From: Bruce Ellis <bruce.ellis@gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Subject: Re: [9fans] A little more ado about async Tclunk
Date: Fri, 29 Oct 2010 20:56:56 +1100 [thread overview]
Message-ID: <AANLkTimCcAOncHe4xOwWCioe2eq7W=6Frcpk7Nrs3TqW@mail.gmail.com> (raw)
In-Reply-To: <AANLkTi=JChj6HGiPptpw9PdJ6PwpkMOMyLTXgsb4uQY7@mail.gmail.com>
right journal. i'll do some unusual tests and report on the results.
wing-commander has to awake early to attend his mother's 80th birthday party.
back to you on this one.
brucee
On Fri, Oct 29, 2010 at 8:45 PM, Gorka Guardiola <paurea@gmail.com> wrote:
> On Fri, Oct 29, 2010 at 11:01 AM, Charles Forsyth <forsyth@terzarima.net> wrote:
>>>What you are saying is that the problem could be something like:
>>
>>>-> Tclunk
>>>(do not wait for response)
>>>-> Topen (the file is exclusive)
>>
>> no, because what actually happens is closer to
>> A: Topen
>> ...
>> queue request to *another process* to send Tclunk
>> ...
>> A: Topen
>>
>> leading to either case 1
>> A: Topen
>> B: Tclunk
>> A: Topen
>>
>> or case 2
>> A: Topen
>> A: Topen
>> B: Tclunk
>>
>> depending on the scheduling of the two processes A and B
>>
>> introducing that other process introduces a new race, that simply did not exist before, because
>> that other process didn't exist before. you always got case 1.
>>
>> and indeed, with the test i ran on the modified system usually it gets case 1
>> because of timing, but eventually a case 2 shows up and the second open fails.
>>
>>
>
> That is why I said, (my second message in the thread) that at least
> you should send the Tclunks synchronously.
> Meaning the close system call sends the Tclunk
> and then fires another thread/writes down in a list for another thread
> for the fid to be recollected when the
> answer comes asynchronously.
> This prevents the fid being reused and simultaneously tries
> to keep the operations in the order the client intended.
> Still there are no guarantees, because the
> messages can be reordered in the server but...
> Still, it does not work in general (the server...) and I thought it was not
> profitable enough to be worth the hassle, but then...
>
>> my froggie has been running for four years 24/7. if you haven't read
>> the BLTJ then styx runs over the PCI ... 4 ixp1200s. asynch clunks.
>> the only problem is it's 10 times quicker.
>
> I haven´t read the Journal, I thought it was referring to the Dylan´s song
> putting froggie as an archetype of decency. You are talking about this?:
> http://onlinelibrary.wiley.com/doi/10.1002/bltj.1/pdf
>
> Do you do completely asynch clunks or just the wait for the response?.
> 10 times?, that is some latency. Can you break up exactly where the gain
> comes from?.
>
> G.
>
>
next prev parent reply other threads:[~2010-10-29 9:56 UTC|newest]
Thread overview: 101+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-27 3:02 Venkatesh Srinivas
2010-10-27 4:23 ` erik quanstrom
2010-10-27 6:27 ` Bruce Ellis
2010-10-27 13:50 ` Charles Forsyth
2010-10-27 13:53 ` Gorka Guardiola
2010-10-27 13:55 ` Francisco J Ballesteros
2010-10-27 14:12 ` John Floren
2010-10-27 14:13 ` Eric Van Hensbergen
2010-10-27 14:48 ` Venkatesh Srinivas
2010-10-27 15:45 ` erik quanstrom
2010-10-27 15:54 ` Venkatesh Srinivas
2010-10-27 21:04 ` Bruce Ellis
2010-10-28 12:12 ` Charles Forsyth
2010-10-28 13:32 ` Bruce Ellis
2010-10-28 14:30 ` Gorka Guardiola
2010-10-28 14:39 ` Gorka Guardiola
2010-10-28 14:58 ` erik quanstrom
2010-10-28 14:41 ` Francisco J Ballesteros
2010-10-28 14:45 ` Eric Van Hensbergen
2010-10-28 14:56 ` Francisco J Ballesteros
2010-10-28 15:37 ` Nathaniel W Filardo
2010-10-28 15:40 ` erik quanstrom
2010-10-28 15:53 ` Francisco J Ballesteros
2010-10-28 19:17 ` Venkatesh Srinivas
2010-10-28 19:24 ` Venkatesh Srinivas
2010-10-28 20:18 ` Charles Forsyth
2010-10-28 20:32 ` Venkatesh Srinivas
2010-10-28 20:55 ` Nemo
2010-10-28 21:22 ` Venkatesh Srinivas
2010-10-28 22:17 ` Francisco J Ballesteros
2010-10-28 23:51 ` Gorka Guardiola
2010-10-29 5:37 ` Bruce Ellis
2010-10-29 7:44 ` Gorka Guardiola
2010-10-29 8:51 ` Bruce Ellis
2010-10-29 9:01 ` Charles Forsyth
2010-10-29 8:58 ` Bruce Ellis
2010-10-29 16:38 ` Charles Forsyth
2010-10-29 9:45 ` Gorka Guardiola
2010-10-29 9:56 ` Bruce Ellis [this message]
2010-10-29 16:01 ` Charles Forsyth
2010-10-29 15:58 ` Venkatesh Srinivas
2010-10-29 16:02 ` roger peppe
2010-10-29 16:17 ` Bruce Ellis
2010-10-29 16:47 ` roger peppe
2010-10-29 16:04 ` Charles Forsyth
2010-10-29 14:14 ` Eric Van Hensbergen
2010-10-29 15:21 ` roger peppe
2010-10-29 15:36 ` erik quanstrom
2010-10-29 15:39 ` Eric Van Hensbergen
2010-10-29 15:49 ` Charles Forsyth
2010-10-29 15:50 ` Eric Van Hensbergen
2010-10-29 15:59 ` Bruce Ellis
2010-10-29 16:28 ` Venkatesh Srinivas
2010-10-29 17:16 ` erik quanstrom
2010-10-29 16:47 ` David Leimbach
2010-10-29 17:13 ` Charles Forsyth
2010-10-29 17:22 ` David Leimbach
2010-10-29 17:26 ` erik quanstrom
2010-10-29 17:47 ` Bruce Ellis
2010-10-29 19:33 ` roger peppe
2010-10-29 20:27 ` Bruce Ellis
2010-10-29 20:41 ` Gorka Guardiola
2010-10-29 20:44 ` Bruce Ellis
2010-10-29 22:31 ` Bakul Shah
2010-10-29 22:39 ` roger peppe
2010-10-29 22:43 ` Bruce Ellis
2010-10-29 23:07 ` roger peppe
2010-10-29 23:28 ` Bruce Ellis
2010-10-30 9:08 ` Nemo
2010-10-31 5:39 ` Bruce Ellis
2010-10-31 8:52 ` roger peppe
2010-10-31 8:57 ` Bruce Ellis
2010-10-31 10:25 ` fgergo
2010-10-31 11:11 ` fgergo
2010-10-31 11:21 ` Bruce Ellis
2010-10-31 11:23 ` roger peppe
2010-10-31 13:06 ` Gorka Guardiola
2010-10-31 13:07 ` Gorka Guardiola
2010-10-31 13:22 ` Gorka Guardiola
2010-11-02 14:29 ` Richard Miller
2010-10-31 14:21 ` Eric Van Hensbergen
2010-10-31 22:17 ` Bruce Ellis
2010-10-31 22:32 ` EBo
2010-10-31 22:41 ` Steve Simon
2010-11-02 13:49 ` Richard Miller
2010-11-02 14:02 ` Eric Van Hensbergen
2010-10-29 18:27 ` David Leimbach
2010-10-29 17:18 ` Charles Forsyth
2010-10-29 17:17 ` erik quanstrom
2010-10-29 17:25 ` David Leimbach
2010-10-29 18:03 ` Eric Van Hensbergen
2010-10-29 18:54 ` Gorka Guardiola
2010-10-29 19:30 ` David Leimbach
2010-10-29 20:20 ` erik quanstrom
2010-10-29 15:55 ` Venkatesh Srinivas
2010-10-29 13:12 ` roger peppe
2010-10-29 13:20 ` Lucio De Re
2010-10-29 15:51 ` Charles Forsyth
2010-10-29 16:36 ` Charles Forsyth
2010-10-27 15:09 ` David Leimbach
2010-10-28 21:39 erik quanstrom
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='AANLkTimCcAOncHe4xOwWCioe2eq7W=6Frcpk7Nrs3TqW@mail.gmail.com' \
--to=bruce.ellis@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).