From: "Zack Weinberg" <zack@codesourcery.com>
To: Jamie Lokier <jamie@shareable.org>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Garbage collectors and VM
Date: Fri, 04 Jul 2003 09:13:33 -0700 [thread overview]
Message-ID: <87d6gqs21e.fsf@egil.codesourcery.com> (raw)
In-Reply-To: <20030704120732.GD22105@mail.jlokier.co.uk> (Jamie Lokier's message of "Fri, 4 Jul 2003 13:07:32 +0100")
Jamie Lokier <jamie@shareable.org> writes:
> Zack Weinberg wrote:
>> Thus, a new pseudo-device, with the semantics:
>
> I like it! I'd use it, too.
Thanks. Implementing it is way beyond me, but it's good to hear that
other people might find it useful.
>> * Reading from the descriptor produces a list of user-space pointers
>> to all the pages that have been reset to read-write since the last
>> read.
>
> Would it be appropriate to have a limit on the number of pages which
> become writable before a signal is delivered instead of continuing to
> make more pages writable? Just like the kernel, sometimes its good to
> limit the number of dirty pages in flight in userspace, too.
Maybe. The GC I wanted to use this with is rather constrained in some
ways - it can't walk the stack, for instance - so it can only collect
when the mutator tells it it's okay. I suppose it could just copy the
list to its own buffer and continue, which saves the kernel from
having to store a potentially very large list.
>> * I never decided what to do if the program forks. The application I
>> personally care about doesn't do that, but for a general GC like
>> Boehm it matters.
>
> It should clone the state, obviously, and COW should be invisible to
> the application :)
Well, yeah, that would be the cleanest thing, but it does require two
dirty bits.
> Btw, are these pages swappable?
They certainly ought to be.
> On a different but related topic, it would be most cool if there were
> a way for the kernel to request memory to be released from a userspace
> GC, prior to swapping the GC's memory. Currently the best strategy is
> for each GC to guess how much of the machine's RAM it can use, however
> this is not a good strategy if you wish to launch multiple programs
> each of which has its own GC, nor is it a particularly good balance
> between GC application pages and other page-cache pages.
Again, this is not particularly useful to me since the collector can't
collect at arbitrary points - but it would be a good thing to have for
a general GC, and maybe I ought to be using a general GC anyway...
zw
next prev parent reply other threads:[~2003-07-04 15:59 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-07-01 1:39 What to expect with the 2.6 VM Mel Gorman
2003-06-30 17:43 ` Daniel Phillips
2003-07-01 20:10 ` Martin J. Bligh
2003-07-01 21:41 ` Mel Gorman
2003-07-01 21:51 ` Davide Libenzi
2003-07-01 21:58 ` Martin J. Bligh
2003-07-02 9:01 ` Mel Gorman
2003-07-01 2:25 ` Andrea Arcangeli
2003-07-01 3:02 ` Andrew Morton
2003-07-01 3:22 ` Andrea Arcangeli
2003-07-01 3:25 ` Andrea Arcangeli
2003-07-01 3:29 ` Rik van Riel
2003-07-01 4:04 ` Andrea Arcangeli
2003-07-01 11:01 ` Hugh Dickins
2003-07-01 3:25 ` William Lee Irwin III
2003-07-01 4:39 ` Andrea Arcangeli
2003-07-01 6:33 ` William Lee Irwin III
2003-07-01 7:49 ` Andrea Arcangeli
2003-07-01 8:59 ` William Lee Irwin III
2003-07-01 9:27 ` Andrea Arcangeli
2003-07-01 12:09 ` severe problem of linux 2.4.21 usb tu guangxiu
2003-07-01 14:24 ` What to expect with the 2.6 VM Martin J. Bligh
2003-07-01 16:22 ` William Lee Irwin III
2003-07-01 17:54 ` Martin J. Bligh
2003-07-02 3:04 ` Andrea Arcangeli
2003-07-01 21:45 ` Mel Gorman
2003-07-01 22:06 ` Martin J. Bligh
2003-07-01 21:46 ` Mel Gorman
2003-07-02 3:08 ` Andrea Arcangeli
2003-07-02 15:57 ` Mel Gorman
2003-07-02 17:11 ` Andrea Arcangeli
2003-07-02 17:10 ` Martin J. Bligh
2003-07-02 17:47 ` Andrea Arcangeli
2003-07-02 17:52 ` Martin J. Bligh
2003-07-02 18:13 ` Andrea Arcangeli
2003-07-02 18:05 ` Rik van Riel
2003-07-02 20:05 ` Martin J. Bligh
2003-07-02 21:40 ` William Lee Irwin III
2003-07-02 21:48 ` Martin J. Bligh
2003-07-02 22:14 ` William Lee Irwin III
2003-07-02 22:02 ` Andrea Arcangeli
2003-07-02 22:15 ` William Lee Irwin III
2003-07-02 22:26 ` Andrea Arcangeli
2003-07-02 23:11 ` William Lee Irwin III
2003-07-02 23:30 ` Andrea Arcangeli
2003-07-02 23:55 ` William Lee Irwin III
2003-07-03 11:31 ` Andrea Arcangeli
2003-07-03 11:46 ` William Lee Irwin III
2003-07-03 12:58 ` Andrea Arcangeli
2003-07-03 13:06 ` Rik van Riel
2003-07-03 13:48 ` Andrea Arcangeli
2003-07-03 18:53 ` William Lee Irwin III
2003-07-03 19:27 ` Andrea Arcangeli
2003-07-03 19:32 ` Rik van Riel
2003-07-03 20:16 ` William Lee Irwin III
2003-07-04 0:40 ` Andrea Arcangeli
2003-07-04 1:46 ` William Lee Irwin III
2003-07-04 2:34 ` Andrea Arcangeli
2003-07-04 4:10 ` William Lee Irwin III
2003-07-04 5:54 ` Andrea Arcangeli
2003-07-04 8:15 ` William Lee Irwin III
2003-07-04 23:44 ` Andrea Arcangeli
2003-07-05 0:05 ` William Lee Irwin III
2003-07-05 0:08 ` Andrea Arcangeli
2003-07-03 18:48 ` Jamie Lokier
2003-07-03 18:54 ` William Lee Irwin III
2003-07-03 19:33 ` Andrea Arcangeli
2003-07-03 22:21 ` William Lee Irwin III
2003-07-04 0:46 ` Andrea Arcangeli
2003-07-04 1:33 ` Jamie Lokier
2003-07-04 1:36 ` William Lee Irwin III
2003-07-04 6:54 ` Garbage collectors and VM (was Re: What to expect with the 2.6 VM) Zack Weinberg
2003-07-04 12:07 ` Jamie Lokier
2003-07-04 16:13 ` Zack Weinberg [this message]
2003-07-03 19:06 ` What to expect with the 2.6 VM Andrew Morton
2003-07-03 19:34 ` Andrea Arcangeli
2003-07-02 18:07 ` Rik van Riel
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=87d6gqs21e.fsf@egil.codesourcery.com \
--to=zack@codesourcery.com \
--cc=jamie@shareable.org \
--cc=linux-kernel@vger.kernel.org \
/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