From: Jay Ashworth <jra@baylink.com>
To: xfs <xfs@oss.sgi.com>
Subject: Re: XFS...Windows?
Date: Mon, 9 Dec 2013 11:42:14 -0500 (EST) [thread overview]
Message-ID: <31871244.254.1386607334674.JavaMail.root@benjamin.baylink.com> (raw)
In-Reply-To: <8D3FA7645C1CFC4E9E783D22B4C708644A595E@P-EXMB2-DC21.corp.sgi.com>
Just to be clear about what's being discussed here:
The proposal is to create a *windows* filesystem driver to allow Windows
machines to read and write locally-attached XFS filesystems, correct?
Presumably, such filesystems would be device-portable to Linux or other
machines also running XFS?
Cheers,
-- jr 'will ask stupid questions for food' a
----- Original Message -----
> From: "Shaun Gosse" <sgosse@sgi.com>
> To: "Yuji Saeki" <yuji1@mail.com>, xfs@oss.sgi.com
> Sent: Monday, December 9, 2013 11:16:38 AM
> Subject: RE: XFS...Windows?
> Well…at the risk of jumping into something here:
>
> “would SGI be okay with that”: keep in mind this a community project.
> There are contributors to XFS here at SGI. There is also a lot of work
> done by employees of Redhat as well as other companies (and a few
> freelance / people interested regardless of current position). Anyhow,
> just a quibble to note that it’s not a question of what SGI dictates.
> Although I will note that I’ve actually raised the idea of XFS on
> Windows internally (here at SGI) and while it’s not a priority I would
> say there’s at least some interest. At the least, if you build it,
> we’ll certainly take a look at it. ☺
>
> I saw an email a while back on this list that implied someone was
> trying to work on that iirc; I didn’t see any follow-up from it. There
> are also some projects that look abandoned which tried to do this I
> recall from previous searches on this.
>
> One of the suggestions I heard when I was considering this was to look
> at how the ext port on Windows works. That looks mildly inactive
> itself but should be good for comparison.
>
> Internally, SGI has a proprietary system, CXFS, which does clustered
> XFS including Windows as client nodes. So we’ve got code that actually
> does XFS data writes from Windows nodes. Of course, the tricky bit is
> managing all of the metadata/logging, which isn’t touched from Windows
> in that system, but it’s just another tidbit to note “yes, this type
> of system should be possible”.
>
> As far as licensing, I’m certainly no expert, but AFAIK it should be
> possible to use the same GPL license as you note and make an
> open-source Windows driver for this using as much of the current XFS
> code as is convenient. There are nuances in the license though. I can
> try to find the article if you’re interested: I read a post previously
> by someone who had developed a GPL driver on Windows and was noting
> concerns about linking against closed source libraries (essentially
> decided it was okay because it (a) could be considered a system
> library and (b) they were starting a fresh project so they could add
> extra exemptions specifically allowing that).
>
> I am a novice at XFS and, frankly, at Windows kernel device
> development as well. But I’m certainly interested in seeing a GPL
> XFS-on-Windows developed at some point. If you go ahead with this, I’d
> definitely appreciate any information and would be glad to assist in
> any way I can. And myself personally, I’d absolutely appreciate
> someone else taking the lead on this, because with other issues I’m
> working on, I would be unlikely to put serious solo effort into this
> anytime soon.
>
> Cheers,
> -Shaun Gosse
> CXFS Windows
> SGI
>
> From: xfs-bounces@oss.sgi.com [mailto:xfs-bounces@oss.sgi.com] On
> Behalf Of Yuji Saeki
> Sent: Sunday, December 08, 2013 5:25 PM
> To: xfs@oss.sgi.com
> Subject: XFS...Windows?
>
> Congratulations on your file system, it’s quite nice. I’ve benchmarked
> it myself, it’s very impressive. I’m curious though, is there are
> thought to a Windows driver? If someone ported the most current-stable
> version to Windows (free) under the same license ‘GNU Lesser General
> Public License’ with entire credit to XFS, would SGI be okay with
> that? A ‘no-warranty implied’ (etc) kind of protection of course.
> Quite a few devs that I work with enjoy the XFS filesystem and we dev
> in both Linux and Windows. We’re very interested in making this happen
> with respect to the XFS project.
>
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs
--
Make Election Day a federal holiday: http://wh.gov/lBm94 100k sigs by 12/14
Jay R. Ashworth Baylink jra@baylink.com
Designer The Things I Think RFC 2100
Ashworth & Associates http://baylink.pitas.com 2000 Land Rover DII
St Petersburg FL USA #natog +1 727 647 1274
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2013-12-09 16:42 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-12-08 23:24 XFS...Windows? Yuji Saeki
2013-12-09 16:16 ` XFS...Windows? Shaun Gosse
2013-12-09 16:42 ` Jay Ashworth [this message]
2013-12-09 16:39 ` XFS...Windows? Eric Sandeen
2013-12-09 20:33 ` [humor] XFS...Windows? Jeffrey Hundstad
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=31871244.254.1386607334674.JavaMail.root@benjamin.baylink.com \
--to=jra@baylink.com \
--cc=xfs@oss.sgi.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.