Commit Graph

7007 Commits

Author SHA1 Message Date
jmcneill
6df1267b6f Enable interrupts in the error path also. 2006-06-19 02:33:19 +00:00
jmcneill
dda9d480db Don't try to fixup PCI interrupt routing here; we're going to use ACPI to
do this for us instead. This prevents us from having to write chipset
specific support -- yay!
2006-06-19 02:30:35 +00:00
nonaka
3b9b5a75bb Added 'Low Voltage Intel Pentium M processor 1.30 GHz' entry. 2006-06-18 16:39:56 +00:00
xtraeme
0b27695a68 Fix AMD 766 PMC entry. 2006-06-18 10:34:34 +00:00
xtraeme
9552b18ed1 Match the following products:
* SIS 962 and SIS 963
* AMD 766 and AMD 768
* ALI M1533

... syncing the code with openbsd.
2006-06-18 09:57:37 +00:00
xtraeme
c8a267e9c8 Sync with openbsd and apply fix from mickey@openbsd rev 1.5:
"It seems that via has 7, not 4 pci interrupt lines.
this fixes unmapped interrupts on some via-based boards."
2006-06-18 09:55:11 +00:00
christos
99d15da99a Don't dick with cr3. It pisses pmap.c off. 2006-06-17 20:14:26 +00:00
mrg
911a6956b3 add another GCC4 uninitialised variable warning. it only shows with -O3. 2006-06-17 19:00:03 +00:00
jmcneill
e1b3711024 More ACPI wakecode changes:
* Enable VBIOS hack by default
	* Beep while calling VBIOS, for debugging.
	* Fix a DIAGNOSTIC error in the sleep path.
	* Remove the erroneous mutex not locked error message.
2006-06-17 17:11:53 +00:00
jmcneill
6d05bd66ec Add powerhook to restore IRQ routing on resume. 2006-06-16 23:04:25 +00:00
jmcneill
2689ddd4e7 Enable the piixpcib(4) device driver by default. This enabled speedstep as
well as sufficient powerhooks to resume from ACPI S3 suspend.
2006-06-16 22:13:35 +00:00
jmcneill
686aef2890 Restore standard PCI configuration registers as well as PIRQX route control
registers on the PIIX4. This fixes up interrupt routing, and makes ACPI S3
suspend/resume work on my Sony Vaio.
2006-06-16 22:10:36 +00:00
simonb
8323964d84 Move ksyms to the "miscellaneous pseudo-devices" - it's not a wscons
thing.
2006-06-16 06:41:13 +00:00
jmcneill
d651f09fd7 Regen. 2006-06-15 18:09:14 +00:00
jmcneill
fe984d2ec9 Do not restore task registers in wakecode. 2006-06-15 18:07:58 +00:00
perry
6b3d64621c edit a couple of new items so that they fit on a line and do the tab
to space dance so that they can be cleanly commented out.
2006-06-13 17:17:15 +00:00
dogcow
b6d29ce0bd Add -Wno-attributes for amd64, too. 2006-06-12 19:25:20 +00:00
perry
639e2af40a aprint_normal's for the "(slowidle)" and bios management messages 2006-06-12 16:42:05 +00:00
perry
42d7bd62c9 Always print out if slowidle, bios mgmt, etc. are on. It is better
that users have more information rather than less to tell us when we
are asking them for dmesgs for debugging purposes
2006-06-12 16:39:38 +00:00
perry
34440af2fc ANSIfy function decls 2006-06-12 15:56:04 +00:00
hubertf
b6eda27950 Rename XEN kernel config files to be more consistent:
INSTALL_XEN3_U        -> INSTALL_XEN3_DOMU
  INSTALL_XENU          -> INSTALL_XEN2_DOMU
  XEN0                  -> XEN2_DOM0
  XEN3_U                -> XEN3_DOMU
  XENU                  -> XEN2_DOMU

OK'd by bouyer@
2006-06-08 22:17:07 +00:00
kardel
09b51ec920 convert to timecounters (from branch simonb-timecounters) 2006-06-07 22:37:14 +00:00
christos
74653dfe15 Make this work with objdirs. 2006-06-06 15:41:50 +00:00
perry
5da686584b convert a tab to a space so commenting out of piixpm is cleaner 2006-06-04 15:48:08 +00:00
perry
18e9385d27 Wrap the UPAGES definition with an #ifndef so that (in emergencies) it
can be put into a kernel config file.

Stack overflows should be fixed, not patched around, but this makes it
easier to test if one is causing a panic.
2006-06-04 15:41:53 +00:00
bouyer
19d40b6b73 -mcpu=i686 is implied by -march=i686 and gcc4 doens't like -mcpu=i686 so
remove it. From Sergey Lungu on current-users.
2006-06-01 19:39:59 +00:00
rpaulo
c1ee12dd17 Don't check for SSE/SSE2 twice. 2006-05-29 17:35:41 +00:00
pavel
ad560fa074 Give Xen its own linker script, which does not set the physical load address
specially. In Xen, the system is already in virtual mode when the kernel is
started, so the old behavior of setting physical address to virtual address
was actually correct, and the previous change made NetBSD/Xen kernels
unbootable.

Pointed out, explained and tested by <bouyer>.
2006-05-28 14:00:13 +00:00
pavel
3b5ae171d2 New linker scripts for the i386 kernel. The main improvement is setting the
physical load address correctly. (Before, it was set to the same very high
value as the virtual address.)

Provided by Jason Thorpe in
http://mail-index.netbsd.org/port-i386/2006/02/10/0008.html
with some modifications by me, mainly to make the output more similar to
the current one, and to not hardwire the physical and virtual addresses.

See also:
http://mail-index.netbsd.org/port-i386/2006/05/19/0001.html

Now an unpatched grub can load a NetBSD Multiboot kernel.
2006-05-26 09:22:42 +00:00
yamt
7445825948 i386 and xen: move struct user to the highest address in uarea
and reduce UPAGES from 4 to 3 in the case of !NOREDZONE.
2006-05-22 13:44:53 +00:00
cube
5536c0b46c Fix tyop reported by Yoshito Komatsu. 2006-05-21 11:39:27 +00:00
gdt
42ecd6e67d Enable ugensa. 2006-05-20 22:03:56 +00:00
christos
d05d5214fe Add another fixup entry Arthur Dimitrelis 2006-05-20 20:32:16 +00:00
dogcow
e04236d6b3 remove bitrotted functions that have never been used, and have been
cargo-culted from one port to another.
2006-05-15 12:47:42 +00:00
christos
b70f726885 XXX: GCC uninitialized 2006-05-15 00:36:55 +00:00
elad
8ccb6c9341 integrate kauth. 2006-05-14 21:55:09 +00:00
christos
103d2f520c XXX: GCC uninitialized. 2006-05-14 05:30:31 +00:00
lukem
c75cb560f1 support MAKEVERBOSE 2006-05-13 12:02:54 +00:00
lukem
e7526fa7ea support MAKEVERBOSE 2006-05-13 09:14:14 +00:00
lukem
261d50393a Convert to using CC instead of LD, as LDFLAGS is for CC not LD. 2006-05-13 09:13:11 +00:00
xtraeme
33a96be9f2 Driver for iTE IT87xxF and compatibles hardware monitor. Add
commented out entries on GENERIC and GENERIC_LAPTOP.

Adapted from OpenBSD.
2006-05-13 09:02:18 +00:00
skrll
214053461e Fix some more comments. 2006-05-13 08:56:08 +00:00
lukem
f2e1749b26 support MAKEVERBOSE 2006-05-13 04:12:33 +00:00
lukem
dff2a78d7a Convert to using CC instead of LD, as LDFLAGS is for CC not LD. 2006-05-13 04:11:04 +00:00
lukem
176b81e4ff Convert to using CC instead of LD, as LDFLAGS is for CC not LD. 2006-05-13 03:37:47 +00:00
lukem
032741abf4 support MAKEVERBOSE 2006-05-13 03:19:28 +00:00
lukem
a6f0ba44b5 When creating BASE.rom, don't use BASE.bin as a temporary filename,
since the latter is a valid target name.
2006-05-13 02:47:03 +00:00
lukem
667bf36cad support MAKEVERBOSE 2006-05-13 02:33:40 +00:00
dogcow
dcc4a379f8 as mrg@ points out, including <bsd.own.mk> is the proper fix. 2006-05-12 16:45:12 +00:00
dogcow
c242ce750a deal sanely(?) with when HAVE_GCC is undefined. 2006-05-12 06:43:55 +00:00