Commit Graph

23677 Commits

Author SHA1 Message Date
tsubai
2e72bb2b72 Add fr key layout. (From Manuel Bouyer, modified by me) 2000-09-01 16:00:38 +00:00
tsubai
a964ecd1a9 Correct iosize. 2000-09-01 10:43:33 +00:00
eeh
67ca76fd68 Use casx instead of stx to change entries in the page table trees. 2000-08-31 20:14:55 +00:00
eeh
65f1bb68b5 Add support for multiple memory segments. 2000-08-31 19:12:45 +00:00
eeh
a17d71cd4b Fix register sheduling bug in st?a() routines and add casxa(). 2000-08-31 19:08:47 +00:00
pk
710295327e Scheduler lock update. 2000-08-31 16:59:12 +00:00
jdolecek
3fd457414e constify first arg to parseopts() 2000-08-31 07:21:46 +00:00
jhawk
a022cf9d37 Use
${MAKE}
instead of
  make
2000-08-30 23:51:46 +00:00
jeffs
f8d9f59ed9 Correct _KERN_MCOUNT restoration of $t9.
From Ethan Solomita (ethan@geocast.com).
2000-08-30 22:31:12 +00:00
is
0a7489b1e8 Enlarge RAMDISK to make it fit the installation fs again.
By Mike Pumford per port-arm32 mailing list..
2000-08-30 21:29:16 +00:00
tsubai
c6b969cd89 Cosmetic changes. 2000-08-30 19:16:15 +00:00
tsubai
f286397ff1 Scheduler locking changes. 2000-08-30 18:41:16 +00:00
jdolecek
ad4cbd94a4 Switch to newvers_stand.sh-based boot block versioning. Add sensible 'version'
template, the boot block changes were picked from cvs annotates for Makefile
and boot.c.

Reviewed by Ragge.
2000-08-30 18:16:09 +00:00
soren
55dbdb6091 Typo. 2000-08-30 14:59:48 +00:00
tsubai
c6615ed380 Initialize pv_head locks. 2000-08-30 09:55:28 +00:00
is
9ec45c470a Use an INSTALL kernel, instead of GENERIC. We want this to be as small as
possible so that it can boot even on low memory Amiga machines.
2000-08-30 09:43:04 +00:00
is
e83ee64031 Stripped down Install kernels (no audio, no OSI or Xerox networking, no
scanners).
INSTWDCPCM has the wdc instead of the idesc, and the PCMCIA support, so
that people can test it easily.
2000-08-29 21:09:28 +00:00
leo
6c521b3fcc Now that formatting works (Thanks Nathan!), it shows that spec_read() is
issueing a 'DIOCGPART' at every read. Avoid re-reading the label at
every format-verify (or read/write on the block-device) by fetching the label
at open-time and caching it until we close.
2000-08-29 20:00:16 +00:00
takemura
212b3fa9d8 The bootloader has been improved by Kazuma Arino.
- dialog style main window
  - support to read compressed kernel image
  - automatic boot with timer
  - human readable configrarion file
version no gets '1.15.00 2000.08.29'
2000-08-29 15:10:14 +00:00
wdk
35871e7996 Report memory sizes using format_bytes(9) kernel function 2000-08-29 12:12:59 +00:00
wdk
149a5fc5f2 Engage the clutch before changing gears.
If the machine uses a Z85230 ESCC device with deep buffers, we observe
output glitches when printing the zstty{0,1} probe lines when the device
is reset & reprogrammed during startup.

There is no easy 'hook' in the MI Z8530 driver, so we wait for output
buffer to drain before changing the baud rate generator prescaler value.
2000-08-29 11:25:08 +00:00
wdk
5656bdea79 Don't trust the realtime clock if the data is before 1988. This is
consistant with other ports.

Prevent uninitialized time from being written back to the RTC (1-Jan-1970)
if the machine is halted from the root device name prompt
(bootflags & RB_ASKNAME)
2000-08-29 10:34:13 +00:00
wdk
aaa3f11967 Fix an obscure bug with dump corrupting files during raw reads.
If the DMA chaning interrupt couldn't be serviced immediately (higher spl
level when kernel doing something else) a few microseconds later the NCR
controller will fill its FIFO and also interrupt the CPU.

The SCSI interrupt sees the terminal count has been reached, calls
asc_dma_intr to finish the job off.   The FIFO cannot be flushed because
the block count hasn't been setup for the last dma segment (DMA chaining
still wasn't serviced).

Since the NCR 53c94 FIFO is only 16 bytes in size, any short DMA in this
size combined with the machine 'doing something else' causes the problem
to occur.

Servicing the DMA chaining interrupt before the NCR SCSI interrupt solves
this problem.

Add tests to ensure the DMA FIFO has been flushed correctly at the end of
each DMA operation just to be on the safe side.
2000-08-29 08:24:06 +00:00
pk
55af96281e Pull down from 1.4 branch: default relocation address = 3E0000 2000-08-28 19:10:31 +00:00
eeh
bbede65f5b Fix E250 identification. 2000-08-28 18:01:23 +00:00
tsubai
f671e87235 Add an argument to need_resched macro. 2000-08-28 11:52:59 +00:00
tsubai
077c9ca312 Enable hash filter mode. 2000-08-28 11:27:44 +00:00
sato
2a9cfc2e99 add pwctl PWCTL_LCD, PWCTL_LCDLIGHT for NEC_MCR_FORDOCOMO, NEC_MCR_500. 2000-08-27 22:56:31 +00:00
matt
4027af140c Make the pmap: enter on myself panic conditional under DIAGNOSTIC 2000-08-27 17:14:44 +00:00
ragge
fa53886b69 Wiring of a page were done at the wrong place, so pages in kernel space
never got the wired bit set. This caused panics if a swapped out process
was swapped in again and the kernel stack had not yet been unmapped.
While here, add a forgotten lock initializer.
2000-08-27 14:14:50 +00:00
is
115bd3b350 Replace one callout handler by using a timeout in tsleep().
1.4.x version of it submitted by Ingolf Koch; translated to 1.5-ish by
myself.
2000-08-27 10:35:43 +00:00
matt
98e5e148fa Since cpu_switch (aka Swtch) is now called at splsched() with sched_lock
locked, cpu_exit needs to do that too.  Since in the lock debug case we
have to use a CALLS which wipes out R0-R6, change the convention for Swtch
so that the proc is passed in R6 and that R6 is clobbered.  This is so
Swtch itself doesn't have to save/restore the proc pointer explicitly.
2000-08-27 00:21:46 +00:00
minoura
cbac1a91d9 Add #ifdef _KERNEL ~ #endif to protect kernel variables from userland. 2000-08-26 23:23:43 +00:00
minoura
33f305090e Set fputype earlier, before proc0 initialization. 2000-08-26 23:04:44 +00:00
eeh
a0f2ab32f7 Make 64-bit clean. 2000-08-26 16:02:06 +00:00
matt
5c62e0cab4 Fix the idle loop. 2000-08-26 15:13:23 +00:00
thorpej
9eed6192a2 Un-transpose some characters in an abbreviation. 2000-08-26 06:58:25 +00:00
matt
7e0ad8b0c2 Update to reflect new processor switching conventions w/ LOCKDEBUG
and MULTIPROCESSOR
2000-08-26 03:38:46 +00:00
thorpej
0ba350214f Thinko in previous. 2000-08-26 03:33:49 +00:00
thorpej
2af5e3861c Fix a couple of typos in previous. 2000-08-26 03:32:36 +00:00
thorpej
0e04909346 Snapshot of TLB shootdown bugfixes. 2000-08-26 03:27:44 +00:00
matt
fc1511e910 add a COMPAT_15 and remove TCP_COMPAT_42 2000-08-26 02:32:57 +00:00
matt
3fec8167d3 Reverse polarity on INSN_EMULATION to NO_INSN_EMULATION. Thus by default
all kernels will have instruction emulation.
2000-08-26 02:30:59 +00:00
bjh21
7c9498f7cb Move setsoftast() and friends into cpu.h and make them into macros, so arm26
is the same as every other port (except arm32).
2000-08-25 16:43:46 +00:00
tsubai
308d7eb3e4 Add (experimental) awacs audio driver. Output (play) only. 2000-08-25 12:44:42 +00:00
bjh21
36f4367186 <machine/spl.h> -> <machine/intr.h> (missed this one last time) 2000-08-25 10:37:58 +00:00
pk
baa6bfe332 The `fb' device is optional. 2000-08-25 08:12:49 +00:00
matt
3351f4cf4b Need <bsd.own.mk> for OBJECT_FMT. 2000-08-25 03:51:34 +00:00
thorpej
4db6fc7542 Make need_resched() take a "struct cpu_info *" argument. This
causes gives a primitive form of processor affinity.  Its use in
roundrobin() still needs some work.
2000-08-25 01:04:06 +00:00
eeh
61197aa4f0 Add workaround for E250s. 2000-08-24 22:00:04 +00:00