From: Paul Jackson <pj@sgi.com>
To: Pavel Machek <pavel@ucw.cz>
Cc: mrmacman_g4@mac.com, serue@us.ibm.com,
linux-kernel@vger.kernel.org, frankeh@watson.ibm.com,
haveblue@us.ibm.com
Subject: Re: [RFC] [PATCH 00/13] Introduce task_pid api
Date: Wed, 16 Nov 2005 13:07:26 -0800 [thread overview]
Message-ID: <20051116130726.2412fbc9.pj@sgi.com> (raw)
In-Reply-To: <20051116203603.GA12505@elf.ucw.cz>
> Could we switch
> to 128-bits so that pids are never reused or something like that?
Not easily. We've got a very cool pid-dispenser at this point that
has excellent performance and scalability, but requires a bit map,
one bit per potential pid. That bitmap can't exceed a small percentage
of main memory on most any configuration, constraining us to perhaps
20 to 30 bits. The code currently has a 22 bit arbitrary limit.
Something like 30 bits would usually only make sense on the terabyte
NUMA monster boxes.
128-bit UUID technology scales fine, but adds quite a few compute
cycles per allocation, and would blow out a whole lot of user code
expecting to put a pid in a machine word.
--
I won't rest till it's the best ...
Programmer, Linux Scalability
Paul Jackson <pj@sgi.com> 1.925.600.0401
next prev parent reply other threads:[~2005-11-16 21:07 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-11-14 21:23 [RFC] [PATCH 00/13] Introduce task_pid api Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 01/13] Change pid accesses: drivers Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 02/13] Change pid accesses: most archs Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 03/13] Change pid accesses: filesystems Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 04/13] Change pid accesses: include/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 05/13] Change pid accesses: ipc Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 06/13] Change pid accesses: kernel/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 07/13] Change pid accesses: lib/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 08/13] Change pid accesses: mm/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 09/13] Change pid accesses: net/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 10/13] Change pid accesses: security/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 11/13] Change pid accesses: sound/ Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 12/13] Change pid accesses: ia64 and mips Serge E. Hallyn
2005-11-15 23:08 ` Keith Owens
2005-11-16 11:58 ` Serge E. Hallyn
2005-11-16 13:53 ` Serge E. Hallyn
2005-11-14 21:23 ` [RFC] [PATCH 13/13] Define new task_pid api Serge E. Hallyn
2005-11-14 23:36 ` [RFC] [PATCH 00/13] Introduce " Paul Jackson
2005-11-15 1:01 ` Serge E. Hallyn
2005-11-15 1:35 ` Paul Jackson
2005-11-15 1:51 ` Paul Jackson
2005-11-15 2:29 ` Serge E. Hallyn
2005-11-15 3:37 ` Paul Jackson
2005-11-15 5:15 ` Serge E. Hallyn
2005-11-15 6:35 ` Paul Jackson
2005-11-15 8:11 ` Serge E. Hallyn
2005-11-15 9:06 ` Paul Jackson
2005-11-15 10:07 ` Dave Hansen
2005-11-15 18:10 ` Paul Jackson
2005-11-15 11:59 ` Robin Holt
2005-11-15 13:32 ` Serge E. Hallyn
2005-11-15 14:37 ` Hubertus Franke
2005-11-15 18:39 ` Paul Jackson
2005-11-15 18:54 ` Hubertus Franke
2005-11-15 19:00 ` Serge E. Hallyn
2005-11-15 19:17 ` Hubertus Franke
2005-11-15 22:11 ` Paul Jackson
2005-11-15 23:15 ` Cedric Le Goater
2005-11-15 23:28 ` Paul Jackson
2005-11-15 16:47 ` Greg KH
2005-11-15 17:08 ` Serge E. Hallyn
2005-11-15 17:33 ` Dave Hansen
2005-11-15 5:51 ` Serge E. Hallyn
2005-11-13 15:22 ` Pavel Machek
2005-11-16 19:36 ` Kyle Moffett
2005-11-16 20:36 ` Pavel Machek
2005-11-16 20:48 ` Dave Hansen
2005-11-19 23:30 ` Pavel Machek
2005-11-20 22:38 ` Serge E. Hallyn
2005-12-07 14:53 ` Eric W. Biederman
2005-11-20 23:29 ` Nix
2005-11-16 21:07 ` Paul Jackson [this message]
2005-11-16 20:24 ` Dave Hansen
2005-11-15 13:34 ` Serge E. Hallyn
2005-11-15 11:17 ` Robin Holt
2005-11-15 12:01 ` Dave Hansen
2005-11-15 19:21 ` Ray Bryant
2005-11-15 19:41 ` Serge E. Hallyn
2005-11-15 20:30 ` Ray Bryant
2005-11-15 21:05 ` Serge E. Hallyn
2005-11-15 22:43 ` Paul Jackson
2005-11-15 22:55 ` Cedric Le Goater
2005-11-16 1:12 ` Paul Jackson
2005-12-07 14:46 ` Eric W. Biederman
2005-12-07 17:47 ` Dave Hansen
2005-12-07 17:55 ` Arjan van de Ven
2005-12-07 18:09 ` Dave Hansen
2005-12-07 19:00 ` Arjan van de Ven
2005-12-07 19:42 ` Eric W. Biederman
2005-12-07 22:13 ` Dave Hansen
2005-12-07 22:20 ` Arjan van de Ven
2005-12-12 10:55 ` Dave Airlie
2005-12-19 14:04 ` Eric W. Biederman
2005-12-07 19:19 ` Eric W. Biederman
2005-12-07 21:40 ` Dave Hansen
2005-12-07 22:17 ` Eric W. Biederman
2004-12-14 15:23 ` Pavel Machek
2005-12-14 13:40 ` Arjan van de Ven
2005-12-14 16:29 ` Serge E. Hallyn
2005-12-07 22:31 ` Dave Hansen
2005-12-07 22:51 ` Eric W. Biederman
2005-12-08 5:42 ` Jeff Dike
2005-12-08 10:09 ` Andi Kleen
2005-12-07 22:17 ` Cedric Le Goater
-- strict thread matches above, loose matches on Subject: below --
2005-11-16 2:24 Hua Zhong (hzhong)
2005-11-16 17:52 ` Bernard Blackham
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=20051116130726.2412fbc9.pj@sgi.com \
--to=pj@sgi.com \
--cc=frankeh@watson.ibm.com \
--cc=haveblue@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mrmacman_g4@mac.com \
--cc=pavel@ucw.cz \
--cc=serue@us.ibm.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