openbsd
16 years ago+I-O DATA WN-G54/CF to the malo(4) list;
jmc [Sun, 27 Jul 2008 16:36:09 +0000 (16:36 +0000)]
+I-O DATA WN-G54/CF to the malo(4) list;

16 years agos/IEEE80211_DPRINTF/DPRINTF/
damien [Sun, 27 Jul 2008 14:21:15 +0000 (14:21 +0000)]
s/IEEE80211_DPRINTF/DPRINTF/
automatically prepend __func__ to output.

deobfuscates debug messages a bit.
no binary change unless compiled with IEEE80211_DEBUG.

16 years agoremove unused structures and duplicated definitions.
damien [Sun, 27 Jul 2008 14:15:42 +0000 (14:15 +0000)]
remove unused structures and duplicated definitions.
remove clause 3 of license while I'm here (from FreeBSD).

16 years agoan enum specifier is more elegant than a set of #defines;
sobrado [Sun, 27 Jul 2008 13:15:31 +0000 (13:15 +0000)]
an enum specifier is more elegant than a set of #defines;
storing the program mode variable (pmode) as a global let us have
a more consistent prototype for usage().

changes suggested by pyr@.

ok pyr@

16 years agoSome devices take longer to load the firmware. Increase the firmware
mglocker [Sun, 27 Jul 2008 13:02:37 +0000 (13:02 +0000)]
Some devices take longer to load the firmware.  Increase the firmware
load timeout from 50000us to 500000us.

Reported and tested by William Dunand with an I-O DATA WN-G54/CF device.
Since this device works now, we add it to the man page.

16 years agofix a typo in a comment.
mbalmer [Sun, 27 Jul 2008 11:28:17 +0000 (11:28 +0000)]
fix a typo in a comment.

ok mglocker

16 years agobad comma;
jmc [Sun, 27 Jul 2008 10:24:27 +0000 (10:24 +0000)]
bad comma;

16 years agoUse __progname instead of hand rolled version using argv[0].
claudio [Sun, 27 Jul 2008 10:06:10 +0000 (10:06 +0000)]
Use __progname instead of hand rolled version using argv[0].
Diff from Gleydson Soares gsoares at gmail dot com
OK millert@

16 years agoGet TAILQ_REPLACE and CIRCLEQ_REPLACE prototypes right.
stefan [Sun, 27 Jul 2008 07:06:42 +0000 (07:06 +0000)]
Get TAILQ_REPLACE and CIRCLEQ_REPLACE prototypes right.
ok jaredy

16 years agoRecognize an explicit + sign in exponent of hex float constants, needed to
miod [Sat, 26 Jul 2008 19:46:56 +0000 (19:46 +0000)]
Recognize an explicit + sign in exponent of hex float constants, needed to
compile current libm. ok martynas@

16 years agoSilence scsi_inquire_vpd() so devices that don't understand the necessary
krw [Sat, 26 Jul 2008 18:55:31 +0000 (18:55 +0000)]
Silence scsi_inquire_vpd() so devices that don't understand the necessary
INQUIRY options don't spam dmesg during probe. Seen by miod@ and kettenis@
for SCSI CD's.

ok miod@

16 years agoAdd M$ NX-6000 cam, ok mglocker@
ian [Sat, 26 Jul 2008 16:20:17 +0000 (16:20 +0000)]
Add M$ NX-6000 cam, ok mglocker@

16 years agoshrink ieee80211_set_link_state().
damien [Sat, 26 Jul 2008 12:56:06 +0000 (12:56 +0000)]
shrink ieee80211_set_link_state().

16 years agoremove useless #includes.
damien [Sat, 26 Jul 2008 12:42:57 +0000 (12:42 +0000)]
remove useless #includes.

16 years agoproperly handle management frames (required for future work).
damien [Sat, 26 Jul 2008 12:41:34 +0000 (12:41 +0000)]
properly handle management frames (required for future work).
remove useless #includes.

16 years agocall Phase1 every 2**16 frames instead of calling it for every frame.
damien [Sat, 26 Jul 2008 12:36:15 +0000 (12:36 +0000)]
call Phase1 every 2**16 frames instead of calling it for every frame.
properly handle frames with an address 4 field (required for future work).
remove useless #includes.

16 years agoSince we are able to change the device image resolution on the fly in the
mglocker [Sat, 26 Jul 2008 11:42:43 +0000 (11:42 +0000)]
Since we are able to change the device image resolution on the fly in the
meantime, the memory allocation for the read(2) method for video(4)
is not right anymore, and can cause a buffer overflow.

We fix this by queuering the maximum available image size for a device at
attach time.  If the image size should exceed our video(4) buffer after a
video format change (which shouldn't happen), uvideo(4) will gracefully
fail.

Also tested by kettenis@

16 years agonote that fetch hooks are a real bad idea
espie [Sat, 26 Jul 2008 11:31:03 +0000 (11:31 +0000)]
note that fetch hooks are a real bad idea

16 years agodocument update-or-install
espie [Sat, 26 Jul 2008 11:05:32 +0000 (11:05 +0000)]
document update-or-install

16 years agoadd -b to SYNOPSIS and usage();
jmc [Sat, 26 Jul 2008 10:57:09 +0000 (10:57 +0000)]
add -b to SYNOPSIS and usage();

16 years agoMake sure the test array is 64 bit aligned, this now makes sparc fail this
miod [Sat, 26 Jul 2008 10:25:04 +0000 (10:25 +0000)]
Make sure the test array is 64 bit aligned, this now makes sparc fail this
test.

16 years agoAdd -b switch for matching on the last part of the path only.
pyr [Sat, 26 Jul 2008 09:48:00 +0000 (09:48 +0000)]
Add -b switch for matching on the last part of the path only.

ok landry@, otto liked the idea and the original diff, ``do it'' deraadt@

16 years agodeivce -> device typo fix in comment.
krw [Sat, 26 Jul 2008 04:39:51 +0000 (04:39 +0000)]
deivce -> device typo fix in comment.

16 years agosync
todd [Sat, 26 Jul 2008 03:59:41 +0000 (03:59 +0000)]
sync

16 years agoOnly full-dupex is supported for 1000baseT.
brad [Sat, 26 Jul 2008 01:25:08 +0000 (01:25 +0000)]
Only full-dupex is supported for 1000baseT.

16 years agono .Pp before/after .Sh;
jmc [Sat, 26 Jul 2008 01:12:58 +0000 (01:12 +0000)]
no .Pp before/after .Sh;

16 years agoAlso mention half-duplex for 1000baseT.
brad [Sat, 26 Jul 2008 01:11:36 +0000 (01:11 +0000)]
Also mention half-duplex for 1000baseT.

16 years agokill trailing whitespace;
jmc [Sat, 26 Jul 2008 01:11:30 +0000 (01:11 +0000)]
kill trailing whitespace;

16 years agoslight wording tweak to make the media selection entry consistent with
brad [Sat, 26 Jul 2008 00:58:43 +0000 (00:58 +0000)]
slight wording tweak to make the media selection entry consistent with
the entry further down.

16 years agoPull out some mostly useless info from these pages.
brad [Sat, 26 Jul 2008 00:51:38 +0000 (00:51 +0000)]
Pull out some mostly useless info from these pages.

16 years agoUpdate for the fibre support.
brad [Sat, 26 Jul 2008 00:47:27 +0000 (00:47 +0000)]
Update for the fibre support.

16 years agophyio -> physio
miod [Fri, 25 Jul 2008 21:24:49 +0000 (21:24 +0000)]
phyio -> physio

16 years agoImprove eisa bus probe by deciding the number of logical eisa slots from
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).

16 years agomissing .Pp
damien [Fri, 25 Jul 2008 20:54:49 +0000 (20:54 +0000)]
missing .Pp

16 years agoThe PIE address range defaults are too big for alpha and can land on or
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@

16 years agoadd a note in the CAVEATS section about the hardware crypto engine not
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@

16 years agotweak the rt2700 stuff; ok damien
jmc [Fri, 25 Jul 2008 20:36:16 +0000 (20:36 +0000)]
tweak the rt2700 stuff; ok damien

16 years agogroff does not like ultra-long lines without whitespace, since it cannot
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;

16 years agoMention @bin plist element, ok espie@
ian [Fri, 25 Jul 2008 19:39:40 +0000 (19:39 +0000)]
Mention @bin plist element, ok espie@

16 years agoWork around a problem on the t5120 where top(1) would lock up the system for
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@

16 years agolink to updated firmware packages that remove the obsolete @ignore
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@

16 years agodon't redefine INFINITY -- math.h defines it now
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@

16 years agoAfter our recent changes, VIDIOC_S_FMT didn't always return the correct
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.

16 years agodocument RT2700 (tested by bernd@) and try documenting WPA support.
damien [Fri, 25 Jul 2008 16:18:54 +0000 (16:18 +0000)]
document RT2700 (tested by bernd@) and try documenting WPA support.

16 years agoIn gdt_grow, always map the gdt on the secondary cpus too. This far
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

16 years agomuch more correct way of dealing with nfs pending reads/writes
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@

16 years agoSwitch to a temporary stack early on such that we don't lose if the stack
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.

16 years agosync
deraadt [Fri, 25 Jul 2008 14:38:33 +0000 (14:38 +0000)]
sync

16 years agoi have to crank this for a ridiculous reason, to save me about 4 hours of work
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

16 years agodon't undefine C99 math macros, if !_GLIBCPP_USE_C99, so that we
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@.

16 years agodoc fix; from David Higgs
jmc [Fri, 25 Jul 2008 12:29:33 +0000 (12:29 +0000)]
doc fix; from David Higgs

16 years agoCorrect printing of the pg_flags for ddb.
art [Fri, 25 Jul 2008 12:05:04 +0000 (12:05 +0000)]
Correct printing of the pg_flags for ddb.

16 years agosome splassert paranoia.
art [Fri, 25 Jul 2008 12:02:09 +0000 (12:02 +0000)]
some splassert paranoia.

16 years agoflock
art [Fri, 25 Jul 2008 11:57:53 +0000 (11:57 +0000)]
flock

16 years agofile advisory locking tests from FreeBSD. We fail to detect
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.

16 years agoNot all devices support GET_DEF, fall back on GET_CUR if so.
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@

16 years agoshrink MINCLSIZE to MHLEN + 1 again for the release, we're not confident
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

16 years agofix apps that use isinff, isnanf; use ieee 754 mi code, as all
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

16 years agoIn random art visualization, make sure to use the end marker only at the
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@

16 years agoAdd VisualHostKey to example file, ok djm@
grunk [Fri, 25 Jul 2008 06:56:35 +0000 (06:56 +0000)]
Add VisualHostKey to example file, ok djm@

16 years agorepair .Pq use; Alan R. S. Bueno
deraadt [Fri, 25 Jul 2008 05:09:55 +0000 (05:09 +0000)]
repair .Pq use; Alan R. S. Bueno

16 years agosync
deraadt [Fri, 25 Jul 2008 01:29:10 +0000 (01:29 +0000)]
sync

16 years agoAdd "ssh-keygen -F -l" to synopsis (displays fingerprint from
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@

16 years agorelease the bind lock before resetting the signal mask
guenther [Thu, 24 Jul 2008 23:42:26 +0000 (23:42 +0000)]
release the bind lock before resetting the signal mask

ok kurt@

16 years agosync
deraadt [Thu, 24 Jul 2008 22:34:58 +0000 (22:34 +0000)]
sync

16 years agothe softc sc_lock lockmgr lock is unused, so remove it.
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@

16 years agodont need malloc.h anymore; plus two tiny whitespace fixes
thib [Thu, 24 Jul 2008 19:01:28 +0000 (19:01 +0000)]
dont need malloc.h anymore; plus two tiny whitespace fixes

16 years agoremove a bunch of bogus ARGSUSED
thib [Thu, 24 Jul 2008 18:48:18 +0000 (18:48 +0000)]
remove a bunch of bogus ARGSUSED

16 years agoUse memcpy() and memcmp() like everywhere else in ospfd. OK norby@
claudio [Thu, 24 Jul 2008 18:46:59 +0000 (18:46 +0000)]
Use memcpy() and memcmp() like everywhere else in ospfd. OK norby@

16 years agosng_exponent should be sng_exp (MI code now cares); ok martynas
deraadt [Thu, 24 Jul 2008 17:36:52 +0000 (17:36 +0000)]
sng_exponent should be sng_exp (MI code now cares); ok martynas

16 years agoAdd timeout_add_sec() etc. to .Nm and, on jmc's request, tweak
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

16 years agoSpecify a cdroot entry that will actually work on a CD.
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@

16 years agoAdd code to decode machine checks on Avanti, providing a description of the
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.

16 years agobump because of __nan in math.h
deraadt [Thu, 24 Jul 2008 15:51:07 +0000 (15:51 +0000)]
bump because of __nan in math.h

16 years agobump
martynas [Thu, 24 Jul 2008 15:20:24 +0000 (15:20 +0000)]
bump

16 years agoRemove some obsolete XXX comments.
mglocker [Thu, 24 Jul 2008 14:59:44 +0000 (14:59 +0000)]
Remove some obsolete XXX comments.

16 years agokill whitespace at eol;
jmc [Thu, 24 Jul 2008 14:19:24 +0000 (14:19 +0000)]
kill whitespace at eol;

16 years agoisinf/isnan are macros now, therefore ap_config_auto.h differ. regen
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@.

16 years agoSimplify uvideo_enum_fmt(), no functional changes.
mglocker [Thu, 24 Jul 2008 13:30:10 +0000 (13:30 +0000)]
Simplify uvideo_enum_fmt(), no functional changes.

16 years agoreserve a few OpenBSD MIBs that might be implemented later or to avoid
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.

16 years agoipsec is glued into the stack in a very weird way, violating all kinds
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

16 years agocheck sysctl return value
henning [Thu, 24 Jul 2008 10:52:43 +0000 (10:52 +0000)]
check sysctl return value
From: Gleydson Soares <gsoares@gmail.com>, ryan ok

16 years agounXXX prototypes: exp2, remquo, nan, exp2f, remquof, nanf
martynas [Thu, 24 Jul 2008 09:41:58 +0000 (09:41 +0000)]
unXXX prototypes: exp2, remquo, nan, exp2f, remquof, nanf

16 years ago- remove isinff, isnanf. this has been moved to libc
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@

16 years ago- add NAN definition (except on VAX)
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@

16 years ago- move isinf, isnan dups to gen, since most is ieee 754
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

16 years agoDo not define vax-inspired LOWPAGES and HIGHPAGES anymore, they make no
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.

16 years agoIf /dev/video* is already used by an application, return EBUSY to other
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@

16 years agowe really need unsigned in arithmetic operations.
av [Wed, 23 Jul 2008 21:33:32 +0000 (21:33 +0000)]
we really need unsigned in arithmetic operations.
comments by fgsch

16 years agoAvoid an occasional double free in getpwent() which corrupted memory. It
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

16 years agoForgot to add rebranded AS800 names there.
miod [Wed, 23 Jul 2008 19:28:27 +0000 (19:28 +0000)]
Forgot to add rebranded AS800 names there.

16 years agoAttempt to sync with reality, and make the supported/unsuported status of
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.

16 years agoDo not rely on an <ichc>invisible cast</ichc> to return failure in
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@

16 years agoregen
brad [Wed, 23 Jul 2008 18:35:31 +0000 (18:35 +0000)]
regen

16 years agoAttansic L1E PCI id.
brad [Wed, 23 Jul 2008 18:34:29 +0000 (18:34 +0000)]
Attansic L1E PCI id.

16 years agoclear sign bit, instead of comparing to zero and setting x=-x.
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

16 years agosync
deraadt [Wed, 23 Jul 2008 17:59:25 +0000 (17:59 +0000)]
sync

16 years agomake this real C
deraadt [Wed, 23 Jul 2008 17:40:29 +0000 (17:40 +0000)]
make this real C

16 years agoSo it turns outthe HP engineers changed the PA-RISC 2.0 architecture
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.

16 years agofix stack abuse in the `time' commmand, using alloc()'d memory instead.
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@