From: Corey Minyard <minyard@acm.org>
To: Buddy Lumpkin <b.lumpkin@attbi.com>
Cc: "'Bill Davidsen'" <davidsen@tmr.com>,
"'Peter Waechtler'" <pwaechtler@mac.com>,
"'Larry McVoy'" <lm@bitmover.com>,
linux-kernel@vger.kernel.org, "'ingo Molnar'" <mingo@redhat.com>
Subject: Re: [ANNOUNCE] Native POSIX Thread Library 0.1
Date: Mon, 30 Sep 2002 09:54:45 -0500 [thread overview]
Message-ID: <3D9865B5.7020006@acm.org> (raw)
In-Reply-To: 000001c2680f$a13af930$0472e50c@peecee
Buddy Lumpkin wrote:
>Sun introduced a new thread library in Solaris 8 that is 1:1, but it did
>not replace the default N:M version, you have to link against
>/usr/lib/lwp.
>
>http://supportforum.sun.com/freesolaris/techfaqs.html?techfaqs_2957
>http://www.itworld.com/AppDev/1170/swol-1218-insidesolaris/
>
>I was at a USENIX BOF on threads in Boston year before last and Bill
>Lewis was ranting about how the N:M model sucks. Christopher Provenzano
>was right there and didn't seem to add any feelings one way or the
>other.
>
>Regards,
>
>--Buddy
>
I heard this a while ago, and talked with someone I knew who had inside
information about this. According to that person, Sun will be switching
the default threads library to 1:1 (It looks like from the document
referenced below it is Solaris 9). In various benchmarks, sometimes M:N
won, and sometimes 1:1 won, so performance was a wash. The main problem
was that they could never get certain things to work "just right" under
an M:N model, the complexity of M:N was just too high to be able to get
it working 100% correctly. He didn't have specific details, though.
Having implemented a threads package with prority inheritance, I expect
that doing that with an M:N thread model will be extremely complex.
With activations is possible, but that doesn't mean it's easy. It's
hard enough with a 1:1 model. A scheduler with good "global" properties
(for example, a scheduler that guaranteed time share to classes of
threads that occur in different processes) would be difficult to
implement properly, too.
Complexity is the enemy of reliability. Even if the M:N model could get
slightly better performance, it's going to be very hard to make it work
100% correctly. I personally think the NPT is going in the right
direction on this one.
-Corey
>
>-----Original Message-----
>From: linux-kernel-owner@vger.kernel.org
>[mailto:linux-kernel-owner@vger.kernel.org] On Behalf Of Bill Davidsen
>Sent: Monday, September 23, 2002 12:15 PM
>To: Peter Waechtler
>Cc: Larry McVoy; linux-kernel@vger.kernel.org; ingo Molnar
>Subject: Re: [ANNOUNCE] Native POSIX Thread Library 0.1
>
>On Mon, 23 Sep 2002, Peter Waechtler wrote:
>
>
>
>>Am Montag den, 23. September 2002, um 12:05, schrieb Bill Davidsen:
>>
>>
>>
>>>On Sun, 22 Sep 2002, Larry McVoy wrote:
>>>
>>>
>>>
>>>>On Sun, Sep 22, 2002 at 08:55:39PM +0200, Peter Waechtler wrote:
>>>>
>>>>
>>>>>AIX and Irix deploy M:N - I guess for a good reason: it's more
>>>>>flexible and combine both approaches with easy runtime tuning if
>>>>>the app happens to run on SMP (the uncommon case).
>>>>>
>>>>>
>>>>No, AIX and IRIX do it that way because their processes are so
>>>>
>>>>
>bloated
>
>
>>>>that it would be unthinkable to do a 1:1 model.
>>>>
>>>>
>>>And BSD? And Solaris?
>>>
>>>
>>Don't know. I don't have access to all those Unices. I could try
>>
>>
>FreeBSD.
>
>At your convenience.
>
>
>
>>According to http://www.kegel.com/c10k.html Sun is moving to 1:1
>>and FreeBSD still believes in M:N
>>
>>
>
>Sun is total news to me, "moving to" may be in Solaris 9, Sol8 seems to
>still be N:M. BSD is as I thought.
>
>
>>MacOSX 10.1 does not support PROCESS_SHARED locks, tried that 5
>>
>>
>minutes
>
>
>>ago.
>>
>>
>
>Thank you for the effort. Hum, that's a bit of a surprise, at least to
>me.
>
>
>
next prev parent reply other threads:[~2002-09-30 14:49 UTC|newest]
Thread overview: 124+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-09-22 18:55 [ANNOUNCE] Native POSIX Thread Library 0.1 Peter Waechtler
2002-09-22 21:32 ` Larry McVoy
2002-09-23 10:05 ` Bill Davidsen
2002-09-23 11:55 ` Peter Waechtler
2002-09-23 19:14 ` Bill Davidsen
2002-09-29 23:26 ` Buddy Lumpkin
2002-09-30 14:54 ` Corey Minyard [this message]
2002-09-23 15:30 ` Larry McVoy
2002-09-23 19:44 ` Olivier Galibert
2002-09-23 19:48 ` Bill Davidsen
2002-09-23 20:32 ` Ingo Molnar
2002-09-24 0:03 ` Andy Isaacson
2002-09-24 0:10 ` Jeff Garzik
2002-09-24 0:14 ` Andy Isaacson
2002-09-24 5:53 ` Ingo Molnar
2002-09-24 20:34 ` David Schwartz
2002-09-24 7:12 ` Thunder from the hill
2002-09-24 7:30 ` Ingo Molnar
2002-09-23 22:35 ` Mark Mielke
2002-09-23 19:59 ` Peter Waechtler
2002-09-23 20:36 ` Ingo Molnar
2002-09-23 21:08 ` Peter Wächtler
2002-09-23 22:44 ` Mark Mielke
2002-09-23 23:01 ` Bill Huey
2002-09-23 23:11 ` Mark Mielke
2002-09-24 0:21 ` Bill Huey
2002-09-24 3:20 ` Mark Mielke
2002-09-23 23:57 ` Andy Isaacson
2002-09-24 6:32 ` 1:1 threading vs. scheduler activations (was: Re: [ANNOUNCE] Native POSIX Thread Library 0.1) Ingo Molnar
2002-09-25 3:08 ` Bill Huey
2002-09-24 18:10 ` [ANNOUNCE] Native POSIX Thread Library 0.1 Christoph Hellwig
2002-09-23 21:32 ` Bill Huey
2002-09-23 21:41 ` dean gaudet
2002-09-23 22:10 ` Bill Huey
2002-09-23 22:56 ` Mark Mielke
2002-09-24 10:02 ` Nikita Danilov
2002-09-23 21:22 ` Bill Huey
2002-09-23 21:03 ` Bill Huey
2002-09-24 12:03 ` Michael Sinz
2002-09-24 13:40 ` Peter Svensson
2002-09-24 14:20 ` Michael Sinz
2002-09-24 14:50 ` Offtopic: (was Re: [ANNOUNCE] Native POSIX Thread Library 0.1) Peter Svensson
2002-09-24 15:19 ` Mark Veltzer
2002-09-24 17:29 ` Rik van Riel
2002-09-25 18:57 ` Mark Mielke
2002-09-25 19:04 ` Rik van Riel
2002-09-25 19:29 ` Mark Veltzer
2002-09-25 19:23 ` Rik van Riel
2002-09-24 16:31 ` Rik van Riel
2002-09-24 18:49 ` Michael Sinz
2002-09-24 19:12 ` PATCH: per user fair scheduler 2.4.19 (cleaned up, thanks hch) (was: Re: Offtopic: (was Re: [ANNOUNCE] Native POSIX Thread Library 0.1)) Rik van Riel
2002-09-24 20:19 ` [ANNOUNCE] Native POSIX Thread Library 0.1 David Schwartz
2002-09-24 21:10 ` Chris Friesen
2002-09-24 21:22 ` Rik van Riel
2002-09-24 21:35 ` Roberto Peon
2002-09-24 21:35 ` Chris Friesen
2002-09-25 19:02 ` David Schwartz
2002-09-24 23:16 ` Peter Waechtler
2002-09-24 23:23 ` Rik van Riel
2002-09-25 19:05 ` David Schwartz
[not found] <987738530@toto.iv>
2002-09-24 2:48 ` Peter Chubb
2002-09-24 3:37 ` Mark Mielke
-- strict thread matches above, loose matches on Subject: below --
2002-09-23 16:36 Matthias Urlichs
2002-09-20 7:46 Joerg Pommnitz
2002-09-20 0:41 Ulrich Drepper
2002-09-20 0:51 ` William Lee Irwin III
2002-09-20 1:35 ` Ulrich Drepper
2002-09-20 1:42 ` William Lee Irwin III
2002-09-20 1:56 ` Larry McVoy
2002-09-20 2:01 ` Rik van Riel
2002-09-20 2:15 ` Benjamin LaHaise
2002-09-20 2:40 ` Dave Hansen
2002-09-20 2:47 ` William Lee Irwin III
2002-09-20 2:17 ` Larry McVoy
2002-09-20 2:24 ` Rik van Riel
2002-09-20 2:32 ` Ulrich Drepper
2002-09-20 6:01 ` Linus Torvalds
2002-09-20 8:02 ` Ingo Molnar
2002-09-20 2:23 ` Anton Blanchard
2002-09-20 9:53 ` Padraig Brady
2002-09-20 13:28 ` Robert Love
2002-09-20 16:01 ` Bill Davidsen
2002-09-20 9:54 ` Adrian Bunk
2002-09-20 10:53 ` Ingo Molnar
2002-09-20 19:04 ` Ulrich Drepper
2002-09-20 23:06 ` J.A. Magallon
2002-09-20 23:33 ` Ulrich Drepper
2002-09-20 23:42 ` J.A. Magallon
2002-09-20 10:20 ` Bill Huey
2002-09-20 10:47 ` Ingo Molnar
2002-09-20 12:06 ` Bill Huey
2002-09-20 16:20 ` Ingo Molnar
2002-09-20 21:50 ` Bill Huey
2002-09-20 22:30 ` dean gaudet
2002-09-20 23:11 ` Bill Huey
2002-09-21 3:38 ` dean gaudet
2002-09-21 4:01 ` Bill Huey
2002-09-21 5:06 ` Ingo Molnar
2002-09-20 23:45 ` Bill Huey
2002-09-21 4:58 ` Ingo Molnar
2002-09-22 2:51 ` Bill Huey
2002-09-21 4:48 ` Ingo Molnar
2002-09-22 1:38 ` Bill Huey
2002-09-22 13:38 ` Bill Davidsen
2002-09-22 18:41 ` Eric W. Biederman
2002-09-22 22:13 ` dean gaudet
2002-09-26 17:21 ` Alan Cox
2002-09-23 0:11 ` Bill Huey
2002-09-24 16:07 ` Eric W. Biederman
2002-09-24 23:21 ` Bill Huey
2002-09-25 3:06 ` Eric W. Biederman
2002-09-23 21:12 ` Bill Huey
2002-09-20 10:35 ` Luca Barbieri
2002-09-20 11:19 ` Ingo Molnar
2002-09-20 18:40 ` Roland McGrath
2002-09-20 21:21 ` Luca Barbieri
2002-09-20 12:37 ` jlnance
2002-09-20 16:42 ` Ingo Molnar
2002-09-24 0:40 ` Rusty Russell
2002-09-24 5:47 ` Ingo Molnar
2002-09-24 6:15 ` Rusty Russell
2002-09-20 15:43 ` Bill Davidsen
2002-09-20 16:15 ` Jakub Jelinek
2002-09-20 17:16 ` Bill Davidsen
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=3D9865B5.7020006@acm.org \
--to=minyard@acm.org \
--cc=b.lumpkin@attbi.com \
--cc=davidsen@tmr.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lm@bitmover.com \
--cc=mingo@redhat.com \
--cc=pwaechtler@mac.com \
/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).