Commit Graph

22734 Commits

Author SHA1 Message Date
mark
f24b0a3f91 Initial commit of the header file defining the architecture specific
syscalls.
Declare ARM32_SYNC_ICACHE call that ensures that main memory and any
instruction cache are in sync.
1996-10-14 22:47:31 +00:00
mark
135f110149 Added an ioctl acknowledgement bit to the buffer record flags field.
Added a new ioctl to select either relative or absolution mouse position
reports.
1996-10-14 22:41:59 +00:00
mark
44faa9154f Added a blanktime field to the vconsole structure so the blanktime time
is managed per virtual console rather than globally.
1996-10-14 22:35:03 +00:00
mark
c4d76a3f02 Added a field for cpu specific data in the cpu_t structure. 1996-10-14 22:32:45 +00:00
mark
60c2014d86 Added magic number of bootconfig structure passed from the bootloader.
Added a field for the physical address of the display memory used
by the bootloader if VRAM is not fitted.
Declare the boot_args variable extern here.
1996-10-14 22:30:19 +00:00
mark
d7004d5a55 Commit of notes on interrupts and their usage. 1996-10-14 22:27:03 +00:00
mark
114e806627 Added note that the current development history can be found
on source-changes.
1996-10-14 22:23:50 +00:00
mark
2c6e3deb80 Not needed as this file has been merged into the main TODO list. 1996-10-14 22:22:12 +00:00
mark
77cd383ae6 Updated. Now includes things that I hope users will work on as well. 1996-10-14 22:21:15 +00:00
thorpej
d2daeab10d Garbage collect the "kernel jump to zero" stuff. VA 0 is now invalid,
so the kernel will fault if the PC gets trashed like that.
1996-10-14 22:11:16 +00:00
thorpej
a5461e87e1 Delete spurious #endif 1996-10-14 21:16:55 +00:00
thorpej
ac35089824 Couple of small changes:
- Fix up usage of MBD_ISPID().
- Ensure that we never deal with a NULL proc, and if our proc has
  no pcb, punt.  Suggested by Gordon Ross.
- Eliminate some redundant NULL pointer checks in the T_MMUFLT cases;
  proc0.p_addr is now initialized early, and we make a single test
  for sanity at the top of trap() now.
1996-10-14 20:06:31 +00:00
thorpej
a72edba413 Couple more small changes:
- Initialize proc0.p_addr just after setting up the kernel stack, to avoid
  getting NULL pointers in trap().  Change suggested by Gordon Ross.
- Panic if main() returns.
- Tidy up a couple of comments.
1996-10-14 20:02:48 +00:00
cgd
b7dcf10e80 If gethostbyaddr() fails, use inet_ntoa() to turn the IP address into
a dotted-quad string.  Make this compile with -Wunused and -Wnested-externs.
Kill unnecessary and trivial-to-remove global variables.
1996-10-14 19:28:09 +00:00
is
8c0a8c37fe KNF patrol. 1996-10-14 18:40:15 +00:00
is
4f8689c804 - increase the / and /usr recommended partition sizes by 5 MB for
crash dumps and the increased kernel source size, respectively
- mention the newly available IBM (720/1440k) floppy encoding and the
  floppy driver write-protect bug.
1996-10-14 18:20:37 +00:00
jonathan
19c816fd6b Delete #ifdef'ed-out TC support (for 5000/200). 1996-10-14 17:28:46 +00:00
jonathan
f2c95d6cce Use kernel driver for serial console. PROM console output no longer used at all. 1996-10-14 17:15:42 +00:00
mrg
e6b86e3b9c fix bogon in last change. 1996-10-14 10:41:07 +00:00
thorpej
ffc4c833ea Be careful not to deref bad pointers in the MMU fault handler! (yow!) 1996-10-14 08:07:23 +00:00
thorpej
7744dc270c Couple of small changes to how kernel is mapped:
- kernel text is now always read-only.
- page 0 is invalid.
1996-10-14 08:05:37 +00:00
thorpej
4fa7362d31 Add a db_memrw.c that understands how to write to read-only kernel text
(remaps page read/write/cache-inhibit, does write, restores previous
mapping).  Kernel text no longer needs to be read/write with DDB/KGDB
is in the kernel.

Based on a similar module written for the Sun3 port by Gordon Ross,
and modified somewhat by me.
1996-10-14 08:02:27 +00:00
thorpej
5b66e34d14 A few small changes:
- Offset kernel text one page.  Stash the PA of this offset page for
  use later.
- Add a few comments.
- Free up some registers earlier in the initialization process.
- Use a `prototc' to set the Translation Control register, rather
  than relying on a2 pointing just past the MMU trampoline in the
  high page.  (Suggested by Charles Hannum.)
- Set VBR to the kernel vector table just before turning on the MMU.
- Just before rebooting, set VBR to 0, which is what the BOOTROM expects
  it to be.
1996-10-14 07:58:54 +00:00
thorpej
27f386f5b2 Garbage collect some now unneeded code, and declare the punit device
driver entry point switch.
1996-10-14 07:34:48 +00:00
thorpej
415248a4f2 Move the machdep_start() code into machdep.c. No need for different
versions of this function anymore, now that we have a sane kernel setroot().
1996-10-14 07:33:43 +00:00
thorpej
18c2b0460a Garbage collect some hackish code that was needed when we had a lame
kernel setroot().  The device type for network boot is set like any
other device now.  Also, call the punit entry point for device drivers
to properly set `bootdev'.
1996-10-14 07:31:47 +00:00
thorpej
9ff0348a49 Implement ctpunit(), a "set physical unit" entry point. 1996-10-14 07:29:57 +00:00
thorpej
9d51a5dcf7 Add a standalone driver entry point that sets the physical unit (HP-IB's
equivalent of a LUN) in `bootdev'.  This is needed to properly detect the
boot device if booted from the tape unit of an HP-IB disk/tape combo.
1996-10-14 07:29:15 +00:00
thorpej
84da57f13b Clean up the code that sets `bootdev' based on information provided
by the BOOTROM.  Also, give network devices the attention they deserve.
1996-10-14 07:26:06 +00:00
thorpej
23640d012f New root device detect/set code, lifted mostly from NetBSD/alpha.
The "options GENERIC" entry in kernel config files is not longer necessary
for "swap generic" kernels.  Uses new config constructs which work with
some glue in an old config environment.  This code will support new config
with minimal changes.
1996-10-14 07:20:26 +00:00
thorpej
8acf9f0215 Set the device class on non-dull devices. This is a transitional thing. 1996-10-14 07:14:11 +00:00
thorpej
7d13691de1 Add some new config glue to the legacy hp300 device structures. This is
a transitional thing.
1996-10-14 07:11:50 +00:00
thorpej
01d74572dd Default the HIL keyboard to "US ASCII". This should fix a bug where
an unrecognized keyboard produced garbage on keypresses in the kernel,
but worked in the boot program (which has this default).  The bug is
that the keymap pointers are unitialized.
1996-10-14 07:09:41 +00:00
scottr
534c6caa6b Add EXEC_AOUT and EXEC_SCRIPT 1996-10-14 07:01:10 +00:00
thorpej
04f65ff081 Need to include <sys/exec.h> 1996-10-14 06:53:23 +00:00
thorpej
7ee4744ed2 Need to include <sys/exec.h>. 1996-10-14 06:51:50 +00:00
cgd
385de32c08 Trade one bit of lossage for another. Now that groff has been upgraded,
the Alpha no longer needs -fno-for-scope.  However, NetBSD/Alpha's
(cygnus snapshot) g++ doesn't allow virtual functions which aren't
declared inline to be later defined as inline without a warning (-> error).
Therefore, on the Alpha, define BROKEN_GXX_VIRTUAL_INLINE (which will
go away after the Alpha is upgraded to use a more recent toolchain), and
use it to make sure that declarations of virtual functions which will
be defined inline are actually marked as inline when they're declared.
1996-10-14 05:24:23 +00:00
mhitch
c376b68e9d Now needs sys/proc.h since compat/common/compat_util.h doesn't anymore. 1996-10-14 05:15:39 +00:00
jonathan
9d781ce37f Back out second inclusion of <sys/poll.h> from duplicated poll(2) changes. 1996-10-14 04:55:26 +00:00
cgd
40bcaca9ac fix a bogon in some new functionality apparently added in the last
commit (which was supposed only to back out the printf->kprintf changes).
The new functionality is: if the user enters control-D, return an error.
1996-10-14 04:49:21 +00:00
jonathan
02fb4a1255 Back out previous revision's change to use mips family-independent locore
initialization.
Locore doesn't (yet) set up cpu_id early enough for mach_init() to use it.
1996-10-14 04:31:57 +00:00
mhitch
b4355fb150 More fallout from printf->kprintf->printf: printf in sys/systm.h now
conflicted with stdio.h.  Get rid of sys/systm.h, as it isn't needed.
1996-10-14 03:04:32 +00:00
mhitch
be3a168305 Removed stray ";" so conditional return is really conditional. Fixes
autoconfig display for xcfb.
1996-10-14 01:39:57 +00:00
cgd
0e4ddf059b deal more sanely with ELF binaries with only a single program header
section.  Patch come up with by Bob Baron <rvb+@cs.cmu.edu> and myself.
This entire bit of code (the code which sets daddr/dsize and taddr/tsize)
is very bogus, but it's not clear what the 'right' way to fix it is
and this patch fixes a problem preventing some ELF executables from
being run.
1996-10-13 22:54:52 +00:00
christos
898d3e0b34 regen 1996-10-13 22:21:15 +00:00
christos
b3de746da6 Apply PR/2840: Recognize more 3com models... 1996-10-13 22:19:19 +00:00
veego
de5f30a2ec Import GNU rcs 5.7 1996-10-13 21:56:51 +00:00
jonathan
27ec6276b8 Rename mips CPU-family locore prefixes for case-consistency:
mips_r2000_, mips_R2000_ -> mips1_
    mips_r4000_, mips_R4000_ -> mips3_
(which are also, for mnemonic reasons,  consistent with gcc flag usage,
 rather than using mipsI_ and mipsIII_).
1996-10-13 21:37:35 +00:00
mhitch
bc4aec6c5b Fix error from in_addr_t changes by christos: htonl() takes in_addr_t
parameter, not in_port_t.
1996-10-13 20:59:02 +00:00
cgd
a23bd4576c if ${MACHINE_ARCH} is "alpha" build shared libs with the 'normal' GNU
linker's command line args, rather than the NetBSD linker's command
line args.  This is a hack, and should go away in favor of something
more general when that something is developed.  In the mean time, it
might be appropriate for the various mips ports to do the same thing.
1996-10-13 20:39:36 +00:00