918 Commits

Author SHA1 Message Date
gwr
626bde2f9b Add bus_subr.c 1997-02-18 15:10:40 +00:00
gwr
1af5733370 New implementation of the findroot/setroot business. Eliminates
dependency on wired-down SCSI target/LUN for boot devices.
1997-02-18 15:07:29 +00:00
gwr
201c1f178e Move bus_peek(), bus_mapin() here from autoconf.c so the latter
may be shared with the sun3 port.
1997-02-18 14:58:32 +00:00
gwr
48da5b3d44 Move bus_peek(), bus_mapin() here from autoconf.c so the latter
may be shared with the sun3x port.
1997-02-18 14:57:41 +00:00
gwr
7190ad9936 Fix a major bug in pmap_remove_kernel() that causd the removal of
one page too many (worse, it unmapped our stack page... Ouch!)
Also, use loadcrp() and such instead of calling pmap_activate
in various places so we have more control over TLB flush.
Now pmap_activate() is called only by locore (and that call is
done only if PMAP_DEBUG is defined).  Other minor clean-up...
1997-02-16 19:38:08 +00:00
gwr
2053794556 Add sun3/regdump.c 1997-02-14 20:10:50 +00:00
gwr
ae0ec6bc84 Moved regdump() and friends to sun3/regdump.c 1997-02-14 20:09:14 +00:00
gwr
f01121ece4 Moved sendsig(), sys_sigreturn() to sun3/sig_machdep.c
Moved regdump() and friends to sun3/regdump.c
1997-02-14 20:00:51 +00:00
gwr
780e636b2d Pull regdump() out of machdep.c and correct a problem where it
would try to print things that are beyond the end of the stack.
1997-02-14 19:55:18 +00:00
gwr
7275a8976f Add pmap_count(), fix some bugs... (from Jeremy). 1997-02-14 03:56:49 +00:00
gwr
dc9ad25881 Define pmap_resident_count(), pmap_wired_count() as
macros that call pmap_count().  (from Jeremy).
1997-02-14 03:55:33 +00:00
gwr
fb373dd41e Add sun3/sig_machdep.c 1997-02-13 22:21:22 +00:00
gwr
905e296504 Moved sendsig() and sys_sigreturn() to sig_machdep.c
Rename some trapframe variables for consistency.
1997-02-13 22:16:53 +00:00
gwr
91c7291256 Move sendsig() and sys_sigreturn() from machdep.c to here.
Should probably move to m68k/m68k eventually...
1997-02-13 22:04:03 +00:00
gwr
443faad4e1 Reorganize/simplify the way the memory layout is done.
Update to the latest sys/dev/isa/if_ie.c (new ieget).
1997-02-13 20:48:36 +00:00
gwr
5997a69a3e Tune MXFRAMES and MXFRAMES down a little.
Reorganize the softc for convenience in ddb.
1997-02-13 20:43:18 +00:00
gwr
6a83772a4e Replace caddr_t with u_int in device structs.
Reduce MAXMCAST by a few to be safe.
1997-02-13 20:40:11 +00:00
gwr
45ff882a54 Replace sun3/kgdb_stub.c with sun3/kgdb_machdep.c 1997-02-13 18:14:31 +00:00
jeremy
8869e895a8 Convert meaningless '0' into a more human readable 'IOMMU_PDE_DT_INVALID'. 1997-02-13 07:38:21 +00:00
gwr
27b8c79679 Latest version from Jeremy. Changes too numerous to describe here. 1997-02-12 23:09:32 +00:00
gwr
c4fa121f14 Reinstate call to pmap_activate() for now, because it is helpful
for debugging the pmap code.  The in-line version of the same is
still here and will be used if PMAP_DEBUG is not defined.
1997-02-12 23:06:27 +00:00
gwr
3476b1d4a4 Adapt to changes in pmap.h 1997-02-12 23:03:14 +00:00
gwr
bb35ee4ae8 In struct pmap, replace pm_mmucrp with pm_a_phys.
Define pmap_resident_count(), pmap_wired_count()
to call pmap_resident_pages(), pmap_wired_pages()
respectively, so we can keep those statistics on
lower-level structures instead of here.
1997-02-12 23:01:28 +00:00
gwr
6cb54de668 Avoid dereferencing null curproc in dumpmem(). 1997-02-12 16:06:18 +00:00
gwr
bb7dc881a3 Force RB_SINGLE for now (in addition to RB_KDB). 1997-02-12 16:04:33 +00:00
gwr
05a09b938e Must convert from/to BCD when reading/writing the clock chip. 1997-02-12 16:00:31 +00:00
gwr
20c8082e79 Now using <kern/kgdb_stub.c> 1997-02-12 01:32:39 +00:00
gwr
9b3ad7dd7e Now using <sys/kgdb.h> 1997-02-12 01:30:36 +00:00
gwr
292e8e784f The KGDB include is now <sys/kgdb.h>
Add the kgdb_recover trick.
1997-02-12 01:10:21 +00:00
gwr
a4b09086fe Actually works now! 1997-02-12 01:07:56 +00:00
gwr
cfb896375f Machine-dependent part of the KGDB remote "stub" 1997-02-12 01:03:20 +00:00
gwr
6ef591a70b Add a few more coments. (i.e. why init a6=0) 1997-02-11 19:06:08 +00:00
gwr
f914f723de Add NBPG 1997-02-11 19:02:30 +00:00
gwr
9536056d54 Fix some comments 1997-02-11 18:59:23 +00:00
gwr
7bd7620f61 Sync with sun3. 1997-02-11 00:56:08 +00:00
gwr
aebbe3f083 Change most occurrences of "struct frame" (union of all frames formats)
to "struct trapframe" (the minimal frame) because in most cases, all we
care about is the minimal, invariant part.  Further, if the declaration
says we have an arg of type "frame" (big union), then KGDB will try to
print the whole bloody thing out whether you want to see it or not.
1997-02-11 00:18:25 +00:00
gwr
f13b3f57fd This is obsolete (and has been for a while) 1997-02-11 00:10:35 +00:00
gwr
2bdded1c83 Move the logic about entering the kernel debugger (whichever one) to
trap.c:trap_kdebug() so it can be C code instead of assembly.
Make sure the initial frame pointer is zero so KGDB will know when to
stop trying to follow frame pointers during backtrace.
1997-02-11 00:05:01 +00:00
gwr
1e0e399b63 Change most occurrences of "struct frame" (union of all frames formats)
to "struct trapframe" (the minimal frame) because in most cases, all we
care about is the minimal, invariant part.  Further, if the declaration
says we have an arg of type "frame" (big union), then KGDB will try to
print the whole bloody thing out whether you want to see it or not.

Also add the function trap_kdebug() which takes care of calling one of:
kgdb_trap (for KGDB), or kdb_trap, (for DDB) or the Sun PROM monitor.
This logic used to be in locore, but it is better as C code.
1997-02-11 00:01:15 +00:00
gwr
4bcf6b565f Change most occurrences of "struct frame" (union of all frames formats)
to "struct trapframe" (the minimal frame) because in most cases, all we
care about is the minimal, invariant part.  Further, if the declaration
says we have an arg of type "frame" (big union), then KGDB will try to
print the whole bloody thing out whether you want to see it or not.
1997-02-10 23:57:34 +00:00
gwr
569d97a2f9 Save/restore the monitor "v" command handler. 1997-02-10 23:44:23 +00:00
gwr
b31965da5a Set cputype,mmutype at compile-time. 1997-02-10 23:42:05 +00:00
gwr
b7d7f667f3 minor nit in bounds_check_with_label() 1997-02-10 23:40:24 +00:00
gwr
c70de0b625 Actually works now. 1997-02-10 23:36:09 +00:00
gwr
2f46cb47e2 update comments 1997-02-10 23:35:27 +00:00
gwr
dfddf9c39b Actually works now. 1997-02-10 23:34:38 +00:00
gwr
f55c7652b7 Export zs_get_chan_addr() for zs_kgdb.c
Print the "console on ..." in cninit(), and
avoid reference to kbd/kd drivers if NKBD==0
1997-02-10 23:29:52 +00:00
gwr
196f91101a Make some more things static, and cosmetic stuff. 1997-02-10 23:26:21 +00:00
gwr
9f9e8281e0 Add zs_get_chan_addr(), zs_kgdb_init() 1997-02-10 23:24:18 +00:00
gwr
7ceee3c729 Add -fno-defer-pop to COPTS (helps ddb backtrace).
Add AFLAGS=-m68020 and use it with ${AS} commands.
Depend kgdb_stub.o on Makefile (KGDBDEV, KGDBRATE).
1997-02-10 19:23:41 +00:00