hvozda [Mon, 15 Jan 1996 00:08:50 +0000 (00:08 +0000)]
Register defs for Intel 82365 PCMCIA controller
hvozda [Mon, 15 Jan 1996 00:05:03 +0000 (00:05 +0000)]
ISA <-> PCMCIA glue for PCMCIA support.
hvozda [Sun, 14 Jan 1996 23:10:30 +0000 (23:10 +0000)]
New devices for PCMCIA support
briggs [Sun, 14 Jan 1996 21:44:26 +0000 (21:44 +0000)]
From NetBSD: (required for last change to scsi_base.c in case we get
"command aborted" status)
Handle cases like the following:
- controller calls scsi_done() with error XS_TIMEOUT
- scsi_done() calls sddone()
- sddone() calls disk_unbusy()
- scsi_done() calls controller to retry command (missing the
call to disk_busy())
- controller calls scsi_done()
- scsi_done() calls sddone()
- sddone() calls disk_busy(), which panics because of the imbalance.
Bug noticed by Leo Weppleman, who also suggested this fix; pass an additional
boolean argument ("complete") to the device's "done" routine, with a
value of `0' passed from the previous call to "done", and add an additional
call to "done" when the xfer resources are freed.
briggs [Sun, 14 Jan 1996 21:34:43 +0000 (21:34 +0000)]
Retry a command when we get a command aborted message for it.
This might be a transient error. Do complain about it, though.
briggs [Sun, 14 Jan 1996 21:29:28 +0000 (21:29 +0000)]
Preliminary cut at SONIC driver. Quadra only at this point.
Still apparently not getting interrupts for some reason, so this is more
or less just a checkpoint.
niklas [Sun, 14 Jan 1996 21:20:10 +0000 (21:20 +0000)]
Use dma_cachectl instead of cachectl when writing into the kernel text
segment. Ordinary cachectl is for userlevel virtual addresses.
briggs [Sun, 14 Jan 1996 21:15:21 +0000 (21:15 +0000)]
from netbsd: more of PR 1931. Fix code that's not compiled unless DEBUG.
briggs [Sun, 14 Jan 1996 21:11:46 +0000 (21:11 +0000)]
from netbsd -- remove some more dead code (PR #1931)
briggs [Sun, 14 Jan 1996 21:06:29 +0000 (21:06 +0000)]
from netbsd--remove bogus code.
briggs [Sun, 14 Jan 1996 20:58:54 +0000 (20:58 +0000)]
from netbsd: Fixes from PR #1931.
briggs [Sun, 14 Jan 1996 20:57:35 +0000 (20:57 +0000)]
from netbsd:
Patch to at least recognise FOCUS EtherLAN. From Erik Bertelson
<erik@sockdev.uni-c.dk>
briggs [Sun, 14 Jan 1996 20:51:08 +0000 (20:51 +0000)]
sync w/ netbsd: splzs -> spl4 from splhigh. Remove bogus DEBUG.
niklas [Sun, 14 Jan 1996 16:05:23 +0000 (16:05 +0000)]
Add hooks for custom audio players. This is the original untouched
kernel diffs from Tim Newsham <newsham@hookomo.aloha.net> found in his
well-known audio LKM kit.
niklas [Sun, 14 Jan 1996 12:21:33 +0000 (12:21 +0000)]
netbsd -> bsd
niklas [Sun, 14 Jan 1996 01:27:17 +0000 (01:27 +0000)]
Redid formatting with mandoc macros. Added an "ERRORS" section.
davem [Sun, 14 Jan 1996 00:39:18 +0000 (00:39 +0000)]
The sun mmu is very broken, and we all can thank crashme for
helping me find this bug. On execution of an atomic load/store instruction
the chip will only say that a read fault is happening, we then load up a
readonly translation to the accessed page, and we get the fault again still
showing a read-fault. We end up faulting in a loop forever and the process
appears to be completely stuck. The algorithm to fix this problem goes like
this. If we get a non-text fault, and the fault type is VM_PROT_READ, and
the SER_PROT bit is set in the syncronous fault error register, we take
a peek at the instruction at pc. If this instruction is indeed an ldstub
or a swap variant we or in VM_PROT_WRITE to the fault type.
niklas [Sun, 14 Jan 1996 00:17:18 +0000 (00:17 +0000)]
Fix typo
briggs [Sat, 13 Jan 1996 16:14:22 +0000 (16:14 +0000)]
Don't issue a linked command to check for a device's support of them.
Instead, check the inquiry_flag. Also allow enabling and disabling
linked command issuance on target-by-target basis.
chuck [Sat, 13 Jan 1996 03:52:44 +0000 (03:52 +0000)]
sync with sparc:
- move disk_attach() to before reading the disk label as per Jason.
otherwise we are reading into an unallocated buffer (oops!)
chuck [Sat, 13 Jan 1996 03:45:01 +0000 (03:45 +0000)]
- call mapiodev() with sizeof(struct xdc/xyc) rather than ra_len (which is
zero since xdcmatch/xycmatch no longer sets it). [if you call
mapiodev() with a zero size it will reuse the KVA it returns, swiping
the device out from under you!]
- make xdc/xyc->iopbase point in the kernels DVMA space rather than at the
normal malloc'd KVA. this isn't compatable with sun4m [doesn't
have a kernel DVMA space] and will need to be changed later (XXX).
- move disk_attach() to before reading the disk label as per Jason.
otherwise we are reading into an unallocated buffer (oops!)
chuck [Sat, 13 Jan 1996 03:12:12 +0000 (03:12 +0000)]
have mapdev() panic if size == 0 to be safe (catches config errors)
chuck [Fri, 12 Jan 1996 23:09:08 +0000 (23:09 +0000)]
we no longer need to add in the offset to the mapped VA to get the correct
device address (bus_tmp, bus_map, and mapiodev now do this for us).
also, we handle all our mappings, so don't have obio.c do any for us
(i.e. don't set ra->ra_len in xycmatch).
nuke uneeded variable in match function.
chuck [Fri, 12 Jan 1996 22:02:15 +0000 (22:02 +0000)]
dmamatch never added in the offset for the dma registers to the VA,
and thus probed the wrong address on the 4/300 (but it worked because
it was probing the esp0 registers!).
now that bus_tmp() adds the offset in for us [as of obio 1.15] i
discoved that a byte access to the dma registers is not allowed.
so, i've change probeget to use a word access.
chuck [Fri, 12 Jan 1996 21:43:09 +0000 (21:43 +0000)]
improved handling of mapping of devices who's registers do not reside on
page boundaries:
- change bus_tmp() to include the offset from the start of page in the
returned KVA [rather than forcing each driver to add it back in
individually]
- changed bus_map() to include the offset from the start of page in the
the returned value if a mapping is found in the PROM's KVA area
- clarified a few comments
chuck [Fri, 12 Jan 1996 21:21:54 +0000 (21:21 +0000)]
change mapdev() so that it includes the offset from the start of the
page in the virtual address it returns.
deraadt [Fri, 12 Jan 1996 21:15:05 +0000 (21:15 +0000)]
remove DEBUG
deraadt [Fri, 12 Jan 1996 21:13:41 +0000 (21:13 +0000)]
from netbsd:
Kludge around a case where a flaky HP-IB disk might be slow to respond
to the identification request in rdmatch(). Similar in spirit to a
patch from Jason Downs (written eons ago), but limited to the broken
device we're trying to reach. My patch tested (and fixed :-) by
Herb Peyerl.
deraadt [Fri, 12 Jan 1996 21:12:59 +0000 (21:12 +0000)]
from netbsd; Fix declaration of hilqfree(), from <hag@wizardz.com>,
PR #1918.
deraadt [Fri, 12 Jan 1996 21:11:52 +0000 (21:11 +0000)]
Correct some typo's, add some virtual consoles and the tun devices
deraadt [Fri, 12 Jan 1996 21:11:12 +0000 (21:11 +0000)]
from netbsd; Correct test for ECHONL (from der Mouse; PR#1922).
niklas [Fri, 12 Jan 1996 21:08:46 +0000 (21:08 +0000)]
Completed incomplete edit and corrected history
deraadt [Fri, 12 Jan 1996 21:02:45 +0000 (21:02 +0000)]
no luns on Tandberg 3600 w/ fake Archive Viper emulation roms; from
raeburn@raeburn.org; netbsd pr#1934
deraadt [Fri, 12 Jan 1996 20:19:48 +0000 (20:19 +0000)]
from netbsd;
New generic disk framework. Highlights:
New metrics handling. Metrics are now kept in the new `struct disk'.
Busy time is now stored as a timeval, and transfer count in bytes.
Storage for disklabels is now dynamically allocated, so that the size
of the disk structure is not machine-dependent.
Several new functions for attaching and detaching disks, and handling
metrics calculation.
Old-style instrumentation is still supported in drivers that did it
before. However, old-style instrumentation is being deprecated, and
will go away once the userland utilities are updated for the new
framework.
For usage and architectural details, see the forthcoming disk(9)
manual page.
deraadt [Fri, 12 Jan 1996 20:14:22 +0000 (20:14 +0000)]
update to new cdio.h
deraadt [Fri, 12 Jan 1996 20:00:43 +0000 (20:00 +0000)]
from netbsd; Document hint file v.2
deraadt [Fri, 12 Jan 1996 19:59:53 +0000 (19:59 +0000)]
from netbsd; Note search mechanism
deraadt [Fri, 12 Jan 1996 19:57:26 +0000 (19:57 +0000)]
from netbsd:
Make dlopen() search for libraries if given a name of the form
`lib<name>.so[.xx[.yy]]' (per Erik M. Theisen; PR#1763; patch from PR
used in simplified form).
Deal with v.2 hint files (hint srch dirs not yet used).
deraadt [Fri, 12 Jan 1996 19:56:20 +0000 (19:56 +0000)]
from netbsd:
Create v.2 hint files.
Note: don't bother to understand v.1 files..
hannken [Fri, 12 Jan 1996 17:52:48 +0000 (17:52 +0000)]
Remove CCB from hash list when it gets unmapped, not the first time looked up.
deraadt [Fri, 12 Jan 1996 16:55:47 +0000 (16:55 +0000)]
clean better
deraadt [Fri, 12 Jan 1996 16:45:00 +0000 (16:45 +0000)]
SIGBUS is an alignment fault. SIGSEGV is an access violation
deraadt [Fri, 12 Jan 1996 14:45:42 +0000 (14:45 +0000)]
proto rfork()
deraadt [Fri, 12 Jan 1996 03:13:42 +0000 (03:13 +0000)]
from netbsd
deraadt [Fri, 12 Jan 1996 02:08:49 +0000 (02:08 +0000)]
disable ddb, enable vnd
niklas [Thu, 11 Jan 1996 22:27:45 +0000 (22:27 +0000)]
Remove obsolete GCC
niklas [Thu, 11 Jan 1996 19:38:24 +0000 (19:38 +0000)]
Remove obsolete GCC
deraadt [Thu, 11 Jan 1996 18:19:06 +0000 (18:19 +0000)]
from leo; Fix the same typo in all files. Yanking is soooo beautiful....
deraadt [Thu, 11 Jan 1996 17:58:59 +0000 (17:58 +0000)]
from netbsd; VM86 support, by John Kohl, touched up a bit by charles
deraadt [Thu, 11 Jan 1996 17:54:00 +0000 (17:54 +0000)]
from netbsd; Fix typo and make some enum. lists more compact
(James Jegers; PR#592).
deraadt [Thu, 11 Jan 1996 17:52:34 +0000 (17:52 +0000)]
from netbsd; Fetch `-p' documentation from a previous version of
the man page
deraadt [Thu, 11 Jan 1996 17:49:42 +0000 (17:49 +0000)]
from netbsd; fix small documentation bug -- we have no bftp(1)
deraadt [Thu, 11 Jan 1996 17:48:48 +0000 (17:48 +0000)]
from netbsd; sync with tzdata96a
deraadt [Thu, 11 Jan 1996 17:48:19 +0000 (17:48 +0000)]
from netbsd; new pages
deraadt [Thu, 11 Jan 1996 17:47:30 +0000 (17:47 +0000)]
sync with netbsd
deraadt [Thu, 11 Jan 1996 17:46:15 +0000 (17:46 +0000)]
from netbsd; sync with tzcode96a
deraadt [Thu, 11 Jan 1996 17:44:35 +0000 (17:44 +0000)]
from netbsd:
Define LD_HINTS_VERSION_2:
- no longer hash on minor version number in hints file.
- add list of ldconfig(8) search directories.
deraadt [Thu, 11 Jan 1996 17:41:53 +0000 (17:41 +0000)]
login names starting with "t" were trashed; original bug report by
<tommy@mdc.tijp.ti.com>. from soda@sra.co.jp; netbsd pr#1926
deraadt [Wed, 10 Jan 1996 22:15:13 +0000 (22:15 +0000)]
merge error
briggs [Wed, 10 Jan 1996 19:08:04 +0000 (19:08 +0000)]
Save inquiry flags in sc_link so low-level drivers can use it.
deraadt [Wed, 10 Jan 1996 18:02:12 +0000 (18:02 +0000)]
plan9-like rfork() implimentation by me based on some earlier incomplete
work by rminnich@Sarnoff.COM. still needs RFNOWAIT support.
deraadt [Wed, 10 Jan 1996 17:10:04 +0000 (17:10 +0000)]
describe actual openbsd rfork() details. this should be converted to a
mandoc manpage by some brave individual (prettier because they fit more
data on the same page). please?
chuck [Wed, 10 Jan 1996 16:26:16 +0000 (16:26 +0000)]
Imported from FreeBSD:
Add support for the SMC8416 (EtherEZ) ISA ethernet card.
The 8416 has an 8K shared mem (the old driver assumed 16K
and failed at attach time).
deraadt [Wed, 10 Jan 1996 09:38:36 +0000 (09:38 +0000)]
original rfork(2) manpage from plan9. This will be modified to describe
the actual implimentation in OpenBSD.
> From: "Rob Pike" <rob@plan9.att.com>
> feel free to borrow the text of the plan 9 rfork page, with attribution.
dm [Tue, 9 Jan 1996 16:24:18 +0000 (16:24 +0000)]
make in ipf directory (oops)
briggs [Tue, 9 Jan 1996 16:14:28 +0000 (16:14 +0000)]
Go ahead and look for the sonic--no problem if we don't find it.
briggs [Tue, 9 Jan 1996 16:11:51 +0000 (16:11 +0000)]
Add first guesses for a few more machines.
Also set sonic flag for MACH_CLASSQ.
briggs [Tue, 9 Jan 1996 16:01:03 +0000 (16:01 +0000)]
Add some more machine defines.
dm [Tue, 9 Jan 1996 09:29:24 +0000 (09:29 +0000)]
added IP filter to netstat/rc and put examples in /usr/share/ipf
deraadt [Tue, 9 Jan 1996 09:24:32 +0000 (09:24 +0000)]
from netbsd:
Update a kludge from 4.4bsd/pmax: DECstation PROMs force the clock to
be in 1972 or 1973. If the clock is set to the actual time, the PROM
appears to reset the chip date to 1972 at each reboot.
Change the "year offset" from 22 years to 24 years (good till 1997)
until this code is replaced with something that uses the chip as a
seconds-since-beginning-of-year clock (encoded as a date, relative to
1972-01-01 00:00:00).
deraadt [Tue, 9 Jan 1996 09:24:01 +0000 (09:24 +0000)]
from netbsd; Fix setcontext call and sendsig
deraadt [Tue, 9 Jan 1996 09:20:36 +0000 (09:20 +0000)]
from netbsd; compile hpux_machdep.c if COMPAT_HPUX
deraadt [Tue, 9 Jan 1996 09:14:38 +0000 (09:14 +0000)]
from netbsd:
Part of the tostools reorganization. All directories now contain a Makefile
and the user interface to the various tools has been standarized.
deraadt [Tue, 9 Jan 1996 09:13:45 +0000 (09:13 +0000)]
from leo:
Common library for all tostools. (Waldi Ravens and myself)
deraadt [Tue, 9 Jan 1996 09:13:10 +0000 (09:13 +0000)]
from leo:
Atari partion table checker. Checks the validity of AHDI and/or
NetBSD/Atari partition tables. (Waldi Ravens)
briggs [Tue, 9 Jan 1996 04:06:49 +0000 (04:06 +0000)]
spltty shouldn't need to block serial hard interrupts--just soft.
splimp can drop, too.
niklas [Mon, 8 Jan 1996 22:16:07 +0000 (22:16 +0000)]
Add OpenBSD configuration, it's exactly like NetBSD.
niklas [Mon, 8 Jan 1996 21:28:57 +0000 (21:28 +0000)]
As it is highly unlikely OpenBSD's object file format will divert from
NetBSD's, alias it instead of implementing a distinct but identical
configuration.
niklas [Mon, 8 Jan 1996 21:20:43 +0000 (21:20 +0000)]
As it is highly unlikely OpenBSD's object file format will divert from
NetBSD's, alias it instead of implementing a distinct but identical
configuration.
deraadt [Mon, 8 Jan 1996 19:31:06 +0000 (19:31 +0000)]
from netbsd; Correct name of sysi86 function
deraadt [Mon, 8 Jan 1996 19:30:21 +0000 (19:30 +0000)]
from netbsd:
Add '/emul/ultrix' pathname processing to Ultrix emulation, and move
Ultrix syscalls that require that processing out of ultrix_misc.c
to ultrix_pathname.c.
deraadt [Mon, 8 Jan 1996 18:55:02 +0000 (18:55 +0000)]
from ragge; Changed to work correctly with both K&R and ANSI cpp.
deraadt [Mon, 8 Jan 1996 18:54:33 +0000 (18:54 +0000)]
from ragge:
Device driver for TS11/TSV05 tape devices.
Written by Bertram Barth, based on TMSCP device driver.
deraadt [Mon, 8 Jan 1996 18:52:51 +0000 (18:52 +0000)]
update from netbsd
deraadt [Mon, 8 Jan 1996 18:52:01 +0000 (18:52 +0000)]
from netbsd; Hand-code softintr()
deraadt [Mon, 8 Jan 1996 18:50:24 +0000 (18:50 +0000)]
from leo; Remove a stray MAXPARTITIONS definition
deraadt [Mon, 8 Jan 1996 18:49:06 +0000 (18:49 +0000)]
from ragge; Changed to only include <machine/asm.h>, as the other ports
deraadt [Mon, 8 Jan 1996 18:48:21 +0000 (18:48 +0000)]
from ragge:
Added 'all' option to create a reasonable amount of standard devices.
Tape devices now called after their kernel names, (ie. ts0, ht0) and
not mt? for all of them. mt? names are kept as symbolic links anyway
so that it don't break anything.
deraadt [Mon, 8 Jan 1996 18:46:19 +0000 (18:46 +0000)]
from ragge; Change to be more consistent with device naming
deraadt [Mon, 8 Jan 1996 18:45:54 +0000 (18:45 +0000)]
from ragge; Updated to conform to reality
niklas [Mon, 8 Jan 1996 18:16:01 +0000 (18:16 +0000)]
Support OpenBSD configuration, by cloning NetBSD config entries.
niklas [Mon, 8 Jan 1996 15:54:12 +0000 (15:54 +0000)]
Tweaks for OpenBSD:
1) recognize OpenBSD at various places
2) m68k-cbm-* is *NOT* ELF always
3) clean-info must be called before distclean in the GAS
realclean target
4) don't install unformatted ma-pages
5) add a Makefile.bsd-wrapper
6) just build gas through Makefile.bsd-wrapper yet
There are still other things to do...
deraadt [Mon, 8 Jan 1996 11:36:02 +0000 (11:36 +0000)]
add rfork.o and minherit.o
niklas [Mon, 8 Jan 1996 11:03:33 +0000 (11:03 +0000)]
Initial GNU binutils 2.6 import
deraadt [Mon, 8 Jan 1996 09:33:31 +0000 (09:33 +0000)]
from netbsd:
Deal with GCC's dead code elimination being suboptimal.
Modify splraise() to allow better optimization.
Make cpl, ipending, and astpending volatile.
Make sure interrupts are disabled before jumping to a resume point,
to prevent races.
Make FPU faults use INTRFASTEXIT, and remove INTREXIT.
Build the frame for recursive interrupts manually, and make sure to
disable interrupts to avoid races.
VS: ----------------------------------------------------------------------
deraadt [Mon, 8 Jan 1996 07:44:56 +0000 (07:44 +0000)]
from netbsd; Make the block size always a whole number of samples
dm [Mon, 8 Jan 1996 07:36:23 +0000 (07:36 +0000)]
graichen@freebsd.org: fixed -type:=direct mounts in amd
deraadt [Mon, 8 Jan 1996 07:32:12 +0000 (07:32 +0000)]
Produce less output on -q as promised by the man page;
from jhawk@mit.edu; netbsd pr#998
deraadt [Sun, 7 Jan 1996 17:38:21 +0000 (17:38 +0000)]
add fdshare() and fdinit() calls for dealing with filedesc structures
deraadt [Sun, 7 Jan 1996 17:20:35 +0000 (17:20 +0000)]
add minherit() system call