Commit Graph

32070 Commits

Author SHA1 Message Date
bjh21
7fa10fa259 Rather than explicitly masking and sign-extending 16-bit integers, use
u_int16_t and int16_t for the X and Y count registers.  GCC produces better
code this way.
2002-06-20 19:33:36 +00:00
bjh21
c62bc841d0 Rather than explicitly masking and sign-extending 16-bit integers, use
u_int16_t and int16_t for the X and Y count registers.  GCC produces better
code this way.

Also, initialise the stored state in wsqms_enable(), so that the mouse doesn't
warp to a random position on open.
2002-06-20 19:33:20 +00:00
eeh
5c6bf5ad03 Overhaul streaming buffer handling:
sysio streaming buffer flushes write a single 8-byte aligned 8-byte
	value.

	psycho streaming buffer flushes write a 64-byte aligned 64-byte block.

So separate out the streaming buffers into their own structure and pass those
in to all iommu operations.  This also means we only flush the correct
streaming buffer for psycho rather than needing to flush both just in case.
2002-06-20 18:26:23 +00:00
christos
c5ccbb71e5 remove syscall debugging code; thanks eduardo for fixing the code! looks
much nicer now.
2002-06-20 00:55:39 +00:00
bjh21
4244560bbc More wsqms cleanups:
sc_flags was never read.  G/C it.
wsqms_attach() took two arguments that pointed to the same structure.  G/C one
  of them
Since wsqms controls the same device as qms, have it match the same attach
  args.
2002-06-19 23:49:14 +00:00
eeh
dcabee4011 Get ktrace working once more. 2002-06-19 23:32:01 +00:00
bjh21
06eecc6ca0 $Id$ -> $NetBSD$ (oops). 2002-06-19 23:27:48 +00:00
bjh21
cccb2d97a7 ANSIfy, un-__P, clean comments a little. 2002-06-19 23:12:14 +00:00
bjh21
26f6ad6038 Un-__P. Clean up comments. 2002-06-19 23:05:07 +00:00
bjh21
f793a16211 Substantial overhaul of the wsqms driver:
Use a callout rather than hanging off the VSYNC interrupt.
Don't emit WSMOUSE_INPUT_ABSOLUTE events, since this isn't an absolute device.
Handle counter wrap-around sensibly, rather than limiting counts.
Don't gratuitously copy sc->sc_dev onto itself at attach time.
2002-06-19 23:02:58 +00:00
bjh21
a7f527777e Move over to using a 6:5:5 R:G:B palette in 16-bit display modes, and abstract
the palette generation to work with arbitrary numbers of bits.
This allows X to work after a fashion, since it tries to put the VIDC into
a 6:5:5 mode itself (which we ignore).  Anything that actually tries to take
advantage of the DirectColor visual it offers will still be screwed, but I
hope such applications are rare.
2002-06-19 22:42:02 +00:00
briggs
029a7e8e9a - The 8240 and 8245 have a 603 core -- add them to the 603 lists...
- Use atop(dcache_size / assoc) for uvm_page_recolor(ing) instead
  of  atop(dcache_line_size / assoc).
2002-06-19 17:43:30 +00:00
briggs
d6cfd2cc8c Include the Processor ID for the MPC8245. 2002-06-19 17:39:26 +00:00
wrstuden
2eb3dc82d8 Fix recent bugs seen on Performa 4400 macppc's by
Makoto Fujiwara <makoto@ki.nu> and Manuel Bouyer <bouyer@netbsd.org>.
Help from Allen Briggs, Jason Thorpe, and Matt Thomas.

We need to call cpu_cache_probe() early in boot (machdep.c).
Add 603 info for completeness, and use NBPG not PAGESIZE, as the
latter relies on uvm being setup (cpu_subr.c).
Let uvm_page_recolor() be called before uvm has been set up; just
note the page coloring value (uvm_page.c).
2002-06-19 17:01:18 +00:00
agc
254fefb729 Fix a typo introduced in the last commit. 2002-06-19 10:13:43 +00:00
grant
30dfd0761f remove extraneous #endif from last commit, make this compile again. 2002-06-19 04:58:32 +00:00
itojun
24bd6d41d9 add netdock ether driver. as discussed on port-mac68k 2002-06-19 04:30:46 +00:00
itojun
decb8a7bae add netdock driver. 2002-06-19 03:48:31 +00:00
itojun
4c1feea9d8 correct writedisklabel() on MBR-partitioned disk.
while here, correct labelsector/offset mismatch in writedisklabel,
when readdisklabel() have found the label at nonstandard location
(due to magic number search).  a lot of other ports have the same problem.
2002-06-19 03:10:56 +00:00
itojun
ef879bfb42 netdock* - ethernet driver for Asante NetDock/Newer Ether MicroDock.
by Daishi Kato <daishi@axlight.com>
test/stabilization by Takeo Kuwata <tkuwata@virus.kyoto-u.ac.jp>
2002-06-19 01:54:26 +00:00
jdolecek
6710074860 update comment 2002-06-18 20:23:08 +00:00
thorpej
ae0f07dd51 Fix syscall_intern after systrace (HI CHRISTOS!) 2002-06-18 18:30:15 +00:00
lukem
a0d38da2c9 Regenerate with new m4 mechanism (vs old cpp mechanism).
New files are same as old (except for whitespace differences), except
that the new GENERIC doesn't have melody, because GENERIC defines
DRACO and the latter can't have it.
2002-06-18 17:45:13 +00:00
lukem
c13ae5b7a9 Use m4 instead of cpp to generate the various kernel configurations,
since cpp is the "C compiler pre-processor", and as such, isn't really
suited to parsing non-C-language files...
2002-06-18 17:39:39 +00:00
itojun
463d764d56 write/readdisklabel are not consistent in MBR-partitioned case;
forbid writes for now (real fix should follow).
2002-06-18 17:25:21 +00:00
christos
0b3fc0988b Regen. 2002-06-18 17:22:10 +00:00
christos
88612aefe1 ${CC} 2002-06-18 17:21:53 +00:00
christos
48b4953ba1 simplify: use hexdump to produce the output in the desired form instead
of massaging it with sed.
2002-06-18 17:21:33 +00:00
wiz
a266aad8d4 Fix typo in comment. 2002-06-18 12:35:09 +00:00
drochner
8d0aee4e8d make it compile with ACPI_DEBUG
(removing the half-done debugging support, sorry)
2002-06-18 10:32:02 +00:00
tron
cee165577f Fix typo which caused a build failure for kernels with diagnostic checks. 2002-06-18 09:56:33 +00:00
fvdl
0b5350133b Add MTRR support. 2002-06-18 08:35:14 +00:00
fvdl
7a777905af Make this compile again. 2002-06-18 08:34:57 +00:00
fvdl
ff5489fd50 MTRR includes. 2002-06-18 08:31:42 +00:00
fvdl
4622ad6f14 MTRR support for x86_64. Will be shared with i386 later. 2002-06-18 08:30:33 +00:00
tshiozak
2d5fc6d0a9 add MD support for ACPI sleep/wakeup. 2002-06-18 07:56:12 +00:00
tshiozak
0c0fe6076b use AcpiHwRegister{Read,Write} functions instead of (obsoleted)
AcpiHwBitRegister{Read,Write}.
2002-06-18 07:53:05 +00:00
tshiozak
c6d5efc480 auto-generated code from acpi_wakecode.S.
standard update process of this file :
  1. fix acpi_wakecode.S .
  2. commit acpi_wakecode.S .
  3. make -f Makefile.wakecode
  4. commit acpi_wakecode.h .
2002-06-18 07:41:15 +00:00
tshiozak
4ba1b15b75 add a code to embed RCS tag of acpi_wakecode.S in acpi_wakecode.h . 2002-06-18 07:37:37 +00:00
tshiozak
834397e5a9 ACPI wake-up code for i386. but, we have no interface to use it efficiently. 2002-06-18 07:25:12 +00:00
itojun
524e044ee5 apm emulation, from openbsd 2002-06-18 05:22:50 +00:00
itojun
ddd8d90dc2 KNF (whitespace) 2002-06-18 04:35:02 +00:00
itojun
b00d5fb87a extern decl for adbHardware 2002-06-18 04:33:36 +00:00
itojun
7eba810efa s/options\t\t/options \t/ 2002-06-18 03:06:58 +00:00
christos
1fb8545c60 Copyright and comment clarification from Masanori-san. 2002-06-18 02:42:16 +00:00
itojun
0c74b798cb make macppc install floppy IPv6 ready
(we have two floppies anyways, and there's plenty of space in 2nd floppy)
2002-06-18 02:02:20 +00:00
itojun
73941c5096 avoid hardcoding MBR_MAGICOFF and MBR_MAGIC. 2002-06-18 00:37:25 +00:00
itojun
980d7c811b LABELSECTOR is redefined to be 0 on macppc, so it is not usable for MBR
disk label offset.  use hardcoded "1" for now, i guess we need a new #define
in sys/disklabel_mbr.h.  fixes ofwboot to boot kernel off from MBR-formatted
disk.
2002-06-17 23:59:01 +00:00
simonb
4cf248e04b Fix tyop. 2002-06-17 22:48:45 +00:00
christos
d0497fdbca glue for MD acpi devices. 2002-06-17 22:38:31 +00:00