From: Warner Losh <imp@bsdimp.com>
To: Larry McVoy <lm@mcvoy.com>
Cc: Bakul Shah <bakul@iitbombay.org>, TUHS main list <tuhs@minnie.tuhs.org>
Subject: Re: [TUHS] FreeBSD behind the times? (was: Favorite unix design principles?)
Date: Fri, 5 Feb 2021 11:16:26 -0700 [thread overview]
Message-ID: <CANCZdfqCKdBT--WEnfAkH4Xnu7nyPAvvDfTmBmardEjon7goRg@mail.gmail.com> (raw)
In-Reply-To: <20210205141820.GO13701@mcvoy.com>
[-- Attachment #1: Type: text/plain, Size: 1540 bytes --]
On Fri, Feb 5, 2021, 7:19 AM Larry McVoy <lm@mcvoy.com> wrote:
> On Thu, Feb 04, 2021 at 09:17:54PM -0800, Bakul Shah wrote:
> > On Feb 4, 2021, at 4:33 PM, Larry McVoy <lm@mcvoy.com> wrote:
> > >
> > > Ignoring the page cache and make their own cache has big problems.
> > > You can mmap() ZFS files and doing so means that when a page is
> referenced
> > > it is copied from the ZFS cache to the page cache. That creates a
> > > coherency problem, I can write via the mapping and I can write via
> > > write(2) and now you have two copies of the data that don't match,
> > > that's pretty much OS no-no #1.
> >
> > Write(2)ing to a mapped page sounds pretty dodgy. Likely to get you
> > in trouble in any case. Similarly read(2)ing.
>
> The entire point of the SunOS 4.0 VM system was that the page you
> saw via mmap(2) is the exact same page you saw via read(2). It's
> the page cache, it has page sized chunks of memory that cache
> file,offset pairs.
>
> There is one, and only one, copy of the truth. Doesn't matter how
> you get at it, there is only one "it".
>
> ZFS broke that contract and that was a step backwards in terms of
> OS design.
>
The double copy is the primary reason we don't use it to store videos we
serve. It's a performance bottleneck as well.
And fixing it is... rather involved... possible, but a lot of work to teach
the ARC about the buffer cache or the buffer cache about the ARC...
But for everything else I do, I accept the imperfect design because of all
the other features it unlocks.
Warner
>
[-- Attachment #2: Type: text/html, Size: 2396 bytes --]
next prev parent reply other threads:[~2021-02-05 18:17 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-25 11:10 [TUHS] Favorite unix design principles? Tyler Adams
2021-01-25 12:32 ` Steve Nickolas
2021-01-26 2:06 ` M Douglas McIlroy
2021-01-26 2:53 ` Steve Nickolas
2021-01-26 10:22 ` Tyler Adams
2021-01-26 12:26 ` John P. Linderman
2021-01-26 15:23 ` Clem Cole
2021-01-26 16:00 ` Niklas Karlsson
2021-01-26 16:13 ` Adam Thornton
[not found] ` <CAKH6PiXKjksEpQOMMMQTbcsMvX2thz3WzqjoRWJAsXnZ4Eq_iQ@mail.gmail.com>
2021-01-30 19:01 ` Tyler Adams
2021-01-30 19:50 ` Jon Steinhart
2021-01-30 20:06 ` Tyler Adams
2021-01-30 21:28 ` Clem Cole
2021-01-30 21:42 ` Dave Horsfall
2021-01-30 21:45 ` Tyler Adams
2021-01-30 22:31 ` Larry McVoy
2021-01-30 22:28 ` Larry McVoy
2021-01-30 23:11 ` [TUHS] FreeBSD behind the times? (was: Favorite unix design principles?) Greg 'groggy' Lehey
2021-01-30 23:17 ` Larry McVoy
2021-01-30 23:22 ` Warner Losh
2021-01-30 23:31 ` [TUHS] [SPAM] " Larry McVoy
2021-01-30 23:37 ` Jon Steinhart
2021-01-30 23:54 ` Larry McVoy
2021-01-31 12:23 ` [TUHS] [SPAM] Re: FreeBSD behind the times? Dermot Tynan
2021-01-31 0:00 ` [TUHS] [SPAM] Re: FreeBSD behind the times? (was: Favorite unix design principles?) Bakul Shah
2021-02-09 2:15 ` [TUHS] " Will Senn
2021-02-09 2:16 ` Will Senn
2021-02-09 2:30 ` Greg 'groggy' Lehey
2021-01-31 0:39 ` Steve Nickolas
2021-01-31 1:47 ` Will Senn
2021-01-31 2:25 ` Larry McVoy
2021-01-31 2:52 ` Will Senn
2021-01-31 3:00 ` Larry McVoy
2021-01-31 3:06 ` Will Senn
2021-01-31 3:32 ` John Cowan
2021-02-04 5:43 ` Dave Horsfall
2021-02-04 6:10 ` Angus Robinson
2021-02-04 7:46 ` Andy Kosela
2021-02-04 22:25 ` Dave Horsfall
2021-02-04 15:45 ` Will Senn
2021-02-04 16:03 ` Henry Bent
2021-02-04 16:32 ` Dan Cross
2021-02-04 16:49 ` Will Senn
2021-02-04 17:46 ` Larry McVoy
2021-02-04 18:41 ` Bakul Shah
2021-02-04 22:28 ` George Michaelson
2021-02-04 22:41 ` Bakul Shah
2021-02-05 0:33 ` Larry McVoy
2021-02-05 5:17 ` Bakul Shah
2021-02-05 14:18 ` Larry McVoy
2021-02-05 18:16 ` Warner Losh [this message]
2021-02-05 18:21 ` ron minnich
2021-02-06 0:03 ` Bakul Shah
2021-02-06 2:06 ` Dan Cross
2021-02-06 3:01 ` Bakul Shah
2021-02-06 1:18 ` John Gilmore
2021-02-06 1:43 ` joe mcguckin
2021-02-06 1:55 ` Bakul Shah
2021-02-05 20:50 ` Dave Horsfall
2021-02-06 0:21 ` Brad Spencer
2021-02-06 2:22 ` Rico Pajarola
2021-02-06 2:55 ` Larry McVoy
2021-02-06 3:07 ` Will Senn
2021-02-27 8:54 ` Stuart Remphrey
2021-02-06 4:55 ` John Cowan
2021-02-04 7:46 ` Chris Torek
2021-02-04 15:47 ` Will Senn
2021-02-11 21:01 ` Angel M Alganza
2021-01-30 23:09 ` [TUHS] Favorite unix design principles? John Cowan
2021-01-30 23:22 ` Jon Steinhart
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=CANCZdfqCKdBT--WEnfAkH4Xnu7nyPAvvDfTmBmardEjon7goRg@mail.gmail.com \
--to=imp@bsdimp.com \
--cc=bakul@iitbombay.org \
--cc=lm@mcvoy.com \
--cc=tuhs@minnie.tuhs.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;
as well as URLs for NNTP newsgroup(s).