98fc96ea40
Add some commentary on the previous.
1995-01-15 01:15:04 +00:00
ecef8fddfa
Remove PSL_USERCLR, and add a PSL_USERSTATIC. Change the bit descriptions to
...
more closely match the 386 manual. (Probably the mnemonics should be changed,
too.)
1995-01-15 00:57:50 +00:00
c7c817b91a
sigreturn() and ptrace() security violations will now trap. Deal with those
...
traps by looking at the instruction stream and the stack to see if it looks
like we're returning to user mode, and if so pop to the hooks in locore.s.
This will fail in the case where a pop %ds or pop %es faults when returning
from a fast interrupt, but this should not be possible.
1995-01-15 00:55:25 +00:00
3e8850eab2
Enable CR0_AM on 486s. (We already handled the fault.) This would be more
...
useful if it applied to the kernel.
Avoid btrl when checking AST flag.
Add hooks to recover from errors during any INTRFASTEXIT and simulate a user-
level protection fault.
1995-01-15 00:52:21 +00:00
c26ad3af41
Cosmetic changes, and use db_printf() rather than printf().
1995-01-15 00:46:38 +00:00
8476bee7b5
Do the eflags security check in a cheaper way. Just check that the privilege
...
level of the return cs is correct; don't check the other selectors. (Will
fault later if they are invalid.)
1995-01-15 00:44:19 +00:00
da79b43eae
Move the existing frame pointer calculation into a separate helper function.
...
Eventually this should cause the child to be swapped in if necessary.
Do the eflags security check in cheaper way.
1995-01-15 00:42:01 +00:00
a2dafd0821
Avoid using btrl, and check astpending before the return cs, because it's
...
faster and usually false.
1995-01-15 00:36:37 +00:00
87d587e640
Simplify some sanity checks.
1995-01-13 14:46:44 +00:00
4740006541
Oops.
1995-01-13 14:05:17 +00:00
705f6e0337
*** empty log message ***
1995-01-13 13:19:34 +00:00
1e5cd9212d
(Almost) allow sector sizes other than 512 bytes. Reduce some expensive operations.
1995-01-13 12:30:59 +00:00
7bb384b66c
Avoid some repeated multiplications.
1995-01-13 11:14:27 +00:00
7650bd97fc
Always boundary check I/O.
1995-01-13 10:46:32 +00:00
c541b630fc
Add a sc_bcount, similar to wd.c.
1995-01-13 10:35:58 +00:00
a0303980c2
Calculate b_cylin correctly for different block sizes.
1995-01-13 10:30:08 +00:00
0033c975c6
Fix another bug in the DMA code.
1995-01-13 10:22:58 +00:00
fda9731199
Further cleanup. Back out part of the previous; it's the device's
...
responsibility to check for partial blocks.
1995-01-13 09:57:39 +00:00
4abdb375bc
Minor rearrangement.
1995-01-13 09:40:21 +00:00
c2bc310a37
Change the way sc_skip is used. Fix a bug in the (incomplete) DMA code. Add a
...
new sc_blkno (which will be useful later).
1995-01-13 08:58:16 +00:00
da7d74e3dc
Fix typos.
1995-01-13 08:37:25 +00:00
1f758ca55c
Fix typo.
1995-01-13 08:31:46 +00:00
ca84c10497
Only allow opening one density at a time.
1995-01-13 08:29:25 +00:00
04a1fc8816
Change the way b_cylin and sc_track (now sc_cylin) are used slightly.
...
If transfer goes past end of disk, truncate it.
1995-01-13 07:57:01 +00:00
dc63598a7f
Clean up bounds_check_with_label(). Round request to sector size in label,
...
and return EINVAL if bcount is less than the sector size or negative.
1995-01-13 06:51:38 +00:00
8ba4c5fca1
Add boca, makeoptions.
1995-01-11 20:59:50 +00:00
0a22930d97
Explicitly test the RXRDY status bit on incoming characters.
1995-01-11 09:56:40 +00:00
d7a8b2964f
Only define sig_atomic_t when _ANSI_SOURCE is defined.
1995-01-10 19:01:00 +00:00
3e0b9f0694
Implement symbolic links (plus a minor fix for %x printing)
1995-01-09 22:13:10 +00:00
7af86b7eb7
Copy physlock code from SPARC port, and some minor cleaning.
1995-01-09 09:13:09 +00:00
b006bf7e68
Added missing forward declaration
1995-01-09 01:05:24 +00:00
e1c97cde06
New file for signal handling and {g,s}etcontext().
1995-01-08 21:23:31 +00:00
7c056bd334
New svr4_machdep.c for svr4 signal handling.
...
Made check_selectors global, to be used by the
svr4 code.
Removed unused variable.
Modified sendsig() to call the svr4_sendsig().
1995-01-08 21:22:16 +00:00
66cc79dbc4
New svr4_machdep.c
1995-01-08 21:18:48 +00:00
e06bc8c13e
Don't issue a controller self-test...
1995-01-07 22:48:29 +00:00
06b0414491
Actually, do the ID reset in elink_reset().
1995-01-07 21:37:54 +00:00
37e16c5747
Don't print line numbers in most of the diagnostics.
1995-01-07 06:49:35 +00:00
d4f91d51a9
Fix a race condition between normal I/O and wd_get_parms().
1995-01-07 03:07:22 +00:00
7f61bec5cd
Cosmetic changes, to look more like the new esp.c will.
1995-01-07 01:51:07 +00:00
ba288b4ae1
Do the reset in the individual drivers, not in elink_idseq().
1995-01-06 21:41:26 +00:00
290ed068ab
Remove many of the `static's, to facilitate debugging.
1995-01-04 09:02:32 +00:00
6125e1b551
Fix oversight in previous.
1995-01-04 00:47:53 +00:00
9e499d41b3
\
...
These files have been 99% rewritten from the original ast driver. Add an
appropriate copyright notice.
1995-01-03 22:38:54 +00:00
0349c5b9e0
Add a multiplexer for BOCA serial boards.
1995-01-03 22:26:36 +00:00
09834a5bd3
This is correct according to the docs on both cards.
...
Doesn't seem to make any difference to the known problems though.
1995-01-03 05:04:04 +00:00
a987158712
Add interrupt sharing types.
1995-01-03 03:56:09 +00:00
b6c8a9a4e9
Update intr_establish() prototype.
1995-01-03 03:14:35 +00:00
93a0c4a258
Add interrupt sharing types.
1995-01-03 01:42:11 +00:00
b5cf1b5ddd
Add interrupt sharing types.
1995-01-03 01:30:14 +00:00
d16dcced99
Add interrupt sharing types.
1995-01-03 01:19:19 +00:00
326830c1ba
Use direct config for subdevices.
1995-01-02 22:27:46 +00:00
e148dc7a64
Move setting of sc_maddr so soft config works.
1995-01-02 22:01:05 +00:00
a774fa6fe5
Use soft configuration when possible.
1995-01-02 21:57:04 +00:00
3a967fcbb6
More cleanup.
1995-01-02 21:27:27 +00:00
72dec3b548
Various cleanup.
1995-01-02 20:54:01 +00:00
3a74796b3e
Need isareg.h.
1995-01-02 20:39:35 +00:00
894bb9145f
Use soft configuration where possible.
1995-01-02 20:32:55 +00:00
2d5b04744a
Don't assume ia_maddr is in the hole.
1995-01-02 20:06:29 +00:00
76b0b02cc3
Fix non-serious typo.
1994-12-31 05:34:00 +00:00
7b8535b554
do the right thing with strange-sized varargs objects.
1994-12-28 22:45:42 +00:00
ff315ba9a4
Don't turn on the SCE bit in the enable register.
1994-12-28 19:45:41 +00:00
8fd135209c
Numerous changes. Many bugs fixed, better autoconfig, a few new features.
1994-12-28 19:42:47 +00:00
262b425866
temporary warning until I get new docs.
1994-12-28 16:47:50 +00:00
e5049f4d59
make the definition of _JBLEN mach. dep. header-dependent.
1994-12-20 10:36:27 +00:00
ce06e718ad
Fix flags to isa_dmastart().
1994-12-17 18:45:11 +00:00
ab77439747
Clean up constab stuff a little.
1994-12-16 04:19:13 +00:00
52d8d259fd
Add a bt.
1994-12-15 16:14:50 +00:00
a7393f95f3
Remove fp arg to d_open.
1994-12-14 18:52:32 +00:00
ca65f2ce5d
Remove dkbad.h.
1994-12-14 15:17:22 +00:00
aeda11fc40
Remove the d_subtype frobnication. It was clearly broken, and not very
...
useful.
1994-12-13 15:31:54 +00:00
05e78f1424
Add empty pcstop().
1994-12-13 13:42:53 +00:00
e694663b46
Make the diagnostics less cryptic.
1994-12-12 04:43:51 +00:00
3e4a358e28
Patch from Thomas Eberhardt to make PCnet-32 chips work.
1994-12-12 04:39:52 +00:00
995e676d90
Add `flags 1' to ast subdevices, as pointed out by John Kohl.
1994-12-11 16:20:48 +00:00
303841c92e
3c509 fixes to avoid adapter getting stuck in high traffic conditions.
...
I don't particularly like the fixes, but at least now the 3c509 never
gets stuck. On the other hand, performance still sucks in some cases.
1994-12-10 14:52:02 +00:00
793e542f89
Fix format of attach message.
1994-12-10 05:55:26 +00:00
1acd9ddb90
Speed up FPU task switching some more.
1994-12-10 00:28:18 +00:00
957694159f
Correct a couple of things...
1994-12-08 20:55:05 +00:00
2227b302cf
update
1994-12-06 06:59:01 +00:00
8d87220928
Don't ignore the exit status of ld(1).
1994-12-06 06:07:39 +00:00
6d6a24357a
Compile with -g.
1994-12-06 05:02:26 +00:00
1e0e6d72c9
Conditionalize the audio driver on NAUDIO, not NSB, now that the former
...
exists.
1994-12-05 05:45:38 +00:00
e92233c07c
Deal correctly with esym being 0.
1994-12-04 07:37:27 +00:00
c3b944cb15
Use correct select routine for audio.
1994-12-04 06:32:39 +00:00
8e6f464987
fdclose -> Fdclose
1994-12-03 15:17:06 +00:00
5c3d80060d
Reset the entire PSL, not just part of it.
1994-12-03 14:55:51 +00:00
73afee1973
We don't need the extra space to preserve the BIOS variables any more.
1994-12-02 03:02:49 +00:00
8fd1f92ae7
Don't futz with the cursor shape unless we really, really need to.
1994-12-01 11:12:04 +00:00
211b002017
Ignore ERR during diagnostics.
1994-12-01 10:45:26 +00:00
434d451672
Adjust dump calculation to avoid sign problems.
1994-12-01 09:53:38 +00:00
6686f7f263
Rework detection a bit, and check for the Pentium fdiv bug.
1994-11-30 04:42:07 +00:00
c72f58d401
Don't need to turn on emulation when npx not present; we leave it off always.
1994-11-30 04:26:06 +00:00
867ebf1d84
Clear the controller active flag after a timeout.
1994-11-30 02:32:03 +00:00
5584ff927f
Don't stall the queue on CHECK SENSE if we get a reselect while selecting.
1994-11-30 02:08:01 +00:00
639a64de08
Fix a null pointer reference.
1994-11-29 21:35:13 +00:00
51bed8d804
Add an assertion mechanism to check for more abnormal conditions.
1994-11-29 21:29:56 +00:00
58cb5339d6
For consistency...
1994-11-29 20:57:28 +00:00
41b26a6dfd
Handle BUS FREE phase a little differently, to get better error detection. Other
...
cleanup.
1994-11-29 20:08:27 +00:00
bb90feebac
Disable debugging breakpoints by default.
1994-11-29 17:56:49 +00:00
5ac2484d95
* Redo most of the message and phase state machines. Multiple pending
...
message outs should now work correctly (including retransmissions).
Multiple messages ins should be more efficient. Missing REQ after
reselection phase should now (correctly) cause a wait with ENREQINIT
set.
* Abstract out the reselection out sequence, and use it directly to
get sense info. (This is somewhat faster than the generic done/sched
sequence.)
* If there are no ACBs available, wait interruptibly for one.
* Leave the interrupt registers alone during message in/out; just
watch for phase change. Don't (incorrectly) reenable interrupts
during data in/out; defer that to aicintr().
* Handle command pointer per SCSI 2 spec.
* Other minor coding improvements.
1994-11-29 17:50:39 +00:00
6a2c060851
fix unary plus typo :-)
1994-11-25 23:19:03 +00:00
cba32aa25f
Minor update for new autoconfig.
1994-11-25 08:17:21 +00:00
0649f37219
Officially credit Onno.
1994-11-23 22:00:08 +00:00
8bacafc13c
Clean up deleted files.
1994-11-23 09:10:13 +00:00
2058abab15
Forgot to commit this as part of the LBA changes.
1994-11-23 07:56:52 +00:00
cc44b4600c
There's no point in bothering to unlock what we didn't lock.
1994-11-23 07:54:15 +00:00
6dda55519e
Fix an outdated comment.
1994-11-23 03:11:01 +00:00
a3fc466734
Use logical block addressing if available. Be a little more careful about
...
maintaining timeouts. Clean up wddump() a bit.
1994-11-23 01:35:43 +00:00
525fc36e19
Implement 32-bit reads and writes, but for now assume 16-bit mode.
1994-11-22 10:20:16 +00:00
3a99ddf172
Add a missing newline in some error messages.
1994-11-22 09:36:56 +00:00
db3bde9525
Implement EIDE multi-sector PIO and DMA modes, based on code from Onno van
...
der Linden. Also, various structural cleanup, and fix two bogons that crept
in, and remove all of the old clustering nonsense.
1994-11-22 09:33:59 +00:00
1e01855683
Minor fix to debugging code.
1994-11-22 05:50:48 +00:00
85826f0fbe
Fudge a couple more fields for ancient disks.
1994-11-22 05:39:16 +00:00
e1a7bac628
Change some constant names, etc., to better match the ATA spec.
1994-11-22 05:34:49 +00:00
95edd299b0
Add still more locking.
1994-11-22 03:23:49 +00:00
dcb90851c2
Minor change.
1994-11-22 01:21:11 +00:00
5bd4115690
Add some missing locking, and some general cleanup.
1994-11-20 22:36:43 +00:00
d2d4dc4654
Use immediate constants for port numbers < 0x100.
1994-11-20 21:36:44 +00:00
f63115591a
Included <sys/cdefs.h> since this file is now using __signed.
...
It broke non gnu compilers.
1994-11-20 18:50:18 +00:00
9a3840f2f0
Stylistic changes.
1994-11-18 22:25:12 +00:00
79a1309e3c
Remove Xwild; it's no longer used.
1994-11-18 22:22:40 +00:00
e075f8c13d
Convert port, IRQ, and DRQ numbers to ints.
1994-11-18 22:18:35 +00:00
61800cfa25
Convert port, IRQ, and DRQ numbers to ints.
1994-11-18 22:07:32 +00:00
35cdd801e6
Convert port, IRQ, and DRQ numbers to ints.
1994-11-18 21:57:40 +00:00
fc8e694056
Fixed error svr4 error code mapping; my old attempt was a noop.
1994-11-16 20:14:42 +00:00
a2727ec558
Check IRQ 9, not IRQ 2.
1994-11-15 00:00:24 +00:00
b341f7a8c0
Map IRQ 2 to IRQ 9.
1994-11-14 23:58:56 +00:00
e91e570cfd
Shorten the Cyrix handling code a little.
1994-11-14 23:35:48 +00:00
423bfa0764
Garbage collect unused variable.
1994-11-14 08:53:06 +00:00
a96cb110a3
Added new svr4 files
1994-11-14 05:54:30 +00:00
55845b1901
changed open prototype in conf.c; added bsd->svr4 errno mapping in trap.c
1994-11-14 05:53:48 +00:00
1c5ac2e88c
pcb.h requires segments.h now.
1994-11-08 23:09:21 +00:00
f875a9efe6
Allocate pv_pages differently.
1994-11-08 01:17:19 +00:00
d254e95680
Garbage collect some variables which are no longer used.
1994-11-07 10:19:28 +00:00
21e911d911
Update for new autoconfig.
1994-11-07 09:03:48 +00:00
cfae52f542
It's legitimate, if foolish, to mask SIGCONT.
1994-11-07 05:57:28 +00:00
297c06de67
Comment previous change.
1994-11-07 05:32:50 +00:00
86a2079aef
When restarting syscalls, make sure we adjust the PC by the correct amount
...
depending on what instruction we used to enter the kernel. Push the
instruction length into tf_err on entry.
1994-11-07 05:26:08 +00:00
ed993b3930
Two trivial optimizations.
1994-11-07 05:09:04 +00:00
d5f99262a9
Since npxdna() and npxsave() are only ever called at spl0 now, we can get rid
...
of all the overhead of frobbing the IDT, and just use a simple flag to tell
npxintr() to ignore interrupts. This allows lots of code reduction.
Also, update some comments and add some DIAGNOSTIC code to detect is these
assumptions ever become wrong.
1994-11-07 03:39:37 +00:00
5133a6cfd7
Only save the FPU state if somebody else needs the FPU, or when forking.
1994-11-06 23:43:17 +00:00
f3ef31e4aa
Oops; forgot to adjust stack offset.
1994-11-06 22:05:08 +00:00
37863af7bb
Forgot two underscores.
1994-11-06 21:51:47 +00:00
58b3010e87
Use %eax slightly more effectively in cpu_switch().
1994-11-06 20:51:46 +00:00
915336b97c
Approximate the divide by multiplying by 3433/4096. This is faster than
...
the previous code, and more accurate. Also avoid saving and restoring
registers.
(The multiplication algorithm was generated by GCC.)
1994-11-06 20:33:35 +00:00
ddb6e42460
Actually, don't bother checking p == 0 at all.
1994-11-06 20:30:09 +00:00
5cf38ebdb4
The case of curproc == 0 `can't happen' unless we're in the idle loop, so
...
allow it to panic by way of a reference to 0 in other cases. Explicitly test
for p == 0 in the specific cases where it's needed, and only calculate pcb if
needed.
1994-11-06 20:28:57 +00:00
9a84611b06
Move the spl0() from the idle loop to the top of cpu_switch(). (This also
...
avoids needing to call spllower() below.) Set curproc immediately after
removing the new process for the run queue, before allowing interrupts again,
for the benefit of schedcpu().
1994-11-06 20:21:29 +00:00
205d425f2f
As pointed out by Bruce, splsoftclock() needs to process pending interrupts
...
when lowering the priority. Do this by simply calling splx(). For the
benefit of NFS (for now), modify splx() to return a value so that
splsoftclock() does. This work is optimized out for every use of splx()
that ignores the value, so it only affects relevant code.
Also, clean up this file a little.
1994-11-06 01:37:47 +00:00
c2191a2ad0
Fully prototype ssdtosd() and sdtossd().
1994-11-05 03:20:36 +00:00
mycroft