non-cachable for use 
with mmap*() and pmap_enter()
Mainly for allowing userland to mmap() the O2's framebuffer uncached
 2007-07-17 04:01:34 +00:00
macallan
8bad22882a add definitions for non-cached pages 2007-07-17 03:53:56 +00:00
macallan
5a3795a6c0 add options MACE_NEEDS_DELAYS since many machines need those delays in 
bus_space*() operations on macs
 2007-07-17 00:26:57 +00:00
macallan
a1457f5905 change pmap_phys_address()s parameter to paddr_t since that's what it gets 
fed from mmap*() anyway
approved by gimpy
 2007-07-16 23:48:03 +00:00
macallan
03d6f84d6b fix a typo - if we pass no parameters to OF_interpret we better don't say 
we do...
 2007-07-16 22:17:03 +00:00
xtraeme
22d9c7eda3 battery_gtredata: there's no need to set ENVSYS_SVALID on each case, 
just set it after the switch.
 2007-07-16 18:24:29 +00:00
xtraeme
6897610c87 Disable monitoring in the Integer sensors, they are just informative 
and shouldn't need it anyway.
 2007-07-16 17:57:41 +00:00
macallan
c1d4259805 turn pmap_phys_address() into a macro since all the function ever did was 
to cast its parameter to paddr_t and return it
approved by mrg
 2007-07-16 16:36:06 +00:00
marcus
48461668c0 Fixed incorrect logic in mkbd_cngetc() (fixes port-dreamcast/28711). 2007-07-16 12:46:03 +00:00
marcus
34fbc04104 Clear the correct amount of audio RAM in AICA driver (fixes port-dreamcast/35022). 2007-07-16 12:22:53 +00:00
isaki
d9167e130a Clean up header file. 2007-07-16 07:29:59 +00:00
he
3bb4d9ddb8 Regenerate configurations after adding COMPAT_40 to GENERIC.in. 2007-07-16 06:35:53 +00:00
he
13a06ef4a8 The COMPAT_AOUT_M68K option now requies COMPAT_40 after the changes 
to mount(), so define COMPAT_40 too.
 2007-07-16 06:26:54 +00:00
he
040448641c Regenerated files after adding COMPAT_40. 2007-07-16 06:23:58 +00:00
he
42a743dc4a The option COMPAT_AOUT_M68K now requires COMPAT_40 (recent mount changes), 
so make sure to define COMPAT_40 too.
 2007-07-16 06:23:28 +00:00
macallan
bd5e7f39e5 make the delay()s and spl*()s in bus_space*() functions optional 
( enable with options MACE_NEEDS_DELAYS )
mec apparently was the only device that really needed them and that's no
longer the case, at least on my O2.
 2007-07-15 23:27:55 +00:00
macallan
01a8ddf94a sprinkle delay() and mec_mii_wait() in MII access functions 
with this I no longer get spontaneous reboots during network traffic
 2007-07-15 23:24:47 +00:00
dsl
de11d360a1 Add the FILECORE filesystem so it gets test compiled. 2007-07-15 19:36:53 +00:00
dsl
af56a15a3c Remove "" from around option filename. 2007-07-15 11:56:04 +00:00
macallan
d292430251 add hardware cursor support 2007-07-15 05:42:13 +00:00
ad
9e6e490e2b Generic soft interrupts are mandatory. 2007-07-14 22:02:52 +00:00
ad
63811f5275 Generic soft interrupts are mandatory. 2007-07-14 21:48:17 +00:00
ad
adb899846a Revert accidentally commited change. 2007-07-14 21:46:36 +00:00
ad
ad84bcbe91 Remove ref to pdp10. Noted by yamt. 2007-07-14 21:00:40 +00:00
ad
45d83b0ef5 This has been stuffed since 2006-04-17 - unbreak it. 2007-07-14 17:23:58 +00:00
kiyohara
0f2145f0b3 Alphabetical reorder. 2007-07-14 16:16:17 +00:00
isaki
61a446b076 Make it compilable after the merge of some vmlocking branch. 
OK'ed by ad@.
 2007-07-14 14:13:21 +00:00
ad
d21a1f43dc - Restore com.c 1.45, par.c 1.31. They were trashed due to file system 
corruption on my machine.
- Adjust to match callout_init signature change.
 2007-07-14 10:46:25 +00:00
isaki
b9fb40c553 Declare candbtimer_ch earlier. 2007-07-14 09:50:04 +00:00
he
c36e288b04 Add a missing comma, and a missing return NULL if you fall off the 
end of plumpcmcia_event_alloc().
 2007-07-13 08:22:31 +00:00
kiyohara
8e5eb16fe0 There is a possibility that TLB-miss is caused in mq200_mdsetup(). 2007-07-13 04:03:55 +00:00
he
f39d28c4bf Adapt to the new signature of callout_init(). 2007-07-12 22:02:38 +00:00
he
c73835c125 Adapt to the new signature of callout_init(). 2007-07-12 20:39:01 +00:00
he
a82c668826 Adapt to new signature for callout_init(). 2007-07-12 20:33:02 +00:00
skrll
c2c660d00f Finish up the _lock_cas implementation by adding the RAS check. 
While I'm here make sure the owner gets set for mutex_enter if we get
interrupted between obtaining the interlock and setting the owner.
 2007-07-12 19:31:01 +00:00
skrll
7db093553a Like I said - don't futz with the trapframe in pmap_activate. Instead 
update the trapframe space and protection register in cpu_lwp_fork and
setregs.
 2007-07-12 14:15:36 +00:00
skrll
3a162b9c55 Initialise the callout before use. 2007-07-12 11:49:48 +00:00
he
fa013716c8 Convert event_thread from a proc to an lwp, so that the use of 
kthread_create() in sa11xx_pcic.c again matches with the prototype.
 2007-07-12 09:47:17 +00:00
kiyohara
5f9e0be38f s/Gumstix, Inc./Gumstix. Inc./ 
Sync to i386's GENERIC 1.837
 2007-07-12 02:37:47 +00:00
kiyohara
7e68bd52d9 This file supports Gumstix. Inc.'s gumstix. 2007-07-12 01:58:32 +00:00
he
3630eb1d24 Adapt to new signature for callout_init(). 2007-07-11 23:57:02 +00:00
uwe
1158881562 Grr, revert unrelated changes that are traces of an unfinished experiment. 2007-07-11 22:17:55 +00:00
uwe
2af9604e01 Catch up with callout_init signature change. 2007-07-11 22:15:41 +00:00
uwe
aa22c77a5b Fix previous: sc_event_thread should be a pointer to lwp_t. 2007-07-11 22:13:30 +00:00
dsl
3a1264a004 Add '0' to three more random callout_init() calls. 2007-07-11 21:57:28 +00:00
he
e8108f3e5b Remove a spurious brace. 2007-07-11 19:50:27 +00:00
he
ebf92f146a Adapt to the new signature of callout_init(). 2007-07-11 19:37:22 +00:00
he
150caa6d8c Adapt to new signature for callout_init(). 2007-07-11 19:13:23 +00:00
he
6cfc4427ed Move the declaration of repeat_ch up before its first use, 
so that this builds again.
 2007-07-11 18:59:18 +00:00
he
0b6cdc57be Bump SYMTAB_SPACE so that it fits again. 2007-07-11 16:18:39 +00:00
njoly
03ccc07594 Display RDTSCP bit on AMD processors (Read Serialized TSC Pair). 
ok by xtraeme
 2007-07-11 11:56:36 +00:00
kiyohara
05e3df98c5 Add nfsmb(4). 2007-07-11 08:01:50 +00:00
dogcow
1367780514 Looks like this was part of a botched merge; apply he@'s patch to make 
everything compile/link again.
 2007-07-11 04:07:08 +00:00
he
ff92ad95e2 Protect the initialization of xen_timepush_co with #ifdef DOM0OPS, 
like its declaration and other uses.
 2007-07-10 16:27:58 +00:00
he
bad1d53b6a Add an IPL_VM argument to comply with the new vmem_create() prototype, 
as per hint from ad@.
 2007-07-10 16:23:26 +00:00
nonaka
faf0f43647 Fix compile failure occured kthread API changes. 2007-07-10 13:55:20 +00:00
tsutsui
12b13f1335 Fix more pesudo. 2007-07-10 12:15:00 +00:00
xtraeme
44af43c54b Add wpi(4). 2007-07-10 11:23:21 +00:00
ad
88ab7da936 Merge some of the less invasive changes from the vmlocking branch: 
- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements
 2007-07-09 20:51:58 +00:00
bouyer
afc9294f4e Add a bootstatic callback for dom0, which allows to take NFS boot informations 
from the command line (nfsroot=, ip=) in the same way domUs did.
While there convert a for (;;) loop to TAILQ_FOREACH().
From Christoph Egger in private mail.

Make all bootstatic callbacks return the new NFS_BOOTSTATIC_NOSTATIC flag
when no nfs boot information is found on command line.
 2007-07-08 21:14:14 +00:00
ws
a3d6358dce Fix "Fix a garbage in previous commit." in 1.10. 
Hi, junyoung!
 2007-07-08 11:44:08 +00:00
pooka
0bc323f5cd Initialize the link context in a signal frame to the receiving lwp's 
link context instead of NULL.  Otherwise, if we got a signal while the
lwp had a link context set, the link context would be set to NULL upon
return from signal delivery.

christos@tech-kern: "I think you are right."
 2007-07-08 10:19:21 +00:00
scw
9e1f0b7978 Bump SYMTAB_SPACE 2007-07-08 10:05:13 +00:00
pooka
91f8a6cda5 RCS Id police 2007-07-08 09:54:50 +00:00
scw
61366a1bd8 Make this build again now that sched_whichqs no longer exists. 2007-07-08 08:37:25 +00:00
jmcneill
887c79e434 Enable gcscehci(4) 2007-07-08 03:01:19 +00:00
jmcneill
424b422ccd Add gcscehci(4), an EHCI glue driver for the AMD CS5536 companion 
controller. This driver enables USB 2.0 high speed support on the
Soekris net5501 (comBIOS ver. 1.31m).
 2007-07-08 01:13:26 +00:00
tsutsui
3682cc2396 - make this compile again 
- remove an unneeded include file

XXX: Avalon A12 devices are not in alpha/conf/GENERIC.
 2007-07-07 21:04:29 +00:00
tsutsui
69dbd2e45e Move x86 common cpu_counter functions into <x86/cpu_counter.h>. 2007-07-07 17:38:26 +00:00
christos
2c9f17f0fc Fix VIA Esther from Heron Gallegos 2007-07-07 17:35:18 +00:00
tsutsui
3a6c410df9 Don't update hz by the hardware RPB field. 
Leave it default HZ since we always use 1024Hz in alpha/mcclock.c.
 2007-07-07 17:08:15 +00:00
tsutsui
d4f1dfc2c4 Add "options HZ=1024" by default for alpha. 
All alpha machines use arch/alpha/alpha/mcclock.c for hardclock(9)
and it always sets MC_RATE_1024_Hz.
The default HZ value also affects tick and tickadj in conf/param.c.
 2007-07-07 17:03:28 +00:00
tsutsui
6a8a584d1a Remove leftover netns stuff. 2007-07-07 05:13:14 +00:00
martin
a698ff2d49 A few nits for 32bit SMP kernels, noticed by mrg. 
While there, simplify the fpstate IPIs a bit.
 2007-07-06 07:36:46 +00:00
martin
43ebaf27df sparc64_ipi_save_fpstate is not a leaf function - create a full grown 
stack frame.
 2007-07-05 20:21:11 +00:00
xtraeme
8957a299a2 - Adapt to envsys2. 
- Use one Indicator sensor for battery charge and set it to 1 or 0
  accordingly.
 2007-07-05 14:49:59 +00:00
isaki
265ca0b1f3 Add cache information of, at least, Celeron M (no processor number). 
Reviewed by tsutsui@ on port-i386.
 2007-07-05 13:51:37 +00:00
bouyer
7135c1aad2 Start with all sensors in ENVSYS_SINVALID state, and switch to ENVSYS_SVALID 
(or other, depending on result) once the sensor has been read.
This way envstat(8) won't show sensors which have not yet their correct value.
 2007-07-04 17:36:17 +00:00
xtraeme
6aec6c2829 s/cur_tre/edata/ => should fix the build error reported by rjs@. 2007-07-04 00:32:35 +00:00
xtraeme
d3cbee92fb ipmi_sensor_status: if state is ok return ENVSYS_SVALID and not 
ENVSYS_WARN_OK, the latter is deprecated on envsys2.
 2007-07-03 23:13:12 +00:00
briggs
fd02762f07 Add (commented-out) support for IPMI on amd64--pretty much copied straight 
from i386.
 2007-07-03 23:05:26 +00:00
briggs
c42d6c37ad Use PRIx64 for a 64-bit quantity instead of llx in a debug print. 2007-07-03 22:58:07 +00:00
christos
2d14013f2c Support for VIA Esther (From FreeBSD) 2007-07-03 17:07:54 +00:00
tsutsui
b3b62c0dff Don't clear BSS even if ksyms(4) is not configured. 2007-07-03 10:31:57 +00:00
tsutsui
de6d77dffd Use C99 initializers. 2007-07-03 10:11:49 +00:00
tsutsui
23e3926bbd No need to clear BSS if running kernel is loaded by the native bootloader. 2007-07-03 10:07:27 +00:00
tsutsui
c6f6d9c3c9 Print NEWS model name right after copyright and version 
as other mips ports.
 2007-07-03 09:55:30 +00:00
xtraeme
dcbea1b1dc - On Intrusion sensors, if something is not ok return a critical event. 
- On Power Supply sensors:
	* if power supply is not installed, return a critical event.
	* if power supply is installed but not powered on, return a
	  warnover event.
 2007-07-02 15:29:47 +00:00
jld
855403caf7 Correct regression in xen3 wall-time reading accidentally committed in r1.35. 
(Discussed on port-xen.)
 2007-07-02 06:26:35 +00:00
xtraeme
41fe337f40 Add support for the VIA C7-M and Eden processors in the 
Enhanced Speedstep driver.

Tested by Heron Gallegos <gallegos at csxxi dot net dot mx>
 2007-07-01 20:12:35 +00:00
xtraeme
19b01fa46a Use PSWITCH_EVENT_{PRESSED,RELEASED} for consistency with other drivers 
using the sysmon_power framework.
 2007-07-01 20:01:44 +00:00
toshii
df9ef2605f Quick and dirty fix for hpcarm build failure, using hpcarm specific 
softintr code for now.
 2007-07-01 16:04:57 +00:00
xtraeme
4ae780df6b Move gcscide(4) to INSTALL_LARGE. 2007-07-01 11:29:22 +00:00
xtraeme
2efc038f73 Enable gcscide(4). 2007-07-01 11:15:14 +00:00
xtraeme
da3f163bd2 Imported envsys 2, a brief description of the new features: 
(Part 2: drivers)

* Support for detachable sensors.
* Cleaned up the API for simplicity and efficiency.
* Ability to send capacity/critical/warning events to powerd(8).
* Adapted all the code to the new locking order.
* Compatibility with the old envsys API: the ENVSYS_GTREINFO
  and ENVSYS_GTREDATA ioctl(2)s are supported.
* Added support for a 'dictionary based communication channel' between
  sysmon_power(9) and powerd(8), that means there is no 32 bytes event
  size restriction anymore.
* Binary compatibility with old envstat(8) and powerd(8) via COMPAT_40.
* All drivers with the n^2 gtredata bug were fixed, PR kern/36226.

Tested by:

blymn: smsc(4).
bouyer: ipmi(4), mfi(4).
kefren: ug(4).
njoly: viaenv(4), adt7463.c.
riz: owtemp(4).
xtraeme: acpiacad(4), acpibat(4), acpitz(4), aiboost(4), it(4), lm(4).
 2007-07-01 07:37:12 +00:00
jmcneill
1a446b6f11 Move ral@pci from INSTALL to INSTALL_LARGE, as requested by Manuel Bouyer. 2007-06-30 12:30:34 +00:00
jmcneill
cbc58c76f0 Enable ral@pci by default in the INSTALL kernel. 2007-06-30 03:11:03 +00:00
rumble
6912898dc8 Add read-only support for SGI's Extent File System. 
Reviewed by pooka@.
 2007-06-29 23:30:16 +00:00
tsutsui
446e7b7d8c __FUNCTION__ -> __func__ 2007-06-29 16:41:31 +00:00
tsutsui
9905ba64da Add options COMPAT_40. 2007-06-29 16:23:29 +00:00
tsutsui
bad1146f72 Fix botch in cnpollc(9) for keyboard devices. 2007-06-29 16:09:04 +00:00
tsutsui
1681c91ced We have to use _splnone() instead of spl0() to enable interrupts on mips 
because spl0() is an alias of _spllower(0) and it doesn't set MIPS_SR_INT_IE.

Fixes ews4800mips kernel getting stuck in cpu_idle(9)
after cpu_configure(9) since yamt-idlelwp merge.
(previous mips_idle() implicitly set MIPS_SR_INT_IE
 but new cpu_idle(9) for mips no longer does it)
 2007-06-29 16:03:04 +00:00
xtraeme
5cd016ef40 For 586 based National Semiconductor CPUs, use amd_cpu_cacheinfo to 
print cache information.

My AMD Geode GX2 500/1W reports good values:

cpu0 at mainbus0: (uniprocessor)
cpu0: National Semiconductor Geode (586-class), 365.26 MHz, id 0x552
cpu0: features 80a13d<FPU,DE,PSE,TSC,MSR,CX8>
cpu0: features 80a13d<PGE,CMOV,MMX>
cpu0: "Geode(TM) Integrated Processor by National Semi"
cpu0: I-cache 16 KB 32B/line 4-way, D-cache 16 KB 32B/line 4-way
cpu0: ITLB 8 4 KB entries fully associative
cpu0: DTLB 8 4 KB entries fully associative
 2007-06-28 21:02:40 +00:00
bouyer
69219b1efe copy from the right offset when handling root= on the command line. 
Pointed out by Pavel Cahyna.
 2007-06-28 20:54:47 +00:00
bouyer
6be5ba1e96 Make sure to check for pending interrupts on the final STI. For this 
reorganise the IPL vectors a bit so that they can be used from more
places than splllower and doreti.
Include patch from Kazushi Marukawa (fixed to handle pending interrupts),
which should fix the read_psl() == 0 assertion failure reported by
several users.
 2007-06-28 20:50:30 +00:00
alc
c82e20912d - add ichlpcib to XEN2 config file 
- few cosmetics

tested with XEN2 and XEN3 kernel
ok bouyer@
 2007-06-28 20:07:00 +00:00
nonaka
e02b41111a cpu_reboot(): 
- fix data write-back to hard disk is failed when reboot/halt.
 2007-06-28 16:07:12 +00:00
nonaka
bc90321cbe Use tpcalib_trans() instead of own function. 2007-06-28 15:44:01 +00:00