Commit Graph

18601 Commits

Author SHA1 Message Date
cgd
20af97b2c0 don't pass max device number for a given bus in, use
PCI_MAX_DEVICE_NUMBER, which can be defined on a per-machine basis.
(defaults to 32.  on i386, it's 32 if pci conf mode == 1, 16 if 2.)
1996-03-14 02:37:28 +00:00
cgd
7fda077700 (1) provide #defines for cf_loc[] entries for devices that attach to
pcibus and pci.
(2) remove the #ifdef i386 from pci.c, and provide a machine-dependent
    hook (pci_md_attach_hook()) to do any machine-dependent attachment
    gunk, e.g. on the i386 printing out the configuration mode (if bus 0)
(3) don't pass max device number for a given bus in, use
    PCI_MAX_DEVICE_NUMBER, which can be defined on a per-machine basis.
    (defaults to 32.  on i386, it's 32 if pci conf mode == 1, 16 if 2.)
1996-03-14 02:35:32 +00:00
cgd
68b00a8e52 RCS Id police 1996-03-14 02:30:30 +00:00
pk
485ccae478 Insert child_return() for the benefit of child return path from cpu_fork(). 1996-03-14 00:55:59 +00:00
pk
0b1cba4b0b Implement cpu_set_kpc() and use its machinery to by-pass a child out
of the kernel from cpu_fork().
1996-03-14 00:54:34 +00:00
pk
d48d8e1512 Install machinery for cpu_set_kpc(). 1996-03-14 00:54:06 +00:00
pk
808df5543f Remove cpu_set_init_frame(). 1996-03-14 00:49:20 +00:00
pk
894fde6bdd Remove __FORK_BRAINDAMAGE. 1996-03-14 00:48:30 +00:00
pk
dc3c7fb2ba Actually enter executables where promised... 1996-03-14 00:11:29 +00:00
scottr
4d30db274f Back out previous change. 1996-03-13 23:42:45 +00:00
veego
b9f085f78f Restore a local change from our old libg++ to fix a compile warning 1996-03-13 23:07:35 +00:00
mark
46e6b251f1 Raise the spl level while modifing the keyboard autorepeat and delay
rates.
Added missing = to test for alphabetic characters so that caps lock
applies to Z as well as A-Y.
1996-03-13 21:51:32 +00:00
mark
a2ff705689 Added code to handle indents in monitor definition files. 1996-03-13 21:47:15 +00:00
mark
cb6f3bb55c Removed references to the kstack double mapping from the old debug
shell.
1996-03-13 21:45:54 +00:00
mark
32f0b6e092 curpcb is declared in pcb.h so don't declare it here.
removed debug printing when faulting in new page tables for the user
process stack.
1996-03-13 21:41:41 +00:00
mark
ba893940d3 savectx() now justs syncs the pcb, it no longer needs to copy the stack.
cpuswitch() simplified, now pushes and pulls a switchframe structure
on entry and exit. Added extra comments to indicate the states of
various registers throughout cpuswitch().
Added proc_trampoline function.
1996-03-13 21:39:10 +00:00
mark
5c857a37ef Use the value returned from initarm() to set the stack pointer before
calling main().
1996-03-13 21:34:17 +00:00
mark
bbf2aa16b5 Removed kstack references. Updated code that sets up the pcb for proc0
now that double mapping has gone.
initarm() now returns the address to be loaded into the stack pointer
before calling main() to kick things off.
1996-03-13 21:32:39 +00:00
mark
d253d345a2 Since kernel stack double mapping no longer exists there is no
point in trying to perform a postmortem on it following an panic.
1996-03-13 21:26:52 +00:00
mark
52f490ccf6 No longer need to allocate hooks for kernel stack double mapping during
pmap bootstrapping.
1996-03-13 21:25:04 +00:00
mark
7045f47891 The kstack pte fields no longer exist so remove them. 1996-03-13 21:22:32 +00:00
mark
caf9dd3edc Removed external declaration of kstack. No longer relevant now kstack
double mapping has been removed.
No longer need to fixup the trap frame address to take double mapping
into consideration.
1996-03-13 21:21:00 +00:00
mark
043ef25c23 Remove cpu_set_init_frame macro. This is no longer needed now
cpu_set_kpc() has been implemented.
1996-03-13 21:18:06 +00:00
mark
f9f45be67c Removed all the double mapping code from cpu_fork().
Arrange for the child process to return to user land via child_return()
on leaving cpuswitch().
Added function cpu_set_kpc().
1996-03-13 21:16:15 +00:00
mark
8fbba52c87 Removed option __FORK_BRAINDAMAGE at last. 1996-03-13 21:10:43 +00:00
mark
ffa6eab259 Removed the kstack pte fields. These are no longer needed as the kernel
stack is no longer double mapped.
1996-03-13 21:08:36 +00:00
christos
b92fe0f4c4 Kernel version of longjmp does not take the return value of setjmp as
an argument. It always returns 1.
1996-03-13 21:08:09 +00:00
christos
b4a31b1fc7 Kernel version of longjmp() does not take the return value of setjmp() as
an argument. setjmp() always returns 1.
1996-03-13 21:06:38 +00:00
mark
f96ae702d7 Added switchframe structure to describe the stack frame pushed on the
stack at the beginning of cpuswitch().
1996-03-13 21:06:34 +00:00
mark
c353f32f10 _mcount() can now be declared inline. 1996-03-13 21:04:55 +00:00
mark
fb4eacf0a5 Removed comments describing debugging code that has long since gone.
Added new comments for parts of the undefined vector entry point.
1996-03-13 21:00:26 +00:00
mark
1e97524220 SCSI scanner device added to the list of SCSI devices configured. 1996-03-13 20:57:45 +00:00
mark
f6c46bc20c Added hooks for the SCSI scanner device driver. 1996-03-13 20:55:21 +00:00
mark
d389b21913 Added boot time override for the memory test during probing. This
memory test apparently causes hangups on some etherH interfaces.
Until this can be debugged with an offending interface the memory
test can be bypassed.
1996-03-13 20:52:35 +00:00
mark
c117a7f133 Only call the debugger if DDB is defined. 1996-03-13 20:28:50 +00:00
scottr
eb473e57bd Add second argument to longjmp() call. 1996-03-13 19:04:04 +00:00
is
3725e6e117 Fix a typo and mention -P/--absolute-paths in the man page.
From Greg Earle, PR 1585.
1996-03-13 14:55:00 +00:00
jonathan
dc6fdf6cdc First commit of Per Fogelstrom's port to the Acer pica r4400/isa machine. 1996-03-13 04:58:04 +00:00
fvdl
f5e6df1ed0 Disable invalidating of directory offsets cookies. Should fix one or two
directory problems.

XXX There is no clean solution to the cookie/cookieverifier validity mess.
    Together with the disabled strict cookie check, this puts us back at
    what v2 did in this case. Slightly better solution possible by
    consequently storing 64bit cookies in other places too.
1996-03-13 00:44:17 +00:00
fvdl
146f96c2ce Make readdirsize default to rsize if rsize is explicitly specified,
and readdirsize isn't.
1996-03-13 00:38:55 +00:00
scottr
b4225290a4 Use 851-specific instructions only #if defined(M68020). 1996-03-12 23:46:32 +00:00
is
905b10d4bc M68060 additional frame formats and bits. Closes PR 1174.
This shouldn't break anything; I find ... grep ... all the relevant
kernel sources subdirecotry trees I could think of for the defines
that don't use the new FSLW or FPF6 prefix. But in case it does anyway,
tell me immediately.
1996-03-12 22:21:51 +00:00
scottr
6208752b50 Oops; fix typo. 1996-03-12 21:34:57 +00:00
phil
6867e1fda1 Sync with previous changes to Makefiles ... 1996-03-12 21:28:52 +00:00
scottr
d504ec531e Add "options M68020" so that 68851-specific code is only compiled when
we want it (we don't support the 020 without an 851 anyway).  Note that
M68030 no longer implies support for 020+851.

This also allows us to compile with -m68030 iff M68020 isn't specified
in the config file.
1996-03-12 21:06:22 +00:00
scottr
89f7d665d5 KNF. 1996-03-12 19:18:49 +00:00
phil
d99c63563c Uncomment the %expect. 1996-03-12 19:01:05 +00:00
phil
2fe7ea6745 Uncomment the %expect. 1996-03-12 18:46:21 +00:00
christos
6750270dc6 data and date aren't the same... 1996-03-12 18:04:27 +00:00
mrg
d55f4dd0c3 oops; back out previous change and add comment describing what the weird goto does. 1996-03-12 13:07:52 +00:00