Commit Graph

18993 Commits

Author SHA1 Message Date
mycroft d0742ed3c5 Several things:
1) If we get an unexpected disconnect, issue a REQUEST SENSE, as recommended
by the SCSI-2 spec.  If the target created a contingent allegiance condition,
this will clear it.  Also, if it happened while sending a SDTR or WDTR message,
disable negotiation for that target.
2) Since some lame devices still don't deal correctly, make sure we deassert
ATN if our last message out is interrupted.  If we get a MESSAGE PARITY ERROR,
we'll reassert ATN anyway.  This should ensure that we never have to send a
MESSAGE NO OPERATION.
3) Set AIC_ABORTING only when actually sending a BUS DEVICE RESET or ABORT,
so we get better error detection.
4) Other internal reorganization of no consequence.
1996-04-01 07:24:37 +00:00
scottr 47b5b87d08 Clean up, primarily KNF. 1996-04-01 06:39:17 +00:00
pk 60188b5bd2 Catch up with system PMAP. 1996-04-01 06:27:54 +00:00
scottr bbb6251b0b Make CLK_INTERVAL depend on HZ; update comments to match. 1996-04-01 05:16:52 +00:00
scottr 5f20924433 Move the I/O map kludge for II-class machines into the MRG init code. No
effective change, but this is more correct and averts some potential
problems.
1996-04-01 04:30:23 +00:00
briggs 1fa595a2f1 Clean up for higher warning level. 1996-04-01 01:35:48 +00:00
pk 25b6ab86fe Update device list. 1996-04-01 00:20:31 +00:00
pk 3cf5ed1f30 Various cleanup; mostly trailing spaces/tabs. 1996-03-31 23:44:53 +00:00
pk 8347d3f584 The sun4m name for this device is "auxio". 1996-03-31 23:43:21 +00:00
pk 301e978cf8 Various sun4m-related changes straight from Aaron. 1996-03-31 23:38:29 +00:00
pk e0344005b4 Changes to match 4m-aware locore.s. Note changed `trapbase' and
`sparc_interrupt_{44c.4m}' labels.
1996-03-31 23:35:20 +00:00
pk e743dda2a9 Update for 4m locore.s. Remove some items, since locore.s now #includes
param.h
1996-03-31 23:28:03 +00:00
pk cf4009e56c Include support for sun4m timers and counters: hardware register layout
has changed. There is a counter for each CPU (currently, just one supported).
1996-03-31 23:24:59 +00:00
pk 8de1fd22bc Add sun4m support. Many, many changes..
Note on multi-architecture kernels: most architecture dependent traps are
forked off directly by having a separate trap table for each (sun4,sun4c,sun4m)
This reduces the number of tests on `cputyp' significantly, and it's easier
on the mind. There are still some entry points that need to test `cputyp',
most notably microtime().. yuck.
1996-03-31 23:19:22 +00:00
pk 6d76616bf0 The sun4m IOMMU "device". 1996-03-31 23:09:32 +00:00
pk 7b55c61c27 Include support for sun4m timers and counters: hardware register layout
has changed. There is a counter for each CPU (currently, just one supported).
1996-03-31 23:07:59 +00:00
pk df2a68f25b Add sun4m VA definitions (for clock and interrupt).
Make these generic for all architectures (i.e. avoid `#if defined(SUN4*)'s).
1996-03-31 23:03:31 +00:00
pk ff109a19a3 Accomodate sun4m CPUs and caches (Aaron). Cache handling needs to be
revised.
1996-03-31 23:00:40 +00:00
pk df08150880 Updates for sun4m. 1996-03-31 22:57:09 +00:00
pk 35802bf06b Accomodate sun4m device layout (v3 PROM) and special 4m startup goo.
Heavily cleaned from most `#if defined(SUN4*)' stuff in favour of
CPU_ISSUN4* macros.
1996-03-31 22:55:33 +00:00
pk 678fbb4806 sun4m device definitions and bus structure. 1996-03-31 22:52:38 +00:00
pk b2d12c182b Aaron's sun4m-related trap handling (memory faults).
Note that fork() no longer return in syscall() in the child.

Add a couple of `#ifdef DEBUG's.
1996-03-31 22:51:58 +00:00
jtk 55960650f7 add include file for function prototypes 1996-03-31 22:50:11 +00:00
pk d15de4b09b Initialize DMA addresses properlier in cpu_startup().
dumpsys(): do not try to dump first physical page, this fucks up pmap_extract().
1996-03-31 22:48:24 +00:00
pk 3c89c0901d Aaron's SRMMU/sun4m pmap implementation.
Note: multi-architecture kernels use function pointers for several PMAP
entry points.

Cut down the `#if defined(SUN4*)' mess severely by using CPU_ISSUN4*
macros (see machine/param.h) wherever possible.
1996-03-31 22:42:59 +00:00
pk 2f3ec93250 Various cleanup; mostly trailing spaces/tabs. 1996-03-31 22:38:29 +00:00
pk 66539854e8 Diddling of the `power' bit in machine control register (Aaron). 1996-03-31 22:34:40 +00:00
pk 954302ae70 Handle sun4m `ledma' and `espdma'.
DMA setup requires services of the IOMMU.
1996-03-31 22:32:45 +00:00
pk 7e45b3449f Re-arrange a bit to stear clear of gcc `-Wall' bites. 1996-03-31 22:30:50 +00:00
pk 25aa56ae29 Handle sun4m `obio' devices.
Cleanup `#if defined(SUN4*)' mess.
1996-03-31 22:28:38 +00:00
pk 96f175d477 Add sbus_translate() to deal with slot address translations, including
new v3 PROM style "range" properties. This routine is also called from obio.c
(which can be viewed upon as just a another Sbus slot).
1996-03-31 22:27:15 +00:00
pk 345295ff08 Deal with sun4m le/ledma. Work around spurious interrupt problem (aaron). 1996-03-31 22:22:52 +00:00
pk bb0897eec6 Various cleanup; mostly trailing spaces/tabs. 1996-03-31 22:21:21 +00:00
pk 1c8b452515 Add a non-interrupt-priority-lowering version of SPL(). Used by splpmap()
at the moment.
1996-03-31 22:20:14 +00:00
pk 76404b0bbe Add sun4m CPU/MMU identifications. 1996-03-31 22:18:16 +00:00
pk 5703b83d00 Move autoconf subroutine prototypes to autoconf.h
Deal with sun4m `software interrupts'.
1996-03-31 22:17:14 +00:00
pk 37a292dea0 Make prototyping of most functions conditional on `PMAP_EXCLUDE_DECLS'.
This allows MD pmap.h to implement access to those functions via pointers.
1996-03-31 22:15:32 +00:00
pk 8d9fbd0949 Add some v3 PROM glue.
Move autoconf subroutine prototypes here (from cpu.h).
1996-03-31 22:12:34 +00:00
pk 0ab4a52862 Add SRMMU/sun4m support. Size and layout of `struct pmap' has changed. 1996-03-31 22:09:13 +00:00
pk 922201e06a Add SRMMU/sun4m definitions.
Cleanup (i.e. mostly delete) `#if defined (SUN4*)' in here.
1996-03-31 22:06:55 +00:00
christos e86e25ba54 Add __kprintf__ attributes to the printf functions 1996-03-31 21:43:17 +00:00
christos 9352303b25 Recognize %p formats so that ps/m works in ddb 1996-03-31 21:42:42 +00:00
christos 96776962f7 Fix PR2277; make lkmenodev() be the same type as enodev(). 1996-03-31 21:40:27 +00:00
christos ccf759ccff Change printf attributes to __kprintf__ attributes 1996-03-31 21:37:15 +00:00
christos c119793167 - Make lkmenodev() be the same type as enodev.
- Change printf attributes to __kprintf__ attributes.
1996-03-31 21:36:40 +00:00
christos c7f6ddc626 make emalloc and erealloc return void *; closes PR/2276 1996-03-31 21:30:05 +00:00
mycroft 004f5e8d39 Fix a slight biff with 16-bit DMA in last change. 1996-03-31 20:51:43 +00:00
scottr 8b3fdd9943 Make this compile again by including <sys/types.h> 1996-03-31 14:21:03 +00:00
mrg 3fd9d3347e use local copy of resolver, with -DDEBUG. fixes pr#2275 (ken hornstein) 1996-03-31 13:55:34 +00:00
scottr 755cc6f914 getopt() returns -1 when the arg list is exhausted, not EOF. Ironic
that it should show up here...
1996-03-31 04:58:32 +00:00