> What could it be doing for that long? I wonder if you have a first
> example of the oft-rumored ACPI virus :-)
we've seen hangups of 1s for usb emulation.
- erik
> i don't really see why xargs (the idea, not the usual unix implementations)
> is inherently such a bad idea. years ago i wrote an ultra simple version
> with no options, and it's about 80 lines of code, which i use to grep
> through all of /sys/src for example.
that's interesting. my objection
> because the limit is big enough that cases that break the
> limit almost never happen except in this case?
we can easily fit all the files in most any system in memory.
why shouldn't that be the limit? see below.
> > i'm not sure i understand when and why this would be useful. nobody
> > has
> On Fri, Jan 01, 2010 at 09:02:28PM -0500, erik quanstrom wrote:
> > > you've got a fast system.
> > > in at least one system i use, du -a of /sys/src takes about 25s.
> >
> > i have a humble 2y.o. single-core 35w celeron as a fileserver.
> >
>
> In addition to adding in new bindings to running processes like rio,
> plumber, dossrv, and exportfs, this mechanism is also fully network
> transparent and useful when importing /proc from remote machines.
> Rewriting the namespace of remote processes is a powerful mechanism
> for fine-grained i
> non-kernel resources. Consequently, if the non-kernel 'root' is lost,
> the dead processes can be killed off, another root acquired, and work
> resumed. If no local resources are available, a srv and mount of
> sources allows access to the tools of the full distribution.
would the dead processes
On Sat Jan 2 12:26:46 EST 2010, aku...@mail.nanosouffle.net wrote:
> > can you give an example of a use of this feature that can't be
> > accomplished by plumbing "Local 9fs $server"?
>
> Wanting to provide access to sources repo or data on external
> hdd, to a program running in background (say,
there are two potential reasons for this error
1. the underlying fileserver doesn't support it. e.g. 9vx #Z.
(recently discussed.)
2. perhaps the mount options are confused in your
namespace file. we have "mount -b #s/wiki /usr/web/wiki"
- erik
i should have started out by stating that i find
what you've done here interesting. i'm just not
sure how it's useful yet. at least to me.
i'm also not convinced that changing the namespace
of a running proecss is a safe operation, in general.
for example, wikifs creates lock files. suppose you
> i'm not saying it can't be passed in an argument list, just that
> xargs gives you a lazy evaluation of the walk
> of the file tree which can result in a faster result
> when the result is found earlier in the file list.
i have no problem with breadth-first.
my beef with xargs is only that it i
> A very good point, and I hope you don't think the response "trust the
> user to administer their system and accept that it is possible to do
> broken things" is trying to dodge the issue.
no, i don't. i think that is a reasonable answer. however,
it does change the use case considerably. and
it now installs in /$objtype/bin/upas
and /acme/$objtype/bin/Mail,
so you may wish to do the following
before installing
mv /sys/src/cmd/upas /sys/src/cmd/upas-
mv /386/bin/upas /386/bin/upas-
- erik
> And can eat up a lot of memory or even run out of it. On a
> 2+ year old MacBookPro "find -x /" takes 4.5 minutes for 1.6M
> files and 155MB to hold paths. My 11 old machine has 64MB
> and over a million files on a rather slow disk. Your solution
> would run out of space on it.
modern cat woul
> Given the way Unix programs
> behave you can't replace arg list with an arg fd (I used to
didn't know this was "unixfans". will keep that in mind.
- erik
> cpu% auth/secstore -g factotum
> secstore password:
> auth/secstore: error: account rtr expired at Sat Jan 2 03:59:59 GMT 2010
> secstore password:
see secstore(8). auth/secuser $user is what you want. on
the console of the auth server.
- erik
i submitted a patch to secuser. i noticed that the
month printed was off by one. according to secuser
one year from today would be (in the odd ddmmy
format it used) 04002011.
i changed that to mmdd and fixed a handful of
bugs. one year from today is now 20110104.
- erik
On Mon Jan 4 16:23:56 EST 2010, st...@quintile.net wrote:
> > (in the odd ddmmy format it used)
>
> We, on this side of the pond, find that format rather modern and exciting;
> not in the least bit old... ☺
i suppose that it has the benefit that it conflicts
with the fileserver's dump format
> Erik has an ethernet (not tcpip, raw ethernet) console kernel driver and
> command line
> tool which is easier these days (EIA interfaces are a dieing breed). You can
> even use
> these tools over a loopback interface, from the machine back to itself.
a version of cec(8), the client, is availa
> Just curious: can an 9P server cleanly differenciate between clients ?
> This would be a great help for transaction isolation, IMHO.
>
> w/o having looked at cookiefs yet, but I would do it like that:
>
> * get cookies by reading /site-cookies/
> * set cookies by writing ": foo=bar" to
it appears that we can mistakenly declare a packet's
checksum to be ok in the following drivers
- ether82563.c
- ether82598.c
- etherigbe.c
- etherm10g.c (saved by hw that eats packets with bad checksums)
- ethervt6105m.c
the problem is that the Block->flags are not cleared.
so if a Block receives
i thought this would be interesting to some.
please send questions directly to me, rather than
spamming the list.
coraid is looking for some engineers. here are
two job postings in the original job-postese:
support:
Coraid, a network data storage company, is looking for
candidate
> The thing that none of the VM monitors seem to offer (though I'd love
> to be proven wrong) is debugging tools for the guest operating
> systems. This is odd, as it was one of the major uses of VM/370. So
> if a guest kernel goes off into space, the VM monitor shuts down the
> virtual machine o
> > it's unfortunate that computer history isn't a bigger
> > component of a computer science degree.
>
> History and Philosophy of Science was slow in becoming a legitimate
> academic pursuit of great practical value. It will probably not be
> quite as long before the analogous subject will mate
> Perhaps the go 8l is not honoring the "-l " option? I just tried to
> create a mbr from assembly and it reported the same runtime
> requirements. A diff between the various sources might be useful since
> they are similar.
see /sys/src/boot/pc*.
; mk -n mbr.install
8l -o mbr -H3 -T0x0600 -l m
> "runtime·morestack" (how does one compose that "dot" on the
> conventional Plan 9 keyboard?)
compose+..
- erik
On Fri Jan 8 23:31:23 EST 2010, pau...@gmail.com wrote:
> On Fri, Jan 8, 2010 at 10:40 AM, anonymous wrote:
> > Why libthread has threadcreate instead of something like fork? With
>
> Preemptive vs cooperative.
i think that misses the point of the question. the question
could have just as easi
> I've heard nothing but bad things about VirtualBox. Mostly from Plan 9
> people(where the results have been uniformly bad), but also from folks
> trying to useother OSes. It seems the worst of the breed across the board.
has anyone tried 9atom on virtualbox? i think there's a good chance
vi
> Just to confirm what Geoff said.
>
> Parallels 5 works like a charm with Plan 9.
> Also, it seems to have full ACPI support, not that this is
> important for Plan 9 yet.
>
> Thanks
>
> On Sat, Jan 9, 2010 at 3:53 PM, erik quanstrom wrote:
> >> I&#
On Sun Jan 10 10:42:33 EST 2010, l.ve...@gmail.com wrote:
> i am experiencing the very same problem -- only i get
> the echo every single time, not 2/3 -- and i get this
> even when using pdksh. as previously reported, bash (or
> zsh, for that matter) does not suffer of this problem.
> any help wo
> Which disk controller should I use for best performance and
> driver support?
>
> AHCI based:
> SB600
> IHC9
> Jmicron JMB363 (supported?)
>
> Marvel based:
> Marvell 88SX6081
this isn't a complete list of supported ahci or marvell
parts. i don't know about jmicron support. i tried
to add it
> I believe that Bell Labs are actively involved in the port of GO to
> Plan 9, I'm not sure how much my efforts are likely to contribute to
> that particular project.
could anyone confirm this?
- erik
> >1. unless one needs more ports than the motherboard
> >offers, i would stick with on-board ports.
>
> That makes sense for me. Perhaps I could save money and
> use the onboard network chip too (most RELTEK)? Then I
> could invest more money in the board and the CPU.
some 8169 parts are pre
> >http://www.msi.com/index.php?func=proddesc&maincat_no=1&prod_no=183
>
>
> Nice board. I could get it for 50$. So far I can see this board has
> the ICH7 chip. So AHCI is not working on this board?
i belive that page says ich6.
ich6 is not ahci capable. the contrib version of the ahci
driver
On Mon Jan 11 03:29:06 EST 2010, w...@hush.com wrote:
> >I recently built a new server out of an Intel Dualcore Atom
> >motherboard
> >(D945GCLF2D), it needs some kernel patches from Erik to get
> >everything to
> >work cleanly but it is now great. I added two 500Gb SATA disks
> >mirrored for st
>
> An alternate configuration, which takes more memory, but might offer
> a bit more in the way of survivability, would be to not use fs for venti.
> Instead,
> run one venti daemon per disk, with independent arenas/indexes. Insert a
> little
> Venti proxy between fossil and your daemons; it sh
> > I suggest you install Fede's contrib package (a sort of package managment
> > system),
> >
> > 9fs sources
> > /n/sources/contrib/fgb/root/rc/bin/contrib/install fgb/contrib
> >
> > now you can list packages and install them - see man contrib
> >
> > some stuff is not in contrib
> Venti deals with incompletely written blocks; the arenas and index structures
> are still workable. The situation is even recoverable - a proxy could notice
> that one of the backends failed to return a read, so it rewrite the data from
> an other copy (which it can verify) to the failed one.
w
> There are some newer mini-itx cards from supermicro which also look
> interesting but I have no experience of them:
>
> http://www.supermicro.com/products/motherboard/ATOM
i'm trying to get one to test. fortunately, this thing has an
82574 nic. which should work just fine.
- erik
On Mon Jan 11 12:54:02 EST 2010, w...@hush.com wrote:
> >> After reading many mails here I plan to run a fossil only
> >> fileserver. So far I understand the "Ken" dedicated fileserver
> >> has gone.
>
> >ken's fs works for me.
>
> I am very interested in the old dedicated Ken fileserver. So
> fa
>
> On Mon, 11 Jan 2010 19:14:19 +0100 erik quanstrom
> wrote:
> >fs was removed from the cd but there's a version in
> >/n/sources/extra.
> >
> >kfs and fs(3) are different fileservers. you can also use the
> >source in
> >contrib quans
> with similar results. in that case factotum debug says "no key
> matches proto=p9sk1 role=server dom?". this last message looked a bit
> weird and when i check /dev/hostdomain, it is empty.
/dev/hostdomain empty here, too.
- erik
On Tue Jan 12 17:41:45 EST 2010, aku...@mail.nanosouffle.net wrote:
> From the inspections of Cinap and I, albeit a while back,
> Erik's FS does not take NVR from floppy. That's why Erik
> suggested that you create a small 9fat partition (using
> the Plan 9 Install/Boot CD) on your primary master
> I am using the floppy only for testing plan9.ini parameters.
> This should be possible with qemu or other emulators too. Later
> I will use an IDE disk or IDE flash disk (I would prefer the
> second).
> So I am not sure whether it makes sense to spend time to make this
> possible. I can live wít
should have explained that my fs accepts 3 partition
layouts.
p(h0)x.ywhere x <= y and y <= 100.
partition by percentage. y% of the disk starting at x.
p(h0)w.zwhere z > 100
partition by block offsets. w-z blocks of the disk
ph0"name"
use fdisk/prep
> Would a IDE-Flash module works too? These modules are not expensive
yes. coraid's fileserver boots from ide flash.
- erik
On Tue Jan 12 22:39:27 EST 2010, lyn...@orthanc.ca wrote:
> > From the inspections of Cinap and I, albeit a while back,
> > Erik's FS does not take NVR from floppy.
>
> So is it worth it to try to nail down a driver that can talk to at
> least some of the on-motherboard NVRAM present on today's c
> >
> >I'm running the traditional Ken FS, sans Erik's mods.
> >
pedants would point out that it is unlikely this is
"traditional" ken fs. it is much more likely to be geoff's
reform 63-bitized version.
- erik
> I found a unused harddrive and an old pc. Is only a P3 but it
> should work for testing purposes. To make testing easier I will
> try Xen hvm with the file server too.
until nov 2008 i ran a piii/600 fileserver with 18g scsi
disks. performance was good, but was limited by
some poor choices wh
>The mount command is nothing like what i'm used to on linux or BSD,
>and the /mnt folder is kind of confusing. How can I mount/ummount
>things and approx. where do they end up?
bind(1) describes how bind and mount work.
could you explain why you find /mnt confusing?
it's just a directory that is
> > I understand the "everything is text" idea
> >
>
> except therefore, you don't :) "everything is a file"
>
> unless you can read cat /dev/screen
there's certainly quite a bit of effort into making
everything possible text. kernel devices generally
produce and consume human-readable outp
> > Then you mount the connection,
> > see the examples in the man page mount(1).
> > Some devices need some other extra programs be run
> > first, which provide files abstracting the disks for the
> > fileserver to use, like usb which needs another server.
>
> Is there already some tool for model
> So, the only thing that I think could work
> would be to have abaco plumb *.pdf as soon
> as it gets the URL, instead of trying to show
> it and ending up with mime-type unsupported.
why can't you just use webfsget?
- erik
> Supposed to just print the first line?, instead, it prints:
>
> read: error reading /mnt/wsys/wctl: buffer too small
>
> Several guys from #plan9 in Freenode told me that that was happening
> because of a missing newline.
not correct. it is happening because rio thinks the read
requests too f
> My only machine is a laptop, the ethernet card is broadcom, and from
> searching thru the past threads, appears to be a thorn in everyone's
> side. BCM44x I believe. If I can ever get "familiar" here, I'll have
> to recompile it's kernel because if i'm reading the output right,
> whilst I have a
> Not very mysterious to me. There's not very much science in computer
> science. If we didn't forget it we wouldn't be able to re-invent it, and
> there would go most of the interesting work, not to mention a lot of
> high salary jobs.
s/interesting //
there, fixed that for ya.
- erik
is anyone interested in testing a marvell yukon 2
driver? i am currently working with a 88e5057
(1186/4b00, dge-560t). but most single-port yukon 2
parts should work fine. if so. please contact me
off list.
- erik
On Sun Jan 17 09:27:23 EST 2010, aed...@gmail.com wrote:
> My mainboard got a 88E8001 and 88E8053 onboard, I'd be interested.
>
> Best regards,
> F. Caulier
could you send me lspci -nnk output or pci output if you're
running plan 9.
- erik
thanks for the pci output. the driver should be able
to recognize your card (the yukon ii, not the old-style
yukon) but i don't have exactly that hardware. you can
either try contrib quanstro/yuk or the 9atom cd at
ftp://ftp.quanstro.net/other/9atom.iso.bz2
let me know how it goes. if you have
On Mon Jan 18 06:08:05 EST 2010, climber@gmail.com wrote:
> Is this possible for UNIX philosophy to develop further? Let's say,
> XML-coded trees or graphs instead of one-line strings in stdin/
> stdout.Or LISP S-expressions. New set of utilities for filtering such
> streams, grep for XML trees
Subject floating point puzzle
i'm seeing a crash in in _v2d. unfortunately, i don't
see how this is happening.
8.out 8535: suicide: sys: fp: stack underflow fppc=0x6397 status=0x81e1
pc=0x63a5
however, acid says that the Vlong passed in was all zeros. the code generating
the
call to _v2d is
> > Not very mysterious to me. There's not very much science in computer
> > science. If we didn't forget it we wouldn't be able to re-invent it,
> > and
> > there would go most of the interesting work, not to mention a lot of
> > high salary jobs.
> But how much of this work is actually redundan
On Wed Jan 20 08:27:58 EST 2010, maht-9f...@maht0x0r.net wrote:
> By the end of May, all the root servers should be running DNSSEC
>
> http://royal.pingdom.com/2010/01/19/the-internet-is-about-to-get-a-lot-safer/
>
> Is Plan9 ready for such a move?
there are two answers to this:
yes, if you mea
> > one would likely need to start with a different structure
> > than ndb/dns currently has to get dnssec. but i think that
> > the most of the query logic could be reused.
> As I understand it; It is an extension, the base DNS stuff should not
> change.
> What would need to be changed in ndb,
> > i think your understanding ma be incomplete. dnssec
> > requires that the rrs be chained together in a particular
> > order. and any change to a rr triggers resigning. it
> > may be doable, but i think it would be easier to start
> > with dnssec in mind.
> That makes their use of the word e
On Wed Jan 20 12:49:14 EST 2010, r...@swtch.com wrote:
> > starting over would seem (and probably is) best.
>
> http://www.jwz.org/doc/cadt.html
you have a design then that will
do dnssec without any rewriting?
- erik
> I found your post a bit confusing then. The little bit of script you
> posted won't do the job, and the problem is not related to find at
> all,
>
> Anyway, while working with some hugely messy non-plan 9-software, I
> found I really needed grep -r.
>
> See /n/sources/contrib/rminnich/grep
why
> > du -a / | awk '{print "grep something " $2}' |/bin/rc
> > ron
>
> Try
>
> touch 'x;reboot'
>
> and then see if it still works fine. I don't think I like your
> version on a system with users I don't trust completely.
sneaky. but it won't work.
; touch 'x;reboot'
; du -a .
0 './x;re
On Fri Jan 22 15:24:49 EST 2010, stal...@maths.tcd.ie wrote:
> > sneaky. but it won't work.
> >
> > ; touch 'x;reboot'
> > ; du -a .
> > 0 './x;reboot'
> > 0 .
> >
> > - erik
>
> It worked under 9vx on my Mac. I didn't test on real hardware.
this rebooted your 9vx? sounds wrong to me.
i
> And if you like find, write and put it in contrib.
contrib/pull quanstro/find
- erik
> A) Why aren't there the following (POSIX) file operations: `move`,
> `link` and `symlink` in the original 9p2000 protocol version?
>
> Now I know that by using `wstat` I can change the name of a file,
> but I can't move it outside the current directory.
symlinks weren't wanted. they cr
On Fri Jan 22 18:29:45 EST 2010, 9...@9netics.com wrote:
> in case anyone's wondering, my problem was due to the fact that keyfs
> was started after aux/listen for trusted services; /mnt/keys/* wasn't
> in authsrv's namespace. in my case, i put the trusted services in
> /cfg/bootes/cpurc, while ke
> Another problem with stuff in contrib is, that their software is not
> well documented (i.e. no man pages), so that one probably has to read
> the source in order to be able to use it. This often forms an obstacle
> for usage... Being accepted to the main tree brings along good
> documentation, t
i had some trouble using /dev/reboot with vesa graphics
enabled. i have been using this code rather than the standard
/dev/reboot: http://9fans.net/archive/2009/12/140
i had a brain wave a few weeks ago but i haven't had a
chance to test it until today. and surprisingly, it works.
; a
> > By the end of May, all the root servers should be running DNSSEC
> >
> > http://royal.pingdom.com/2010/01/19/the-internet-is-about-to-get-a-lot-safer/
> >
> > Is Plan9 ready for such a move?
>
> Reading what D. J. Bernstein has to say about DNSSEC is always fun.
> See e.g. this paper http://cr
> if the goal is avoiding ssl mitm attacks,
> dns is the least of your worries. a mitm will
> just take over the connection attempt for the
> actual ip address. the solution there is
> to implement proper ssl certificate chain checking.
doesn't work with the recent renegotiation bug.
it's a serve
> > doesn't work with the recent renegotiation bug.
>
> disable renegotiation.
>
> > but i don't
> > think one can dismiss dns as a non-issue.
>
> dns is a non-issue if the rest of ssl is working.
> dns is irrelevant if it isn't.
the renegotiation bug is a protocol flaw. i'm
not so sure i trus
On Sun Jan 24 08:21:50 EST 2010, 23h...@googlemail.com wrote:
> so you need to have a more "secure" dns because you don't trust your ssl?
do you feel dnssec provides the same or worse security? why?
the simple answer to your question is yes. the renegotiation bug
in ssl requires you to start ta
On Sun Jan 24 17:15:17 EST 2010, news...@lava.net wrote:
> > you are changing the topic.
> >
> > your original mail claimed to be worried
> > about man-in-the-middle attacks. that means
> > the attacker can respond to arbitrary traffic;
> > the fact that you can verify the dns response
> > is irre
does anyone know why the → A is just missing
from the output unless the output is non-text?
i haven't had time to get to the bottom of this.
- erik
context:
The protocol to obtain a ticket pair is:
.TP
.IR C \(-> A
.IR AuthTreq ,
.IR IDs ,
.IR DN ,
.IR CHs ,
.IR IDc ,
.IR IDr
.sp -\n(PDu
On Sun Jan 24 20:41:14 EST 2010, aku...@mail.nanosouffle.net wrote:
> why not just use the arrow symbol
> directly, in the troff file?
>
it was like that originally, and is not realted to the problem.
you can replace the → with and you'll see that the output is
adjusted but the doesn't
i have a little program, aux/cpuid, that gets cpuid
information for x86 processors. clearly it's not going
to run on an arm or mips. is there a standard
trick for preventing such a program from being
built (and failing)? the hack i currently have is
objtype=386 as the first line of the mkfile.
On Mon Jan 25 10:00:19 EST 2010, lyn...@orthanc.ca wrote:
> > i have a little program, aux/cpuid, that gets cpuid
> > information for x86 processors. clearly it's not going
> > to run on an arm or mips. is there a standard
> > trick for preventing such a program from being
> > built (and failing)
On Mon Jan 25 12:03:08 EST 2010, fors...@terzarima.net wrote:
> >perhaps this would be better in /dev/cons/cpuid ?
i don't think so.
> doppio% cat '#P/cputype'
> AMD64 1802
>
> doppio% cat '#P/archctl'
> cpu AMD64 1802 pge
> pge on
> coherence mb586
> cmpswap cmpswap486
> i8253set on
none of th
On Mon Jan 25 13:11:49 EST 2010, rminn...@gmail.com wrote:
> On Mon, Jan 25, 2010 at 9:54 AM, erik quanstrom wrote:
>
> > ; aux/cpuid -v
> > AuthenticAMD
> > ; aux/cpuid -b
> > AMD Athlon(tm) 64 X2 Dual Core Processor 5000+
> > ; aux/cpuid -s
> > 000
On Mon Jan 25 15:15:42 EST 2010, eeke...@fastmail.fm wrote:
>
> On 13 Jan 2010, at 4:23 pm, Enrico Weigelt wrote:
>
> > * erik quanstrom wrote:
> >
> >> i think you misunderstand the problem. cookiefs' fs interface
> >> is not the issue. cookiefs&
> > nothing to do with the concept of a line delimited by newlines. For
> > the movement, such delimited lines are completely irrelevant. For the
> > movement there is no difference whether there is a newline on the
> > previous visible line or the line was broken due to the width of the
> > window
On Tue Jan 26 15:22:22 EST 2010, st...@quintile.net wrote:
> It will take a very long time, espicially if your IDE controller is not
> known or it is and DMA is known to be problematic.
the output of sossd (/n/sources/contrib/quanstro/sossd) is useful
if you suspect a problem. i'll have a look if
> Heh. :)
>
> P9P uses a separate file for each arch: plan9/src/libthread/$os-$arch.s
>
> Inferno uses a separate dir. Some of the dirs have C code instead
> since the asm is only an optimisation in this case.
>
> I'd put a test in the mkfile: if cpuid-$objtype exists build that,
> else.. we
> aux/vga: vgactlw: : bad VGA control message "type radeon"
> ...
> rio: can't open display: initdisplay: /dev/draw/new: no frame buffer
> init: rc exit status: rio 23: display open
you need to add vgaradeon +cur to your plan9.ini.
pull can be pretty antisocial.
- erik
On Wed Jan 27 08:08:42 EST 2010, quans...@quanstro.net wrote:
> > aux/vga: vgactlw: : bad VGA control message "type radeon"
> > ...
> > rio: can't open display: initdisplay: /dev/draw/new: no frame buffer
> > init: rc exit status: rio 23: display open
>
> you need to add vgaradeon +cur to your
On Wed Jan 27 08:42:31 EST 2010, rogpe...@gmail.com wrote:
> i guess that's because it's walking into mnt/acme/new,
> which creates a new window.
>
> i've thought in the past that perhaps the first write
> to a file in mnt/acme/new should create the window,
> rather than just walking to it.
>
> i
> On 1/19, the Radeon driver made it into the Plan 9 distribution; anyone with
> a Radeon R100 (7000, 7500, Mobility M6, M7), some R200 (8500, 9000) or
> some R300s (9500, 9700) is encouraged to try them out, see if they feel faster
> or work for you or whatnot.
>
> The driver also has, disabled-b
> My long-term goal is to eliminate all the vga drivers but vgavesa,
> which make up about 10% of the pc kernel port by line count. This may
> not be possible due to currently-working graphics cards with broken
> vesa bioses nor desirable because the native drivers are vastly faster
> than the ves
On Wed Jan 27 09:37:02 EST 2010, rogpe...@gmail.com wrote:
> fuse is probably just doing a stat of each file, as is
> conventional and necessary in unix.
>
> the 9p fuse converter can't legitimately cache the
> qids from the directory read, so there's probably
> no other way.
why is it walking th
> My worry is that, over time, neither will the graphics vendors. I
> don't trust them not to screw this up ... why would they continue to
> support it?
so windows can display something before it's finished
probing hardware?
- erik
> well that's good to know. If M$ needs it then we're good to go.
always the optimist!
unfortunately, 2d performance appears to be taking a back
seat to fancy 3d accelerated stuff:
http://www.tomshardware.com/reviews/2d-windows-gdi,2539.html
- erik
> A colleague put me on to Plan9, some of whose online documentation I
> have read with interest, in particular the "Hello World" discussion as
> it relates to Unicode/UTF-8.
>
> I'm one of the authors of the Cuneiform proposal now encoded under
> Unicode (see block U+12000), and I'm interesting
On Thu Jan 28 17:09:45 EST 2010, rminn...@gmail.com wrote:
> On Thu, Jan 28, 2010 at 1:51 PM, Karljurgen Feuerherm
> wrote:
> > Well having worked with the Unicode Consortium, I know there's a little
> > more to it than that... :)
>
>
> I'm curious because I don't know much about all this s
On Thu Jan 28 16:22:58 EST 2010, ge...@plan9.bell-labs.com wrote:
> Yes, we only support the 16-bit runes of Unicode plane 0. That really
> should be enough space, except for bungling by the Unicode Consortium.
good point.
at this point only ~21829 codepoints are assigned, depending
on your defi
> Really? They're 32 bits in plan9port and, although there are a few
> things that need to be patched, we know what they are.
>
> Rune should be 32 bits by now.
there is a patch for plan 9. actually 2 that enable one to set
UTFmax = 4 and Runemax = 0x10:
/n/sources/patch/saved/runesize /n/
1101 - 1200 of 6034 matches
Mail list logo
