From: James Bottomley <James.Bottomley@HansenPartnership.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Szabolcs Szakacsits <szaka@ntfs-3g.com>,
Alan Cox <alan@lxorguk.ukuu.org.uk>,
Grant Grundler <grundler@google.com>,
Linux IDE mailing list <linux-ide@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
Jens Axboe <jens.axboe@oracle.com>,
Arjan van de Ven <arjan@infradead.org>
Subject: Re: Implementing NVMHCI...
Date: Sun, 12 Apr 2009 17:23:54 +0000 [thread overview]
Message-ID: <1239557034.3461.14.camel@mulgrave.int.hansenpartnership.com> (raw)
In-Reply-To: <alpine.LFD.2.00.0904120820410.4583@localhost.localdomain>
On Sun, 2009-04-12 at 08:41 -0700, Linus Torvalds wrote:
>
> On Sun, 12 Apr 2009, Szabolcs Szakacsits wrote:
> >
> > I did not hear about NTFS using >4kB sectors yet but technically
> > it should work.
> >
> > The atomic building units (sector size, block size, etc) of NTFS are
> > entirely parametric. The maximum values could be bigger than the
> > currently "configured" maximum limits.
>
> It's probably trivial to make ext3 support 16kB blocksizes (if it doesn't
> already).
>
> That's not the problem. The "filesystem layout" part is just a parameter.
>
> The problem is then trying to actually access such a filesystem, in
> particular trying to write to it, or trying to mmap() small chunks of it.
> The FS layout is the trivial part.
>
> > At present the limits are set in the BIOS Parameter Block in the NTFS
> > Boot Sector. This is 2 bytes for the "Bytes Per Sector" and 1 byte for
> > "Sectors Per Block". So >4kB sector size should work since 1993.
> >
> > 64kB+ sector size could be possible by bootstrapping NTFS drivers
> > in a different way.
>
> Try it. And I don't mean "try to create that kind of filesystem". Try to
> _use_ it. Does Window actually support using it it, or is it just a matter
> of "the filesystem layout is _specified_ for up to 64kB block sizes"?
>
> And I really don't know. Maybe Windows does support it. I'm just very
> suspicious. I think there's a damn good reason why NTFS supports larger
> block sizes in theory, BUT EVERYBODY USES A 4kB BLOCKSIZE DESPITE THAT!
>
> Because it really is a hard problem. It's really pretty nasty to have your
> cache blocking be smaller than the actual filesystem blocksize (the other
> way is much easier, although it's certainly not pleasant either - Linux
> supports it because we _have_ to, but sector-size of hardware had
> traditionally been 4kB, I'd certainly also argue against adding complexity
> just to make it smaller, the same way I argue against making it much
> larger).
>
> And don't get me wrong - we could (fairly) trivially make the
> PAGE_CACHE_SIZE be bigger - even eventually go so far as to make it a
> per-mapping thing, so that you could have some filesystems with that
> bigger sector size and some with smaller ones. I think Andrea had patches
> that did a fair chunk of it, and that _almost_ worked.
>
> But it ABSOLUTELY SUCKS. If we did a 16kB page-cache-size, it would
> absolutely blow chunks. It would be disgustingly horrible. Putting the
> kernel source tree on such a filesystem would waste about 75% of all
> memory (the median size of a source file is just about 4kB), so your page
> cache would be effectively cut in a quarter for a lot of real loads.
>
> And to fix up _that_, you'd need to now do things like sub-page
> allocations, and now your page-cache size isn't even fixed per filesystem,
> it would be per-file, and the filesystem (and the drievrs!) would hav to
> handle the cases of getting those 4kB partial pages (and do r-m-w IO after
> all if your hardware sector size is >4kB).
We might not have to go that far for a device with these special
characteristics. It should be possible to build a block size remapping
Read Modify Write type device to present a 4k block size to the OS while
operating in n*4k blocks for the device. We could implement the read
operations as readahead in the page cache, so if we're lucky we mostly
end up operating on full n*4k blocks anyway. For the cases where we've
lost pieces of the n*4k native block and we have to do a write, we'd
just suck it up and do a read modify write on a separate memory area, a
bit like the new 4k sector devices do emulating 512 byte blocks. The
suck factor of this double I/O plus memory copy overhead should be
mitigated partially by the fact that the underlying device is very fast.
James
next prev parent reply other threads:[~2009-04-12 17:24 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20090412091228.GA29937@elte.hu>
2009-04-12 15:14 ` Szabolcs Szakacsits
2009-04-12 15:20 ` Alan Cox
2009-04-12 16:15 ` Avi Kivity
2009-04-12 17:11 ` Linus Torvalds
2009-04-13 6:32 ` Avi Kivity
2009-04-13 15:10 ` Linus Torvalds
2009-04-13 15:38 ` James Bottomley
2009-04-14 7:22 ` Andi Kleen
2009-04-14 10:07 ` Avi Kivity
2009-04-14 9:59 ` Avi Kivity
2009-04-14 10:23 ` Jeff Garzik
2009-04-14 10:37 ` Avi Kivity
2009-04-14 11:45 ` Jeff Garzik
2009-04-14 11:58 ` Szabolcs Szakacsits
2009-04-17 22:45 ` H. Peter Anvin
2009-04-14 12:08 ` Avi Kivity
2009-04-14 12:21 ` Jeff Garzik
2009-04-25 8:26 ` Pavel Machek
2009-04-12 15:41 ` Linus Torvalds
2009-04-12 17:02 ` Robert Hancock
2009-04-12 17:20 ` Linus Torvalds
2009-04-12 18:35 ` Robert Hancock
2009-04-13 11:18 ` Avi Kivity
2009-04-12 17:23 ` James Bottomley [this message]
[not found] ` <6934efce0904141052j3d4f87cey9fc4b802303aa73b@mail.gmail.com>
2009-04-15 6:37 ` Artem Bityutskiy
2009-04-30 22:51 ` Jörn Engel
2009-04-30 23:36 ` Jeff Garzik
2009-04-11 17:33 Jeff Garzik
2009-04-11 19:32 ` Alan Cox
2009-04-11 19:52 ` Linus Torvalds
2009-04-11 20:21 ` Jeff Garzik
2009-04-11 21:49 ` Grant Grundler
2009-04-11 22:33 ` Linus Torvalds
2009-04-11 23:25 ` Alan Cox
2009-04-11 23:51 ` Jeff Garzik
2009-04-12 0:49 ` Linus Torvalds
2009-04-12 1:59 ` Jeff Garzik
2009-04-12 1:15 ` david
2009-04-12 3:13 ` Linus Torvalds
2009-04-12 14:23 ` Mark Lord
2009-04-12 17:29 ` Jeff Garzik
2009-04-11 19:54 ` Jeff Garzik
2009-04-11 21:08 ` John Stoffel
2009-04-11 21:31 ` John Stoffel
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=1239557034.3461.14.camel@mulgrave.int.hansenpartnership.com \
--to=james.bottomley@hansenpartnership.com \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=arjan@infradead.org \
--cc=grundler@google.com \
--cc=jens.axboe@oracle.com \
--cc=linux-ide@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=szaka@ntfs-3g.com \
--cc=torvalds@linux-foundation.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
all inboxes | Powered by JetHome®