From: "Warren W. Gay VE3WWG" <ve3wwg@NoSpam.cogeco.ca>
Subject: Re: For the AdaOS folks
Date: Wed, 05 Jan 2005 21:34:24 -0500
Date: 2005-01-05T21:34:24-05:00 [thread overview]
Message-ID: <Ts1Dd.15053$0y4.13782@read1.cgocable.net> (raw)
In-Reply-To: <gemini.i9veo20018zi3034w.nick.roberts@acm.org>
Nick Roberts wrote:
> "Warren W. Gay VE3WWG" <ve3wwg@NoSPAM.cogeco.ca> wrote:
>>Mach's roots go back to 1975, which is quite some time ago ;-)
>
> Blimey, I didn't know that. I started designing the microkernel that has
> become Bachar in the mid 1980s. Certainly I had never heard of Mach at that
> time.
There is a little Mach history blurb available at:
http://hurd.gnufans.org/bin/view/Mach/MachHistory
Presumably if the project started in 1975, the ideas must have been
floating around for a time prior to that.
>>However, the "name service" for example, will often have many clients
>>(only one receive port for the name service, but at least one send-right
>>for every client "task" on the system). This is one example where the
>>numbers of send-rights can be large.
...
> Looking at my Windows XP Task Manager just now, I see 47 processes and 446
> threads. Interesting. I'm not sure if it's a problem. Currently, I
> anticipate capacity for 500 processes and 1500 threads (per workstation),
> and for 1000 IPC services and 3000 service keys. Each (normal) thread will
> eat up about 300 bytes of RAM (for the TSS and ring 0, 1, and 2 stacks). The
> RAM used up for an IPC service or service key is negligible.
>
> How would Mach do it better?
The only suggestion I was making is that if you separate the port from
the thread (the "Mach way" instead of the L4 way where port=thread), then
you don't have the "thread memory overhead" for each "port". A port
becomes much like a UNIX file descriptor, where a dup(2)-ed file
descriptor just shares an existing file entry (virtually no overhead
- just a pointer and reference count). A port just becomes a nearly
cost-free handle.
As soon as you say every communication end-point implies a thread, then
I as a designer, have to start counting the cost.
--
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg
next prev parent reply other threads:[~2005-01-06 2:34 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-12-27 5:09 For the AdaOS folks Wes Groleau
2004-12-27 10:56 ` Florian Weimer
2004-12-27 12:50 ` Georg Bauhaus
2004-12-27 13:12 ` Florian Weimer
2004-12-28 1:18 ` Wes Groleau
2004-12-27 13:46 ` Adrien Plisson
2004-12-27 16:28 ` Georg Bauhaus
2004-12-28 6:19 ` Microkernels & Ada (Was for the AdaOS folks) Warren W. Gay VE3WWG
2004-12-28 12:02 ` Adrien Plisson
2004-12-28 15:28 ` Warren W. Gay VE3WWG
2004-12-30 1:19 ` For the AdaOS folks Nick Roberts
2004-12-30 13:58 ` Warren W. Gay VE3WWG
2004-12-30 15:27 ` Dmitry A. Kazakov
2004-12-30 16:30 ` Warren W. Gay VE3WWG
[not found] ` <otb8t09dkjh54e1k5s5ccn23ggkqk6ndui@4ax.com>
2004-12-30 19:06 ` OT: Mach Ports (For the AdaOS folks) Warren W. Gay VE3WWG
2004-12-31 10:03 ` For the AdaOS folks Dmitry A. Kazakov
2004-12-31 11:30 ` Warren W. Gay VE3WWG
2004-12-31 12:31 ` Dmitry A. Kazakov
2004-12-31 16:24 ` Warren W. Gay VE3WWG
2004-12-31 17:57 ` Marven Lee
2004-12-31 18:40 ` Warren W. Gay VE3WWG
2004-12-31 19:22 ` Warren W. Gay VE3WWG
2005-01-02 15:09 ` Marven Lee
2005-01-02 20:06 ` Luke A. Guest
2005-01-03 3:13 ` Warren W. Gay VE3WWG
2005-01-03 6:40 ` Luke A. Guest
2005-01-03 10:30 ` Marven Lee
2005-01-03 15:52 ` Warren W. Gay VE3WWG
2005-01-03 16:48 ` Ad Buijsen
2005-01-03 18:49 ` Warren W. Gay VE3WWG
2005-01-03 13:43 ` Marven Lee
2005-01-04 23:36 ` Nick Roberts
2005-01-03 16:22 ` Warren W. Gay VE3WWG
2005-01-04 23:16 ` Nick Roberts
2005-01-05 3:48 ` Warren W. Gay VE3WWG
2005-01-05 13:14 ` Nick Roberts
2005-01-01 12:53 ` Dmitry A. Kazakov
2005-01-02 0:31 ` Warren W. Gay VE3WWG
2005-01-02 11:50 ` Dmitry A. Kazakov
2005-01-02 22:04 ` Warren W. Gay VE3WWG
2005-01-03 10:30 ` Dmitry A. Kazakov
2005-01-03 16:36 ` Warren W. Gay VE3WWG
2005-01-03 17:05 ` Dmitry A. Kazakov
2005-01-03 19:01 ` Warren W. Gay VE3WWG
2005-01-03 19:55 ` Dmitry A. Kazakov
2005-01-03 20:44 ` Warren W. Gay VE3WWG
2005-01-04 0:02 ` Randy Brukardt
2005-01-04 17:44 ` Warren W. Gay VE3WWG
2005-01-04 20:14 ` Nick Roberts
2005-01-04 9:59 ` Dmitry A. Kazakov
2005-01-04 18:00 ` Warren W. Gay VE3WWG
2005-01-04 19:07 ` Dmitry A. Kazakov
2005-01-04 19:57 ` Warren W. Gay VE3WWG
2005-01-05 0:02 ` Nick Roberts
2005-01-05 4:37 ` Warren W. Gay VE3WWG
2005-01-05 18:54 ` Nick Roberts
2005-01-05 20:04 ` Warren W. Gay VE3WWG
2005-01-06 0:32 ` Nick Roberts
2005-01-06 1:29 ` Wes Groleau
2005-01-06 11:03 ` Dmitry A. Kazakov
2005-01-05 9:39 ` Dmitry A. Kazakov
2005-01-05 11:20 ` Warren W. Gay VE3WWG
2005-01-05 12:18 ` Dmitry A. Kazakov
2005-01-05 14:39 ` Warren W. Gay VE3WWG
2005-01-05 17:16 ` zest_fien
2005-01-05 19:44 ` Larry Kilgallen
2005-01-04 20:09 ` Nick Roberts
2005-01-05 10:19 ` Dmitry A. Kazakov
2005-01-05 18:33 ` Nick Roberts
2005-01-05 20:15 ` Dmitry A. Kazakov
2004-12-31 18:47 ` Nick Roberts
2004-12-31 20:36 ` Warren W. Gay VE3WWG
2005-01-04 18:22 ` Nick Roberts
2005-01-05 5:12 ` Warren W. Gay VE3WWG
2005-01-05 18:02 ` Nick Roberts
2005-01-05 19:55 ` Warren W. Gay VE3WWG
2005-01-06 0:57 ` Nick Roberts
2005-01-06 2:34 ` Warren W. Gay VE3WWG [this message]
-- strict thread matches above, loose matches on Subject: below --
2005-01-05 12:14 Mike Brenner
2005-01-05 18:04 ` Warren W. Gay VE3WWG
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox