]>
CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
obrien [Sun, 2 Sep 2001 18:39:17 +0000 (18:39 +0000)]
Merge our rev 1.2,3 (use or native files for NFSv3 support)
changes into Amd 6.0.7.
obrien [Sun, 2 Sep 2001 18:26:32 +0000 (18:26 +0000)]
We are returning to the stock (6.0.7) file now.
I don't recall why the rev 1.2 hack is needed. It looks like Amd was
already using our local headers. I no longer have a FreeBSD 2.x box to
test this on, and this will never be MFC'ed to RELENG_2_2. So lets just
decrease the maintenance effort.
obrien [Sun, 2 Sep 2001 18:22:46 +0000 (18:22 +0000)]
We are returning to the stock (6.0.7) file now.
+ The rev 1.4 addition was taken from a snapshot previous to 6.0.7, so
it is included in 6.0.7.
+ The vendor sources are now POSIX [me harder] compliant.
obrien [Sun, 2 Sep 2001 18:20:22 +0000 (18:20 +0000)]
We are returning to the stock (6.0.7) file now.
My "nfsv2" mount option hack was added to the vendor sources.
obrien [Sun, 2 Sep 2001 18:17:08 +0000 (18:17 +0000)]
We are returning to the stock (6.0.7) file now.
The vendor sources are now POSIX [me harder] compliant.
obrien [Sun, 2 Sep 2001 18:15:24 +0000 (18:15 +0000)]
We are returning to the stock (6.0.7) file now.
The bufov vulnerability was fixed in the vendor sources.
obrien [Sun, 2 Sep 2001 18:13:17 +0000 (18:13 +0000)]
Merge rev 1.7 (mount directory change) into Am-utils 6.0.7.
sheldonh [Sun, 2 Sep 2001 17:24:19 +0000 (17:24 +0000)]
Extend the functionality offered by the -o option into a new option
-O, which limits the impact of the write-only restriction to guest
users.
*) The existing manual page's SYNOPSIS and option listing in the
DESCRIPTION are already horribly disordered. No attempt has been
made to fix this.
*) The existing source's getopt() optstring and option handling switch
are already horribly disordered. No attempt has been made to fix
this.
Discussed with: nik, -audit
obrien [Sun, 2 Sep 2001 17:19:13 +0000 (17:19 +0000)]
This commit was generated by cvs2svn to compensate for changes in r82794,
which included commits to RCS files with non-trunk default branches.
obrien [Sun, 2 Sep 2001 17:19:13 +0000 (17:19 +0000)]
Virgin import of AMD (am-utils) v6.0.7
ache [Sun, 2 Sep 2001 14:40:51 +0000 (14:40 +0000)]
File positions are off_t nowdays, not long, so:
fseek -> fseeko
ftell -> ftello
NOTE: fseek/ftell not works for >long offsets per POSIX:
[EOVERFLOW] For fseek( ), the resulting file offset would be a value which
cannot be represented correctly in an object of type long.
[EOVERFLOW] For ftell ( ), the current file offset cannot be represented
correctly in an object of type long.
ache [Sun, 2 Sep 2001 14:18:28 +0000 (14:18 +0000)]
long -> off_t
long -> time_t
%ld -> %qd
fseek -> fseeko
NOTE: that fseek not works for >long offsets per POSIX:
[EOVERFLOW] For fseek( ), the resulting file offset would be a value which
cannot be represented correctly in an object of type long.
shiba [Sun, 2 Sep 2001 13:05:00 +0000 (13:05 +0000)]
Always turned on 8bit access card support for the fe driver
both i386/pc98, so options FE_8BIT_SUPPORT was deleted.
Reviewed by: nyan
peter [Sun, 2 Sep 2001 12:24:38 +0000 (12:24 +0000)]
The !RESTARTABLE_PANICS code has some loose ends.
peter [Sun, 2 Sep 2001 12:23:08 +0000 (12:23 +0000)]
For ia64, set the default elf brand to be FreeBSD. This is temporarily
necessary only for as long as we're using a linux toolchain.
peter [Sun, 2 Sep 2001 10:07:09 +0000 (10:07 +0000)]
Sync with i386 / alpha. Whitespace unindent / style prep for kse.
n_hibma [Sun, 2 Sep 2001 09:31:08 +0000 (09:31 +0000)]
Add quirks for Kingbyte USB Pen drive.
Submitted by: Eugene M. Kim <gene@nttmcl.com>
PR: 29530
n_hibma [Sun, 2 Sep 2001 09:26:14 +0000 (09:26 +0000)]
Only clear endpoint stall if status was USBD_STALLED.
This avoids panicing the system by unplugging a hub. The interrupt transfer
would sometimes arrive after the driver had been removed.
peter [Sun, 2 Sep 2001 07:47:47 +0000 (07:47 +0000)]
Merge from i386: various cleanups including moving the map calculations
to MI code. This gets ia64 to compile again.
non [Sun, 2 Sep 2001 07:18:29 +0000 (07:18 +0000)]
Fix bus timeout bug which might happen when nsp do suspend I/O burst write.
A nsp chip does suspend I/O write by 512bytes burst write,
though the chip only has 48 bytes FIFO. The chip assert I/O WAIT
signal to PC-Card bus after the CPU writes more than 48 bytes to
the chip if the SCSI device does not respond immediately in supsend
I/O burst write. If the device does not respond for a while it might
cause PC-Card bus timeout.
The previous work around was to wait the request from SCSI device.
But there are some devices which request bytes for synchronous transfer
immediately. So current work aound is to fill 32bytes FIFO, wait for
FIFO empty and burst write 512-32 bytes for every 512 bytes block.
Submitted-by: Honda-san (the author of the driver)
Obtained-from: NetBSD/pc98
shiba [Sun, 2 Sep 2001 06:43:45 +0000 (06:43 +0000)]
Make the fe driver kernel object module to compile in default.
Reviewed by: imp
non [Sun, 2 Sep 2001 06:42:40 +0000 (06:42 +0000)]
Demand minimum I/O size rather than 0 when it calls
bus_alloc_resource(dev, SYS_RES_IOPORT, ...)
Pointed-out by: Yamamoto-san shigeru@iij.ad.jp
shiba [Sun, 2 Sep 2001 06:37:41 +0000 (06:37 +0000)]
Update cis tuple parser, add a pccarddevs entry,
and improve PCCARD_IVAR_ETHADDR in pccard_read_ivar().
Change points:
(1) Read Function Ext tuple.
(2) Add Ratoc REX-R280 entry(fe driver).
(3) Take ether address from function ext tuple.
Reviewed by: imp
Obtained from: NetBSD
shiba [Sun, 2 Sep 2001 06:28:37 +0000 (06:28 +0000)]
Make the fe driver to compile in default under a NEWCARD kernel.
Reviewed by: imp
iwasaki [Sun, 2 Sep 2001 06:28:20 +0000 (06:28 +0000)]
Fix typo; CTLFLAG_RO -> CTLFLAG_RD.
shiba [Sun, 2 Sep 2001 06:27:07 +0000 (06:27 +0000)]
Update the fe driver to probe and attach under a NEWCARD kernel.
Reviewed by: imp
Obtained from: NetBSD
obrien [Sun, 2 Sep 2001 04:06:48 +0000 (04:06 +0000)]
Document my tweaks better.
jhb [Sun, 2 Sep 2001 01:05:36 +0000 (01:05 +0000)]
Use sched_lock to protect rtp_to_pri() and pri_to_rtp() when needed.
jhb [Sun, 2 Sep 2001 00:52:11 +0000 (00:52 +0000)]
Protect pri_to_rtp() with sched_lock when needed.
logo [Sun, 2 Sep 2001 00:45:41 +0000 (00:45 +0000)]
From Mike Silbersack's message on cvs-commits:
> > Doh! There is a sx man page, the mutex man page just doesn't have a SEE
> > ALSO pointing to sx(9). Could some mdoc madman go add one?
ache [Sat, 1 Sep 2001 23:36:40 +0000 (23:36 +0000)]
strtol -> strtoll (for off_t file size)
ache [Sat, 1 Sep 2001 23:26:40 +0000 (23:26 +0000)]
File positions are off_t nowdays, not long, so:
fseek -> fseeko
ftell -> ftello
NOTE: that fseek/ftell not works for >long offsets per POSIX:
[EOVERFLOW] For fseek( ), the resulting file offset would be a value which
cannot be represented correctly in an object of type long.
[EOVERFLOW] For ftell ( ), the current file offset cannot be represented
correctly in an object of type long.
jedgar [Sat, 1 Sep 2001 23:18:15 +0000 (23:18 +0000)]
Use ACL_PERM_NONE instead of hardcoding 0 when initializing
ACL entry permissions.
Reviewed by: rwatson
jedgar [Sat, 1 Sep 2001 23:16:02 +0000 (23:16 +0000)]
In the case of ACL_OTHER and undefined ACL entry id's, set
ae_id to ACL_UNDEFINED_ID instead of 0.
Reviewed by: rwatson
n_hibma [Sat, 1 Sep 2001 23:06:14 +0000 (23:06 +0000)]
Small nit: Make both prints use 'at device %d.%d'.
ache [Sat, 1 Sep 2001 23:01:29 +0000 (23:01 +0000)]
File positions are off_t nowdays, not long, so:
long -> off_t
fseek -> fseeko
NOTE: that fseek not works for >long offsets files per POSIX:
[EOVERFLOW] For fseek( ), the resulting file offset would be a value which
cannot be represented correctly in an object of type long.
ache [Sat, 1 Sep 2001 22:42:47 +0000 (22:42 +0000)]
File positions are off_t nowdays, not long, so:
strtol -> strtoll
fseek -> fseeko
NOTE: that fseek not works for >long offsets files per POSIX:
[EOVERFLOW] For fseek( ), the resulting file offset would be a value which
cannot be represented correctly in an object of type long.
msmith [Sat, 1 Sep 2001 22:41:53 +0000 (22:41 +0000)]
Add a MODULE_VERSION declaration. This should prevent duplicate loading
of the module, and allows other modules to depend on and link against
the ACPI module.
Add a sysctl that allows us to retrieve the ACPI CA version number as
well.
ache [Sat, 1 Sep 2001 22:22:45 +0000 (22:22 +0000)]
File positions are off_t nowdays, not long, so:
long -> off_t
strtol -> strtoll
fseek -> fseeko
NOTE: that fseek not works for >long offsets files per POSIX:
[EOVERFLOW] For fseek( ), the resulting file offset would be a value which
cannot be represented correctly in an object of type long.
chris [Sat, 1 Sep 2001 21:32:52 +0000 (21:32 +0000)]
Another punctuation fix (missing comma ending preposition).
chris [Sat, 1 Sep 2001 21:31:33 +0000 (21:31 +0000)]
o Mention the fact that specifying a fileid of 0 searches all loaded
modules.
o Properly terminate a preposition.
luigi [Sat, 1 Sep 2001 21:19:15 +0000 (21:19 +0000)]
use correct "compile" directory, which is ../../compile/<filename>
This does not apply to RELENG_4 for which the compile directory is
../compile/<filename>
rwatson [Sat, 1 Sep 2001 21:00:28 +0000 (21:00 +0000)]
o Note that some sites will want to select generally more conservative
permissions on some files, and give hints as to what those permissions
might be. Note also that the current more liberal permissions might
get changed in future revisions.
jhb [Sat, 1 Sep 2001 20:17:43 +0000 (20:17 +0000)]
Remove #if 0'd remnants of the old idle page zeroing.
jhb [Sat, 1 Sep 2001 20:16:30 +0000 (20:16 +0000)]
Process priority is locked by the sched_lock, not the proc lock.
rwatson [Sat, 1 Sep 2001 20:11:05 +0000 (20:11 +0000)]
o At some point, unmounting a non-EA file system with EA's compiled
in got a bit broken, when ufs_extattr_stop() was called and failed,
ufs_extattr_destroy() would panic. This makes the call to destroy()
conditional on the success of stop().
Submitted by: Christian Carstensen <cc@devcon.net>
Obtained from: TrustedBSD Project
dillon [Sat, 1 Sep 2001 19:37:41 +0000 (19:37 +0000)]
Regenerate syscalls
dillon [Sat, 1 Sep 2001 19:36:48 +0000 (19:36 +0000)]
Synchronize syscalls.master(s) with recent Giant pushdown work
dillon [Sat, 1 Sep 2001 19:34:23 +0000 (19:34 +0000)]
Giant Pushdown:
read() pread() readv() write () pwrite() writev() ioctl() select ()
poll() openbsd_poll()
luigi [Sat, 1 Sep 2001 19:16:00 +0000 (19:16 +0000)]
Supply a default value for the kernel config binary.
dillon [Sat, 1 Sep 2001 19:04:37 +0000 (19:04 +0000)]
Giant Pushdown. Saved the worst P4 tree breakage for last.
reboot() getpriority() setpriority() rtprio() osetrlimit() ogetrlimit()
setrlimit() getrlimit() getrusage() getpid() getppid() getpgrp()
getpgid() getsid() getgid() getegid() getgroups() setsid() setpgid()
setuid() seteuid() setgid() setegid() setgroups() setreuid() setregid()
setresuid() setresgid() getresuid() getresgid () __setugid() getlogin()
setlogin() modnext() modfnext() modstat() modfind() kldload() kldunload()
kldfind() kldnext() kldstat() kldfirstmod() kldsym() getdtablesize()
dup2() dup() fcntl() close() ofstat() fstat() nfsstat() fpathconf()
flock()
luigi [Sat, 1 Sep 2001 18:27:39 +0000 (18:27 +0000)]
Bugfix: Use the correct variable for picobsd type (same bug is
present in STABLE). Also add a bit more logging (when enabled).
dillon [Sat, 1 Sep 2001 18:19:21 +0000 (18:19 +0000)]
Giant Pushdown
clock_gettime() clock_settime() nanosleep() settimeofday()
adjtime() getitimer() setitimer() __sysctl() ogetkerninfo()
sigaction() osigaction() sigpending() osigpending() osigvec()
osigblock() osigsetmask() sigsuspend() osigsuspend() osigstack()
sigaltstack() kill() okillpg() trapsignal() nosys()
marcel [Sat, 1 Sep 2001 18:11:45 +0000 (18:11 +0000)]
Speculatively add this file. It's part of the Linuxulator update
to make it emulate Linux kernel version 2.4.2, which is required
in order to upgrade the linux_base port to RH 7.1.
Note that this file is only needed for 32-bit architectures. To
us this means i386 (for now?)
ache [Sat, 1 Sep 2001 15:28:24 +0000 (15:28 +0000)]
Save errno before function call and restore it on success (because many
internal functions there may fail and set (i.e. overwrite) errno in normal
(not error) situation). In original variant errno testing after call
(as POSIX suggest) is wrong when errno overwrite happens.
ache [Sat, 1 Sep 2001 15:01:37 +0000 (15:01 +0000)]
Remove even more unneded checks, original code can't overflows in that place
ache [Sat, 1 Sep 2001 14:48:45 +0000 (14:48 +0000)]
Remove two checks unneeded now (can't happens)
ache [Sat, 1 Sep 2001 14:40:01 +0000 (14:40 +0000)]
Make fseek(... SEEK_CUR) fails if current file-position is unspecified.
ache [Sat, 1 Sep 2001 14:23:30 +0000 (14:23 +0000)]
Note that prev. commit addition is for ftell/ftello
ache [Sat, 1 Sep 2001 14:11:53 +0000 (14:11 +0000)]
Describe ESPIPE as result of unspecified file-position indicator value.
Add more to SEE ALSO section.
ache [Sat, 1 Sep 2001 14:01:52 +0000 (14:01 +0000)]
Describe file-position behaviour from POSIX
ache [Sat, 1 Sep 2001 12:13:33 +0000 (12:13 +0000)]
Strict in the POSIX sence, if file position is unspecified after ungetc() at
0, return that we can't specify it, i.e. error with ESPIPE.
(hint from: "Peter S. Housel" <housel@acm.org>)
Back out sinit() addition, not needed after various code simplifications.
ache [Sat, 1 Sep 2001 11:21:28 +0000 (11:21 +0000)]
If lseek to wrong value sucessfully happens despite all pre-checks, set __SERR
to indicate that stream becomes inconsistent.
ache [Sat, 1 Sep 2001 11:18:53 +0000 (11:18 +0000)]
If position is underflowed, don't try to hide that fact by recovery, just
return EIO and set __SERR to mark stream as inconsistent.
dd [Sat, 1 Sep 2001 10:52:25 +0000 (10:52 +0000)]
New release note: syslogd -b
n_hibma [Sat, 1 Sep 2001 09:38:45 +0000 (09:38 +0000)]
First check whether the USB transfer was cancelled before starting to
use the data structures related to that keyboard.
n_hibma [Sat, 1 Sep 2001 09:35:46 +0000 (09:35 +0000)]
Regen.
n_hibma [Sat, 1 Sep 2001 09:33:15 +0000 (09:33 +0000)]
Spelling nit.
sos [Sat, 1 Sep 2001 08:47:11 +0000 (08:47 +0000)]
Make "atapi less" kernels link again.
Question is if we should allow the ioctl only interface to be
created if atapi devices are present, but no driver linked in....
dd [Sat, 1 Sep 2001 08:42:49 +0000 (08:42 +0000)]
Introduce a -b option that allows the user to specify which address to
bind to. This is useful for hosts running jails that need syslog to
maintain an open socket to log to a remote host.
Reviewed by: sheldonh
dillon [Sat, 1 Sep 2001 08:31:09 +0000 (08:31 +0000)]
Undo last commit. The problem's been fixed for a long time.
kris [Sat, 1 Sep 2001 08:01:06 +0000 (08:01 +0000)]
Update a duplicate quotation.
PR: 30228
MFC after: 2 weeks
kris [Sat, 1 Sep 2001 07:57:35 +0000 (07:57 +0000)]
Fix some unsafe signal handlers, and be careful not to overflow on fd_set.
Fix some string buffer operations.
Based on: OpenBSD
Reviewed by: audit
MFC after: 2 weeks
kris [Sat, 1 Sep 2001 07:40:19 +0000 (07:40 +0000)]
Don't do unsafe activities in the syscall handler.
MFC after: 2 weeks
kris [Sat, 1 Sep 2001 07:35:25 +0000 (07:35 +0000)]
Remove some unsafe function calls from the signal handlers.
Obtained from: OpenBSD
Reviewed by: audit
MFC after: 2 weeks
dillon [Sat, 1 Sep 2001 05:47:58 +0000 (05:47 +0000)]
Pushdown Giant for: profil(), ntp_adjtime(), ogethostname(),
osethostname(), ogethostid(), osethostid()
bmah [Sat, 1 Sep 2001 04:39:16 +0000 (04:39 +0000)]
It is *indeed* possible to run Alpha Linux binaries under emulation.
Reviewed by: gallatin
dillon [Sat, 1 Sep 2001 04:37:34 +0000 (04:37 +0000)]
Giant pushdown sys_exit(), [o]wait(), wait4()
dillon [Sat, 1 Sep 2001 04:33:22 +0000 (04:33 +0000)]
Giant Pushdown ACL syscalls:
__acl_get_file, __acl_set_file, __acl_get_fd, __acl_set_fd,
__acl_delete_file, __acl_delete_fd, __acl_aclcheck_file,
__acl_aclcheck_fd
dillon [Sat, 1 Sep 2001 03:56:12 +0000 (03:56 +0000)]
regenerate syscalls
dillon [Sat, 1 Sep 2001 03:54:09 +0000 (03:54 +0000)]
Make yield() MPSAFE.
Synchronize syscalls.master with all MPSAFE changes to date. Synchronize
new syscall generation follows because yield() will panic if it is out
of sync with syscalls.master.
dillon [Sat, 1 Sep 2001 03:04:31 +0000 (03:04 +0000)]
Pushdown Giant for acct(), kqueue(), kevent(), execve(), fork(),
vfork(), rfork(), jail().
ache [Sat, 1 Sep 2001 01:56:54 +0000 (01:56 +0000)]
Back out disabling ungetc() at 0, use different solution:
keep negative offset internally, but return 0 externally in ftell*()
I.e. use 0 now as 'unspecified value' per POSIX ungetc() description.
jhb [Sat, 1 Sep 2001 00:49:29 +0000 (00:49 +0000)]
Axe stale mp_fixme().
rwatson [Sat, 1 Sep 2001 00:00:50 +0000 (00:00 +0000)]
o Attach cap_cmp.c and cap_copy.c to the build.
o Attach cap_copy_ext.3 and cap_copy_int.3 to the install, and link
cap_size.3 to cap_copy_ext.3.
Submitted by: tmm
Obtained from: TrustedBSD Project
rwatson [Fri, 31 Aug 2001 23:44:51 +0000 (23:44 +0000)]
o More conservative permissions for kerberos.log: 600 instead of 644.
Reviewed by: peter
ache [Fri, 31 Aug 2001 23:06:28 +0000 (23:06 +0000)]
Describe EOVERFLOW I implement per POSIX.
Sort ERRORS section
msmith [Fri, 31 Aug 2001 22:59:04 +0000 (22:59 +0000)]
Don't claim memory resources owned by a PNP0C01 device ("system memory")
as some systems claim the entire physical address space is owned by it.
murray [Fri, 31 Aug 2001 22:49:22 +0000 (22:49 +0000)]
Add an entry for the Zip 250.
PR: i386/29639
Submitted by: David Yeske <dyeske@yahoo.com>
dillon [Fri, 31 Aug 2001 22:39:36 +0000 (22:39 +0000)]
Pushdown Giant for nfs syscalls (nfssvc())
dillon [Fri, 31 Aug 2001 22:34:40 +0000 (22:34 +0000)]
Make various posix4 system calls MPSAFE (will fixup syscalls.master later)
sched_setparam()
sched_getparam()
sched_setscheduler()
sched_getscheduler()
sched_yield()
sched_get_priority_max()
sched_get_priority_min()
sched_rr_get_interval()
murray [Fri, 31 Aug 2001 22:18:50 +0000 (22:18 +0000)]
Add entries for 3wire terminals. (carrier not supplied, so we set nc).
PR: conf/26420
Submitted by: Jon Passki <cykyc@yahoo.com>
dillon [Fri, 31 Aug 2001 22:15:37 +0000 (22:15 +0000)]
make swapon() MPSAFE (will adjust syscalls.master later)
dillon [Fri, 31 Aug 2001 22:12:32 +0000 (22:12 +0000)]
Add FD_SETSIZE warning to types.h
dillon [Fri, 31 Aug 2001 22:10:03 +0000 (22:10 +0000)]
mark obreak() and ovadvise() as being MPSAFE
murray [Fri, 31 Aug 2001 21:51:00 +0000 (21:51 +0000)]
Add ID for the Best Data Smart One 56SPS Voice modem.
PR: kern/24851
Submitted by: Daniel Rudy <dcrudy@aol.com>
mjacob [Fri, 31 Aug 2001 21:50:09 +0000 (21:50 +0000)]
add 2300/2312 notes
mp [Fri, 31 Aug 2001 21:50:06 +0000 (21:50 +0000)]
Exit gracefully when a SIGHUP is received. This prevents ee from going into
an infinite spin loop when the terminal window is forcibly blown away.
PR: 29553
Reported by: Sung N. Cho <sucho2@vt.edu>
MFC after: 1 day
rwatson [Fri, 31 Aug 2001 21:44:12 +0000 (21:44 +0000)]
o Screw over users of the kern.{security.,}suser_permitted sysctl again,
by renaming it to kern.security.suser_enabled. This makes the name
consistent with other use: "permitted" now refers to a specific right
or privilege, whereas "enabled" refers to a feature. As this hasn't
been MFC'd, and using this destroys a running system currently, I believe
the user base of the sysctl will not be too unhappy.
o While I'm at it, un-staticize and export the supporting variable, as it
will be used by kern_cap.c shortly.
Obtained from: TrustedBSD Project
mjacob [Fri, 31 Aug 2001 21:43:35 +0000 (21:43 +0000)]
add 2300/2312 references