From: David Leimbach <leimy2k@gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Subject: Re: [9fans] Plan9 - the next 20 years
Date: Tue, 21 Apr 2009 09:52:44 -0700 [thread overview]
Message-ID: <3e1162e60904210952y6d875e97s4f92f2ded6d06707@mail.gmail.com> (raw)
In-Reply-To: <df49a7370904210119t1a2f8afegfd314211567c3568@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 2962 bytes --]
On Tue, Apr 21, 2009 at 1:19 AM, roger peppe <rogpeppe@gmail.com> wrote:
> 2009/4/20 andrey mirtchovski <mirtchovski@gmail.com>:
> >> with 9p, this takes a number of walks...
> >
> > shouldn't that be just one walk?
> >
> > % ramfs -D
> > ...
> > % mkdir -p /tmp/one/two/three/four/five/six
> > ...
> > % cd /tmp/one/two/three/four/five/six
> > ramfs 640160:<-Twalk tag 18 fid 1110 newfid 548 nwname 6 0:one 1:two
> > 2:three 3:four 4:five 5:six
> > ramfs 640160:->Rwalk tag 18 nwqid 6 0:(0000000000000001 0 d)
> > 1:(0000000000000002 0 d) 2:(0000000000000003 0 d) 3:(0000000000000004
> > 0 d) 4:(0000000000000005 0 d) 5:(0000000000000006 0 d)
>
> that depends if it's been gated through exportfs or not
> (exportfs only walks one step at a time, regardless of
> the incoming walk)
>
> i'm sure something like this has been discussed before,
> and this idea somewhat half-baked, but one could get
> quite a long way by allowing the notion of a sequence
> of related 9p actions - if one action fails, then all subsequent
> actions are discarded.
>
> one difficulty with using multiple concurrent requests
> with 9p as it stands is that there's no way to force
> the server to process them sequentially. fcp works
> because the reads it sends can execute out of order
> without changing the semantics, but this only works
> on conventional files.
>
> suppose all 9p Tmsgs were given an sid (sequence id)
> field. a new 9p message, Tsequence, would start
> a sequence; subsequent messages with the same sid
> would be added to a server-side queue for that sequence
> rather than being executed immediately.
>
> the server would move sequentially through the queue,
> executing actions and sending each reply when complete.
> the sequence would abort when one of:
> a) an Rerror is sent
> b) a write returned less than the number of bytes written
> c) a read returned less than the number of bytes requested.
>
> this mechanism would allow a client to "program" a set of
> actions to perform sequentially on the server without
> having to wait for each reply in turn, i.e. avoiding the
> usual 9p latency.
>
> some use cases:
>
> the currently rather complex definition of Twalk could
> be replaced by clone and walk1 instead, as
> in the original 9p: {Tclone, Twalk, Twalk, ...}
>
> {Twrite, Tread} gives a RPC-style request - no need
> for venti to use its own protocol (which i assume was invented
> largely because of the latency inherent in doing two
> separate 9p requests where one would do).
>
> streaming - send several speculative requests, and keep
> adding a request to the sequence when a reply arrives.
> still probably not as good as straight streaming TCP,
> but easier than fcp and more general.
>
> there are probably lots of reasons why this couldn't
> work, but i can't think of any right now...
>
Roger... this sounds pretty promising. 10p? I'd hate to call it 9p++.
[-- Attachment #2: Type: text/html, Size: 3499 bytes --]
next prev parent reply other threads:[~2009-04-21 16:52 UTC|newest]
Thread overview: 101+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-04-17 22:08 Francisco J Ballesteros
2009-04-17 22:15 ` ron minnich
2009-04-17 22:35 ` J.R. Mauro
2009-04-17 23:01 ` ron minnich
2009-04-18 2:06 ` J.R. Mauro
2009-04-18 2:39 ` ron minnich
2009-04-18 2:43 ` J.R. Mauro
2009-04-18 5:55 ` lucio
2009-04-18 3:37 ` erik quanstrom
2009-04-18 4:04 ` J.R. Mauro
2009-04-18 4:16 ` erik quanstrom
2009-04-18 5:57 ` J.R. Mauro
2009-04-18 13:50 ` erik quanstrom
2009-04-18 14:53 ` lucio
2009-04-18 15:07 ` ron minnich
2009-04-18 15:11 ` erik quanstrom
2009-04-18 16:13 ` J.R. Mauro
2009-04-18 16:10 ` J.R. Mauro
2009-04-18 16:20 ` ron minnich
2009-04-18 16:26 ` erik quanstrom
2009-04-18 16:36 ` J.R. Mauro
2009-04-18 16:53 ` tlaronde
2009-04-18 17:12 ` andrey mirtchovski
2009-04-18 17:37 ` ron minnich
2009-04-18 23:31 ` Charles Forsyth
2009-04-18 23:26 ` J.R. Mauro
2009-04-18 17:35 ` J.R. Mauro
2009-04-18 5:58 ` lucio
2009-04-18 11:59 ` tlaronde
2009-04-18 14:31 ` tlaronde
2009-04-18 15:05 ` ron minnich
2009-04-18 15:33 ` tlaronde
2009-04-23 16:56 ` tlaronde
2009-04-24 15:33 ` ron minnich
2009-04-24 16:43 ` tlaronde
2009-04-18 15:16 ` Latchesar Ionkov
2009-04-19 19:34 ` Enrico Weigelt
2009-04-19 19:52 ` ron minnich
2009-04-19 7:12 ` Skip Tavakkolian
2009-04-19 15:26 ` David Leimbach
2009-04-20 2:14 ` Skip Tavakkolian
2009-04-20 14:58 ` Uriel
2009-04-20 17:18 ` maht
2009-04-20 18:15 ` J.R. Mauro
2009-04-20 18:30 ` maht
2009-04-20 19:02 ` Charles Forsyth
2009-04-20 18:03 ` Skip Tavakkolian
2009-04-20 18:07 ` erik quanstrom
2009-04-23 5:07 ` sqweek
2009-04-23 5:36 ` Nathaniel W Filardo
2009-04-23 11:51 ` erik quanstrom
2009-04-20 18:18 ` David Leimbach
2009-04-20 18:35 ` erik quanstrom
2009-04-20 18:55 ` David Leimbach
2009-04-20 19:03 ` erik quanstrom
2009-04-20 20:17 ` David Leimbach
2009-04-20 20:33 ` erik quanstrom
2009-04-20 21:18 ` David Leimbach
2009-04-20 21:28 ` andrey mirtchovski
2009-04-21 8:19 ` roger peppe
2009-04-21 12:00 ` roger peppe
2009-04-21 16:52 ` David Leimbach [this message]
2009-04-21 17:06 ` roger peppe
2009-04-21 17:11 ` David Leimbach
2009-04-21 7:38 ` Bakul Shah
2009-04-20 19:13 ` Steve Simon
2009-04-20 19:22 ` erik quanstrom
2009-04-20 18:39 ` Francisco J Ballesteros
2009-04-21 9:52 ` maht
2009-04-21 10:23 ` roger peppe
2009-04-21 12:04 ` erik quanstrom
2009-04-21 14:03 ` roger peppe
2009-04-21 14:09 ` erik quanstrom
2009-04-21 14:33 ` Fco. J. Ballesteros
2009-04-21 14:50 ` erik quanstrom
2009-04-21 16:03 ` roger peppe
2009-04-21 16:09 ` erik quanstrom
2009-04-21 17:12 ` roger peppe
2009-04-21 17:43 ` erik quanstrom
2009-04-21 18:14 ` roger peppe
2009-04-21 16:38 ` Bakul Shah
2009-04-21 16:59 ` roger peppe
2009-04-21 16:10 ` Bakul Shah
2009-04-21 16:25 ` erik quanstrom
2009-04-21 17:03 ` David Leimbach
2009-04-21 17:23 ` roger peppe
2009-04-21 16:53 ` David Leimbach
-- strict thread matches above, loose matches on Subject: below --
2009-04-17 16:32 [9fans] VMs, etc. (was: Re: security questions) blstuart
2009-04-17 19:16 ` [9fans] Plan9 - the next 20 years Steve Simon
2009-04-17 19:39 ` J.R. Mauro
2009-04-17 19:43 ` tlaronde
2009-04-17 19:56 ` J.R. Mauro
2009-04-17 20:14 ` Eric Van Hensbergen
2009-04-17 20:18 ` Benjamin Huntsman
2009-04-18 4:26 ` erik quanstrom
2009-04-17 20:29 ` J.R. Mauro
2009-04-18 3:56 ` erik quanstrom
2009-04-18 4:12 ` J.R. Mauro
2009-04-18 4:16 ` erik quanstrom
2009-04-18 5:51 ` J.R. Mauro
2009-04-18 12:52 ` Steve Simon
2009-04-17 20:20 ` John Barham
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=3e1162e60904210952y6d875e97s4f92f2ded6d06707@mail.gmail.com \
--to=leimy2k@gmail.com \
--cc=9fans@9fans.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).