Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
28,095 Commits 606 Branches 0 Tags 3.1 GiB
a155715d44
Commit Graph

11507 Commits

Author SHA1 Message Date
jonathan
d6611ebcef * Add fixes for possibly-uninitialized variables from rev 1.11. 
(The operator-precdence changes in 1.11 were pulled up in rev 1.14).

* If the chip asserts a SCSI-bus error (SII_BER) in sii_intr,
  reset the SII_BER bit instead   of ignoring it.
  XXX SII_BER May need checking and resetting elsewhere.
 1997-04-29 01:58:38 +00:00
gwr
42f8cf900b Use unsigned when doing copy/xor check. 1997-04-28 23:43:56 +00:00
gwr
cb96a536e4 Add a comment about obio_mapin(). 1997-04-28 23:38:47 +00:00
gwr
1046c915b6 Drivers that use obio_mapin should enforce OBIO attachment 
(just to be paranoid).
 1997-04-28 23:30:19 +00:00
gwr
304d49b554 include <machine/leds.h> 1997-04-28 23:21:01 +00:00
gwr
712f2e8d93 Moved leds_* declarations into leds.h 1997-04-28 23:19:39 +00:00
gwr
589bdf49d5 si_intr: if our CSR indicates an interrupt, return 1. 1997-04-28 22:10:41 +00:00
gwr
445e244536 Do console initialization earlier (see obio_init). 
Document initialization order dependencies.
 1997-04-28 22:08:36 +00:00
gwr
2e1c5c36ad Be quiet in _save_symtab() unless there is a problem. 
Improve some comments (and sync with sun3x code).
 1997-04-28 22:06:20 +00:00
gwr
2223431f0a No longer needs an early init call (just normal autoconfig). 
Only animate the LEDs when we were idle (like SunOS).
 1997-04-28 22:04:29 +00:00
gwr
df35245ac0 Fix a bug where OBIO locators were not presented to the children. 
Rename obio_alloc to obio_mapin; add comments, etc.  (like sun3x)
 1997-04-28 22:02:43 +00:00
gwr
388bf0e0e0 Rename obio_alloc/obio_mapin and check for obio in match functions 
for attachments that will use obio_mapin.
 1997-04-28 21:59:19 +00:00
gwr
a5efc65699 Sync. up with the sun3x version. 1997-04-28 21:55:59 +00:00
gwr
479d3fa66d No longer needs an early init call (just normal autoconfig). 1997-04-28 21:51:49 +00:00
gwr
3c841dcb79 Export the symbol _Idle for clock.c (and fix some comments). 1997-04-28 21:49:38 +00:00
gwr
dcd4fae823 Use AUTOVEC_BASE instead of 0x18. 1997-04-28 21:48:25 +00:00
gwr
46cd1d0fec Add a kernel declarations section. 1997-04-28 21:47:10 +00:00
gwr
52e455e8ba Fix declaration of get_control_byte(). 1997-04-28 21:46:31 +00:00
gwr
236a126750 Declare peek_long(). 1997-04-28 21:45:47 +00:00
gwr
a0d8cf3b65 Kill an obsolete comment. 1997-04-28 21:44:53 +00:00
scottr
c69d32662f From Mycroft: Oops; forgot to GC the last mbuf allocated when 
out of clusters.
 1997-04-28 20:31:16 +00:00
mycroft
df082e3c8d A handle of 0 means the open failed... 1997-04-28 18:46:15 +00:00
mycroft
37cb09ba66 DTRT if OF_read() returns 0 when no data is available. 1997-04-28 18:33:57 +00:00
scottr
70dcf0b5f6 Undo a hack added in the last overhaul (which caused an extraneous byte 
to be written to the target).  Also, use the sc_min_dma_len field
to determine whether to use PDMA or PIO.
 1997-04-28 15:59:20 +00:00
pk
547d95a029 Remove esp FLUSH when residual bytes are present in FIFO, the ncr53c9x 
driver will flush when necessary. Doing the FLUSH here interferes
with DMA-enabled target selection.
 1997-04-27 22:12:40 +00:00
thorpej
97a6c32e58 Garbage-collect a bunch of _very_ hp300-specific code, most of which 
doesn't work anyhow (due to current and future fundamental differences
between NetBSD and HP-UX device semantics).  This can probably be shared
with other m68k ports now.
 1997-04-27 21:38:57 +00:00
thorpej
ba0a5d5c58 Bump version: Better SPU detection code 1997-04-27 21:18:31 +00:00
thorpej
8b4cae0516 CPU -> SPU 1997-04-27 21:17:13 +00:00
thorpej
3cb561ed9f Don't define our own machineid values - pull in <machine/hp300spu.h> 1997-04-27 21:15:14 +00:00
thorpej
fbad957c52 Update getmachineid() for new machineid values. 1997-04-27 21:13:55 +00:00
thorpej
767517edcc Don't even look for a Domain keyboard if we're not a 400, 425, or 433. 1997-04-27 21:12:43 +00:00
thorpej
5b511796c1 Uniquely identify 345 vs. 375 vs. 400 and 380 vs. 425. Adapted by me from 
code provided by Mike Smith.  While I'm here, convert this file to use
macros from <machine/asm.h> rather than making assumptions about how
C labels are handled.
 1997-04-27 21:10:06 +00:00
thorpej
cdb54b8d65 Update for new SPU options. 1997-04-27 21:06:29 +00:00
thorpej
b27a863b02 Generate option headers for SPU type options, so that we may generate 
dependencies on them.
 1997-04-27 21:05:49 +00:00
thorpej
772891accd Use information provided by <machine/hp300spu.h> to determine if we 
might have a physical address cache.
 1997-04-27 21:02:34 +00:00
thorpej
650df29d4d These files depend on things in <machine/hp300spu.h> 1997-04-27 20:58:55 +00:00
thorpej
18a931000e Use <machine/hp300spu.h> to determine if we might have a physical address 
cache.
 1997-04-27 20:54:39 +00:00
thorpej
544f501d29 Rewrite identifycpu() to be partially table-driven, and to be more 
sane in general.  Since we can uniquely identify each SPU type
(except for 318/319/330, because there _are_ no differences, really),
require that the appropriate SPU option be configured into the kernel.
 1997-04-27 20:53:30 +00:00
thorpej
d62e267a0c Uniquely identify 345 vs. 375 vs. 400 and 380 vs. 425, useful when 
looking for hp4xx-only hardware during autoconfiguration.  Code
mostly from Mike Smith, with a few changes by me.
 1997-04-27 20:50:01 +00:00
thorpej
9d4c5c9106 - Define new machineid and mmuid constants. 
- Rearrange slightly.
 1997-04-27 20:47:23 +00:00
thorpej
fc79829eab #include <machine/hp300spu.h> 1997-04-27 20:43:37 +00:00
thorpej
80ff0f0f2f - Move a bunch of hp300 SPU-specific stuff out of cpu.h, and place it 
in new hp300spu.h file.  This includes the code that associates
  model number with cpu/mmu/cache combinations.
- Define "mmuid" constants for several models, derived from code
  provided by Mike Smith.
 1997-04-27 20:37:07 +00:00
thorpej
f5ccce44c8 srt0.s renamed to srt0.S via repository copy Magick - we want this 
file to be preprocessed before it is assembled.
 1997-04-27 20:29:43 +00:00
briggs
d25f63a7bf What was I thinking? 1997-04-27 19:26:48 +00:00
is
5c869c4c5b The delay loop needs 8 cycles, not 3, on the 68030; so use _that_ value 
for the pre-calibration.
 1997-04-27 18:14:34 +00:00
tls
c96effb7f8 don't allow i386_set_ioperm if securelevel > 1, either (and fix a really silly bug) 1997-04-26 21:33:42 +00:00
tls
3d9b09a843 If in highly-secure mode, don't ever allow access to IO ports 1997-04-26 20:30:02 +00:00
jonathan
26adc250c2 Stop ignoring DCD transitions on ioasic SCC serial lines. 
From entropy@zippy.bernstein.com>.
 1997-04-26 02:26:54 +00:00
jonathan
45d369d5fb Ted Lemon's (RR 3540) cross-compilation patch: use ${SIZE} instead of size, 
Also add "SIZE=? size",  so that native kernel compiles work, too.
 1997-04-26 02:20:59 +00:00
jonathan
55a34d0cf4 Add "options MIPS1" and delete commented-out lines inherited from GENERIC. 1997-04-26 01:35:41 +00:00
gwr
5d8f130bb6 Be quiet in sunmon_abort(). 1997-04-25 22:26:14 +00:00
veego
560544f47c Add a -I${.CURDIR} to build netbsd.sa with a obj dir 1997-04-25 22:17:33 +00:00
leo
7c2c0c846c - Provide the reasonable size that can/may be mmap-ed. This might differ 
from the current frame buffer size.
- The et-driver now allows for a 16Kb io-region (was 8Kb) that can be
  mapped. The XFree server uses this.
 1997-04-25 19:25:38 +00:00
leo
d5733f126e Allow formatting of 360Kb floppies. 1997-04-25 19:18:04 +00:00
leo
d1c63a4095 Changes to bounds_check_with_label: 
- Take care of the case where d_secsize < DEV_BSIZE (From Waldi Ravens)
- Don't check if an unsigned value <= 0. This was the cause of various
  panics when savecore(1) was reading at bogus offsets because the crash
  dump format was changed. Thanks to Jason for unintently revealing the
  bug ;-)
 1997-04-25 19:16:30 +00:00
gwr
92efd222d0 Hack for now: if fpu is there, report it as an mc68882. 1997-04-25 19:16:10 +00:00
gwr
dfc2e0f30d Add devices: idprom, oclock. Add file leds.c. 1997-04-25 19:12:57 +00:00
gwr
1676e3a212 Add idprom, oclock, and rearrange somewhat. 1997-04-25 19:11:48 +00:00
leo
ef06b2e55d Use info on the actual RAM-segments to check the validity of /dev/mem 
accesses. Reformat some comments.
 1997-04-25 19:07:45 +00:00
gwr
cd6ee4737e Renamed obio_alloc/obio_mapin. 1997-04-25 19:02:46 +00:00
gwr
678f326c7f Make idprom_init() figure out where the IDPROM resides. 
Add autoconfig glue, which currently just shows the
address where the IDPROM was found.
 1997-04-25 19:02:06 +00:00
gwr
49f095a342 No longer need eeprom_init(). 1997-04-25 18:57:49 +00:00
gwr
e2ca51d2a1 Fixed OBIO locators bug, so this can do the normal thing now. 
Also improved handling of spurious interrupts, and sync up with
the sparc version.
 1997-04-25 18:55:05 +00:00
gwr
4c64781365 Moved the actual model identification work into identifycpu(), 
which used to be in _startup.c:_verify_hardware().
Use sunmon_abort() in cpu_reboot() for now...
 1997-04-25 18:52:11 +00:00
gwr
9ae1b69554 Rename "Idle" to "_Idle" so clock.c can reference it. 1997-04-25 18:48:39 +00:00
gwr
25d7b6b348 Provide a struct representing only the clock. 1997-04-25 18:47:41 +00:00
gwr
885a1fc439 Add /dev/leds support. 1997-04-25 18:46:10 +00:00
gwr
2dd63e3378 Use bus_mapin, and minor tweaks. 1997-04-25 18:45:39 +00:00
gwr
f7749f47fe Add "oclock" support for the Sun3/470. 1997-04-25 18:31:37 +00:00
gwr
8f9276c727 Move the _verify_hardware() work into machdep.c:identifycpu(). 
Call obio_init() after _vm_init() so we can use peek_byte() to
probe the idprom.  Be quiet in _save_symtab() when all is well.
 1997-04-25 18:29:58 +00:00
gwr
b41693a034 Moved bus_scan() and bus_print() from here to bus_subr.c 
because they are now different in sun3 vs sun3x.
 1997-04-25 18:07:36 +00:00
gwr
477a320277 Moved bus_scan() and bus_print() here from autoconf.c 
because they are now different in sun3 vs sun3x.
 1997-04-25 18:02:47 +00:00
gwr
3619a4da3d Rename obio_alloc to obio_mapin, update *_init declarations. 
Remove OBIO_P4_REG (not a fixed address, but per-device).
 1997-04-25 18:00:49 +00:00
gwr
fd42058232 Add kernel function declarations. 1997-04-25 17:52:54 +00:00
gwr
23f821bcc5 Add idprom_init(). 1997-04-25 17:42:34 +00:00
gwr
963c410824 Change the OBIO address probe order a little, to allow determination of 
which kind of idprom and clock we have (3/80 vs 3/470 differences).
Fix a bug where obio_submatch() failed to copy some locators.
Rename obio_alloc to obio_mapin.  Add lots of comments.
 1997-04-25 15:35:27 +00:00
gwr
20ac88e4b1 Add /dev/leds support for the 3/470. 1997-04-25 15:15:45 +00:00
briggs
e861dbf8c7 A couple of performance improvements from Denny Gentry <denny1@home.com> 1997-04-25 03:40:09 +00:00
thorpej
1a3df57d81 Preprocess .s files before assembling them. 1997-04-25 02:26:48 +00:00
thorpej
ffd2acc819 Minimalistic changes to not make assumptions about how C labels 
are handled.  This file's ".s" counterpart must now be pre-processed.
 1997-04-25 02:26:04 +00:00
thorpej
dcca39c0f1 - Don't make assumptions about how C labels are handled. 
- Use PANIC() where appropriate.
 1997-04-25 02:22:00 +00:00
thorpej
d8b7a2f037 - Don't make assumptions about how C labels are handled, use macros 
provided by <machine/asm.h> instead.
- Adjust for <m68k/m68k/trap_subr.s>
 1997-04-25 02:17:38 +00:00
thorpej
722ead42ca Don't make assumptions about how C labels are handled, use macros 
provided by <machine/asm.h> instead.
 1997-04-25 02:16:38 +00:00
thorpej
48fa8154e7 Adjust for <m68k/m68k/trap_subr.s> 1997-04-25 01:49:05 +00:00
thorpej
82f0cd80c8 Sync with hp300. 1997-04-25 01:46:53 +00:00
thorpej
6649814b0e Machine-dependent crash dump header - just pulls in <m68k/kcore.h> 1997-04-25 01:45:31 +00:00
thorpej
6df98d905a Sync with hp300: 
- Clean up cpu_startup()
- Convert to new crash dump format.
 1997-04-25 01:44:09 +00:00
thorpej
1b1a15a81f Adjust for <m86k/m68k/trap_subr.s> 1997-04-25 01:42:51 +00:00
thorpej
6a00a9aa75 Adjust for <m68k/m68k/trap_subr.s> 1997-04-25 01:35:45 +00:00
thorpej
deb6fcac46 Add several common trap vector stubs. Not all of them are here (yet), 
but several are.
 1997-04-25 01:33:18 +00:00
cgd
f5bfcda4fe comment out text after #endif. "Pasto!" pointed out by Kevin Sullivan 
in PR 3529.
 1997-04-24 23:58:24 +00:00
thorpej
658ce85446 Add several useful macros: 
- ENTRY_NOPROFILE() and ASENTRY_NOPROFILE(), similar to ENTRY() and
  ASENTRY(), but without the profiling prologue.
- GLOBAL()/ASGLOBAL() and LOCAL()/ASLOCAL(), for defining global and
  local variables with C and ASM labels.
- BSS()/ASBSS(), for defining items in the BSS segment, with C or
  ASM labels.
- PANIC("panic message") - shorthand for calling panic() from assembly code.
- VECTOR(), ASVECTOR(), VECTOR_UNUSED - shorthand for defining entries
  in the vector table.
Also, change RCSID() to pad out the string to even boundary.
 1997-04-24 22:49:39 +00:00
gwr
021d7c2fe5 s/REGENTS/FOUNDATION/ 1997-04-24 22:37:11 +00:00
mycroft
bba8a4e9b1 GC. 1997-04-24 20:45:53 +00:00
mycroft
0b3a169a66 This is no longer needed. 1997-04-24 20:45:10 +00:00
thorpej
5fddec888d Add Domain keyboard support source files (oops, forgot to commit this 
file when I integrated the rest of it!).  Fixes PR #3523, from
Thorsten Frueauf <frueauf@ira.uka.de>.
 1997-04-24 20:39:26 +00:00
scottr
72111ae284 From Mycroft: If we fail to allocate a cluster to hold a large packet, 
simply drop it rather than using a chain of tiny mbufs.
 1997-04-24 16:52:05 +00:00
veego
3b947046b9 remove a duplicated : for the install rule. 1997-04-24 08:22:45 +00:00
is
412fee4faf - Minor brushup of man page. 
- Try to fix manual installation.
 1997-04-24 08:02:29 +00:00
mycroft
d1c236b1e4 Tweak to improve performance slightly on very small packets and headers. 
Suggested by Jonathan Stone.
 1997-04-24 05:46:20 +00:00
cgd
1b16a9577a note eb64+/AlphaPC 64 support 1997-04-23 20:52:45 +00:00
scottr
c1c244e0a3 Implement new crash dump format. Mostly taken from hp300, extended to 
support multiple physical RAM segments by me.  Garbage collect functions
obsoleted by this change.
 1997-04-23 19:15:16 +00:00
briggs
ed74d8f653 On video devices, if we're loading non-default resource values, save the 
resource ID.  Thanks to Paul Goyette <paul@pgoyette.bdt.com> for spotting
this one.
 1997-04-23 13:37:43 +00:00
scottr
9784a8d3cf space vs. tab formatting, KNF 1997-04-22 21:00:04 +00:00
scottr
b09a74e8b9 If we can't map SONIC register space, just print a diagnostic message and 
continue.  Make formatting consistent and KNF while here.
 1997-04-22 20:56:15 +00:00
scottr
a653364eb1 Remove obsolete drsw diagnostic; make formatting consistent; KNF. 1997-04-22 20:32:08 +00:00
scottr
65673e5ec2 Make formatting consistent; inline -> __inline__; KNF. 1997-04-22 20:26:24 +00:00
scottr
2111f73a90 Remove obsolete drsw diagnostic 1997-04-22 20:21:59 +00:00
scottr
e733e600d3 #ifdef DIAGNOSTIC, print the NuBus type fields for unrecognized boards. 1997-04-22 20:20:32 +00:00
briggs
d13ebf0e72 Make sure that we unmap the space that we mapped if we can't identify the card. 1997-04-22 13:45:38 +00:00
mrg
00e4b647d9 add NetBSD RCSid. 1997-04-22 07:48:31 +00:00
briggs
d2a4acec48 Initialize 'offset' to make gcc happy. 1997-04-22 03:04:33 +00:00
ragge
749ec2a297 Remove #define LEDEBUG. 1997-04-21 22:04:23 +00:00
mycroft
58fbadc150 Put back missing label. 1997-04-21 18:20:14 +00:00
scottr
43f35682d8 April 18 update from John Wittkoski, w/space vs. tab KNFing by me. 1997-04-21 18:04:28 +00:00
matthias
9dd24ca41c * Print device information in md_attach_hook. Since this is a pseudo device, 
we have to output the device name as well.
 1997-04-21 16:20:39 +00:00
matthias
237a4629c4 * Somehow the definition of ieee_handler_disable got lost in my last commit. 1997-04-21 16:17:44 +00:00
matthias
7bc1bddc96 * Do not use Debugger(), use breakpoint(). 1997-04-21 16:17:39 +00:00
matthias
16fbb47606 * Do only check against sd* if we did boot from a scsi disk. 1997-04-21 16:17:36 +00:00
matthias
34bffb6338 * Omit Debugger() if DDB is not defined. 1997-04-21 16:17:32 +00:00
matthias
7fe4f42d0a * Add a comment describing movs[bdw]nu. 
* Add breakboint().
 1997-04-21 16:16:31 +00:00
matthias
5a23a56e6b * Apply a fix from Ian Dall to reduce system load and fifo/silo overflows 
when running at high speeds. This works by using hardware RTS again and
  using the receive fifo threshold as well as the transmit fifo provided
  by the sc26c92.
 1997-04-21 16:16:16 +00:00
matthias
61d420149f * Use higher timeouts for PLIP. 
* Do some cosmetic reformatting.
* If the interface is considered dead, re-enable receive interrupts from
  time to time.
 1997-04-21 16:16:10 +00:00
matthias
12994a40ab * Configure ncr*, not ncr0. That allows the user to specify flags for ncr0 
in his config file.
 1997-04-21 16:15:59 +00:00
cgd
fc8625db4f minimal changes to make hacked wd driver work under a different name. 
It is configured (in config files) as 'awdc'/'awd', but shows up as
'wdc'/'wd', so that a minimal amount of code had to be modified to make
the name change work.  This is only intended to be temporary, anyway.
 1997-04-21 14:47:33 +00:00
cgd
dbb6d5170a Convert to use bus_space_* macros. The resulting code isn't quite correct, 
but works on the alpha (since there's currently no floppy drive support).
 1997-04-21 13:38:38 +00:00
cgd
ba574b6f20 copy ISA wd driver from /sys/dev/isa/wd.c. This rev is unmodified. 
This is temporary, until /sys/dev/isa/wd.c is properly bus_space_*-ified.
 1997-04-21 13:36:49 +00:00
mycroft
89205383a7 Make microtime() work correctly with a 4- or 4C-only kernel. 1997-04-21 05:47:08 +00:00
briggs
fc43576e8d Take care when accessing DAFB--some quadra-class machines use other video 
controllers.
 1997-04-21 00:57:59 +00:00
scottr
7e00ee54d8 Fix itematch() for case where fboff is not zero. From Dave Huang, 
closes PR #3501.
 1997-04-20 20:31:18 +00:00
pk
43445623fd mmmmap() should return -1, not ENOTSUP. 
Optimize the case of writing to /dev/zero etc, like the alpha,i386 and m68k.
 1997-04-19 21:28:53 +00:00
ragge
400a0d512e Fix so that these files compile again after recent ARP changes. 1997-04-19 15:02:27 +00:00
ragge
12c25bea6a Update floppy boot files to confirm to new boot syntax. 1997-04-19 11:42:14 +00:00
jonathan
902f17af37 Correct sense of fi->fi_blanked in bt455_video_off(). 1997-04-19 08:25:31 +00:00
jonathan
22d3d9819d Correct sense of FBIO[GS]VIDEO ioctl as noted in PR #3515. 1997-04-19 08:21:59 +00:00
thorpej
2c8a12761d Fix a pasto 1997-04-19 01:47:24 +00:00
thorpej
a53aafbf1a ...and again. 1997-04-19 01:40:08 +00:00
thorpej
28d3c2e2d3 Replay history.... 1997-04-19 01:39:44 +00:00
pk
d3028dd4a2 * Remove ineffective and incorrect cache flush in viking_mmuenable(). 
* A bunch of cosmetic changes.
 1997-04-18 19:51:05 +00:00
cgd
09b3ed084e add ep at isa and eisa 1997-04-18 19:41:17 +00:00
ragge
df9ea3ac75 Updates to KA43 (VS3100/76) support by Bertram Barth. 1997-04-18 18:53:38 +00:00
ragge
36b2211d14 Change implementation of TOY clock addressing. Gives us only 
one routine for three different HW implementations.
 1997-04-18 18:49:33 +00:00
scottr
19c5134870 For small writes, use PIO rather than PDMA; this increases the reliability 
of the driver (considerably, on some systems/targets).
 1997-04-18 17:38:08 +00:00
briggs
bbc4732fd1 Fix from Denny Gentry--initialize receive seq. counter. 1997-04-18 12:04:37 +00:00
scottr
11b0f01c17 Make sure to call adbop() on Power Manager-based machines. Noted by 
Takashi Hamada, who also provided the code I based this modifcation
on.
 1997-04-18 05:53:41 +00:00
cgd
0f72c17209 try a slightly different heuristic to differentiate between ISA and PCI 
(VGA) consoles.
 1997-04-18 00:07:53 +00:00
thorpej
84c938e92f oops, forgot to commit this yesterday - newvers.sh script for boot 
program.
 1997-04-17 20:24:17 +00:00
thorpej
d76ea6e415 STRIP -> STRIPFLAG 1997-04-17 07:46:24 +00:00
thorpej
5b733ef58a STRIP -> STRIPFLAG 1997-04-17 07:41:05 +00:00
thorpej
6a2fc99857 STRIP -> STRIPFLAG 1997-04-17 07:27:46 +00:00