miod [Fri, 25 Jul 2008 21:24:49 +0000 (21:24 +0000)]
phyio -> physio
miod [Fri, 25 Jul 2008 21:11:14 +0000 (21:11 +0000)]
Improve eisa bus probe by deciding the number of logical eisa slots from
the motherboard eisa id, instead of using a ``one size fits all'' value which
is too large on more than half the eisa-capable alpha designs.
The id -> slot # logic is based on the alpha ECU configuration files, so we
should not perform worse than ECU itself (and see all slots ECU sees too).
damien [Fri, 25 Jul 2008 20:54:49 +0000 (20:54 +0000)]
missing .Pp
kurt [Fri, 25 Jul 2008 20:51:56 +0000 (20:51 +0000)]
The PIE address range defaults are too big for alpha and can land on or
above the stack. Fix by mapping PIE into the first quarter of the address
space before stack. "will do for now" miod@
damien [Fri, 25 Jul 2008 20:49:23 +0000 (20:49 +0000)]
add a note in the CAVEATS section about the hardware crypto engine not
being used on the RT2500, RT2501 and RT2600 chipset.
+ Xr wpa-psk(8) as pointed out by jmc@
jmc [Fri, 25 Jul 2008 20:36:16 +0000 (20:36 +0000)]
tweak the rt2700 stuff; ok damien
jmc [Fri, 25 Jul 2008 20:09:05 +0000 (20:09 +0000)]
groff does not like ultra-long lines without whitespace, since it cannot
format them properly; thus i've replaced the .Pa markup on the very
lengthy URL with an unfilled display. this also has the benefit of not
splitting the URL with a hyphen;
ian [Fri, 25 Jul 2008 19:39:40 +0000 (19:39 +0000)]
Mention @bin plist element, ok espie@
kettenis [Fri, 25 Jul 2008 19:37:16 +0000 (19:37 +0000)]
Work around a problem on the t5120 where top(1) would lock up the system for
several minutes by making pmap_resident_count() return 0 for the kernel pmap.
discussed with miod@
damien [Fri, 25 Jul 2008 18:41:16 +0000 (18:41 +0000)]
link to updated firmware packages that remove the obsolete @ignore
directive. firmware itself does not change.
requested by brad@
martynas [Fri, 25 Jul 2008 17:43:44 +0000 (17:43 +0000)]
don't redefine INFINITY -- math.h defines it now
pointed out by, and ok david@, go ahead henning@
mglocker [Fri, 25 Jul 2008 17:20:27 +0000 (17:20 +0000)]
After our recent changes, VIDIOC_S_FMT didn't always return the correct
maximum image size. Fix it. Makes read(2) method work again.
damien [Fri, 25 Jul 2008 16:18:54 +0000 (16:18 +0000)]
document RT2700 (tested by bernd@) and try documenting WPA support.
art [Fri, 25 Jul 2008 15:01:33 +0000 (15:01 +0000)]
In gdt_grow, always map the gdt on the secondary cpus too. This far
we depended on the pageability of the kernel_map to map them which worked
most of the time, until we hit the most improbable case - when cpu_switchto
wanted to fault in the missing gdt page, which at this moment is Verboten.
kettenis@ ok
beck [Fri, 25 Jul 2008 14:56:47 +0000 (14:56 +0000)]
much more correct way of dealing with nfs pending reads/writes
ok thib@
kettenis [Fri, 25 Jul 2008 14:53:38 +0000 (14:53 +0000)]
Switch to a temporary stack early on such that we don't lose if the stack
provided by the PROM gets flushed out of the TLB before we install our own
trap table. Makes RAMDISK kernels work on T2 systems.
deraadt [Fri, 25 Jul 2008 14:38:33 +0000 (14:38 +0000)]
sync
deraadt [Fri, 25 Jul 2008 14:37:09 +0000 (14:37 +0000)]
i have to crank this for a ridiculous reason, to save me about 4 hours of work
martynas [Fri, 25 Jul 2008 14:07:05 +0000 (14:07 +0000)]
don't undefine C99 math macros, if !_GLIBCPP_USE_C99, so that we
can use these functions in C++.
ok millert@. also looked over by espie@, fgsch@.
jmc [Fri, 25 Jul 2008 12:29:33 +0000 (12:29 +0000)]
doc fix; from David Higgs
art [Fri, 25 Jul 2008 12:05:04 +0000 (12:05 +0000)]
Correct printing of the pg_flags for ddb.
art [Fri, 25 Jul 2008 12:02:09 +0000 (12:02 +0000)]
some splassert paranoia.
art [Fri, 25 Jul 2008 11:57:53 +0000 (11:57 +0000)]
flock
art [Fri, 25 Jul 2008 11:41:22 +0000 (11:41 +0000)]
file advisory locking tests from FreeBSD. We fail to detect
two deadlocks at the moment.
mglocker [Fri, 25 Jul 2008 11:12:34 +0000 (11:12 +0000)]
Not all devices support GET_DEF, fall back on GET_CUR if so.
Reported and tested kettenis@
henning [Fri, 25 Jul 2008 08:53:39 +0000 (08:53 +0000)]
shrink MINCLSIZE to MHLEN + 1 again for the release, we're not confident
enough yet that all drivers handle this correctly. intended to be grown
again after release. discussed with claudio, thib and theo
martynas [Fri, 25 Jul 2008 08:00:01 +0000 (08:00 +0000)]
fix apps that use isinff, isnanf; use ieee 754 mi code, as all
other ieee fp archs do; pointed out by theo. tested by theo
grunk [Fri, 25 Jul 2008 07:05:16 +0000 (07:05 +0000)]
In random art visualization, make sure to use the end marker only at the
end. Initial diff by Dirk Loss, tweaks and ok djm@
grunk [Fri, 25 Jul 2008 06:56:35 +0000 (06:56 +0000)]
Add VisualHostKey to example file, ok djm@
deraadt [Fri, 25 Jul 2008 05:09:55 +0000 (05:09 +0000)]
repair .Pq use; Alan R. S. Bueno
deraadt [Fri, 25 Jul 2008 01:29:10 +0000 (01:29 +0000)]
sync
sthen [Thu, 24 Jul 2008 23:55:30 +0000 (23:55 +0000)]
Add "ssh-keygen -F -l" to synopsis (displays fingerprint from
known_hosts). ok djm@
guenther [Thu, 24 Jul 2008 23:42:26 +0000 (23:42 +0000)]
release the bind lock before resetting the signal mask
ok kurt@
deraadt [Thu, 24 Jul 2008 22:34:58 +0000 (22:34 +0000)]
sync
thib [Thu, 24 Jul 2008 19:03:08 +0000 (19:03 +0000)]
the softc sc_lock lockmgr lock is unused, so remove it.
ok krw@, dlg@
thib [Thu, 24 Jul 2008 19:01:28 +0000 (19:01 +0000)]
dont need malloc.h anymore; plus two tiny whitespace fixes
thib [Thu, 24 Jul 2008 18:48:18 +0000 (18:48 +0000)]
remove a bunch of bogus ARGSUSED
claudio [Thu, 24 Jul 2008 18:46:59 +0000 (18:46 +0000)]
Use memcpy() and memcmp() like everywhere else in ospfd. OK norby@
deraadt [Thu, 24 Jul 2008 17:36:52 +0000 (17:36 +0000)]
sng_exponent should be sng_exp (MI code now cares); ok martynas
mk [Thu, 24 Jul 2008 17:31:59 +0000 (17:31 +0000)]
Add timeout_add_sec() etc. to .Nm and, on jmc's request, tweak
some other stuff.
ok art jmc blambert
jsing [Thu, 24 Jul 2008 16:44:56 +0000 (16:44 +0000)]
Specify a cdroot entry that will actually work on a CD.
ok krw@
miod [Thu, 24 Jul 2008 16:34:22 +0000 (16:34 +0000)]
Add code to decode machine checks on Avanti, providing a description of the
problem and lots of register dumps to help the skilled user to fix the hardware.
Not compiled if option SMALL_KERNEL.
From NetBSD.
deraadt [Thu, 24 Jul 2008 15:51:07 +0000 (15:51 +0000)]
bump because of __nan in math.h
martynas [Thu, 24 Jul 2008 15:20:24 +0000 (15:20 +0000)]
bump
mglocker [Thu, 24 Jul 2008 14:59:44 +0000 (14:59 +0000)]
Remove some obsolete XXX comments.
jmc [Thu, 24 Jul 2008 14:19:24 +0000 (14:19 +0000)]
kill whitespace at eol;
martynas [Thu, 24 Jul 2008 14:11:42 +0000 (14:11 +0000)]
isinf/isnan are macros now, therefore ap_config_auto.h differ. regen
from sthen@. discussed w/ millert@. "yep" sthen@.
mglocker [Thu, 24 Jul 2008 13:30:10 +0000 (13:30 +0000)]
Simplify uvideo_enum_fmt(), no functional changes.
reyk [Thu, 24 Jul 2008 12:59:06 +0000 (12:59 +0000)]
reserve a few OpenBSD MIBs that might be implemented later or to avoid
conflicts wird 3rd parties.
henning [Thu, 24 Jul 2008 10:55:44 +0000 (10:55 +0000)]
ipsec is glued into the stack in a very weird way, violating all kinds
of expected semantics. thus, for return packets coming out of an ipsec
tunnel, we need to clear the pf state key pointer in the mbuf header
to prevent a state for encapsulated traffic to be linked to the
decapsulated traffic one.
problem noticed by Oleg Safiullin <form@pdp-11.org.ru>, took me some
time to understand what the hell was going on. ok ryan
henning [Thu, 24 Jul 2008 10:52:43 +0000 (10:52 +0000)]
check sysctl return value
From: Gleydson Soares <gsoares@gmail.com>, ryan ok
martynas [Thu, 24 Jul 2008 09:41:58 +0000 (09:41 +0000)]
unXXX prototypes: exp2, remquo, nan, exp2f, remquof, nanf
martynas [Thu, 24 Jul 2008 09:40:16 +0000 (09:40 +0000)]
- remove isinff, isnanf. this has been moved to libc
- remove never-enabled signbit. libc has is now
- add C99 implementations, from freebsd, for nan(), nanf() (needed
to write _digittoint for it), exp2(), exp2f(), remquo(), remquof(),
needed STRICT_ASSIGN macro for math_private.h
- bump major
man pages will follow
exp2{,f} has been requested by chl@
ok millert@
martynas [Thu, 24 Jul 2008 09:35:30 +0000 (09:35 +0000)]
- add NAN definition (except on VAX)
- add C99 real-floating macros fpclassify isfinite isinf isnan
isnormal signbit
- add C99 macros isgreater isgreaterequal isless islessequal
islessgreater isunordered
- add function prototypes, __fpclassify __fpclassifyf __fpclassifyl
__isfinite __isfinitef __isfinitel __isinf __isinfl __isnan __isnanl
__isnormal __isnormalf __isnormall __signbit __signbitf __signbitl
ok millert@
martynas [Thu, 24 Jul 2008 09:31:06 +0000 (09:31 +0000)]
- move isinf, isnan dups to gen, since most is ieee 754
- is{inf,nan} should be macros for real-floating, so rename to
__is{inf,nan}, per C99
- implement C99 __fpclassify(), __fpclassifyf(), __isfinite(),
__isfinitef(), __isnormal(), __isnormalf(), __signbit(), __signbitf()
- long functions added, but not yet enabled, till ieee.h is fixed
- implement vax equivalents of the functions
- reimplement isinff, isnanf in a better way, and move to libc
- add qnan bytes for all archs
- bump major
man pages will follow
ok millert@. arm bits looked over by drahn@
discussed w/ theo, who showed the right direction, to put these
functions in libc
miod [Thu, 24 Jul 2008 04:54:27 +0000 (04:54 +0000)]
Do not define vax-inspired LOWPAGES and HIGHPAGES anymore, they make no
sense on m68k. Give USRSTACK its real (HP-UX compatible) value instead of
computing it from HIGHPAGES.
No functional change.
mglocker [Wed, 23 Jul 2008 22:10:21 +0000 (22:10 +0000)]
If /dev/video* is already used by an application, return EBUSY to other
applications. Fixes a kernel panic.
Reported by ian@
av [Wed, 23 Jul 2008 21:33:32 +0000 (21:33 +0000)]
we really need unsigned in arithmetic operations.
comments by fgsch
deraadt [Wed, 23 Jul 2008 19:36:47 +0000 (19:36 +0000)]
Avoid an occasional double free in getpwent() which corrupted memory. It
happened only in the case of a + record indicating to use YP, but with ypbind
not actually running. Problem found by pyr; solution found with millert
ok millert pyr
miod [Wed, 23 Jul 2008 19:28:27 +0000 (19:28 +0000)]
Forgot to add rebranded AS800 names there.
miod [Wed, 23 Jul 2008 19:24:25 +0000 (19:24 +0000)]
Attempt to sync with reality, and make the supported/unsuported status of
the various AlphaServer references less ambiguous.
miod [Wed, 23 Jul 2008 19:14:13 +0000 (19:14 +0000)]
Do not rely on an <ichc>invisible cast</ichc> to return failure in
dino_intr_map(); rather than adding the missing cast, make the intent of
the code clearer by explicitenly testing for PCI_INTERRUPT_LINE being ff.
While there, enable the out-of-extent-range checks in dino_memmap() and
dino_memalloc() even if no option DEBUG, but return failure instead of
panicing.
discussed with and ok kettenis@ marco@
brad [Wed, 23 Jul 2008 18:35:31 +0000 (18:35 +0000)]
regen
brad [Wed, 23 Jul 2008 18:34:29 +0000 (18:34 +0000)]
Attansic L1E PCI id.
martynas [Wed, 23 Jul 2008 18:11:13 +0000 (18:11 +0000)]
clear sign bit, instead of comparing to zero and setting x=-x.
fixes special cases, such as neg. zero, and makes C99 conformant
ok miod@, millert@
since there's nothing else in arm's fabs.c, replace 4-clause license
w/ the one at /usr/share/misc/license.template
deraadt [Wed, 23 Jul 2008 17:59:25 +0000 (17:59 +0000)]
sync
deraadt [Wed, 23 Jul 2008 17:40:29 +0000 (17:40 +0000)]
make this real C
kettenis [Wed, 23 Jul 2008 17:39:35 +0000 (17:39 +0000)]
So it turns outthe HP engineers changed the PA-RISC 2.0 architecture
after it was published. In particular, they changed the maximum cache
aliasing boundary from 1MB to 16MB.
It turns that on the PA-8700 the aliasing boundary is actually 4MB
(reported as such by the firmware at least). There are some comments
in the Linux code that suggest that HP never actually built PA-RISC
CPUs with an 8MB or 16MB aliasing boundary.
So raise the aliasing boundary to 4MB. This fixes the weird ps(1) problem
where it didn't print its own arguments correctly.
jaredy [Wed, 23 Jul 2008 16:34:38 +0000 (16:34 +0000)]
fix stack abuse in the `time' commmand, using alloc()'d memory instead.
reported by Thorsten Glaser, thanks.
ok millert@, earlier version miod@
beck [Wed, 23 Jul 2008 16:24:42 +0000 (16:24 +0000)]
Correct cases of mishandling of pending reads and writes to prevent
them going negative - this consists of identifying a number of cases of
IO not going through the buffer cache and marking those buffers with
B_RAW - as well as fixing nfs_bio to show pending writes and reads through
the buffer cache via NFS
still has a problem with mishandling the counters I believe in the
async/sync fallback case where counters stay positive which will be
addressed seperately.
ok tedu@ deraadt@
sthen [Wed, 23 Jul 2008 16:05:47 +0000 (16:05 +0000)]
Prevent warning about insecure hostnames where no /etc/hostname.*
exists. From wcmaier@.
Check target of symbolic links to avoid noise at boot and in
seucrity output where you have several interfaces symlinked to one
config file.
"If you think this is the right thing to do" deraadt@
damien [Wed, 23 Jul 2008 15:55:46 +0000 (15:55 +0000)]
Fix a blatant misuse of MINCLSIZE I introduced in ieee80211_output.c r1.59
The bug was added on 04/16 but it started to show up only after 06/12 when
the value of MINCLSIZE was changed to something different from MHLEN + 1.
Thanks to dlg@ and giovanni (qgiovanni at gmail dot com) for putting me
on the right track.
Tested by giovanni.
Should fix system/5881 too.
kettenis [Wed, 23 Jul 2008 15:39:52 +0000 (15:39 +0000)]
Stop printing the size of BARs; the way the code calculates it is wrong.
We'll need kernel support to do it properly.
ok dlg@
mglocker [Wed, 23 Jul 2008 14:10:58 +0000 (14:10 +0000)]
- Fix potential uvm_fault crash in
uvideo_vs_parse_desc_frame_uncompressed().
- If a VS alternate interface isn't valid (e.g. doesn't contain a
UE_ISOCHRONOUS endpoint), abort attachment with a corresponding error
message instead crashing the kernel.
Tested and OK maja@
kettenis [Wed, 23 Jul 2008 12:18:40 +0000 (12:18 +0000)]
Clear correctable errors after handling them. From Bjorn Andersson.
art [Wed, 23 Jul 2008 11:53:07 +0000 (11:53 +0000)]
- in nfs_mknodrpc - initialize the node we pass to nfs_lookitup. It's
not just a return value, nfs_lookitup treats it specially when *npp
is not NULL.
- nfs_lookitup - When we match the file handle to the directory
filehandle (lookup on '.'), initialize np so that we don't return
stack garbage to the caller. The function shouldn't ever be
called with '.' at this moment, but that's no reason to not do the
right thing in case future callers use it that way.
blambert@ thib@ ok
art [Wed, 23 Jul 2008 11:45:21 +0000 (11:45 +0000)]
Document the new timeout_add_* functions.
reyk [Wed, 23 Jul 2008 10:05:18 +0000 (10:05 +0000)]
validate packet length in debug dns packet logging before printing the header.
mglocker [Wed, 23 Jul 2008 07:56:45 +0000 (07:56 +0000)]
sync
mglocker [Wed, 23 Jul 2008 07:55:31 +0000 (07:55 +0000)]
Unbreak the tree.
djm [Wed, 23 Jul 2008 07:36:55 +0000 (07:36 +0000)]
do not try to print options that have been compile-time disabled
in config test mode (sshd -T); report from nix-corp AT esperi.org.uk
ok dtucker@
deraadt [Wed, 23 Jul 2008 04:59:27 +0000 (04:59 +0000)]
sync
deraadt [Wed, 23 Jul 2008 04:59:18 +0000 (04:59 +0000)]
new ati goo; giovanni
fgsch [Wed, 23 Jul 2008 00:20:35 +0000 (00:20 +0000)]
move the aml_register_notify() call below for better debugging output.
marco@ ok.
reyk [Tue, 22 Jul 2008 23:17:37 +0000 (23:17 +0000)]
Add dynamic IPv6-to-IPv4 and IPv4-to-IPv6 translation inspired by
faithd(8) by doing a similar mapping of IPv4/6 addresses with
relayd(8) and pf(4) redirections without the need of the faith(4)
interface. The trick works in both directions, it can accept IPv6
connections and relay them to IPv4 hosts by extracting the last 4
octets from the IPv6 destination (like faithd(8)), and it can accept
IPv4 connections and relay them to IPv6 hosts by prepending the 4
octets of the original IPv4 destination to a configured IPv6 prefix.
An access list is not needed because the classification is done in
pf.conf(5). It helps to get more faith in relayd.
manpage bits ok jmc@
yes, sounds good todd@
deraadt [Tue, 22 Jul 2008 21:47:45 +0000 (21:47 +0000)]
use arc4random_uniform(); ok djm millert
espie [Tue, 22 Jul 2008 20:53:13 +0000 (20:53 +0000)]
FAKE (the variable) died about 2 years ago...
miod [Tue, 22 Jul 2008 20:06:01 +0000 (20:06 +0000)]
None of the pceb EISA alpha designs has more than 8 physical slots, so don't
bother trying to probe more.
martynas [Tue, 22 Jul 2008 19:58:40 +0000 (19:58 +0000)]
remez -> Remes here, too. for consistency
mglocker [Tue, 22 Jul 2008 19:45:51 +0000 (19:45 +0000)]
Add another working cam reported by Jerome Pinot.
miod [Tue, 22 Jul 2008 18:47:24 +0000 (18:47 +0000)]
Proper interrupt swizzling for devices behind bridges; matches what SRM
configures when it can. ok kettenis@
miod [Tue, 22 Jul 2008 18:45:50 +0000 (18:45 +0000)]
Modify pci_intr_map() implementation to pass the pci_attach_args to the
per-platform implementation, instead of selected members of it; this allows
us to get rid of some globals, and paves the way for better bridge support
on some models.
miod [Tue, 22 Jul 2008 18:44:25 +0000 (18:44 +0000)]
Use cgbase() instead of doing arithmetic on fs_fpg, for the former does
proper casts to ensure the result is not truncated.
From FreeBSD via NetBSD. ok thib@
miod [Tue, 22 Jul 2008 18:15:48 +0000 (18:15 +0000)]
Do not define VM_MIN_ADDRESS as __LDPGSZ, but provide its value, to prevent
uvm_map.c from requiring <machine/exec.h>.
mglocker [Tue, 22 Jul 2008 16:24:40 +0000 (16:24 +0000)]
There are devices out there which are UVC compatible, but do not show
up as this (we love standards). To make those devices attach anyway
we add an explicit quirk device list. Of course this list still needs
to be extended.
Thanks to Jerome Pinot for reporting about this issue and testing this
diff with his cam, which is such a device.
OK yuo@
martynas [Tue, 22 Jul 2008 16:01:46 +0000 (16:01 +0000)]
- added HUGE_VALF, HUGE_VALL, INFINITY, FP_INFINITE, FP_NAN,
FP_NORMAL, FP_SUBNORMAL, FP_ZERO, FP_ILOGB0, FP_ILOGBNAN definitions,
per C99
- classify functions into sections, add missing definitions, add
and some '#if 0' functions, and long double definitions, so we see
what's needed to be done
- other stuff is XXX'd, and will be uncommented soon
ok millert@
henning [Tue, 22 Jul 2008 12:31:35 +0000 (12:31 +0000)]
after pf_state_key_atach nothing must use the state keys passed to it any
more, since they might have been invalidated and free'd.
one synproxy errorpath did so, however. just get the state keys from the
state itself. ok david mcbride
martynas [Tue, 22 Jul 2008 11:20:10 +0000 (11:20 +0000)]
more negotation -> negotiation; ok sthen@
mglocker [Tue, 22 Jul 2008 11:02:17 +0000 (11:02 +0000)]
Spelling, negotation -> negotiation.
mglocker [Tue, 22 Jul 2008 10:57:15 +0000 (10:57 +0000)]
Fix uvm_fault crash in uvideo_vs_negotation() which was reported by some
users.
Thanks to mpf@ for his help to debug this bug down.