ade7d52fc9
Use powerpc/pmap.h.
1999-06-22 01:38:38 +00:00
17900cd587
Install ofwboot.elf in addition to ofwboot. Thanks to Chris for info on how
...
to do this!
1999-06-21 21:02:49 +00:00
7a71d12944
Clear out 'struct fbdevice' cruft; it represents bitrot on pmax. Add
...
#ifdef'ed out definition for 'struct fbsoftc', the way forward until
[nisimura-pmax-wscons] is merged.
1999-06-21 19:21:10 +00:00
5756acc071
Cleanup.
1999-06-21 19:07:16 +00:00
ebd6cdffde
px.c requires that qvss_compat.c be compiled in.
1999-06-21 19:05:10 +00:00
51dd099ef5
Add DW750 and DW780 UBA support. Still no BDP's.
1999-06-21 16:23:00 +00:00
5425cbfa41
fix compile warning.
1999-06-21 06:28:33 +00:00
0c32dc3995
Clean this up a little. More later. (Gee, this really is becoming a huge
...
mess).
1999-06-21 01:44:14 +00:00
d326454d31
Nuke the last vestiges of a single DVMA map. This stuff should have
...
been completely migrated to individual bus drivers.
1999-06-21 01:42:36 +00:00
2c283b9149
Improve TRAPTRACE. Add FLTRACE that only tracks MMU faults.
1999-06-21 01:39:06 +00:00
e0972091c8
Remove moved files.
1999-06-20 18:19:19 +00:00
e410ff8410
Remove moved files.
1999-06-20 18:14:41 +00:00
ccd6d1e4de
Pull in `sparc/compat_13_machdep.c' for SunOS compatibility too.
1999-06-20 16:30:22 +00:00
7a1f65022b
Remove more unused files.
1999-06-20 16:00:14 +00:00
dc65be6cf8
Remove old unused files.
1999-06-20 15:58:28 +00:00
92a492d5ea
Some 4000/200 fixes. From Michael Kukat.
1999-06-20 15:52:51 +00:00
459e986089
PR7812: add prototype for sysarch(); fixes build of libarch on arm32
1999-06-20 15:26:38 +00:00
8d68bdec13
Make this compile.
1999-06-20 05:42:09 +00:00
9ac9208434
More initialization by calling Open Firmware.
1999-06-20 05:34:59 +00:00
e2fb3def19
Use uba & mscp code from dev/ instead.
1999-06-20 01:05:19 +00:00
cc43d8766d
Fix of bug causing allocations of physical pages to get wrong address.
1999-06-20 00:59:54 +00:00
abcc4bed15
Better support for tracing stacks. From Chuck McManis.
1999-06-20 00:58:23 +00:00
cb57cab9b1
Add macros to round/trunc vax pages.
1999-06-20 00:53:58 +00:00
76b9763345
Add some small checks to avoid unwanted panic types.
1999-06-20 00:53:02 +00:00
9cf2782159
Move the DVMA mapin/mapout functions from the SBus driver to the iommu driver.
1999-06-20 00:51:29 +00:00
069858b1e2
Want device-register, to detect qbus boot devices.
1999-06-20 00:50:08 +00:00
1b5eb98937
Sync with Atari pmap: support for the UVM changes
1999-06-19 19:44:08 +00:00
656c580511
Don't try to disconnect SCSI disks. From Michael Kukat.
...
Now DEC disks also works in VAXstations :-)
1999-06-19 15:46:05 +00:00
0e64bf3732
Add ess to isa attachment. The entry is commented out for the moment since
...
I'm not sure how accurate the probe is.
1999-06-18 20:27:59 +00:00
43ddb951ea
Whitespace police.
1999-06-18 18:11:56 +00:00
471cb34fb2
Fix a harmless typo.
1999-06-18 18:08:52 +00:00
170b78e476
Reduce TLB operations.
1999-06-18 14:27:09 +00:00
1c949badf4
- Fix an large error I made last month in TLB mod improvement, still
...
wondering why not I was immediately blown away.
- Continuing invastigations on VM related panics on very high loads.
1999-06-18 08:17:50 +00:00
bfdcd7d519
Update pmap_check_wiring(). Also attach TNF copyright because of the
...
pmap_remove_mapping() changes.
1999-06-18 07:13:16 +00:00
88f71d9598
Update a uvm_map_pageable() call that I forgot to commit previously.
1999-06-18 06:32:05 +00:00
4eb92b494a
Sync up with the following hp300 trap.c change:
...
revision 1.41
date: 1996/09/07 22:26:50; author: mycroft; state: Exp; lines: +5 -20
Use SIGBUS iff we get a legitimate bus fault. Use SIGSEGV for page protection
violations (per Solaris, SVR4, AIX, Linux, Irix, and SunOS).
...which happens to eliminate an unsafe call to uvm_map_checkprot().
1999-06-18 06:31:01 +00:00
e6455c6e13
less views, and comment them
1999-06-18 05:58:25 +00:00
60bb9923a8
COMPAT_SVR4 does not compile witout EXEC_ELF32
1999-06-18 05:57:31 +00:00
d66f459df0
COMPAT_SVR4 does not compile without EXEC_ELF32.
1999-06-18 05:56:45 +00:00
9e9f068f43
Add the guts of mlockall(MCL_FUTURE). This requires that a process's
...
"memlock" resource limit to uvm_mmap(). Update all calls accordingly.
1999-06-18 05:13:45 +00:00
7c02d93c04
nuke current definitions provided by __BUS_SPACE_COMPAT_OLDDEFS.
...
I'll do what I can to make new ones soon. 8-)
1999-06-18 04:49:24 +00:00
0288ffb53a
pmap_change_wiring() -> pmap_unwire().
1999-06-17 19:23:20 +00:00
f5a527bb4e
Remove pmap_pageable(); no pmap implements it, and it is not really useful,
...
because pmap_enter()/pmap_change_wiring() (soon to be pmap_unwire())
communicate the information in greater detail.
1999-06-17 18:21:21 +00:00
ffd3da04d4
Comply with the changes Jason suggested yesterday. A copy of his log message:
...
Don't (ab)use uvm_map_pageable() to allocate PT pages. Instead, do
some internal reference counting on PT pages. We still allocate them
with the page fault routine (a wire-fault, now), but no longer free
PT pages from pmap_pageable().
1999-06-17 07:59:16 +00:00
a8d7da0d21
Add "trackpad" definition.
1999-06-17 06:59:05 +00:00
f51688dd8b
PMAP_NEW is no longer optional on the pc532; the old pmap's page table
...
allocation strategy no longer works at all. Move pmap.new.* to pmap.*.
To read the revision history of PMAP_NEW up until this merge, use cvs
rlog of the old pmap.new.* files.
1999-06-17 00:22:41 +00:00
e2442268e9
PMAP_NEW is no longer optional on the i386; the old pmap's page table
...
allocation strategy no longer works at all. Move pmap.new.* to pmap.*.
To read the revision history of PMAP_NEW up until this merge, use cvs
rlog of the old pmap.new.* files.
1999-06-17 00:12:10 +00:00
c28bad1d51
Define PowerBook G3 Japanese keyboard.
1999-06-16 08:48:37 +00:00
a2e8151f8e
Define symbolic name of PMU commands.
1999-06-16 07:43:58 +00:00
602d8e81f7
Don't (ab)use uvm_map_pageable() to allocate PT pages. Instead, do
...
some internal reference counting on PT pages. We still allocate them
with the page fault routine (a wire-fault, now), but no longer free
PT pages from pmap_pageable().
1999-06-15 22:18:07 +00:00
857dfc7058
Slightly rearrange the SSIR inline assembly implementation in order to avoid
...
making assumptions on C symbol name prefixes.
1999-06-15 15:26:34 +00:00
51eb26bf7e
Slightly rearrange the inline assembly in ledcontrol() to avoid frobbing a
...
C symbol name prefix.
1999-06-15 15:18:55 +00:00
cd06c07723
Check file open mode at ioctl.
...
Add debug option SRAM_DEBUG_DONTDOIT.
1999-06-15 15:04:56 +00:00
e33c6c47a2
set bb_flags to zero; who knows what they were (on disk) previously,
...
and SRM is finicky about what values it will allow.
1999-06-14 23:55:29 +00:00
b6e1d0f85a
set bb_flags to 0. Though not strictly necessary (because of an earlier
...
memset()), this is nice for source-level consistency.
1999-06-14 23:52:25 +00:00
4df925b8b0
* Don't use DMA interrupt.
...
* Stop DMA explicitly in finish routine.
1999-06-14 08:53:06 +00:00
86d2f9253a
Consider "have /psuedo-hid and only ADB keyboard" case to make the
...
new PowerBook G3 happy.
1999-06-14 06:22:50 +00:00
3cce22665e
Clear interrupts on a Radius GS/C card. Thanks go to Yasufumi Suzuki
...
<yattin@sco.bekkoame.ne.jp> for doing the MacsBug "leg"work for me.
1999-06-14 03:11:34 +00:00
c867e93fae
Pickup changes from ../biosboot/Makefile and add .PATH so that source
...
files are found. Now this compiles ok (have yet to run the blocks).
1999-06-13 17:21:18 +00:00
f90f079222
Fix typo in previous.
1999-06-13 07:58:20 +00:00
fee9cd9c12
Initialize the heap (for alloc()) explicitly, since the `end' symbol
...
is not necessarily double-word aligned.
1999-06-12 12:49:24 +00:00
2b55a690f8
Change the annoying printf() of 2 backspaces to a space followed by a
...
backspace for aesthetics.
1999-06-12 00:21:13 +00:00
35f356e920
Eliminate more overhead in delay(); there's no point in doing the same
...
thing in both the calibrator function and the actual delay function. At
this point, _delay() does all of the work, so we can garbage collect
dummy_delay() and make _delay() an alternate entry point for delay().
This results in a small (but measurable) improvement on the IIci.
1999-06-11 06:51:39 +00:00
0693ec8379
Don't change protection on the clock for the moment.
1999-06-10 04:31:20 +00:00
a6b22a059a
Note that 0x146 is a Micron Exceed.
1999-06-10 03:25:18 +00:00
3ea7c860b9
Use information from Kensuke Ogawa <kenogawa@festa.or.jp> to disable
...
interrupts on a Micron Exceed color video card.
1999-06-10 03:24:38 +00:00
9a3291290d
- Correct errorneous comments for kn*_wbflush() routines.
1999-06-10 01:37:10 +00:00
0ebace83f4
- Remove local declaration of 'dev_name2blk[]'.
...
- Minor formatting correction and removal of defunct debug lines.
1999-06-10 01:06:40 +00:00
6adfba2698
Tweak previous: use shorter instructions to accomplish the same
...
#task (minimizes overhead).
1999-06-09 15:34:08 +00:00
3c26ad58a1
Someone forgot to tell ddb that SWAP is not PEA.
1999-06-09 07:53:22 +00:00
981a3b4cb2
- Nuke declarations for now obsolete implementations of spl()s.
1999-06-09 07:33:29 +00:00
f2f8426e86
Write a new low-level assembly help for delay() and the calibrator. This
...
eliminates stalls during instruction prefetch and makes the delay
consistent regardless of kernel configuration.
1999-06-09 06:59:53 +00:00
f98c8a0394
Remove some register declarations.
...
Get Q9x0 dual SCSI at least basically working.
Be paranoid about some register reads.
1999-06-09 03:41:34 +00:00
2cfd9ef718
Synchronize with GENERIC
1999-06-09 01:47:14 +00:00
6a949cddfb
Include <pmax/pmax/machdep.h> instead of declaring our own prom_*()
...
prototypes.
1999-06-08 23:42:36 +00:00
b7a2ec3f01
+ Call prom_scsiid() to found out what the host SCSI ID should
...
be instead of hardcoding values.
+ Print the host SCSI ID in siiattach().
XXX - probably should do the same to the ASC driver?? The manuals for
the 5000 series machines don't mention the scsiidN env variable, but it
is present on all the models I've looked at...
1999-06-08 23:41:59 +00:00
7eb995464d
+ Add prototypes from the prom_*() callback routines.
...
+ Remove leading "extern" from function declarations.
+ CG unused latched_cycle_cnt.
1999-06-08 23:40:42 +00:00
661c94c72c
+ Add prom_getenv(), and use it instead off calling callv->_getenv.
...
+ Add prom_scsiid() to read and decode the prom scsiidN environment
variable that says what the host SCSI id should be.
+ Include <pmax/pmax/machdep.h> instead of declaring our own
prom_*() prototypes.
1999-06-08 23:40:19 +00:00
30c351dc59
mmm, missing return types make gcc unhappy.
1999-06-08 10:35:36 +00:00
d9be0d6e6b
When entering a read-only page for MIPS3, va does not need to be adjusted by PAGE_SIZE
...
when flushing the I-cache.
1999-06-08 03:44:18 +00:00
540b02ab76
- Exterminate haunted evil soul of MIPS va_arg(). Verified OK with
...
either endian. Not a perfect solution which would be revealed on
a certain condition when va_arg() is applied to magical struct
arguments passed by value. format_bytes() is now saved. With the
help from Noriyuki Soda and Masanari Tsubai.
1999-06-08 00:46:38 +00:00
b7ee76f1c4
Check for an unterminated bus; if yes: complain and don't try to use this
...
bus. Without this, cbiiisc used to hang if nothing was connected.
Reported by Matthias Scheler. A similar method to avoid this was suggested
by Ralph Schmidt.
1999-06-07 21:30:57 +00:00
2c35c7bd03
Only declare block major numbers for devices which can be the root device.
1999-06-07 20:31:27 +00:00
d76e7b8c6d
Don't pass a nam2blk around at all; just have setroot() and friends reference
...
dev_name2blk[] directly. Addresses PR #7622 (ITOH Yasufumi), although
in a different way.
1999-06-07 20:16:08 +00:00
0525e9c7ef
add ebus attachment for cs4231. this does not work as it requires ebus DMA and will panic if enabled. but it should otherwise be mostly complete.
1999-06-07 14:59:14 +00:00
03c5c00056
remove "EXPERIMENTAL" comment from this file, as it does not exist elsewhere in this directory.
1999-06-07 14:43:34 +00:00
a0faadd7d2
Make traces work with ELF kernels.
1999-06-07 07:00:07 +00:00
aa506f8098
copyright foo. in the future, we will all be binary-redistributable.
1999-06-07 05:40:08 +00:00
1884f05354
Another general cleanup:
...
Remove the entire idea of fasttrap interrupts since V9 traps are really cheap,
the CPUs are really fast, and the completely different trap frames would make
these handlers really difficult to implement.
pmap_changeprot() was only used by the clock and one other place; deprecate it.
probeget() and probeset() now take 64-bit addresses even in 32-bit mode so we
can probe IO locations by physical addresses.
Some pmap cleanup.
Some more copyright cleanup.
1999-06-07 05:28:03 +00:00
92568502cc
Don't dump all the registers at attach time. It works now.
1999-06-06 19:58:31 +00:00
3bb07459cd
Qbus moved in hierarchy.
1999-06-06 19:10:49 +00:00
4756cdb74e
Page tables must be pageable (where did this disappear?)
...
Fix a PGOFSET -> VAX_PGOFSET miss.
Stylistic cleanup.
1999-06-06 19:09:50 +00:00
f6d61fb6ad
Add an define for SGMAP.
1999-06-06 19:07:11 +00:00
d55d0ac4b5
Add some more IPRs for KA670.
1999-06-06 19:06:29 +00:00
db8c0c9c4c
Use gencons if VAX670.
1999-06-06 19:05:21 +00:00
02637941e5
Ops, forgot this file.
1999-06-06 19:03:40 +00:00
134b65fde3
Added support for Qbus DMA.
1999-06-06 19:00:53 +00:00
64572c459a
sgmap files, copied from the Alpha port and modified for vax.
1999-06-06 18:58:50 +00:00
a775803f91
Add adv* entry under ISA devices
1999-06-06 17:57:23 +00:00
600618d977
Add adv0 entry under ISA SCSI controllers
1999-06-06 17:37:03 +00:00
22b5707362
Complete support for VAX 4000/300. From Michael Kukat. (michael@camaronet.de)
1999-06-06 14:23:46 +00:00
e6545647b8
update this to GENERIC rev. 1.84.
1999-06-06 13:06:38 +00:00
440ac638bc
EXEC_ELF32 is now default. enable "rnd"
1999-06-06 13:00:03 +00:00
3d9ae33cfb
Add COMPAT_AOUT.
1999-06-06 09:23:38 +00:00
1428ee6103
note that this is for an u5/u10 (from hubert)
1999-06-06 01:59:40 +00:00
186c3eba2b
Make pbrobeget() and probeset() work for 64-bit values as well. To do this
...
the arguments are changed so the address is first and the ASI second so we
can have the address in %o0:%o1 and not worry about unused registers.
Also a bit of copyright cleanup.
1999-06-05 21:58:16 +00:00
3f7968c736
Fix things up so they compile again.
1999-06-05 20:38:24 +00:00
fcd1351b4c
Add support for AOUT_COMPAT which we'll need when the sparc port cuts to ELF.
1999-06-05 16:55:19 +00:00
14729c2b11
fix lpt definition. add audiocs at ebus. oops, comment broken/incomplete devices.
1999-06-05 14:34:22 +00:00
5e7ea137ac
lpt at ebus. fixed (commented) audiocs attachments.
1999-06-05 14:33:44 +00:00
756c2cde22
map in registers, configure interrupts, this appears to be OK now.
1999-06-05 14:19:44 +00:00
e8ccf39491
find ebus device INO values via the "interrupt-map" and "interrupt-map-mask" properties.
1999-06-05 14:18:26 +00:00
bea0725979
64-bit safety to help sharing compilation tools with sparc64.
1999-06-05 12:54:23 +00:00
3670ff2120
if we are not bypassing the MMU, use ASI_PRIMARY. map PCI memory space
...
non-cached. XXX clean this up by looking at the "non-cacheable" bit of
the full physical address.
avoid having 'nbuf' change between calls to `mdallocsys()' by setting it
in mdallocsys() like the MI allocsys() does. XXX fix this too!
fix some printf lossage.
update for probeget() changes -- though bus_space_probe() appears to be
unused on the sparc64.
1999-06-05 05:35:41 +00:00
cc49a21298
use common IOMMU routines.
1999-06-05 05:30:43 +00:00
da265ce045
clean up a bit, implement pci_conf_{read,write}() with probe[gs]et() (not yet used).
1999-06-05 05:29:50 +00:00
bb0d20abc4
lpt at ebus.
1999-06-05 05:28:36 +00:00
dcb6040569
clock attaches at ebus also now.
1999-06-05 05:10:01 +00:00
4b183190b5
update a commit slightly.
1999-06-05 05:08:25 +00:00
2ee43849be
catch up with probe[gs]et() changes.
1999-06-05 05:05:11 +00:00
6dffb4dff3
use common IOMMU code for sbus and psycho (remove old iommu
...
device declaration, we don't want it). pull in psycho, pci,
ata and pciide code. clock, eeprom and power attach at sbus
*and* ebus now (clock as `eeprom' works). add `lpt at ebus',
wd major number, and other misc. commented devices.
1999-06-05 03:16:19 +00:00
cb392a2e1d
only set LD to /usr/ccs/bin/ld if it exists. avoid empty SFILES.
1999-06-05 01:56:31 +00:00
6b1b77230e
in switch_exit() previous fix forgot to guard REG_L with REG_PROLOGUE and
...
REG_EPILOGUE.
1999-06-04 21:43:12 +00:00
4f92d800a6
fix register restore to be safe for n32 model.
1999-06-04 21:35:47 +00:00
c5e6001ff8
Don't built most of the PCVT utilities; they're not really useful anymore,
...
now that we're trying to deprecate PCVT.
XXX Leave the fonts and font editor for now.
1999-06-04 18:47:33 +00:00
8613b8e39d
avoid uninitalised pointers.
1999-06-04 14:29:38 +00:00
3a1472c106
GC dead code. add support for PCI and IDE boot device detection, plus hme ethernet.
...
should probably deal with be and qe, etc.
1999-06-04 14:00:38 +00:00
27bcf2b6e6
eek. sparc can have "wd" disks.
1999-06-04 13:58:59 +00:00
e40f54e4c7
GC dead code. add support for PCI and IDE boot device detection, plus hme ethernet.
...
should probably deal with be and qe, etc.
1999-06-04 13:55:37 +00:00
a1c100d377
clone this from the macppc port.
1999-06-04 13:51:57 +00:00
10b89455c7
pull iommu code out of sbus code so we can share it with the PCI.
...
define some useful PCI macros.
1999-06-04 13:48:48 +00:00
03adf4aad1
PCI driver for the UltraSPARC. this only works on the Ultra5/10 machines
...
(`SUNW,sabre') for now, and it doesn't really quite work there yet anyway.
the bus space/dma code is cloned from the sbus driver. the IOMMU code also
is cloned from the sbus code, but separated out into iommu.c so that we can
share it with the sbus driver. hopefully, much of the bus space/dma code
can also be re-shared with the sbus driver and the ebus driver but for now
these copies will do.
support for the real UltraSPARC PCI (`SUNW,psycho') is unwritten, though
most of this code is shared with it.
we can probe PCI config space and try to configue devices, but interrupts
don't work yet...
1999-06-04 13:42:14 +00:00
36237341af
support for the UltraSPARC `Extension Bus' found on the PCI machines.
...
the bus space/dma code is cloned from the PCI code to do the same thing
which itself was cloned from the sbus bus space/dma code. the bus dma
code is non-functional at this point.
1999-06-04 13:29:13 +00:00
da08612af4
Ultra5/10 configuration.
1999-06-04 13:22:22 +00:00
bb1f24c221
Could not be compiled without PMAP_NEW.
1999-06-03 14:43:33 +00:00
9a13ebbdd6
Fix lossage with the et4000. Some of the CRT registers were incorrect.
1999-06-03 12:38:50 +00:00
beb12b13b9
Don't include rasops and wscons goo here; we get it already from fbvar.h
1999-06-02 23:24:00 +00:00
59b9a4fb42
Handle RDB ( Amiga native ) partition tables on media with more than 512
...
bytes/sector correctly.
Submitted by Detlev Würkner in PR 7653.
Non-trivial merge with the 1.30 change (handle old AmigaDOS fs partitions
with simulated sectorsperblock > 1) by myself, so don't blame Detlev if it
doesn't work (but I checked!).
XXX We still need changes to the file systems to handle this fully.
1999-06-02 21:09:03 +00:00
7d5f7a34ce
We need wsdisplayvar.h for this to compile now!
1999-06-02 12:11:39 +00:00
80054d22ac
Fix copyright (exchange clauses 3 & 4).
1999-06-01 14:27:39 +00:00
eb947d2f5e
Fix bug in event signalling.
1999-06-01 09:34:06 +00:00
6e95146ec6
Raise spl4 to spl6 in quick_go--I am now experimenting with interrupts
...
at levels 5 (sound) and 6 (via1) on the Q700.
Raise splhigh when reading and writing registers. It's ugly, but it's
possible that we could get an interrupt from a higher-level source,
causing a re-read of a memory location.
This is not good when, e.g., we are reading the interrupt register.
Ideally, we would just raise splhigh() in ncr53c9x_readregs(), but
that's m.i. code, and it may be expensive on other architectures.
1999-06-01 03:40:12 +00:00
29fe96b920
Guard inclusion of opt_cputypes.h.
1999-06-01 03:37:02 +00:00
0ff3b0087c
Add debug routine as an aid for working out new nubus cards.
1999-06-01 03:26:42 +00:00
5e28d99a64
Add define for another ethernet card (Nuvolink)
...
Wrap nubus_scan_slot() prototype in #ifdef DEBUG.
1999-06-01 03:24:25 +00:00
42d2797a4b
Comment out DISABLE_EXT_CACHE as it unnecessarily slows down the IIci
...
w/ external cache. Leave it around so the paranoid can still use it.
1999-06-01 02:39:25 +00:00
a1c15a355d
In PMAP_ISACTIVE(), check curproc->p_vmspace before using it; this macro
...
is called early in startup, long before p_vmspace is set to &vmspace0.
1999-05-31 20:49:00 +00:00
4313734913
After some banging and bashing, commit Jason's new ddb traceback code.
1999-05-31 20:42:15 +00:00
604d4df11e
Convert to traceback-friendly temporary labels.
1999-05-31 20:40:23 +00:00
ffc59b8bad
Be less drastic when running strip(1) on DEBUG=-g kernels.
1999-05-31 18:54:49 +00:00
sakamoto