11ea6e42ce
conf_init() is no more.
1995-04-10 05:44:08 +00:00
3ba14de085
Rewrite in new format.
1995-04-10 05:42:34 +00:00
e957423a98
Move fp_except and fp_rnd definitions to machine/include.
1995-04-09 21:31:58 +00:00
f0dfc32ae9
Sync with net/if.h changes
1995-04-09 05:08:33 +00:00
bcd63a2a47
minor cleanup
1995-04-08 04:46:59 +00:00
f41996e453
Do the I-cache flush in the process switch asm code like other ports.
1995-04-08 04:45:43 +00:00
cf2da3de32
Add some comments.
1995-04-08 04:43:44 +00:00
cd79987a43
Fix CGTWO entry
1995-04-08 04:42:01 +00:00
a1cc6d49c4
It works now.
1995-04-08 04:40:27 +00:00
d0d1dc8a4e
Fix that probe sequence right this time.
1995-04-07 05:19:24 +00:00
8f584db874
Flush the cache when switching contexts (thanks to David Jones).
1995-04-07 04:46:41 +00:00
e7f174fdcd
Make sure curproc->p_addr is set before any faults occur.
...
Put msgbuf in page zero (so its location is predictable).
1995-04-07 04:44:55 +00:00
9a714f87a8
Use vmempage for /dev/mem like other ports.
1995-04-07 04:44:26 +00:00
e1b1e78d15
Allocate vmempage here (for use by /dev/mem).
1995-04-07 04:44:05 +00:00
d047cd4b9c
Announce unexpected interrupts once for each level.
1995-04-07 04:43:25 +00:00
88f350e128
Make sure we do not enable clock interrupts too early.
1995-04-07 04:30:13 +00:00
4a3ad0819d
Look out for null curproc->p_addr which caused a "Watchdog Reset!"
...
when the recursive faults overflowed the stack. (from David Jones)
1995-04-07 03:12:51 +00:00
e5a537ade9
Fix to make "root on sd2a" work (from David Jones)
1995-04-07 03:02:36 +00:00
65725c0bce
Add cgtwo
1995-04-07 02:58:49 +00:00
522e7f03d7
First cut at cgtwo driver. Not quite right yet.
1995-04-07 02:54:43 +00:00
0af73b6fc3
Changed fb_attach() interface
1995-04-07 02:51:21 +00:00
19c01b5c67
Corrected fbtype->fb_size (which x11r6 is picky about...)
1995-04-07 02:47:40 +00:00
637b45dbf2
Changed fb_attach() interface, hack probe to avoid impossible matches.
...
Corrected fbtype->fb_size (which x11r6 is picky about...)
1995-04-07 02:43:25 +00:00
d49502cfa2
Changed fb_attach() interface
1995-04-07 02:37:18 +00:00
1feb639a38
First cut at cgtwo driver. Not quite right yet.
1995-04-07 02:31:45 +00:00
63052ecc6e
Add cgtwo
1995-04-07 02:23:03 +00:00
3de4f27e67
Yet more SYMTAB_SPACE
1995-04-07 02:18:39 +00:00
aa197980e8
Use new core file format in cpu_coredump()
1995-04-03 22:06:11 +00:00
69aa4da3cf
Add FDESC FIFO NULLFS UNION file systems.
1995-04-03 22:03:38 +00:00
8ba211cde0
Added #define _KERNEL
1995-03-28 18:37:27 +00:00
71ab4ed9dc
KERNEL -> _KERNEL
1995-03-28 18:13:48 +00:00
cd70d15138
Avoid passing non-ASCII chars to the PROM.
...
Call PROM at splsoftclock to avoid missing interrupts.
1995-03-28 16:09:41 +00:00
0f9e8841da
Add _KERNEL
1995-03-28 04:30:45 +00:00
507979e02b
Increase ZSSOFT_PRI from 2 -> 3 now that tty ipl is 4.
1995-03-27 01:25:20 +00:00
b7982644e1
Put spltty at level 4 and adjust others appropriately.
...
Also, improve the inline _spl() and move it to psl.h
1995-03-27 01:22:50 +00:00
4916e6ce63
Nuke FPCOPROC
1995-03-26 19:38:23 +00:00
56bbea071a
Add/fix some comments.
1995-03-26 19:37:41 +00:00
1a22be7490
Enable interrupts before autoconfig.
...
The reasons to do it later no longer exist.
1995-03-26 19:32:31 +00:00
58662ede8c
Enable interrupts before autoconfig (like hp300).
...
The reasons to do it later no longer exist.
1995-03-26 19:28:29 +00:00
42d319670c
Add "tun" device.
1995-03-26 19:24:16 +00:00
e2d6f35c09
The splbio()/splx(s) is now appropriate even if SCSI_POLL is set.
1995-03-26 19:23:12 +00:00
0784cab36b
Sync with GENERIC
1995-03-26 19:17:48 +00:00
3316bb27fd
Nuke FPCOPROC
1995-03-26 19:17:23 +00:00
fa2133533d
invoke ktrsyscall with (vp, code, argsize, args) as args.
1995-03-26 08:03:29 +00:00
a1bc8a65f8
Cleanup, and deal with kd != console
1995-03-24 19:48:41 +00:00
d57d569cad
Determine FPU type during autoconfig.
1995-03-24 17:27:12 +00:00
d57fb977bb
Added -D_KERNEL to CFLAGS alongside each -DKERNEL.
1995-03-23 19:46:18 +00:00
04446336b4
correct the file name
1995-03-21 16:22:25 +00:00
a59d9a4ac9
This is needed by the Xsun servers.
1995-03-21 16:19:57 +00:00
b8403d582b
don't be so quick to cast macro arguments to unsigned; if the
...
arguments are really off_t's (e.g. to btodb), then you can lose
information. This was the "> 4G file systems don't work" bug; physio
uses btodb, which was broken.
1995-03-18 07:22:43 +00:00
mycroft