the following should do it from the live cd:
fossil/fossil -c 'srv fossil' –f /dev/sdC0/fossil
mount /srv/fossil /n/fossil
if your fossil was configured correctly during install, you can skip
the '-c 'srv fossil'' part.
or, you can try -c 'srv -AWP fossil' on the command line. change
'fossil' to something else (and as the argument to mount) if it
complains that 'fossil' is already used.
Well, cracking the lottery jackpot happens quite often (if people
would buy as many lotter tickets as we've got disitinct data
blocks as we have in larger data storages or network traffic
over several years, it would happen very regularily).
i think what you fail to take into
standard suggestion is the ibm scrollpoint. three buttons and a more
ergonomic scrolling interface than the wheel:
http://web.mit.edu/atic/www/images/mice/ibm.jpg
the original poster asked about OSX and Plan 9, on which the
scrollpoint works quite fine: i've been using one for close to 6 years
now.
I have a couple of those, but found I didn't like the scroll knob.
Under Linux (sorry) at least, it was way too sensitive. I couldn't
press and release
we don't assume students are all 30-years grizzled veterans.
you just turn them into ones ;)
seriously though, as a former intern of Ron's I can only recommend it!
there's no better way to learn new skills and Ron is a great mentor
and a fine master of the programming sword.
everybody who
well, i checked the source. turns out bash 3.2 drops privileges if uid
!= euid and requires the -p flag to allow itself to run in setuid
mode:
$ cp /bin/bash .
$ sudo chown root bash
$ sudo chmod 4755 bash
$ ./bash -p
# id
uid=500(andrey) gid=500(andrey) euid=0(root) groups=500(andrey)
# whoami
On Feb 19, 2008 10:10 PM, Pietro Gagliardi [EMAIL PROTECTED] wrote:
Or perhaps someone should write a book aimed at newbies on using Plan
9 from Bell Labs. It would be called Using Plan 9 from Bell Labs,
typeset in troff, and not go into technical details.
Oh, that one's easy: the single page
p9p rio doesn't use the Plan 9 file model and is not based on file
descriptors. it was written at a much earlier age and is a straight
X11 program that emulates the rio look and feel. it used to be called
'9wm'.
On Feb 7, 2008 7:52 AM, Lluís Batlle [EMAIL PROTECTED] wrote:
Isn't the plan9port
a non-plan9 solution: run rio with -virtuals num and have a few
different desktops, switching between them with the middle button.
i typically run 'rio -virtuals 4'
andrey
On Feb 7, 2008 10:22 AM, [EMAIL PROTECTED] wrote:
Hi all,
to cut it short: how do you guys do it?
Since I am not to
Or even better to forget drawterm and replace it with Inferno.
inferno also needs to know how to display natively on OSX.
the optional argument is obviously a text description of the desired
priority level (e.g., superbad!, or not in a big hurry, care for a
spot of tea?) accompanied by an optional justification in 150 words
or less.
support for interpretive dance descriptions to follow in the next release :p
http://mirtchovski.com/p9/artwork.jpg
a console that's not associated with anything and can be used for
communication between people. i think the chat option has a few extras
like prepending people's names in front of their text and so on. we
used it for quite a while as a substitute for IRC when that got too
noisy.
i'm not sure how
and then mk aborts. Am I doing anything wrong? Thanks.
there's a script in the source directory called p9config, take a look at it.
echo in cpurc. same for rwm...
I hate changing cpurc, but maybe I'm too squeamish. Is that really the way?
i've always carried my config from way back, when such niceties as
cpurc.local didn't exist and cpurc pretty much had to be edited to get
a local cpu server up.
i just put it in a if(test -w /dev/sdC?/ctl)
here's the offending code in clipread (gui-osx/screen.c):
while (*s) {
if (*s == '\r') *s = '\n';
s++;
}
i don't know the reason this is in there, but if you remove this and
recompile you can test whether anything breaks for
How do you check if your disks are still ok?
i used to run cmp on the two mirrorred partitions to verify weekly
that devfs hadn't missed anything. i would imagine if one of the disks
went south cmp would fail.
Anyone ever done a select on a socket in ape?
the links port does that and it works fine, at least for a while.
the code snippet you gave is suspect, although i don't know how that
relates to the stack trace. libssh2 lacks documentation, but from the
little that i read libssh2_channel_read()
Everything is now there, please update your mirrors :)
New videos: Acid Trips, Van Hensbergen, Guardiola, Nichols, Quanstrom
We apologize if your talk is missing or incomplete.
Sincerely: Ensign Ricky
New videos have been uploaded. Including the keynote speaker :)
Still working on Acid Trips and Van Hensbergen's talk.
New talks:
Cox, Evans, Ganti, Ritchie and the THNX tutorial.
also, i'd like to have a word with Mr. Van Hensbergen after class,
regarding Bulgarians and font sizes ;)
oh, it's all there :P
On Dec 12, 2007 10:11 PM, ron minnich [EMAIL PROTECTED] wrote:
It's just a shame about the video camera failure during Lucho's very
fine talk :-)
ron
http://swtch.com/~rsc/talks/
On Dec 12, 2007 11:46 PM, Skip Tavakkolian [EMAIL PROTECTED] wrote:
any pointers to the paper or slides for Russ' talk?
Lucho and I are slowly going through encoding the videos of the IWP9
talks for those who weren't there. To ease the downloading pain we
will announce the ones we've already completed for your viewing
pleasure, and will let you know when we've added more.
The videos are located at:
oops. i think i know what happened. i'll reencode it.
well, i tried to get iMovie to encode one while importing the other.
turns out iMovie is multitasking only in spirit.
once again, sorry about the inconvenience.
On Dec 11, 2007 6:33 PM, Skip Tavakkolian [EMAIL PROTECTED] wrote:
oops. i think i know what happened. i'll reencode it.
same with
better yet: videos. i will gladly handle the reencoding and hosting
for russ' tutorials if someone videotapes them.
On Nov 27, 2007, at 12:51 PM, Pietro Gagliardi wrote:
I'm not coming to IWP9 this year, but I looked at the transcript and
I was interested in some of them (Acid Trips - does
Use q to quit (and
crash, for some threadexitsall-related reason).
that usually means you need a bigger stack for your threads.
It draws nicely then. I can even boot the live Plan 9 CD, but it will
not boot from the installed system for some reason.
If you have an older kernel lying around try booting with it instead
of the latest kernels. i recently did a pull on my parallels
installation and found that 9load does its
in my case commenting out the consinit line does not fix the problem.
i can still load an older kernel in parallels, but crash when
attempting to load a recent one.
that does the trick. cpu kernel. the cutoff point seems to be
somewhere above 2.5megs (after decompression).
what kernel? cpu or terminal?
if it's cpu build a smaller kernel and try that. i ran into this recently
with a pccpuf kernel.
the best way to catch timesync-related issues is to use a stop watch
to measure the duration of 'sleep 60'. seriously :)
On Nov 20, 2007 5:08 PM, erik quanstrom [EMAIL PROTECTED] wrote:
On Tue Nov 20 17:38:05 EST 2007, [EMAIL PROTECTED] wrote:
I figured maybe something is going on with venti
i don't seem to be able to replicate this with my setup. could you try to
rebuild 9pcf with /n/sources/contrib/quanstro/l.s to see if this fixes your
problem? if it does, i'll submit it.
I tried l.s with both consinit() enabled and disabled in 9load with
identical results: booting an older,
to add a bit more, now that we have printing enabled: the kablamo
appears to be in meminit().
the vm is configured with 512 megs of ram. the code immediately
following lowraminit() is e820init, which works fine (i presume) for
kernels that boot fine. here's the output for a normal kernel, a
modified 9pccpuf 2.8 megs in size:
E820: 0009fc00 memory
E820: 0010 2000 memory
still crashes:
x=4231168; pa=8388608
memset 4157440
but now i'm suddenly much more depressed about it...
that will require BL to give each author in there the ability to
push things to the repository.
i suggest using something like rsync (tra, most likely:
http://swtch.com/tra/) to keep track of contrib updates.
On Nov 16, 2007 10:35 PM, [EMAIL PROTECTED] wrote:
I started to mirror the entirety
I suspect a lot of eyecandy would be easier to implement on
plan9. I have been thinking about openGL on plan9. Has
anyone looked at adding openGL to Plan9?
the free opengl libraries (mesagl) were ported long time ago without
any of the optimizations. they were never tied to the windowing
(gets me 4 parallel interactive fdisk's on the four nodes md,d1,d2,d3
-- it seems weird but worked wonderfully well. cpu is just not set up
to do this type of thing, and neither would your consterm be).
we had a cpu that could talk to multiple nodes at the same time
(multiplex i/o) in 2002, i
What do you think does the Look command do?
On 11/5/07, Pietro Gagliardi [EMAIL PROTECTED] wrote:
I still don't get the Look part. I tried this (on this email)
Look Nov
2-clicked on Look. Nothing. Selected Look Nov and 2-clicked. Still
nothing. Put Nov before and tried again.
of literal text indicated by the
argument or, if none is given, by the selected text in the body.
On 11/5/07, Pietro Gagliardi [EMAIL PROTECTED] wrote:
The same thing as grep(1)
On Nov 5, 2007, at 4:26 PM, andrey mirtchovski wrote:
What do you think does the Look command do?
On 11/5/07, Pietro
I believe this might be a bug. Using `Look' with an argument doesn't
seem to work for me either. However, selecting some text in the body and
executing `Look' alone does work.
look with an argument works, you just need to make sure you've
selected both Look and the argument. a shortcut is to
I pass russ's test with flying colors on OSX (native acme from p9p),
and Plan 9. i can't get the extra mouse buttons to work on the X11
acme :)
now, if you want something a bit more difficult, try opening a file
named test(1).txt from within acme :) it's not exactly trivial.
I believe Russ just fixed the problem, try updating your sources.
On 11/5/07, Noah Evans [EMAIL PROTECTED] wrote:
Hey,
How do you guys deal with the the $DISPLAY variable in leopard? The
default on my machine dies because $DISPLAY contains a directory
name(something like:
I haven't gotten to the point where David is, but i'm trying (leopard
isn't approved for use in our organization yet, they expect it to be
in a few months, but i can still use it at home :).
something to add to the knowledge base (although pretty obvious, it
stunned me this morning): if you are
the problem comes from libmp: the machine-dependent files written in
assembly (/src/libmp/386/*-Darwin.s) are causing it (and they are all
broken, not just the one reported). it's beyond me to figure out why
at the moment. i tried a couple of things at random, but that's not a
way to fix a
this is the last piece of the puzzle that i was missing from the OSX
drawterm: the ability to input text when i'm using a different
keyboard layout than the english one (that's bulgarian for me).
this file will pick up the unicode key from the input event and send
it directly to plan9 without
At this point you're best off putting a packet sniffer on the wire
and taking a look to see what is really happening (or not happening).
...then let us know what happened so we can finally have an answer
when google asks to please describe a situation in which you used
tcpdump to diagnose and
looks like your drawterm can't connect to the authentication server's
port. are you specifying '-a' on the command line? do you have
anything filtering port 567? are you behind a nat and have to forward
it?
On 10/22/07, Martin Neubauer [EMAIL PROTECTED] wrote:
I've set up a new auth/cpu/fossil
is this a google interview question? i love those NOT!
first thing to do is to limit the number from above, so at least we
know how large the playing field is:
assume a microprocessor is 1 cm³ on average. the slabs of silicone
used to make such processors are 1 m³ in volume, so one of
available on sources and here: http://mirtchovski.com/p9/GnuGo/
see README.PLAN9 for installation instructions.
sample game:
9grid% gnugo --boardsize 5
GNU Go 3.6
Copyright 1999, 2000, 2001, 2002, 2003 and 2004 by the Free Software
Foundation, Inc. See http://www.gnu.org/software/gnugo/ or
i did not know that:
http://www.intelligentgo.org/Home/Events/Usenix/Usenix%201984.html
On 10/16/07, Bruce Ellis [EMAIL PROTECTED] wrote:
goanna would kick its ass.
brucee
does it happen when you open this specific file, or any other file
from src/johnny? if it's only wc.c that kills it, does it kill it when
you open the file directly (acme wc.c)? if it's wc.c, then is there
something special that you have in that file, special french
characters, perhaps? can we see
i'll finish the sshv2 port of libssh i was working on. send the money to me :)
On 10/1/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
On Mon, 1 Oct 2007, Enrico Weigelt wrote:
* app [EMAIL PROTECTED] wrote:
Hi,
So, if not X, then how about some higher (or eperhaps lower?) layer
How much you think it's worth? :-)
0
I'm curious as to how far along it is :-)
search the 9fans archives :)
i promise you that we'll look at the edge of criticality from below,
as it speeds away at 9.78ms¯² :)
On 10/1/07, ron minnich [EMAIL PROTECTED] wrote:
Andrey, I hope you will restructure the code completely, and be sure
to use transparent accessor functions, spontaneous self-organization,
and
the right thing to do is to fix timesync to notice
such jumps and write them off as suspend/resume
instead of assuming that the cpu has gotten *very* slow.
timesync still needs to set the correct time (a system which was
suspended for, say, 10 hours will live 10 hours in the past). perhaps
- Can I safely assume I will be able to run Plan9 on a VM like
Parallels on the Mac with OS X Leopard?
yes (it doesn't depend on plan9, but rather on Parallels running under Leopard).
- Any idea when plan9port, Inferno, and drawterm will be available for
Leopard?
right after leopard is
On 9/29/07, Francisco J Ballesteros [EMAIL PROTECTED] wrote:
I´m using both a macbook pro and a macbook regularly.
Plan9 on Parallels works fine, although time does funny things when
you suspend (I do not) and some times parallels crashes (not plan 9) when
you start it.
if you don't mind
the system recalculated the cpu clock rate and forces it down the
cpu's throat, which causes time to run much faster than usual.
that statement is incorrect, the throat belongs to the kernel. sorry :)
how stable is the new venti on plan9? how tested?
we tried it here about a year ago on a 1.7TB single raided partition
and for two weeks it wasn't able to stay up for more than 24 hours...
On 9/5/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
I'm about to change the distribution on sources to
there's f2c: a fortran 77 to C converter...
On 8/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Does Plan 9 support any sort of Fotran programming? Thanks for the
help.
Axel, I've resorted to running a cpu server inside parallels and
connecting to it via the OSX-native drawterm. I need this in order to
get full three-button support on my laptop.
I have not experience the effect that you're seeing under vnc, for me
the graphics have always ran fine (including
You know what I'd really like to see? It would be great if the
organizing committee could invite Rob Pike to give a talk: an update
to Systems Software Research Is Irrelevant (preferably with an
accompanying article, so we're not left with just the talk slides).
The reason for this is that the
Thx.
But I don't see any not whether things like symlinks are
represented by file modes.
when you figure that bit out, you'll be enlightened :)
that was a hack to allow locking of terminals as per management's
requirements: http://mirtchovski.com/lanlp9/rlock/index.html
not really sure that's what Antonin was looking for.
there is blanking for vga enabled displays (man 3 vga) but i'm not
sure consoles can actually be blanked if there's
put it in the hostowner's secstore where factotum will pick it up
every time the machine boots. let me know if you need an example.
On 7/27/07, erik quanstrom [EMAIL PROTECTED] wrote:
the sshserve(1) (sic.) man page mentions how to generate a host
key on-the-fly, but doesn't mention where to
factotum /mnt/factotum/ctl
with the host key in the factotum file, of course.
andrey
On 7/27/07, andrey mirtchovski [EMAIL PROTECTED] wrote:
put it in the hostowner's secstore where factotum will pick it up
every time the machine boots. let me know if you need an example.
On 7/27/07, erik
invalidate the nvram first and make sure you set the secstore password
correctly to the one that you have set for hostowner's user in
secstored when it asks you on the next reboot. if you have done
already then simply cat the nvram partition: you should be able to see
it in full and in plain text
i may not be qualified to answer all those questions for you, but i'll
try my best. i was dealing with FP exceptions last week and
consequently may have some of the answers:
1) Is there a Plan 9 policy concerning the notes (exceptions) on
arithmetic operations, both integer and float?
i don't
i just did a pull: it's not only links that fails, it's everything in
/sys/src/ape/cmd :)
from a first glance it looks like not all libraries are linked (a few
pragmas missing?). for example, by adding libc to the mix you get rid
of all warnings but one. the following example is from
oops, the first line in my paste in the previous message should not
have /386/lib/libc.a in it: it's a leftover from editing. the line
should match the output from mk as it shows in /sys/src/ape/cmd/diff
found the problem. change DIRS in /sys/src/ape/lib/ap/mkfile to
include $objtype (the badly named patch mkfile-broken i just submitted
has the fix). you'll need to recompile libap.
the old mkfile had the $objtype hidden in the 'for' loop in install.
putting it in the DIRS definition makes it
upper management, in a concerted effort with the Plan 9 PR department
and damage mitigation team, have requested that the following picture
be disseminated to quell the disturbance among the ranks:
http://mirtchovski.com/screenshots/UnixRoom2007.jpg
ensign ricky
On 7/12/07, yplan9 [EMAIL PROTECTED] wrote:
What a shame !!!
you don't like young jmk? we ran him through a time machine and now we
can keep him working on plan9 for another 40 years before retirement
;)
High Command and the PR department have released the following image
from the engine room, which replaces all occurrences of the previously
referred to Plan9-1995.gif image effective immediately:
http://mirtchovski.com/screenshots/Plan9-jmk.jpg
Please make all necessary adjustments immediately.
colors are hardcoded for applications, however there's a small
loophole you can use on 8-bpp displays to get something that is
almost, but not quite, entirely unlike X's resources.
the 'getmap' command will let you change to a different colormap from
the default. pre-defined cmaps are in
That decision would depend entirely on whom you are making the
software for. You can't possibly decide for a large majority of the
people in the world, considering they are all different with
different perceptions of what works the best for them.
you can if what you're designing is different
On 6/20/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
current problem is, given an ip address, say 135.104.9.7
and an ip mask, say 255.255.255.0,how do i get the result
of addr mask using only sed and echo?
sitting idly at usenix, not listening to the boring talks. it can be
done, but it's
The file was getting
pretty big, so I deleted it.
i cron a C program^W^Wscript to clean up the logs once a month so they
are kept manageable. i don't care if i miss a few minutes' worth:
cpu% cat /bin/chop
#!/bin/rc
cd /sys/log
for (i in \
aan \
auth
i just found a small program i had forgotten about that mimics unix'
lsof. it's in /contrib/andrey/cmd/lsof.c
you can do lsof | grep 'tcp/25' to get the pid of the program.
But, assuming it's useful, why write it in C when you may as well write
it in rc?
why is ps written in C when it may as well be written in rc?
But your lsof is equivalent to about 3 lines of rc script, done almost
entirely by programs written in C, and is much clearer. I doubt that
there's any noteworthy difference in speed, or that it would matter much
if there were, so I just can't imagine why you'd want to write it in C.
either i
cfs can be a problem for a programmer: compile/run/debug/compile
cycles tend to be unfavourably long since everything has to be copied
back to the server and then read again.
see loadimage for the function that creates an Image * from a
bitstream. if you look at the xscreensaves hacks (xscr.tgz in
/contrib/andrey on sources) you'll find a couple that use that hack,
most notably eruption.
you can fullscreen an application in two ways: use that app instead of
rio (then
further to lucho's comments, he and i just confirmed that a converted
v2 image works fine in v3 (except a crash during reboot which used to
happen with v2 too). i'll try a fresh install now and let you know
how it goes.
definitive guide to installing plan9 on parallels v3, compile 4124:
note: if you've done this before you need not bother reading the rest,
just make sure you say no when asked whether you want DMA enabled
for your disks. if you say yes then after boot you'll get the
following error message:
I just updated to build 4144 and Plan 9 works fine on a Macbook Pro with
that's confusing. my copy, downloaded and installed today, says build
4124 (June 7, 2007)
what hardware? fresh install or a converted disk image?
On 6/8/07, David Leimbach [EMAIL PROTECTED] wrote:
On 6/8/07, David Leimbach [EMAIL PROTECTED] wrote:
On 6/8/07, Anant Narayanan [EMAIL PROTECTED] wrote:
russell wrote:
I tried this the other day (on an older parallels and the most
for the record: i've been using a converted image all morning on a
MacBook Pro (one generation old now, 2.1ghz) with no crashes.
for those that have systems that crash, i suggest recording the panic
messages and sending them either to lionkov or to me. we'll get
together with jmk next week and
i got it to freeze by suspending a machine and then trying to stop it
from the suspended state.
turned on the VT-x extensions and now can not reboot it in 9 cases out
of 10. in the one case when i reboot it i have a severely restricted
set of vesa modes to use.
what's interesting is that i need
i'm now running a stock kernel on Parallels with VT-x on and with full
graphics. it still takes a few tries and errors to get it booted up
but at least i have all the available screen sizes that vesa should
support.
what i did: changed the OS Type to Linux and the OS Version to Red
Hat. at this
http://www.joeyoder.com/papers/patterns/BBOM/mud.html
Why does this paper have SO MANY CAPITALIZED WORDS, and what can we do
to improve it?
come on you two, i was throwing a jab at this sentence from the abstract:
Why are so many existing systems architecturally undistinguished, and
what can we do to improve them?
i find google to be a very good search engine for the wiki:
http://www.google.com/search?q=plan9%20site%3Ahttp%3A%2F%2Fplan9.bell-labs.com%2Fwiki%2Fplan9%2F
if you're bored with nothing to do there's a new hack in
/n/sources/contrib/andrey/xscr.tgz called boxfit.
http://mirtchovski.com/p9/xscr/boxfit.png
I have no problems with parallels and plan9 . i started with david's
changes initially but recompiled a brand new kernel just today (after
a source update) which runs just fine.
i'm using parallels more and more often nowadays to run Plan 9. one of
the things that annoy me is the inability to copy/paste between the
hosted plan9 and the hosting osx. i couldn't find a solution anywhere
so i wrote a file server that exports the osx snarf buffer to the
parallels network via
the perl script i use to generate galleries can be rewritten in rc or
awk or whatever:
http://mirtchovski.com/2007-05-04-Karting/
it can be hosted on plan 9 just fine (it was hosted on 9grid for a year or so)
if anyone wants to use it give me a shout in private and i'll send it
to you together
1 - 100 of 327 matches
Mail list logo
