4f60437127
minor api cleanup, and remove useless VOP_GETATTR() calls.
2006-07-15 20:07:36 +00:00
a904c6895a
update my email on the copyright to @netbsd.org
2006-07-15 16:48:51 +00:00
a890e27076
move veriexec_clear() from dev/verified_exec.c to kern/kern_verifiedexec.c
2006-07-15 16:43:35 +00:00
e3f6ad54bf
dont use magic numbers.
2006-07-15 16:42:12 +00:00
904a157292
some cleanup and fixes:
...
- fix possible panic and vfs refcnt issue
- use log(9) instead of printf(9) where possible
- indent
- stop logging fsid/fileid
2006-07-15 16:33:16 +00:00
631d071e84
FHANDLE_SIZE_MIN is an allowed value for the requested size (it happens
...
to be the old static size on 32bit archs, so the compat_30 code uses it)
2006-07-15 16:32:29 +00:00
73f44d02f7
do_setresuid, do_setresgid: use the suser privilege only when necessary.
2006-07-15 06:31:34 +00:00
bfe8806772
kauth_cred_setgroups: fix an assertion.
2006-07-15 05:54:56 +00:00
7e5391a22a
rename boottimebin to timebasebin as this
...
struct actually keeps the start of the UTC
time scale and not the boot time. the relationship
is: utc-time = up-time + timebase.
background: when doing an ACPI sleep the uptime
freezes and on wakeup the tc_setclock() leads to
a new timebasebin - this had no relationship with
a boottime as the structure was previously called.
discussed on tech-kern@
anomalies (moving boottime, uptime describing running time)
where discovered by Arnaud Lacombe.
2006-07-14 23:01:12 +00:00
97b1c42feb
keep NetBSD boottime semantics:
...
- only set at boot
- only tracking delta of set-time operations
-> will keep boottime stable across ACPI sleeps
uptime(1) will report the time since last boot
2006-07-14 22:44:28 +00:00
9a7410ffff
reduce sleep time by slept time for retrys
2006-07-14 22:35:15 +00:00
ac3f10a8d9
make inittimeleft() and gettimeleft() available (for e. g. kern_event.c)
2006-07-14 22:33:27 +00:00
1c8d298b89
move security.setid_core.* to kern.coredump.setid.*, as requested by yamt@.
2006-07-14 21:55:19 +00:00
b5d09ef065
okay, since there was no way to divide this to two commits, here it goes..
...
introduce fileassoc(9), a kernel interface for associating meta-data with
files using in-kernel memory. this is very similar to what we had in
veriexec till now, only abstracted so it can be used more easily by more
consumers.
this also prompted the redesign of the interface, making it work on vnodes
and mounts and not directly on devices and inodes. internally, we still
use file-id but that's gonna change soon... the interface will remain
consistent.
as a result, veriexec went under some heavy changes to conform to the new
interface. since we no longer use device numbers to identify file-systems,
the veriexec sysctl stuff changed too: kern.veriexec.count.dev_N is now
kern.veriexec.tableN.* where 'N' is NOT the device number but rather a
way to distinguish several mounts.
also worth noting is the plugging of unmount/delete operations
wrt/fileassoc and veriexec.
tons of input from yamt@, wrstuden@, martin@, and christos@.
2006-07-14 18:41:40 +00:00
a1c2fd0906
introduce filehandle size limits:
...
- FHANDLE_SIZE_MAX: refuse unreasonable size allocation, esp. when
it's a user-specified value.
- FHANDLE_SIZE_MIN: pad small filehandles with zero for compatibility.
XXX it might be better to push this into filesystem dependent code so that
new filesystems can choose smaller handles.
2006-07-14 18:30:35 +00:00
cf80fa09d3
- sys___getfh30:
...
- restructure code so that it doesn't try to allocate user-specified
unbound amount of memory.
- don't ignore copyout failure in the case of E2BIG.
- rename vfs_copyinfh to vfs_copyinfh_alloc for consistency.
2006-07-14 18:29:40 +00:00
a2f707bc1b
factor out common code.
2006-07-14 16:02:45 +00:00
2d1ef0388d
- fix buffer overruns in fhopen and friends.
...
- share some code among them.
2006-07-14 15:59:29 +00:00
199b3952c8
- rename sleepts to sleeptv since it is a timeval.
...
- don't initialize it needlessly.
- fix the poll code the same way the select code was fixed, so that it
computes the remaining time to sleep properly.
2006-07-14 15:52:44 +00:00
e221f55a3b
sys___getfh30: fix a vnode lock botch in rev.1.244.
2006-07-14 14:28:58 +00:00
387e1097ba
sys___getfh30: remove unnecessary casts.
2006-07-14 14:00:46 +00:00
a360a47f63
regenerate again to have correct source RCS Id
2006-07-13 23:26:24 +00:00
edf136d748
fix typo
2006-07-13 21:51:50 +00:00
a3b5baed42
Fix alignement problems for fhandle_t, exposed by gcc4.1.
...
While touching all vptofh/fhtovp functions, get rid of VFS_MAXFIDSIZ,
version the getfh(2) syscall and explicitly pass the size available in
the filehandle from userland.
Discussed on tech-kern, with lots of help from yamt (thanks!).
2006-07-13 12:00:24 +00:00
af4af92111
Fix behaviour of files with no signatures on exec.
2006-07-09 10:13:53 +00:00
b36308b48b
when adjusting the left over timeout value in selcommon()
...
do actually subtract the elapsed time instead of adding it.
2006-07-08 16:01:25 +00:00
e472b9091e
fix a bug reported by Steven M. Bellovin regarding oversleeping
...
select(2) timeouts. Introduced via timecounter branch from a
tvtohz() conversion.
The left over timeout was not decremented when re-starting
the sleep in select.
2006-07-08 12:16:09 +00:00
d788cc37f6
report true clock resolution based on the frequency information
...
from the underlying counter in clock_getres(). For frequencies
above 1GHz report a resolution if 1 nsec.
2006-07-08 12:10:33 +00:00
d9530c47ba
add DEBUG code to detect modifications on free memory.
2006-07-08 06:01:53 +00:00
34b5f92bac
Don't define bpendtsleep on vax (gcc4 optimizer will duplicate the asm
...
that contains it result in a multiple symbol definition in gas).
2006-07-08 00:23:29 +00:00
8c61e0077d
Request executable memory for LKM code/data areas. This fixes LKMs
...
on alpha (which were broken for more than a year appearently and noone
noticed). (The other archs didn't suffer because their pmap_kenter_pa()
doesn't support non-executable mappings.)
2006-07-05 14:31:00 +00:00
d145ea66dc
change KMEM_QUANTUM_SIZE from sizeof(void *) to (ALIGNBYTES + 1).
...
the latter is larger on eg. sparc.
noted by Christos Zoulas.
http://mail-index.NetBSD.org/port-sparc/2006/07/02/0001.html
2006-07-03 09:18:35 +00:00
2f4b451441
Make sure we have at least PIPE_BUF bytes available in the socket send buffer.
...
Review and comment by yamt.
2006-07-03 02:34:39 +00:00
5f5ceecc09
Revert previous change to bump the socket low watermark to sock_loan_thresh.
...
With sock_loan_thresh=4096, sb_lowat==sb_hiwat, and sowritable will never
be true (even if only a single byte is pending). Some programs (like screen)
expect select() to return that a socket is writable on a socket when there
is space to write to it. XXX: What is the right thing to do here?
2006-07-01 15:38:28 +00:00
6dc2fb00a7
L_CLR(time_adj) each round so adjtime() corrections don't
...
accumulate over time resulting in a constantly speeding/
slowing clock. found with wiz@ in a timecounter non NTP
kernel configuration.
2006-07-01 05:44:26 +00:00
596d823cde
always call ntp initialisation for timecounter systems as
...
the ntp code degenerates to the adjtime implementation in the
non NTP case
2006-07-01 05:41:10 +00:00
1b7ff51599
regenerate.
2006-06-26 21:30:50 +00:00
e2eb31d3a3
version the socket(2) syscall. for compat30 socket, we use
...
EPROTONOSUPPORT instead of EAFNOSUPPORT.
from pavel@ with a little bit of clean up from myself.
XXX: netbsd32 (and perhaps other emulations) should be able
XXX: to call the standard socket calls for this i think, but
XXX: revisit this at another time.
2006-06-26 21:23:56 +00:00
ffa1c23e58
fix VM_BESTFIT.
2006-06-26 10:23:20 +00:00
9595f19850
wrap long lines.
2006-06-26 10:21:59 +00:00
a3b2d62857
sa_stackused, sa_setstackfree: share some common code.
2006-06-26 10:21:34 +00:00
f374633f81
remove some unnecessary casts.
2006-06-25 08:13:28 +00:00
0fca5c447b
move SA related pools to where they are used. make them static.
2006-06-25 08:12:54 +00:00
44c017beb2
sa_makeupcalls: don't leak kernel stack garbage to userland.
2006-06-25 08:12:10 +00:00
8308eb1f7a
implement kmem_zalloc.
2006-06-25 08:10:04 +00:00
23fcb7c3d3
sa_makeupcalls: cleanup usage of sae_sacopyout.
2006-06-25 08:09:10 +00:00
d038c11b60
sa_makeupcalls: simplify code and fix an sau leak on error.
2006-06-25 08:08:13 +00:00
5df39bb5be
sa_makeupcalls: don't allocate big structures on stack.
2006-06-25 08:05:36 +00:00
bc4977819f
1. implement solaris-like vmem. (still primitive, though)
...
2. implement solaris-like kmem_alloc/free api, using #1 .
(note: this implementation is backed by kernel_map, thus can't be
used from interrupt context.)
2006-06-25 08:00:01 +00:00
6108244889
remove unused M_SA.
2006-06-25 07:46:39 +00:00
8075fe8705
don't put the bpendtsleep handy breakpoint in sun2 kernels as the
...
output asm includes it twice causing multiply-defined symbols.
2006-06-24 05:23:06 +00:00
99f30cd9d2
add a comment telling which structure members are zero-initialized,
...
as requested by YAMAMOTO Takashi
2006-06-23 16:17:23 +00:00
e408053d1b
fix a simonb-timecounters regression.
...
the precision of getnanotime() is not suitable for file timestamps.
esp. when it's nfs-exported.
- introduce vfs_timestamp().
(the name is from freebsd. currently merely a wrapper of nanotime())
- for ufs-like filesystems, use it rather than getnanotime().
XXX check other filesystems.
2006-06-23 14:13:02 +00:00
9b00231778
zero-init callback_head by memset(). code is 12 bytes shorter
2006-06-21 17:16:00 +00:00
ece76dd170
Don't leak memory on success. Allocate only the type of struct that we'll
...
need for efficiency.
2006-06-21 13:46:17 +00:00
300d4b1ed6
bump default so_snd.sb_lowat to increase chance to use loaning.
...
the idea to tweak the watermark from Jonathan Stone.
reviewed by Bill Studenmund.
2006-06-21 12:55:12 +00:00
709b2e6f55
don't allocate too much stuff on the stack.
2006-06-20 03:20:44 +00:00
7d3142aa6a
- introduce vfs_composefh() and use it where appropriate.
...
- fix lock/unlock mismatch in sys_getfh.
2006-06-17 07:06:50 +00:00
7b37f4549b
sysctl_security_setidcorename: don't allocate MAXPATHLEN bytes on stack.
2006-06-17 06:54:58 +00:00
5174aa84b6
The UID_MAX limit is not enforced by syscalls such as setreuid(2),
...
so I remove the assertion uid >= 0 && uid <= UID_MAX. This squashes
a bug where Quagga would panic my machine by passing a UID outside
the range [0, UID_MAX].
AFAICT, this restores the historical (pre-kauth) behavior.
It is likely that GIDs do not satisfy the assertion gid >= 0 &&
gid <= GID_MAX, so remove that, too.
Patch from elad.
2006-06-13 22:56:46 +00:00
d05e0bc3e8
Add EAFNOSUPPORT as a possible error if the address family is not
...
supported. This adds further differentiation between which argument to
socket(2) caused the error. No longer are invalid domain (address family)
errors classified as ENOPROTOSUPPORT errors. This should make socket(2)
conform to current POSIX and X/Open standards. Fixes PR/33676.
2006-06-13 21:19:56 +00:00
f755e9e9b8
remove unnecessary arguments from kauth_authorize_process.
...
ie. make it similar to the one found in apple TN.
2006-06-13 13:56:50 +00:00
4602c1be73
sysctl_unpcblist: don't abuse kauth_authorize_process for non-process object.
2006-06-13 13:52:39 +00:00
c1e6396657
sysctl_kern_file, sysctl_kern_file2: don't abuse kauth_authorize_process
...
for non-process objects.
2006-06-13 13:52:06 +00:00
52e88e8188
sysctl_kern_file2: fix an indent.
2006-06-13 13:23:03 +00:00
67894004ec
Don't allocate > 2K on the stack.
2006-06-12 01:25:05 +00:00
c0160ad4aa
don't allocate statvfs on the stack.
2006-06-12 00:22:47 +00:00
1214d130c2
Always make partitions contiguous even if the mbr has gaps. From someone
...
who wants to remain anonymous.
2006-06-11 23:25:23 +00:00
4ba88d2489
Add includes of opt_multiprocessor.h and opt_lockdebug.h where missing.
2006-06-11 07:32:18 +00:00
54cd6fafa8
re-order initialization sequence to have real counters available during autoconfig
2006-06-09 22:47:56 +00:00
698fb4fb6c
make the public declaration of "hardclock_ticks" signed again; other
...
code (kernel timeout/callout) does comparisions with it
2006-06-08 17:23:11 +00:00
de4337ab21
merge FreeBSD timecounters from branch simonb-timecounters
...
- struct timeval time is gone
time.tv_sec -> time_second
- struct timeval mono_time is gone
mono_time.tv_sec -> time_uptime
- access to time via
{get,}{micro,nano,bin}time()
get* versions are fast but less precise
- support NTP nanokernel implementation (NTP API 4)
- further reading:
Timecounter Paper: http://phk.freebsd.dk/pubs/timecounter.pdf
NTP Nanokernel: http://www.eecis.udel.edu/~mills/ntp/html/kern.html
2006-06-07 22:33:33 +00:00
65a5eb3712
Grr, change the code so that it compiles with gcc-3. It was ok with gcc-4.
2006-06-04 16:44:08 +00:00
95e13e63bd
Introduce SA_NOKERNINFO, a flag for SIGINFO not to print kernel messages.
2006-06-03 18:18:26 +00:00
ad909cf73e
export ntp_gettime() and ntp_timestatus(), for use by compat code
2006-05-29 16:43:05 +00:00
4f2e2f7d77
regen
2006-05-29 09:58:51 +00:00
42a3bd3098
Extend "struct ntptimeval" for the needs of "timecounters".
...
Allocate a new syscall for ntp_gettime() and set up COMPAT_30 for
the old one.
2006-05-29 09:57:54 +00:00
07ddfaead3
systrace_seteuid, systrace_setegid:
...
fix bugs in kauth change. don't forget to update p_cred.
2006-05-28 07:08:41 +00:00
04c3beb7b1
make some internal variables static.
2006-05-28 06:52:17 +00:00
b7da9130d1
remove kauth_cred_destroy, which isn't used anymore. ok'ed by Elad Efrat.
2006-05-28 06:49:27 +00:00
e78022e1d6
Limit the size of any kernel buffers allocated by the VOP_READDIR
...
routines to MAXBSIZE.
2006-05-27 23:46:49 +00:00
4e9ca7aa01
callback_head_init: don't forget to initialize ch_running and ch_flags.
...
fix a problem reported by Jeff Rizzo on tech-kern@.
2006-05-27 07:42:42 +00:00
c24f70bcad
move wait points for kva from upper layers to vm_map. PR/33185 #1 .
...
XXX there is a concern about interaction with kva fragmentation.
see: http://mail-index.NetBSD.org/tech-kern/2006/05/11/0000.html
2006-05-25 14:27:28 +00:00
67c564ace6
sa_upcall_userret: yield cpu on stack starvation so that it's
...
killable at least. PR/28612 from ITOH Yasufumi.
2006-05-25 14:13:29 +00:00
3018e35533
Add kauth header for function prototypes
...
Whitespace clean up.
2006-05-25 11:23:11 +00:00
a53726f2a7
kauth_cred_uucmp: fix inversed return code. PR/33546 from Juan RP.
2006-05-24 23:00:49 +00:00
be46b8e46c
KNF. wrap a long line.
2006-05-23 00:43:30 +00:00
1075c99d89
introduce macros, UAREA_TO_USER and USER_TO_UAREA,
...
to convert uarea VA into a pointer to struct user and vice versa,
so that MD code can change the layout in uarea.
2006-05-22 13:43:54 +00:00
b43dc97b09
use consistent order of function specifiers and type specifiers.
...
(int inline -> inline int)
2006-05-22 12:42:01 +00:00
b3e7e1b010
Better implementation of PaX MPROTECT, after looking some more into the
...
code and not trying to use temporary solutions.
Lots of comments and help from YAMAMOTO Takashi, also thanks to the PaX
author for being quick to recognize that something fishy's going on. :)
Hook up in mmap/vmcmd rather than (ugh!) uvm_map_protect().
Next time I suggest to commit a temporary solution just revoke my
commit bit.
2006-05-20 15:45:37 +00:00
0f0124d1a9
remove NOLOCKF and use normal NULL instead.
2006-05-20 12:20:55 +00:00
b22546608e
remove an debug printf slipped into the previous.
2006-05-20 12:19:30 +00:00
a6518311cc
fix F_SETLKW deadlock detection, which has been broken since lwpify.
...
although this doesn't work for processes with multiple lwps, it's better
than not working at all.
2006-05-20 12:06:20 +00:00
cecc761849
make lockfpool static.
2006-05-20 12:04:21 +00:00
408b7c1e5c
move lockf implementation details from sys/lockf.h to kern/vfs_lockf.c.
2006-05-20 12:02:47 +00:00
56a6a2b09b
don't break lkms; pointed out by hannken@ and he@, thanks!
2006-05-18 17:35:49 +00:00
b6894eda07
CTLFLAG_IMMEDIATE doesn't do what I thought it did. from yamt@, thanks!
2006-05-18 17:33:18 +00:00
b5420599d2
timers_alloc: use PR_WAITOK.
2006-05-18 10:09:12 +00:00
ccd6888699
Don't set mature an fd that has been ffree'd
2006-05-16 21:00:02 +00:00
04d63f90b5
Introduce PaX MPROTECT -- mprotect(2) restrictions used to strengthen
...
W^X mappings.
Disabled by default.
First proposed in:
http://mail-index.netbsd.org/tech-security/2005/12/18/0000.html
More information in:
http://pax.grsecurity.net/docs/mprotect.txt
Read relevant parts of options(4) and sysctl(3) before using!
Lots of thanks to the PaX author and Matt Thomas.
2006-05-16 00:08:24 +00:00
215bd95ba4
integrate kauth.
2006-05-14 21:15:11 +00:00
33b0a10da4
add kauth backend.
2006-05-14 21:12:38 +00:00
103d2f520c
XXX: GCC uninitialized.
2006-05-14 05:30:31 +00:00
99b7478875
Initialize an uninitialized variable gcc 4 found
2006-05-14 05:26:59 +00:00
5eed059930
Add MSG_NOSIGNAL (from FreeBSD)
2006-05-11 15:49:44 +00:00
084c052803
quell GCC 4.1 uninitialised variable warnings.
...
XXX: we should audit the tree for which old ones are no longer needed
after getting the older compilers out of the tree..
2006-05-10 21:53:14 +00:00
7729c57a0f
don't allocate struct statvfs on stack as it's too large.
2006-05-10 11:02:29 +00:00
2977973ac1
Initialize dv_properties in config_attach_pseudo(). PR kern/33438
2006-05-08 01:04:09 +00:00
621d724145
Remove the obsolete (and now unused) propdb API.
2006-05-05 18:26:19 +00:00
fb44a8574b
Remove the devprop API and switch everthing over to the new proplib. Add
...
a new device_properties() accessor for device_t that returns the device's
property dictionary.
2006-05-05 18:04:41 +00:00
89a5572015
Regen
2006-05-04 17:50:28 +00:00
9ec9aed92e
fhstat needs to be versioned too (for ino_t). Pointed out by Izumi Tsutsui
2006-05-04 17:48:56 +00:00
86f30eea43
Regen after making VOP_FCNTL take an unlocked vnode.
2006-05-04 16:49:54 +00:00
935530188d
Change VOP_FCNTL to take an unlocked vnode. Approved by wrstuden@.
2006-05-04 16:48:16 +00:00
69ade2726c
Use for in a forever loop as per KNF.
2006-05-02 13:26:07 +00:00
a0e5478237
lwp_suspend: don't forget to decrement p_nrlwps when suspending ourselves.
...
fix PR/33287 from Gregory McGarry.
2006-04-26 11:44:39 +00:00
261fb8266c
sprinkle some const and static.
2006-04-21 14:00:18 +00:00
80dfcaed5b
iostat_alloc: don't forget to initialize io_busy etc.
...
fix weird "time" in iostat.
2006-04-21 13:58:10 +00:00
b9eb99f927
iostat_find/disk_find: constify and simplify.
2006-04-21 13:53:30 +00:00
c1078a4a2a
sysctl_io_stats_setup: HW_IOSTATS is iostats, not drivestats.
2006-04-21 13:52:23 +00:00
f7fc2f12db
remove some unnecessary #include.
2006-04-21 13:51:24 +00:00
86ddb75ddc
iostat_alloc: use waitok allocation.
2006-04-21 13:50:38 +00:00
1a29b7c5c3
unwrap a short line.
2006-04-21 13:49:32 +00:00
6237cabc7d
whitespace.
2006-04-21 13:48:57 +00:00
10df330c85
Prefix iostat structure elements with io_
2006-04-20 12:13:51 +00:00
51ab7131e8
from Christian Ehrhardt:
...
* physio: The first call to ltsleep should apparently use o &obp->b_interlock
instead of bp->b_interlock (bp is probably NULL here).
2006-04-18 09:54:32 +00:00
45af66f763
Make sure we clear cpi_name; reported by Aymeric Vincent.
2006-04-17 22:05:23 +00:00
ae87f36cbf
Ensure creation of kern node.
2006-04-17 03:45:19 +00:00
7ee081e4bd
Move securelevel-specific stuff to its own file.
2006-04-17 03:39:39 +00:00
727facb192
Add a DEBUG check that panics if pool_init() is called more than
...
once on the same pool.
As discussed on tech-kern a few months ago.
2006-04-15 14:23:11 +00:00
6f0c68c543
Spell coalesce correctly.
2006-04-15 05:07:15 +00:00
1e59f0da77
Coverity CID 2727: Simplify code.
2006-04-15 05:05:44 +00:00
cb70a32776
Coverity CID 848: Protect against NULL deref.
2006-04-15 04:58:14 +00:00
df507c4fbb
Coverity CID 846: Simplify code.
2006-04-15 04:56:14 +00:00
76a19fcdb4
Coverity CID 845: Make it clear that devnullfp != NULL.
2006-04-15 04:50:08 +00:00
d269a5fa4d
Coverity CID 760: Protect against NULL deref.
2006-04-15 04:47:11 +00:00
10c2c730a5
Coverity CID 716: Fix query size vs return results initialization.
2006-04-15 04:41:52 +00:00
e073deb555
Coverity CID 602: Remove always true if statement.
2006-04-15 04:33:48 +00:00
b0c1177d65
Coverity CID 840: Add KASSERT before deref.
2006-04-15 04:26:43 +00:00
6b9fd03a44
Make this compile again.
2006-04-15 02:12:49 +00:00
1ec3fc44a3
Coverity CID 1050, 1051: Correct NULL guards.
2006-04-15 00:34:42 +00:00
31162c28c3
Coverity CID 1739: Plug leak.
2006-04-14 23:54:21 +00:00
ce9b807645
Coverity CID 1089: Add more KASSERTs to prevent NULL deref.
2006-04-14 23:15:21 +00:00
ff57dc92a8
Coverity CID 1088: Add KASSERT to prevent NULL pointer deref.
2006-04-14 23:12:14 +00:00
eb5227b874
Functions to support io statistics collection.
2006-04-14 13:17:20 +00:00
3c0adb7d99
Make i/o statistics collection more generic, include tape drives and
...
nfs mounts in the set of devices that statistics will be reported on.
2006-04-14 13:09:05 +00:00
e26a0c7cb1
Coverity CID 2352: Protect all l2 accesses inside the if statement.
2006-04-13 18:55:49 +00:00
5cf38f035b
Strip the chrooted portion of the path from the reported pty path. Reported
...
and tested by Lasse Kliemann. Thanks!
2006-04-13 17:44:24 +00:00
cda5c405e0
Add a KASSERT to document a condition for the PRU_ABORT case.
2006-04-13 04:58:31 +00:00
32d32e7e2a
sodopendfree/sodopendfreel: remove unused "so" argument.
2006-04-11 09:32:58 +00:00
elad