]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
20 years agoAdd Sandberg USB to Serial Link (model number 133-08) to the list of
simon [Sat, 24 Jan 2004 22:12:06 +0000 (22:12 +0000)]
Add Sandberg USB to Serial Link (model number 133-08) to the list of
supported devices.

MFC after: 1 week

20 years agoDeal with MOD_FREQUENCY before MOD_OFFSET because the latter is the
phk [Sat, 24 Jan 2004 21:48:43 +0000 (21:48 +0000)]
Deal with MOD_FREQUENCY before MOD_OFFSET because the latter is the
one which runs the actual update.  This fixes a bug where there were
a delay in applying the frequency adjustment.  In extreme cases this
could result in marginal stability of the kernel-pll.

20 years ago - Remove local changes that leaked into my last commit.
jeff [Sat, 24 Jan 2004 21:45:25 +0000 (21:45 +0000)]
 - Remove local changes that leaked into my last commit.

Spotted by: juli

20 years ago - Recruit some new ULE users by making it the default scheduler in GENERIC.
jeff [Sat, 24 Jan 2004 21:38:52 +0000 (21:38 +0000)]
 - Recruit some new ULE users by making it the default scheduler in GENERIC.
   ULE will be in a probationary period to determine whether it will be left
   as the default in 5.3 which would likely mean the rest of the 5.x series.

20 years ago - ULE is not exactly experimental anymore. Change some comments and enable
jeff [Sat, 24 Jan 2004 21:32:53 +0000 (21:32 +0000)]
 - ULE is not exactly experimental anymore.  Change some comments and enable
   it in LINT.

20 years ago1. Statically initialize swap_pager_full and swap_pager_almost_full to the
alc [Sat, 24 Jan 2004 21:31:06 +0000 (21:31 +0000)]
1. Statically initialize swap_pager_full and swap_pager_almost_full to the
   full state.  (When swap is added their state will change appropriately.)
2. Set swap_pager_full and swap_pager_almost_full to the full state when
   the last swap device is removed.
Combined these changes eliminate nonsense messages from the kernel on swap-
less machines.

Item 2 submitted by: Divacky Roman <xdivac02@stud.fit.vutbr.cz>
Prodding by: phk

20 years agoRegen after rev. 1.161 of usbdevs.
sobomax [Sat, 24 Jan 2004 21:24:36 +0000 (21:24 +0000)]
Regen after rev. 1.161 of usbdevs.

20 years agoAdd support for Crystalfontz CFA-631 USB LCD (uftdi(4) driver).
sobomax [Sat, 24 Jan 2004 21:23:12 +0000 (21:23 +0000)]
Add support for Crystalfontz CFA-631 USB LCD (uftdi(4) driver).

For some very unclear reason this device contains a FTDI 8U232AM USB->COM
adapter, but reports different device id than original 8U232AM. At the same
time, it reports vendor id of FTDI.

Sponsored by: Porta Software Ltd
MFC after: 2 weeks

20 years ago- Move performance-controlling sysctls into hw.p4tcc.* tree;
sobomax [Sat, 24 Jan 2004 21:13:13 +0000 (21:13 +0000)]
- Move performance-controlling sysctls into hw.p4tcc.* tree;

Suggested by:   nate

- get rid of "magick" values in code and make sysctl's reflecting reality
  on processor versions which have one or another frequency "forbidden"
  due to errata.

MFC after:      2 weeks

20 years ago- Move performance-controlling sysctls into hw.p4tcc.* tree;
sobomax [Sat, 24 Jan 2004 21:13:12 +0000 (21:13 +0000)]
- Move performance-controlling sysctls into hw.p4tcc.* tree;

Suggested by:   nate

- get rid of "magick" values in code and make sysctl's reflecting reality
  on processor versions which have one or another frequency "forbidden"
  due to errata.

PR:
Submitted by:
Reviewed by:
Approved by:
Obtained from:
MFC after:      2 weeks

20 years agoMove the test used to determine whether IPFilter is loaded or not
mux [Sat, 24 Jan 2004 20:40:11 +0000 (20:40 +0000)]
Move the test used to determine whether IPFilter is loaded or not
into its own function to avoid a small duplication of code.

20 years agoRename the makefile variable MIBS to BMIBS so that it does not
harti [Sat, 24 Jan 2004 20:14:03 +0000 (20:14 +0000)]
Rename the makefile variable MIBS to BMIBS so that it does not
conflict with the environment variable MIBS that is used by net-snmp.

20 years agoRename the MIBS makefile variable to BMIBS so that it does not conflict
harti [Sat, 24 Jan 2004 20:12:30 +0000 (20:12 +0000)]
Rename the MIBS makefile variable to BMIBS so that it does not conflict
with the environment variable MIBS that is used by net-snmp.

20 years ago - Move smp_topology to subr_smp.c so that it is defined on all architectures.
jeff [Sat, 24 Jan 2004 19:52:48 +0000 (19:52 +0000)]
 - Move smp_topology to subr_smp.c so that it is defined on all architectures.

20 years agoo Pass a correct argument to errx(3).
maxim [Sat, 24 Jan 2004 19:20:09 +0000 (19:20 +0000)]
o Pass a correct argument to errx(3).

PR: bin/61846
Submitted by: Eugene Grosbein
MFC after: 1 week

20 years agoInstead of bogusly complaining about odd file names, handle them properly
des [Sat, 24 Jan 2004 17:32:43 +0000 (17:32 +0000)]
Instead of bogusly complaining about odd file names, handle them properly
by escaping all suspicious characters.

MFC after: 3 days

20 years agoTry to apply consistent indentation.
des [Sat, 24 Jan 2004 15:35:04 +0000 (15:35 +0000)]
Try to apply consistent indentation.

20 years agoInclude the hostname in the history file.
des [Sat, 24 Jan 2004 15:06:36 +0000 (15:06 +0000)]
Include the hostname in the history file.

20 years agoAdd PFIL_HOOKS to the GENERIC kernel configuration, primarily so
nectar [Sat, 24 Jan 2004 14:59:51 +0000 (14:59 +0000)]
Add PFIL_HOOKS to the GENERIC kernel configuration, primarily so
that one can load the IPFilter module (which requires PFIL_HOOKS).

Requested by: Many, for over a year

20 years agoBump the date.
des [Sat, 24 Jan 2004 14:53:25 +0000 (14:53 +0000)]
Bump the date.

20 years agoBump version number to reflect the addition of cvsup support.
des [Sat, 24 Jan 2004 14:53:11 +0000 (14:53 +0000)]
Bump version number to reflect the addition of cvsup support.

20 years agoIgnore incomplete logs when figuring out what branches and platforms
des [Sat, 24 Jan 2004 14:52:44 +0000 (14:52 +0000)]
Ignore incomplete logs when figuring out what branches and platforms
to display.

20 years agoI won't be using stanley any more.
des [Sat, 24 Jan 2004 14:32:54 +0000 (14:32 +0000)]
I won't be using stanley any more.

20 years agoDrop RELENG_5_1 due to lack of resources.
des [Sat, 24 Jan 2004 14:31:25 +0000 (14:31 +0000)]
Drop RELENG_5_1 due to lack of resources.

20 years agoAdd powerpc, which should build cleanly now.
des [Sat, 24 Jan 2004 14:30:40 +0000 (14:30 +0000)]
Add powerpc, which should build cleanly now.

20 years agoCopy workaround from FFS: open device for write access even if
tjr [Sat, 24 Jan 2004 08:43:06 +0000 (08:43 +0000)]
Copy workaround from FFS: open device for write access even if
the user requests a read-only mount. This is necessary because we
don't do the VOP_OPEN again if they upgrade a read-only mount to
read-write.

Noticed by: bde

20 years agoSync to 1.160 of usbdevs
sanpei [Sat, 24 Jan 2004 07:15:23 +0000 (07:15 +0000)]
Sync to 1.160 of usbdevs

20 years agoAdd Support
sanpei [Sat, 24 Jan 2004 07:14:07 +0000 (07:14 +0000)]
Add Support
- 2nd type of GL641USB USB-IDE bridge [1]
- Plextor Plexwriter 40/12/40U [2]

PR: kern/61671 [1]
kern/61803 [2]
Submitted by: Ralf Wenk <RZ-FreeBSD0401@fh-karlsruhe.de> [1]
Feisal Mohammed <feisal@uwi.tt> [2]
MFC after: 1 week

20 years agoAdd missing newlines to some device_printf()s.
wpaul [Sat, 24 Jan 2004 02:48:22 +0000 (02:48 +0000)]
Add missing newlines to some device_printf()s.

Don't do anything in ndis_get_assoc() if the link isn't up (avoids
spurrious "couldn't get bssid" messages on the console).

20 years agoSync the list of supported SiS chips with the hardware notes
brueffer [Sat, 24 Jan 2004 02:02:31 +0000 (02:02 +0000)]
Sync the list of supported SiS chips with the hardware notes

20 years agoWhen writing/reading longs use explicit 32bit temporary values.
mbr [Sat, 24 Jan 2004 01:23:30 +0000 (01:23 +0000)]
When writing/reading longs use explicit 32bit temporary values.

MFNetBSD rev. 1.15 + rev 1.16

PR: bin/60901
Submitted by: Logan Gabriel <gersh@sonn.com>
Obtained from: NetBSD

20 years agoMake references to using dmesg and ps directly on dumps rather than
grog [Sat, 24 Jan 2004 00:55:15 +0000 (00:55 +0000)]
Make references to using dmesg and ps directly on dumps rather than
running gdb just for these purposes.

Submitted by: simokawa

Add references to dmesg(8) and ps(1).
Sort references.

20 years agoAdd SiI3114 to the list of supported chipsets
brueffer [Fri, 23 Jan 2004 23:55:22 +0000 (23:55 +0000)]
Add SiI3114 to the list of supported chipsets

20 years agoIf we have ISP_ROLE_INITIATOR set, make sure that we clear ICBOPT_INI_DISABLE
mjacob [Fri, 23 Jan 2004 23:23:31 +0000 (23:23 +0000)]
If we have ISP_ROLE_INITIATOR set, make sure that we clear ICBOPT_INI_DISABLE
from the fwoptions. Likewise, we *set* ICBOPT_INI_DISABLE if we don't have
initiator role.

20 years agoadd MAKE_WWN_FROM_NODE_NAME macro
mjacob [Fri, 23 Jan 2004 23:22:35 +0000 (23:22 +0000)]
add MAKE_WWN_FROM_NODE_NAME macro

20 years agoAdd firmware major, minor and micro revsions to the ISP_FC_GETHINFO ioctl
mjacob [Fri, 23 Jan 2004 23:22:11 +0000 (23:22 +0000)]
Add firmware major, minor and micro revsions to the ISP_FC_GETHINFO ioctl
structure.

20 years agoDocument the CVSUP and REPOSITORY configuration variables.
des [Fri, 23 Jan 2004 23:06:40 +0000 (23:06 +0000)]
Document the CVSUP and REPOSITORY configuration variables.

20 years agoAdd support for tinderbox.pl's --cvsup and --repository options.
des [Fri, 23 Jan 2004 23:06:14 +0000 (23:06 +0000)]
Add support for tinderbox.pl's --cvsup and --repository options.
Sort %CONFIG.

20 years agoDocument cvsup support.
des [Fri, 23 Jan 2004 23:04:38 +0000 (23:04 +0000)]
Document cvsup support.

20 years agoAdd support for using cvsup instead of cvs to update the source tree.
des [Fri, 23 Jan 2004 23:04:26 +0000 (23:04 +0000)]
Add support for using cvsup instead of cvs to update the source tree.
Fix the "clean" command: don't try to rmdir symlinks, and run chflags
before trying to delete the chroot tree.
Tweak some error and info messages.

20 years agoMerge vfprintf.c, v1.61 and 1.62. For compatibility with __hdtoa()
das [Fri, 23 Jan 2004 22:48:16 +0000 (22:48 +0000)]
Merge vfprintf.c, v1.61 and 1.62.  For compatibility with __hdtoa()
and to reduce diffs between vfprintf.c and vfwprintf.c, declare xdigs*
to be char arrays rather than wchar_t arrays.

In collaboration with: tjr

20 years agoDeal with keys with descriptions that have empty fields in some of
wpaul [Fri, 23 Jan 2004 22:28:22 +0000 (22:28 +0000)]
Deal with keys with descriptions that have empty fields in some of
their lines.

Properly discard PCMCIA device declarations. I plan to support
PCMCIA cards, but they don't work yet, and it appears some .INF files
declare both PCI and PCMCIA device instances.

20 years agoMake sure to keep track of canceled events.
phk [Fri, 23 Jan 2004 21:09:38 +0000 (21:09 +0000)]
Make sure to keep track of canceled events.

Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>

20 years agoDon't grab Giant in crfree(), since prison_free() no longer requires it.
rwatson [Fri, 23 Jan 2004 21:07:52 +0000 (21:07 +0000)]
Don't grab Giant in crfree(), since prison_free() no longer requires it.
The uidinfo code appears to be MPSAFE, and is referenced without Giant
elsewhere.  While this grab of Giant was only made in fairly rare
circumstances (actually GC'ing on refcount==0), grabbing Giant here
potentially introduces lock order issues with any locks held by the
caller.  So this probably won't help performance much unless you change
credentials a lot in an application, and leave a lot of file descriptors
and cached credentials around.  However, it simplifies locking down
consumers of the credential interfaces.

Bumped into by: sam
Appeased: tjr

20 years agoAdd KASSERTS.
phk [Fri, 23 Jan 2004 21:02:49 +0000 (21:02 +0000)]
Add KASSERTS.

Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>

20 years agoDefer the vrele() on a jail's root vnode reference from prison_free()
rwatson [Fri, 23 Jan 2004 20:44:26 +0000 (20:44 +0000)]
Defer the vrele() on a jail's root vnode reference from prison_free()
to a new prison_complete() task run by a task queue.  This removes
a requirement for grabbing Giant in crfree().  Embed the 'struct task'
in 'struct prison' so that we don't have to allocate memory from
prison_free() (which means we also defer the FREE()).

With this change, I believe grabbing Giant from crfree() can now be
removed, but need to check the uidinfo code paths.

To avoid header pollution, move the definition of 'struct task'
to _task.h, and recursively include from taskqueue.h and jail.h; much
preferably to all files including jail.h picking up a requirement to
include taskqueue.h.

Bumped into by: sam
Reviewed by: bde, tjr

20 years agoPlug an insignificant memoryleak.
phk [Fri, 23 Jan 2004 20:40:25 +0000 (20:40 +0000)]
Plug an insignificant memoryleak.

Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>

20 years agoAdd missing newline in printf.
phk [Fri, 23 Jan 2004 20:36:21 +0000 (20:36 +0000)]
Add missing newline in printf.

Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>

20 years agoAdd missing options from pcvt_conf.h to options.i386. This
philip [Fri, 23 Jan 2004 20:30:52 +0000 (20:30 +0000)]
Add missing options from pcvt_conf.h to options.i386.  This
allows the kernel build to survive config when these options
are specified in the config file.

Reviewed by: hm
Approved by: njl (mentor)

20 years agoFix section number in cross reference.
tjr [Fri, 23 Jan 2004 20:29:03 +0000 (20:29 +0000)]
Fix section number in cross reference.

20 years ago* fwohci_pci.c
simokawa [Fri, 23 Jan 2004 17:37:09 +0000 (17:37 +0000)]
* fwohci_pci.c
Improve error message for attach failure.

* sbp_targ.c
- Add speed in struct sbp_targ_login.
- Remove unnecessary htonl().

20 years agoFix bogus "ffsinfo -c 0" example with "ffsinfo -g 0 -l 4".
blackend [Fri, 23 Jan 2004 17:35:09 +0000 (17:35 +0000)]
Fix bogus "ffsinfo -c 0" example with "ffsinfo -g 0 -l 4".

PR: bin/61472
Submitted by: Alex Popa <razor@ldc.ro>
MFC after: 1 week

20 years agoSync with DFBSD v.1.16. Add new codecs IDs, fix some spelling.
matk [Fri, 23 Jan 2004 16:38:54 +0000 (16:38 +0000)]
Sync with DFBSD v.1.16.  Add new codecs IDs, fix some spelling.

Approved by:    des (interim mentor)

20 years agoMFS: Switch from iso-8859-1 to iso-8859-15.
ale [Fri, 23 Jan 2004 16:37:51 +0000 (16:37 +0000)]
MFS: Switch from iso-8859-1 to iso-8859-15.

Approved by: blackend (mentor)

20 years agoCreate the OID and tree files while building the modules and the daemon
harti [Fri, 23 Jan 2004 16:22:49 +0000 (16:22 +0000)]
Create the OID and tree files while building the modules and the daemon
instead of creating them by hand and storing them in the CVS tree. Add
gensnmptree to the bootstrap tools (it is used to generated these files).
This simplifies the update procedure.

Submitted by: ru

20 years agoDon't panic if there are more than 255 interfaces in the system.
ru [Fri, 23 Jan 2004 15:53:23 +0000 (15:53 +0000)]
Don't panic if there are more than 255 interfaces in the system.

20 years agoAllow the socket buffer sizes to be controlled via sysctl(8).
ru [Fri, 23 Jan 2004 14:35:44 +0000 (14:35 +0000)]
Allow the socket buffer sizes to be controlled via sysctl(8).

MFC after: 3 days

20 years agoRemove the MD5_KEY debugging tool
phk [Fri, 23 Jan 2004 11:47:06 +0000 (11:47 +0000)]
Remove the MD5_KEY debugging tool

20 years agoUpdate generated files.
harti [Fri, 23 Jan 2004 11:08:47 +0000 (11:08 +0000)]
Update generated files.

20 years agoCatch up with import of version 1.5a of bsnmp.
harti [Fri, 23 Jan 2004 11:08:17 +0000 (11:08 +0000)]
Catch up with import of version 1.5a of bsnmp.

20 years agoCatch up with the import of version 1.5a of bsnmp.
harti [Fri, 23 Jan 2004 11:07:55 +0000 (11:07 +0000)]
Catch up with the import of version 1.5a of bsnmp.

Fix the script to insert $FreeBSD$ into the generated file instead
of the script's FreeBSD id.

20 years agoStyle: remove an empty line.
harti [Fri, 23 Jan 2004 11:01:57 +0000 (11:01 +0000)]
Style: remove an empty line.

20 years agoRemove no longer necessary debug printfs
phk [Fri, 23 Jan 2004 10:56:16 +0000 (10:56 +0000)]
Remove no longer necessary debug printfs

20 years agoThis commit was generated by cvs2svn to compensate for changes in r124861,
harti [Fri, 23 Jan 2004 10:44:47 +0000 (10:44 +0000)]
This commit was generated by cvs2svn to compensate for changes in r124861,
which included commits to RCS files with non-trunk default branches.

20 years agoVirgin import of bsnmpd 1.5a
harti [Fri, 23 Jan 2004 10:44:47 +0000 (10:44 +0000)]
Virgin import of bsnmpd 1.5a

20 years agoFix breakage in PIO multisector support.
sos [Fri, 23 Jan 2004 10:38:56 +0000 (10:38 +0000)]
Fix breakage in PIO multisector support.

20 years agoFix traceroute where [number of hops] * [number of packets per hop] > 255.
cperciva [Fri, 23 Jan 2004 06:55:15 +0000 (06:55 +0000)]
Fix traceroute where [number of hops] * [number of packets per hop] > 255.

PR: misc/61336
Submitted by: Mike Hibler <mike@cs.utah.edu>
Approved by: rwatson (mentor)

20 years agogrammar
mtm [Fri, 23 Jan 2004 06:37:19 +0000 (06:37 +0000)]
grammar

20 years agoCopy comment about caller check from ffs_read to ffs_extread, don't
ache [Fri, 23 Jan 2004 06:00:41 +0000 (06:00 +0000)]
Copy comment about caller check from ffs_read to ffs_extread, don't
check for uio_resid < 0 here too.

20 years agoFix various panic() strings to reflect true function name to allow
ache [Fri, 23 Jan 2004 05:52:31 +0000 (05:52 +0000)]
Fix various panic() strings to reflect true function name to allow
easy grep.
Small code reorganization to look more logic.
Copy ffs_write check from prev. commit to ffs_extwrite.

20 years agoffs_read:
ache [Fri, 23 Jan 2004 05:38:02 +0000 (05:38 +0000)]
ffs_read:
Replace wrong check returned EFBIG with EOVERFLOW handling from POSIX:

36708 [EOVERFLOW] The file is a regular file, nbyte is greater than 0, the
starting position is before the end-of-file, and the starting position is
greater than or equal to the offset maximum established in the open file
description associated with fildes.

ffs_write:
Replace u_int64_t cast with uoff_t cast which is more natural for types
used.

ffs_write & ffs_read:
Remove uio_offset and uio_resid checks for negative values, the caller
supposed to do it already. Add comments about it.

Reviewed by:    bde

20 years agoPR: docs/58586
wes [Fri, 23 Jan 2004 05:24:06 +0000 (05:24 +0000)]
PR: docs/58586
Submitted by: Chris S.J. Peron <maneo@bsdpro.com>

20 years agoClarify behaviour of ffsinfo: It appends to outfile without
cperciva [Fri, 23 Jan 2004 05:13:22 +0000 (05:13 +0000)]
Clarify behaviour of ffsinfo: It appends to outfile without
removing any existing contents.

PR: bin/61473
Submitted by: Alex Popa <razon@ldc.ro>
MFC after: 7 days
Approved by: rwatson (mentor)

20 years agoDocument fetch(1) environment variables in fetch(1) man page, instead of
wes [Fri, 23 Jan 2004 04:50:48 +0000 (04:50 +0000)]
Document fetch(1) environment variables in fetch(1) man page, instead of
taking lazy users on the arduous journey through SEE ALSO to fetch(3).

PR: docs/61759
Submitted by: Christian S.J. Peron <maneo@bsdpro.com>

20 years agoRemove leftover FREE() from changes in rev 1.50.
andre [Fri, 23 Jan 2004 01:39:12 +0000 (01:39 +0000)]
Remove leftover FREE() from changes in rev 1.50.

Noticed by: Jun Kuriyama <kuriyama@imgsrc.co.jp>

20 years agoUnbreak amd64: Rename calls from panic to __panic
peter [Fri, 23 Jan 2004 01:04:28 +0000 (01:04 +0000)]
Unbreak amd64: Rename calls from panic to __panic

20 years agoSplit the overloaded variable 'win' into two for their specific purposes:
andre [Thu, 22 Jan 2004 23:22:14 +0000 (23:22 +0000)]
Split the overloaded variable 'win' into two for their specific purposes:
recwin and sendwin.  This removes a big source of confusion and makes
following the code much easier.

Reviewed by: sam (mentor)
Obtained from: DragonFlyBSD rev 1.6 (hsu)

20 years agoMove the reduction by one of the syncache limit after the zone has been
andre [Thu, 22 Jan 2004 23:14:48 +0000 (23:14 +0000)]
Move the reduction by one of the syncache limit after the zone has been
allocated.

Reviewed by:    sam (mentor)
Obtained from:  DragonFlyBSD rev 1.6 (hsu)

20 years agoRemove an unused variable and put the sockaddr_in6 onto the stack instead
andre [Thu, 22 Jan 2004 23:10:11 +0000 (23:10 +0000)]
Remove an unused variable and put the sockaddr_in6 onto the stack instead
of malloc'ing it.

Reviewed by: sam (mentor)
Obtained from: DragonFlyBSD rev 1.6 (hsu)

20 years agoFix SYNOPSIS of manual page, clustering no-arg options correctly.
sheldonh [Thu, 22 Jan 2004 21:26:41 +0000 (21:26 +0000)]
Fix SYNOPSIS of manual page, clustering no-arg options correctly.

Sync usage with manpage.

20 years agoTwo fixes for script(1):
cperciva [Thu, 22 Jan 2004 21:01:14 +0000 (21:01 +0000)]
Two fixes for script(1):
1. Don't do tty stuff to stdin if stdin isn't a tty.
2. When running in non-interactive mode, don't select(2)
on the standard input.

This un-breaks non-interactive portupgrade.

PR: bin/59036 [1]
PR: bin/56166, bin/57414, ports/57415, ports/60534 [2]
MFC after: 7 days
Approved by: rwatson (mentor)

20 years agoSerial murderers shouldn't commit suicide. (killall should
cperciva [Thu, 22 Jan 2004 20:56:06 +0000 (20:56 +0000)]
Serial murderers shouldn't commit suicide.  (killall should
avoid killing itself.)

PR: bin/46107
Submitted by: stacey <stacey_@starwhack.net>

20 years agoIf we're going to "add path 'fd/*' unhide", it only makes
cperciva [Thu, 22 Jan 2004 20:53:15 +0000 (20:53 +0000)]
If we're going to "add path 'fd/*' unhide", it only makes
sense to "add path fd unhide" first.

Requested by: mtm
Approved by: rwatson (mentor)

20 years agoWrite 100 times for tomorrow:
phk [Thu, 22 Jan 2004 19:50:06 +0000 (19:50 +0000)]
Write 100 times for tomorrow:
"Always print time_t as %jd, you never know what width it has"

20 years agoInitialize the (i386) frame pointer when setting up a thread
deischen [Thu, 22 Jan 2004 19:15:08 +0000 (19:15 +0000)]
Initialize the (i386) frame pointer when setting up a thread
context.

Submitted by: Marc Olzheim <marcolz@stack.nl>
Tested by: Marc Olzheim <marcolz@stack.nl>

20 years agoFixed broken logic when parsing double quotes.
ru [Thu, 22 Jan 2004 18:17:44 +0000 (18:17 +0000)]
Fixed broken logic when parsing double quotes.

PR: bin/61673
MFC after: 1 month

20 years agoCatch up with etc/Makefile,v 1.324 -- split "install" and "distribution"
ru [Thu, 22 Jan 2004 17:51:02 +0000 (17:51 +0000)]
Catch up with etc/Makefile,v 1.324 -- split "install" and "distribution"
targets, have "distribution" always install original (as from the release
media) files, whether SENDMAIL_MC is set or not.  Do error handling the
make(1) way.

Reviewed by: gshapiro
Approved by: gshapiro

20 years agoFix the PCI attach routine to properly setup the IRQ and port resource
jhb [Thu, 22 Jan 2004 16:07:03 +0000 (16:07 +0000)]
Fix the PCI attach routine to properly setup the IRQ and port resource
rid's and to deallocate resources if a failure occurs during attach.  This
patch also fixes the driver to return failure if bus_alloc_resource() for
the IRQ fails rather than panic'ing on the next line by passing a NULL
resource to bus_setup_intr().  The other attachments already do all this.

Submitted by: Jun Su <csujun@263.net>

20 years agoo Implement the pthread_spin_* functions in libthr.
mtm [Thu, 22 Jan 2004 15:31:56 +0000 (15:31 +0000)]
o Implement the pthread_spin_* functions in libthr.
o Man pages

20 years agoAdd missing free() in exception handlers.
simokawa [Thu, 22 Jan 2004 14:41:17 +0000 (14:41 +0000)]
Add missing free() in exception handlers.

Reported by:    Stanford Metacompilation research group

20 years agoFix generation of random multicast MAC address.
rse [Thu, 22 Jan 2004 13:34:11 +0000 (13:34 +0000)]
Fix generation of random multicast MAC address.

In case no real/physical IEEE 802 address is available, both the expired
"draft-leach-uuids-guids-01" (section "4. Node IDs when no IEEE 802
network card is available") and RFC 2518 (section "6.4.1 Node Field
Generation Without the IEEE 802 Address") recommend (quoted from RFC
2518):

  "The ideal solution is to obtain a 47 bit cryptographic quality random
  number, and use it as the low 47 bits of the node ID, with the _most_
  significant bit of the first octet of the node ID set to 1. This bit
  is the unicast/multicast bit, which will never be set in IEEE 802
  addresses obtained from network cards; hence, there can never be a
  conflict between UUIDs generated by machines with and without network
  cards."

Unfortunately, this incorrectly explains how to implement this and
the FreeBSD UUID generator code inherited this generation bug from
the broken reference code in the standards draft. They should instead
specify the "_least_ significant bit of the first octet of the node ID"
as the multicast bit in a memory and hexadecimal string representation
of a 48-bit IEEE 802 MAC address.

This standards bug arised from a false interpretation, as the multicast
bit is actually the _most_ significant bit in IEEE 802.3 (Ethernet)
_transmission order_ of an IEEE 802 MAC address. The standards authors
forgot that the bitwise order of an _octet_ from a MAC address _memory_
and hexadecimal string representation is still always from left (MSB,
bit 7) to right (LSB, bit 0).

Fortunately, this UUID generation bug could have occurred on systems
without any Ethernet NICs only.

20 years agoOnly enforce -fno-strict-aliasing for optimization levels that
ru [Thu, 22 Jan 2004 10:01:47 +0000 (10:01 +0000)]
Only enforce -fno-strict-aliasing for optimization levels that
imply -fstrict-aliasing.

Reviewed by: bde

20 years agoSince id(1) is in /usr, it may not be available until after filesystem
mtm [Thu, 22 Jan 2004 08:46:03 +0000 (08:46 +0000)]
Since id(1) is in /usr, it may not be available until after filesystem
mounting.

Reminded by: bde

20 years ago- Run makewhatis(1) in etc/Makefile at the end of "install".
ru [Thu, 22 Jan 2004 08:18:37 +0000 (08:18 +0000)]
- Run makewhatis(1) in etc/Makefile at the end of "install".
- Removed redundant and undocumented NO_MAKEDB_RUN knob.

20 years agoUserland signed char fixes for PPC build. Problems were using a char
grehan [Thu, 22 Jan 2004 07:23:36 +0000 (07:23 +0000)]
Userland signed char fixes for PPC build. Problems were using a char
return for getopt() and comparing to -1, ditto with fgetc() and EOF,
and using the kg_nice value from <sys/user.h>

Submitted by: Stefan Farfeleder <stefan@fafoe.narf.at>
Reviewed by: obrien, bde (a while back)
Tested lightly on: ppc, i386, make universe

20 years agoMake proc's kg_nice/ki_nice explicitly signed for PPC. This is a
grehan [Thu, 22 Jan 2004 06:56:00 +0000 (06:56 +0000)]
Make proc's kg_nice/ki_nice explicitly signed for PPC. This is a
no-op on {i386/alpha/ia64/sparc64} where chars are signed by
default. Should help ARM and S390 which also suffer from this.

Tested on: ppc, i386, objdump disasm before/after diffs
Reviewed by: obrien, bde (a while back)

20 years agoRemove references to cryptotest
imp [Thu, 22 Jan 2004 05:57:48 +0000 (05:57 +0000)]
Remove references to cryptotest

20 years agoFix alignment of size field in `ls -lh` -- the width was being computed
cperciva [Thu, 22 Jan 2004 04:33:00 +0000 (04:33 +0000)]
Fix alignment of size field in `ls -lh` -- the width was being computed
from log[10](largest file size), but when outputting in human-friendly
format the width is always at most 4. (eg. "123K", " 12K", "1.2K".)

PR: bin/59320
Approved by: rwatson (mentor)

20 years agoTwo fixes here:
cperciva [Thu, 22 Jan 2004 04:24:15 +0000 (04:24 +0000)]
Two fixes here:
1. If fgets fails, don't go into an infinite cpu-intensive loop.  Instead,
check to see if the terminal still exists, and sleep(1) otherwise.
2. When we check to see if the terminal still exists, make sure we're not
mislead by EINTR.  This could have been a security issue, but fortunately
the current implementation of tcgetattr doesn't EINTR.

PR: bin/60758
Approved by: rwatson (mentor)

20 years agoFix definition of IFM_MODE, which should be refrected the change of
onoe [Thu, 22 Jan 2004 03:23:28 +0000 (03:23 +0000)]
Fix definition of IFM_MODE, which should be refrected the change of
IFM_IEEE80211_ mode.  Also ifconfig(8) must be recompiled.
Pointed out by Sam Leffeler.