On Thu, Aug 27, 2009 at 9:51 AM, David Leimbach wrote:
> I think there's work going on to use plan 9 to load plan 9 (maybe?) to
> replace 9load.
It's a gsoc project for Iruata to which I just gave a passing grade.
it's doable. It needs a new PBS, which iruata wrote.
ron
On Thu, Aug 27, 2009 at 11:26 AM, Uriel wrote:
> Er, it doesn't need a new PBS, booting Plan 9 from a Plan 9 kernel
> already worked just fine with what russ did years ago.
Hi, to clear up the air, and correct this wrong comment:
we did a few things
- kill the FAT partition
- parse an a.out header
On Thu, Aug 27, 2009 at 12:07 PM, Uriel wrote:
> Because the whole point of the project was to replace 9load, and the
> way plan9 systems tell 9load what kernel to load is using plan9.ini
no, the point of the project was to have a new way to load that did
not require 9load or 9fat or any legacy at
On Thu, Aug 27, 2009 at 11:48 AM, Uriel wrote:
> Can it load and parse plan9.ini?
why do you want to do that? Just wondering.
ron
On Thu, Aug 27, 2009 at 12:27 PM, Uriel wrote:
> plan9.ini is certainly not perfect, and I'm happy to see it changed
> some day, but that was not what the project was about.
I'm glad to know you defined the project. I guess the guy who wrote the code
(Iruata) and the mentor (me) were just confuse
On Thu, Aug 27, 2009 at 1:14 PM, erik quanstrom wrote:
> could someone explain what the disadvantages and problems
> with 9fat are? i'm asking out of ignorance, since 9fat hasn't
> been a problem for me.
8.3. It's burned me time and again.
ron
On Thu, Aug 27, 2009 at 1:35 PM, Uriel wrote:
> Yes, you were confused, and yes, the project was my idea (although
> that was just a restatement of russ original suggestion).
you mean it was russ' original idea which you misunderstood, I expect.
And it's hardly new: we were doing it at LANL 8 yea
Nothing prevents anyone from using 9fat or building on or changing
iruatas work. It is there so go forth and hack. What is nice is his
new pbs goes to 32 bit mode right away so hacking is easier than
before.
Ron
On 8/27/09, Federico G. Benavento wrote:
> the problem comes when you just can't boo
On Thu, Aug 27, 2009 at 10:08 PM, Federico G.
Benavento wrote:
> I really don't get this, I didn't criticize iru's work, I was just pointing
> how
> being able to put the loader in a fat partition was convenient to me,
> just that.
I did not take it that way. Your point is a very important one.
On Thu, Aug 27, 2009 at 6:54 PM, Federico G.
Benavento wrote:
> I could achieve the same as I did by doing "copy 9load E:" on windows
> with this new approach, but I'd need to boot some linux live CD
> and dd my way out to put the new loader there which I'll be too
> hacky and I'd probably need a
On Fri, Aug 28, 2009 at 8:31 AM, ron minnich wrote:
> it was. It is not any longer. But for the first year of our existence,
> 1999-2000, linuxbios really was linux.
plus some special extra bits, of course ...
ron
On Fri, Aug 28, 2009 at 8:19 AM, erik quanstrom wrote:
>> I'm pretty sure Ron has done that too... from LinuxBIOS.
>
> i'm pretty sure that coreboot neé linuxbios is not linux.
it was. It is not any longer. But for the first year of our existence,
1999-2000, linuxbios really was linux. But then li
On Fri, Aug 28, 2009 at 8:24 AM, David Leimbach wrote:
> Could be, I've never had the luxury of trying it all out... however I
> thought a minimal linux from coreboot/linuxbios (I think it was called
> linuxbios when this was tried) could kexec plan 9.
Actually i wrote something in 1999 called lo
On Fri, Aug 28, 2009 at 8:37 AM, David Leimbach wrote:
> Did that pre-date the "two kernel monte" that used to be used on Scyld's
> "Beowulf" thing?
I think Eric Hendriks and I started development at the same time, but
mine was working first. No longer sure.
But Biederman's code won out, and is
I think your first, best bet is to try to find out what the Plan 9
community needs, rather than adding on something that might not be the
that important. I have not heard anyone express a need for zeroconf in
Plan 9, but maybe I'm missing something.
Simple example: file systems on Plan 9 are slow.
I see your point. It does sound like zeroconf would be useful to some
people. I wonder if it could be done with a 9p orientation as eric
suggested.
I don't recall what the security issues are with zeroconf, but, if
it's the microsoft-inspired version I'm thinking of, I would guess
there are many.
One way to make this kind of interesting is to address how you'd do a
reasonable zeroconf effort given that you need to boot 1m+ machines.
We've booted 4400*250 VMs on a machine at sandia, and, let me tell
you, it was a pain. It is amusing to watch the programs traverse
million line /etc/hosts file
Q: "Will C continue to be important into the future?"
(Dave Kirk, Nvidia)A: "No, I think C will die like Fortran has"
ron
On Wed, Sep 2, 2009 at 7:29 AM, ron minnich wrote:
> Q: "Will C continue to be important into the future?"
> (Dave Kirk, Nvidia)A: "No, I think C will die like Fortran has"
let me explain the joke. In HPC circles, people have been predicting
the death of fortran
On Wed, Sep 9, 2009 at 9:27 AM, Fernan Bolando wrote:
> I am trying to port nhc98 the bytecode haskell compiler.
> I am hoping to avoid changing those definition because I think those
> are generated by the
> haskell compiler and then fed to the C compiler. So there are a whole
> bunch of them.
2.7M lines last year
10K lines added a day.
5K lines deleted per day.
I keep thinking this can't be sustained. What happens next?
At the same time, well, as pointed out, we all use it all the time.
I'm sending this from gmail.
Or you can use Linux by googling these stats :-)
ron
There was a GSOC project to add 802.11 support. I mention this because
the code might be simple enough to be useful as a template for Plan 9
wireless drivers.
ron
If I am in 9vx and have imported a file system from somewhere, and do
an ls, and get impatient and hit del, the import dies.
term% grep full *.c
grep: can't open *.c: '*.c' mount rpc error
term% ls
ls: .: clone failed
term%
mnt: proc grep 91: mismatch from /net/tcp/0/data
/n/o/usr/rminnich/9k/bg
On Mon, Sep 21, 2009 at 10:18 AM, Patrick Kelly wrote:
> I was under the impression Plan 9 had a few wireless drivers... am I wrong?
nothing modern. That's the problem.
http://www.etherboot.org/wiki/wirelessboot/drivers
ron
I skipped the panel. Those things are never interesting. I was working
outside and heard lots of laughter however.
ron
Having just read the register article, I'm now doubly glad I wasn't
there I was working on Plan 9 :-)
ron
OK, a little more fooling around.
term% grep full *.c
(wait about a second, hit DEL)
grep: can't open apic.c: 'apic.c' './sys/src/9/pc/grep' does not exist
grep: can't open apm.c: 'apm.c' mount rpc error
grep: can't open archmp.c: 'archmp.c' mount rpc error
grep: can't open bios32.c: 'bios32.c' m
OK, I did this in mntralloc, in the code path in which we reuse the
rpc from the mtnalloc.rpcfree: I just simply always allocated a new
tag, even if we found an old rpc which was nominally free: I always
allocated a new tag, not reusing the old tag.
That fixed the problem.
So, basically, the way
On Tue, Sep 22, 2009 at 11:35 AM, roger peppe wrote:
> surely the correct way to go about this (caveat: i haven't looked at the code)
> is to drop the rpc struct onto the rpcfree list only when the Rflush is
> received?
you just got an Eintr. Did the request get sent?
I don't know. I am not sur
here is a 9p trace of the problem. See line 43. Topen and Tclunk go
out with same tag. This is with a print in the rpc code as suggested
by Russ.
ron
y
Description: Binary data
here's one last "caught in the act" scenario. I have a print in
mntralloc when I reuse something.
The fid is being read and clunked. But the Tclunk goes out before the
Rread comes in. Oops.
Reuse 1
Tread tag 1 fid 454 offset 0 count 8192
Reuse 1
Tclunk tag 1 fid 454
reply Rread tag 1 count 8192
On Tue, Sep 22, 2009 at 2:23 PM, erik quanstrom wrote:
> On Tue Sep 22 17:22:08 EDT 2009, rminn...@gmail.com wrote:
>> here's one last "caught in the act" scenario. I have a print in
>> mntralloc when I reuse something.
>>
>> The fid is being read and clunked. But the Tclunk goes out before the
>>
OK, so what happens in 9vx.
mount sources
cd /n/blah/blah
grep full *.c
hit DEL
devip.c read on Qdata fails, and we do this:
if(r < 0){
oserrstr();
nexterror();
}
So just need to fix oserrstr() or fix this in devip it
sent in via codereview but I missed a debug print I left in. Sorry. I
assume you can reject it so I'm sending the correct patch.
ron
not having done this before, the reference is URL:
http://codereview.appspot.com/122046
ron
I think 9vx is now the equivalent of software tools from the old days.
It's fast. But the big beauty of it for me is that in vx32/src/9vx/a
is pretty much a plan 9 kernel in plan 9 C vernacular. I just spent an
easy short time prototyping some new stuff that I can now drop into a
real plan 9 kerne
http://osel.oregonstate.edu/files/osel_newsletter_200905.pdf
Neat stuff. It looks like something I'd buy, but it's all student
designed, even the lexan case.
ron
On Sat, Sep 26, 2009 at 10:32 AM, John Floren wrote:
> I wonder if
> they'd be willing to go big time and sell these things?
I asked them. "Everyone asks us that". They're not sure.
The zigby is an issue, they did it because the 802.11 chips are really
closed due to FCC fear of unlicensed spec
I think we owe the vx32/9vx guys some help here.
So here's the question. 9vx is running. It breaks. How should we go
about providing
- useful diagnosis
- useful backtracking
There are some tools. Worst case, those of us who see it die from time
to time could elect to run it under gdb for a while.
On Thu, Oct 1, 2009 at 6:23 AM, Sam Watkins wrote:
> I tried to check out v9fs, but the compressed git repo without checkout is
> over
> 300Mb.
That's git for you. When you go to git it, you git ALL of it. Kind of
like deciding to download sources and getting the entire venti arena.
ron
I'm having an issue with replica on 9vx. I keep getting files like this:
d-2 tc staff4096 Oct 4 05:38 bibtex/
d-2 tc staff4096 Oct 4 05:38 bin/
strace shows this:
13293 mkdir("/mnt/sdb1/tc/plan9/9vx-0.12/sys/lib/texmf/bibtex/bst", 0)
= -1 EAC
On Sun, Oct 4, 2009 at 8:22 AM, erik quanstrom wrote:
>> 13293 mkdir("/mnt/sdb1/tc/plan9/9vx-0.12/sys/lib/texmf/bibtex/bst", 0)
>> = -1 EACCES (Permission denied)
>>
>
> perhaps replica is expecting specific error messages
> and they're not getting translated
>
> ; cd /n/sources/plan9/`{pwd}
> ; g
On Sun, Oct 4, 2009 at 10:20 AM, Steve Simon wrote:
>> I don't see this explaining a
>> mkdir with mode of 0 however.
>
> Does the file/dir actually have a mode of zero on the source machine?
no, it has mode 755
This is for the tex install but the problem exists in general when
using applylog on
seems to be a bug in replica on 9vx? I'll look more tomorrow.
mkdir /sys/lib/texmf perm -200
mkdir /sys/lib/texmf/bibtex perm -200
mkdir /sys/lib/texmf/bibtex/bst perm -200
mkdir /sys/lib/texmf/bibtex/bst/base perm -200
mkdir /sys/lib/texmf/bibtex/bst/base/abbrv.bst
On Sun, Oct 4, 2009 at 10:30 PM, erik quanstrom wrote:
>> Do you have the CD-ROM drive attached to the VM while you boot it?
>>
>> I remember Plan 9 didn't start when you have an empty CD-ROM drive
>> connected to the VM on ESX 3.5. So I removed the CD-ROM drive from VM's
>
> if this is the case,
OK, it's an incompatibility between Posix world and Plan 9 world. It's
because mkdir doesn't open the directory, and
create on Plan 9 does.
Here is what I think is the relevant code from applylog.
if(rd.mode&DMDIR){
fd = create(local, OREAD
Should have gone to bed before writing the last note.
On Sun, Oct 4, 2009 at 11:39 PM, ron minnich wrote:
> if (perm&0400 == 0) then do a mkdir with 0400, an open, and an fchmod.
> Tomorrow,
> that is. I think the work-around is pretty easy.
> I figure the 0400 might work bec
I'm just now looking at tex on plan 9 and finding that a 14-year old
release is not that useful with the newer packages.
Has anyone tried newer stuff at all?
thanks
ron
On Mon, Oct 5, 2009 at 2:04 PM, John Stalker wrote:
> `cat' would concatenate 0 files, i.e. ouptut nothing,
cat would copy stdin to stdout
> `chmod 755' would set the permissions of no files to 755,
would read a list of files from stdin and change the modes
> `cp foo/' would move no files to
On Mon, Oct 5, 2009 at 5:40 PM, erik quanstrom wrote:
>
> i just hate how the rush to use every last new feature
> leads to things like gnu configure and autotools, and
> ancient warhorses like tex no longer running.
I last built the C version of TeX quite some time ago. At the time, it
was quite
OK, found this: http://www.math.uni.wroc.pl/~hebisch/tex/
7 years old but 7 years newer than the plan 9 version. I have mailed
the author to see how it is going.
compiles (quickly!) and runs under ape.
ron
On Tue, Oct 6, 2009 at 4:37 AM, erik quanstrom wrote:
>> > `ln foo/' would hardlink no files into the directory foo,
>> would read a list of filenames in and ln them
>
> better: ln: '/bin/ln' directory entry not found
well, yea, but I didn't want to get into that one :-)
ron
http://www.glomationinc.com/
49 bucks!
It's an arm 9 -- anybody know what variety?
ron
On Tue, Oct 6, 2009 at 9:55 AM, wrote:
> The cortex-a8 arms are arm v7-a architecture. L2 page table
> entries have changed format. The a8 includes trustzone, so
> many registers have forked, producing a "secure" and a "nonsecure"
> version of the register. The arm v7-a manual is a 2,150-page
On Tue, Oct 6, 2009 at 11:06 AM, erik quanstrom wrote:
> the one that you didn't mention is the one that bothers me: smm mode.
> this has been around for a very long time. smm mode takes a special
> interrupt and takes over the cpu and runs some real-mode code. things
> like ps/2 emulation for
one last note on this because it is kind of interesting.
Consistency is good. Sometimes it goes too far. The RCA computer OS
guys decided that CR and LF were not really command delimiters, that
rather ETX made the most sense. So to end a command you had to
remember to type ^C.
Consistent, totally
On Tue, Oct 6, 2009 at 11:16 AM, W B Hacker wrote:
>
> Anyone know if the AMD environment is any more 'open'?
way, way, more open. same with via. They regularly contribute chipset
source code to coreboot. That's my measure.
> I hadn't paid much attention to the ARM until the recent '2 GHz' blur
I'd like to have a hack session the wed. morning before IWP9.
What I'd like to propose is a sheeva plugfest. People commit to bringing a plug
and we get them set up to run Plan 9.
Any interest?
ron
On Tue, Oct 6, 2009 at 12:53 PM, Francisco J Ballesteros wrote:
> I'm interested, but don't have one yet :(
I am willing to order a small number if people commit to paying me
back. A SMALL NUMBER, I'm not midas.
ron
On Tue, Oct 6, 2009 at 12:13 PM, Lyndon Nerenberg - VE6BBM/VE7TFX
wrote:
> I don't think DEC deserves this branding. In my experience they were
> one of the most open hardware companies around. Back when they were still
> DEC, of course.
You never dealt with Alpha maybe. The story is long and sa
oh yeah, I assume the first step to hacking these is cracking them open?
Or is there a way via the usb port to get serial console (does not
seem so from block diagram)
ron
There's nothing says we can not hack thursday night as well. I
proposed wed. as I am there at that time and it's an empty slot. By
thursday there may be enough things going on that it's hard to do the
hack time.
Me, I see us sitting in the hotel lobby one evening surrounded by
pitchers and wires
On Tue, Oct 6, 2009 at 2:15 PM, Aharon Robbins wrote:
> I understand all your points, and many of them are good ones. But there
> really are places where you don't want to go, and into the chipset
> is one of them.
Not really the case. People do want to go there, so they can do
interesting thing
On Tue, Oct 6, 2009 at 4:58 PM, Joseph Stewart wrote:
> I have some weird PPC-based IBM devices I'm willing to give away to those
> inclined to hack on (hey, it's either this or the trash).
bring 'em on. let's see.
ron
On Tue, Oct 6, 2009 at 5:19 PM, Dave Eckhardt wrote:
> For something "nobody would want to do", there sure are a
> lot of hits for "pcs750.bin".
It's the difference between "nobody would want to do it" and "we don't
want you do it" ;-)
ron
On Tue, Oct 6, 2009 at 6:26 PM, Latchesar Ionkov wrote:
> Or you can setup Plan9 on my gumstix cluster :)
>
bring that one too!
ron
Thinking about it a bit more ... when systems become more and more
closed, as x86 systems are becoming now, the field of innovation is
reduced to what a single company can think of -- the monopoly
provider, so to speak.
When systems become more closed, you hear stuff like this: "The
percentage of
I'll be in around 9pm or so the 20th, if I can give anyone a ride I will.
ron
I have one request. Cost to you will be $99 plus shipping and I'll
expect cash on delivery :-)
Delivery to athens ga. by me. Again, I can only take a few more orders
but I'll do it. I'm going to order tomorrow.
ron
well, actually, I ordered a sheeva on oct. 6 with 2 day shipping (paid
extra for that).
And, no sheeva. I call today and find out that due to lack of stock,
it won't ship until oct. 14 earliest. This is a 10/6 order for which I
paid $18.66 for fedex
2nd day. They pretty clearly stated when they to
I've confirmed rides with three people, can take no more. You people I
confirmed rides with get back with me ASAP to double-confirm.
ron
And, hey, maybe lantronics will actually sell them ... I still haven't
heard back from ionics.
ron
-- Forwarded message --
Date: Wed, Oct 14, 2009 at 7:24 AM
Subject: world's smallest Linux networking server
To: ron minnich
This is an odd little part, you cou
Now I remember this paper. Was the code ever released anywhere?
ron
But they require paypal ...
I can't find it anyway.
But global did finally ship my sheeva plug today, only about 10 days late.
ron
---
Thank you for your interest in our plug computer.
For your information, we have just activated our on line shopping cart
ordering. You can visit our website at
On Fri, Oct 16, 2009 at 10:03 AM, Sam Watkins wrote:
> So it would only be a problem supposing that a significant part of the program
> is unparallelizable. I can think of many many tasks where "Amdahl's law" is
> not going to be a problem at all, for a properly designed system.
>
> For example
the use of qualitative terms such as "embarassingly parallel" often
leads to confusion.
Scaling can be measured. It can be quantified. Nothing scales forever,
because at some point you want to get an answer back to a person,
and/or the components of the app need to talk to each other. It's
these b
Since we seem to be having the parallel programming discussion again
please look at this:
https://asc.llnl.gov/sequoia/benchmarks/
The summaries are interesting.
ron
On Mon, Oct 19, 2009 at 1:57 AM, Dmitry Golubovsky wrote:
> /bin/sh XXX: cannot execute - Access denied
>
> XXX is any program I am trying to run from ape/psh.
>
> Has anyone else run into this recently?
>
it all works for me.
ron
confusion reigns. People I thought I was giving a ride to are coming
in at different times.
So, let's try again.
I have one rider: maht.
I can take two more. Roger?
Anyway I get in 835PM on the 20th, I can take 2 people besides maht,
so let me know.
ron
On Mon, Oct 19, 2009 at 8:26 AM, Sam Watkins wrote:
> On Fri, Oct 16, 2009 at 12:18:47PM -0600, Latchesar Ionkov wrote:
>> How do you plan to feed data to these 31 thousand processors so they
>> can be fully utilized? Have you done the calculations and checked what
>> memory bandwidth would you ne
On Mon, Oct 19, 2009 at 8:57 AM, Sam Watkins wrote:
> This is only a very rough estimate and does not consider all the issues.
well that part is right anyway.
ron
On Mon, Oct 19, 2009 at 9:34 AM, Sam Watkins wrote:
> The "processors" (actually smaller processing units) would mostly be
> configured
> at load time, much like an FPGA. Most units would execute a single simple
> operation repeatedly on streams of data, they would not read instructions and
> e
I'm here, anyone doing breakfast? Where to go?
ron
I think we converged on 8am at this thing on college?
ron
So, floren and I are meeting in the holiday in lobby at 0730 and then
will go find grill.
I will bring laptop and will happily demo TVX and burn sticks for
anyone who cares.
Also bringing sheevaplug.
ron
SOP for some workshops in the evening for me is to find a lobby with
tables
couches
beer
hardware (we supply that)
tolerant hotel staff who don't threaten to throw you out at 2 am for
not renting a conference room (as happened in Hamburg one year)
and having a hack session. don't know if anyone el
On Wed, Oct 21, 2009 at 8:43 AM, Sam Watkins wrote:
> People do this stuff every day.
> Have you heard of a render-farm?
Yes, and some of them are on this list, and have actually done this
sort of work, as you clearly have not. Else you would understand where
the limits on parallelism are in p
I thought it was just wonderful, and noticed similar reactions from
everyone else. It was a very fine meeting.
And the hack sessions in the evening were extremely useful!
ron
this info needs more info.
what OS? What version? What compiler? What version?
thanks
ron
to use 9vx as a file server for sheeva plug,
cd /rc/bin/service
mv !tcp564 tcp564
and start listen -q tcp.
Works!
ron
On Mon, Oct 26, 2009 at 3:20 AM, Dmitry Golubovsky wrote:
> One more kind of trap (did not cause crash) - not sure when exactly it
> happened:
>
> 76 stats fault 0x33c no segment
> segment 0xf00 0x1000
> segment 0x1000 0x1b000
> segment 0x1b000 0x22000
> segment 0x22000 0x3a000
> 76 stats:
nebula.nasa.gov
and see what you see
ron
On Mon, Oct 26, 2009 at 6:52 PM, Latchesar Ionkov wrote:
> Did anybody come up with cloud management software called Zeus or Jupiter yet?
interesting. I got
arg. you broke it.
and you guys got the web page and I just did.
And, yes, it's another !@@#$! cloud.
But why is USG competing with amazo
How about:
"Clouds: how we reinvented the time sharing bureau, which we all
hated, back in the day, and replaced with our own computers, and
recreated the time sharing bureau in our own company, and hated it,
and now our management has decided they hate us and want to fire our
IT people and trust
true story: some ex-(name withheld) people were telling me they loved
their (name withheld) IT guys. Then (name withheld), for its own
reasons, decided to outsource their IT people.
You'd recognize (name withheld); it's one of the biggest computer
companies out there. They also like to sell their
note to make this work you need to run as root, not an ordinary user ...
ron
>> Equally true story. We used to run our own servers. A (name withheld)
>> sysadmin always felt he knew better than management how servers should
>> be configured and managed even when in fact he did not. So we went to
>> Rackspace, where we are treated as customers and where sysadmins manage
>> t
On Tue, Oct 27, 2009 at 9:16 AM, erik quanstrom wrote:
> there were a lot of wierd os people talking to a lot of package
> maintainers at google this weekend. the solution that made
> sense to *BOTH* was to ditch autoconf and submit a platform
> specific makefile.
I'm impressed. What are the pr
I realize it is early but a neat GSOC project would be to take the
mods I made to 8l and friends and use these as a way to build dtrace
for plan 9.
Had a fun talk with someone from Sun today and for at least part of
dtrace functionality the 8l mods get us part of the way there.
What they can do w
One other thought on this line. The dtrace tools include a kernel
module which understands the dtrace language. Maybe an alternative
plan 9 approach is a kernel driver which understands acid.
ron
101 - 200 of 1324 matches
Mail list logo
