Reiser4 and ACLs

Reiser4 and ACLs

[Marc Perkel]

So - anyone have any ideas when ACLs might appear in Reiser4?

-- 
Marc Perkel - marc@perkel.com

Spam Filter: http://www.junkemailfilter.com
    My Blog: http://marc.perkel.com

Re: Reiser4 and ACLs

[Hans Reiser]

Marc Perkel wrote:

> So - anyone have any ideas when ACLs might appear in Reiser4?
>
We have a lot on our plate now, and no one paying for that work.  Surely
someone will pay for it.

Hans

Reiser4 and ACLs

[Marc Perkel]

So - how do you get Reiser 4 to work with ACLs? Can't find a mount 
option in any docs.

-- 
Marc Perkel - marc@perkel.com

Spam Filter: http://www.junkemailfilter.com
    My Blog: http://marc.perkel.com

Re: Reiser4 and ACLs

[Mark Nipper]

On 13 Aug 2005, Marc Perkel wrote:
> So - how do you get Reiser 4 to work with ACLs? Can't find a mount 
> option in any docs.

	They are not supported yet under Reiser4.  I'm sure this
is keeping just a few people from early adoption at the moment.
:)

-- 
Mark Nipper                                                e-contacts:
4475 Carter Creek Parkway                           nipsy@bitgnome.net
Apartment 724                               http://nipsy.bitgnome.net/
Bryan, Texas, 77802-4481           AIM/Yahoo: texasnipsy ICQ: 66971617
(979)575-3193                                      MSN: nipsy@tamu.edu

-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GG/IT d- s++:+ a- C++$ UBL++++$ P--->+++ L+++$ !E---
W++(--) N+ o K++ w(---) O++ M V(--) PS+++(+) PE(--)
Y+ PGP t+ 5 X R tv b+++@ DI+(++) D+ G e h r++ y+(**)
------END GEEK CODE BLOCK------

---begin random quote of the moment---
"I do not agree with what you have to say, but I'll defend to the
death your right to say it."
 -- Evelyn Beatrice Hall, _The Friends of Voltaire_, 1906
----end random quote of the moment----

Re: Reiser4 and ACLs

[Marc Perkel]

[-- Attachment #1: Type: text/plain, Size: 1239 bytes --]



Mark Nipper wrote:

>On 13 Aug 2005, Marc Perkel wrote:
>  
>
>>So - how do you get Reiser 4 to work with ACLs? Can't find a mount 
>>option in any docs.
>>    
>>
>
>	They are not supported yet under Reiser4.  I'm sure this
>is keeping just a few people from early adoption at the moment.
>:)
>
>  
>
ACLs not supported? That's bizzare! That's insane!

I had been reading that the very reason you want to use Reiser 4 is 
because that's where the ACLs were going to be. It is after all built on 
a database and the Reiser4 page is full of talk about ACLs. It is 
claimed that Reiser4 is done and ready for the real world and no ACL 
support. Reiser 3 has ACL support. EXT3 has ACL support. Every other 
file system has ACL support.

How can there be a claim that Reiser4 is ready to be merged into the 
Kernel without ACL support? Here I am gearing up to switch over and now 
I can't. I just spent the last few week hassling Redhat to make Fedora 
more Reiser friendly. This is moe than just a little disapointing. It's 
just plain crazy.

So - what's the story with that? Why no ACLs in a system built on a 
dadtbase?

-- 
Marc Perkel - marc@perkel.com

Spam Filter: http://www.junkemailfilter.com
    My Blog: http://marc.perkel.com


[-- Attachment #2: Type: text/html, Size: 1946 bytes --]

Re: Reiser4 and ACLs

[michael chang]

On 8/14/05, Marc Perkel <marc@perkel.com> wrote:
>  So - what's the story with that? Why no ACLs in a system built on a
> dadtbase?

Database.  And you *could* ask, I suppose, why there are no
compression or other plugins available in a system based on plugins. 
Why there is no repacker when there is supposed to be an interface for
one and is one of the main reasons Reiser4 is so wonderful.  Why...

The reason? Time.  Money.  Resources.

And indeed, the bare bones of Resier4 (the only part the -mm team
wanted) are ready for vanilla inclusion.  The problem is we can't go
any further until this is done -- otherwise, a huge bug that crops up
could end up going undetected for a long time, and fixing it could end
up being a huge task that could take ages.  Reiser4 is undertested and
underfunded.  Vanilla could change that.

Note, this is all speculation.

-- 
~Mike
 - Just my two cents
 - No man is an island, and no man is unable.

Re: Reiser4 and ACLs

[Marc Perkel]

btw - is Reiser4 still going to get merged into 2.6.13?

Re: Reiser4 and ACLs

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 345 bytes --]

On Sun, 14 Aug 2005 05:38:40 PDT, Marc Perkel said:
> btw - is Reiser4 still going to get merged into 2.6.13?

It's not in 2.6.13-rc6, and I doubt Linus is going to blop *that* big
a chunk of code in this late - it's already well into the "is this 3-liner
too drastic" phase.

What happens when the 2.6.14 tree opens is up to Linus and Andrew.


[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

Re: Reiser4 and ACLs

[Nikita Danilov]

michael chang writes:
 > On 8/14/05, Marc Perkel <marc@perkel.com> wrote:
 > >  So - what's the story with that? Why no ACLs in a system built on a
 > > dadtbase?
 > 
 > Database.  And you *could* ask, I suppose, why there are no
 > compression or other plugins available in a system based on plugins. 
 > Why there is no repacker when there is supposed to be an interface for
 > one and is one of the main reasons Reiser4 is so wonderful.  Why...
 > 
 > The reason? Time.  Money.  Resources.

Not exactly. As a matter of fact, ACL and EA support was already
implemented in reiser4. But it used standard xattrs API to interface to
the user-land, and it was decided that reiser4 should go sys_reiser4()
route instead. So, it was reaped.

Nikita.

Re: Reiser4 and ACLs

[Marc Perkel]

Nikita Danilov wrote:

>Not exactly. As a matter of fact, ACL and EA support was already
>implemented in reiser4. But it used standard xattrs API to interface to
>the user-land, and it was decided that reiser4 should go sys_reiser4()
>route instead. So, it was reaped.
>
>Nikita.
>  
>

OK - now I'm confused. Why not the standard? After all - it is standard. 
And - when is this going to be included?

My background is in Netware. Netware did it right. It's permissions that 
it had 15 years ago were beyond what Linux might ever acheive. Then I 
downgraded to Windows, then I further downgraded to Linux. In fact when 
I started using linux it took me months of denial before I realized - 
this really is all there is. One user, one group - and that's all. 
Further - the ability of a user to delete files that they have no 
read/write rights to is just plain silly. In Netware if you have no 
right to a file, you not only can't delete it - you can't even list it 
in the directory. It's as if that file isn't even there.

ACLs finally restored some of this power although the implementation in 
Linux is crude at best. I thought there is a glimmer of hope that some 
day Linux will catch up to where Novell was 15 years ago. It is so 
frustrating dealing with Linux culture and VI mentality trying to get 
people to think outside the box.


-- 
Marc Perkel - marc@perkel.com

Spam Filter: http://www.junkemailfilter.com
    My Blog: http://marc.perkel.com

Re: Reiser4 and ACLs

[Matt Stegman]

On Sun, 14 Aug 2005, Marc Perkel wrote:
>
> My background is in Netware. Netware did it right. It's permissions that
> it had 15 years ago were beyond what Linux might ever acheive. Then I
> downgraded to Windows, then I further downgraded to Linux. In fact when
> I started using linux it took me months of denial before I realized -
> this really is all there is. One user, one group - and that's all.
> Further - the ability of a user to delete files that they have no
> read/write rights to is just plain silly. In Netware if you have no
> right to a file, you not only can't delete it - you can't even list it
> in the directory. It's as if that file isn't even there.

A minor nitpick: if you wish to delete a file, you need write permission
on the directory - the permission on the file doesn't really matter
(unless the sticky bit set is set on the directory, in which case you have
to own the file to delete it...).

> ACLs finally restored some of this power although the implementation in
> Linux is crude at best. I thought there is a glimmer of hope that some
> day Linux will catch up to where Novell was 15 years ago. It is so
> frustrating dealing with Linux culture and VI mentality trying to get
> people to think outside the box.

Marc, I agree with you.  You might be interested in the Linux Trustees
project:

http://trustees.sourceforge.net/

Still not not the match of NetWare's flexibility, but it's a step in the
right direction.  If it patches correctly into a -mm kernel, I think it
should even work on top of reiser4.  Additionally, Novell has released
Open Enterprise Server, which supposedly supports creation of NSS volumes
on Linux (SUSE Linux Enterprise Server 9), supporting trustees!

-- 
Matt Stegman

Re: Reiser4 and ACLs

[David Masover]

Marc Perkel wrote:
> 
> 
> Nikita Danilov wrote:
> 
>> Not exactly. As a matter of fact, ACL and EA support was already
>> implemented in reiser4. But it used standard xattrs API to interface to
>> the user-land, and it was decided that reiser4 should go sys_reiser4()
>> route instead. So, it was reaped.
>>
>> Nikita.
>>  
>>
> 
> OK - now I'm confused. Why not the standard?

Why aren't you using ext3?  Or Windows?  In your own words:
> After all - it is standard. 

> And - when is this going to be included?

Your guess is as good as mine.

> ACLs finally restored some of this power although the implementation in 
> Linux is crude at best.

That's why the "standard" way isn't happening.  Although I wish they'd 
left it in, and the standard xattrs interface, and created the "better" 
interface later.

> I thought there is a glimmer of hope that some 
> day Linux will catch up to where Novell was 15 years ago. It is so 
> frustrating dealing with Linux culture and VI mentality trying to get 
> people to think outside the box.

Speak for yourself.  After all, you wanted us to do the "standard".

Better yet, go read the whitepaper.  It's on the homepage at 
namesys.com, and it's very human-readable.

Re: Reiser4 and ACLs

[Hubert Chan]

On Sun, 14 Aug 2005 17:24:17 +0400, Nikita Danilov <nikita@clusterfs.com> said:

> Not exactly. As a matter of fact, ACL and EA support was already
> implemented in reiser4. But it used standard xattrs API to interface
> to the user-land, and it was decided that reiser4 should go
> sys_reiser4() route instead. So, it was reaped.

Does this mean that file-as-dir has also been abandoned?

-- 
Hubert Chan <hubert@uhoreg.ca> - http://www.uhoreg.ca/
PGP/GnuPG key: 1024D/124B61FA
Fingerprint: 96C5 012F 5F74 A5F7 1FF7  5291 AF29 C719 124B 61FA
Key available at wwwkeys.pgp.net.   Encrypted e-mail preferred.

Re: Reiser4 and ACLs

[Nikita Danilov]

Hubert Chan writes:
 > On Sun, 14 Aug 2005 17:24:17 +0400, Nikita Danilov <nikita@clusterfs.com> said:
 > 
 > > Not exactly. As a matter of fact, ACL and EA support was already
 > > implemented in reiser4. But it used standard xattrs API to interface
 > > to the user-land, and it was decided that reiser4 should go
 > > sys_reiser4() route instead. So, it was reaped.
 > 
 > Does this mean that file-as-dir has also been abandoned?

Sorry, it seems I was too vague. It's exactly the opposite: standard
xattr API was abandoned in favour of accessing EAs and ACLs through
pseudo files. The latter method is not implemented yet, and I don't know
how stable sys_reiser4() API is currently. Hans is the proper person to
ask.

 > -- 
 > Hubert Chan <hubert@uhoreg.ca> - http://www.uhoreg.ca/

Nikita.

Re: Reiser4 and ACLs

[Jonathan Briggs]

[-- Attachment #1: Type: text/plain, Size: 1202 bytes --]

On Tue, 2005-08-16 at 00:19 +0400, Nikita Danilov wrote:
> Hubert Chan writes:
>  > On Sun, 14 Aug 2005 17:24:17 +0400, Nikita Danilov <nikita@clusterfs.com> said:
>  > 
>  > > Not exactly. As a matter of fact, ACL and EA support was already
>  > > implemented in reiser4. But it used standard xattrs API to interface
>  > > to the user-land, and it was decided that reiser4 should go
>  > > sys_reiser4() route instead. So, it was reaped.
>  > 
>  > Does this mean that file-as-dir has also been abandoned?
> 
> Sorry, it seems I was too vague. It's exactly the opposite: standard
> xattr API was abandoned in favour of accessing EAs and ACLs through
> pseudo files. The latter method is not implemented yet, and I don't know
> how stable sys_reiser4() API is currently. Hans is the proper person to
> ask.

There is no reason not to support the standard xattr calls, whether it
is done in-kernel or in the C library.  The pseudo files would also be
good, but several programs already use the current xattr support.

Namesys may not want to add the support themselves but I hope they do
not reject contributed code for it.
-- 
Jonathan Briggs <jbriggs@esoft.com>
eSoft, Inc.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Reiser4 and ACLs

[Nikita Danilov]

Jonathan Briggs writes:
 > On Tue, 2005-08-16 at 00:19 +0400, Nikita Danilov wrote:
 > > Hubert Chan writes:
 > >  > On Sun, 14 Aug 2005 17:24:17 +0400, Nikita Danilov <nikita@clusterfs.com> said:
 > >  > 
 > >  > > Not exactly. As a matter of fact, ACL and EA support was already
 > >  > > implemented in reiser4. But it used standard xattrs API to interface
 > >  > > to the user-land, and it was decided that reiser4 should go
 > >  > > sys_reiser4() route instead. So, it was reaped.
 > >  > 
 > >  > Does this mean that file-as-dir has also been abandoned?
 > > 
 > > Sorry, it seems I was too vague. It's exactly the opposite: standard
 > > xattr API was abandoned in favour of accessing EAs and ACLs through
 > > pseudo files. The latter method is not implemented yet, and I don't know
 > > how stable sys_reiser4() API is currently. Hans is the proper person to
 > > ask.
 > 
 > There is no reason not to support the standard xattr calls, whether it
 > is done in-kernel or in the C library.  The pseudo files would also be
 > good, but several programs already use the current xattr support.
 > 
 > Namesys may not want to add the support themselves but I hope they do
 > not reject contributed code for it.

Well, this song has been sung many times already. :-) Hans has strong
opinion about xattr API (which, indeed, is not a paragon of design
beauty by a large margin).

 > -- 
 > Jonathan Briggs <jbriggs@esoft.com>
 > eSoft, Inc.

Nikita.

Re: Reiser4 and ACLs

[Marc Perkel]

Nikita Danilov wrote:

>
>Well, this song has been sung many times already. :-) Hans has strong
>opinion about xattr API (which, indeed, is not a paragon of design
>beauty by a large margin).
>  
>
I'm new to this debate. What is his strong opinion?

Re: Reiser4 and ACLs

[David Masover]

Marc Perkel wrote:
> 
> 
> Nikita Danilov wrote:
> 
>>
>> Well, this song has been sung many times already. :-) Hans has strong
>> opinion about xattr API (which, indeed, is not a paragon of design
>> beauty by a large margin).
>>  
>>
> I'm new to this debate. What is his strong opinion?

Well, I'll let him speak for himself, but I think it's basically this:

xattrs suck, for reasons which are varied and irrelevant.

When it's done, the sys_reiser4 api (and file-as-dir/metas/whatever) 
will be able to do everything xattrs did, only much better, faster, and 
more powerfully.

So, it's more worth their time to work on other things, like:
- kernel acceptance
- repacker
- crypto/compression
- security ("views" or ACLs)

Of course, if someone wanted to pay them for xattrs, and they didn't 
have another contract going, they'd probably support xattrs.  But for 
now, it's probably up to us to add xattr support.

Once it's added, these should do the same thing:

openat ("some_attribute")
open ("metas/some_attribute")
// or maybe
open ("metas/xattrs/some_attribute")

Well, there's more to it than that, but the most important thing is, the 
xattrs should be a subset of the metas namespace.

Re: Reiser4 and ACLs

[Marc Perkel]

David Masover wrote:

>
> Well, I'll let him speak for himself, but I think it's basically this:
>
> xattrs suck, for reasons which are varied and irrelevant.
>
> When it's done, the sys_reiser4 api (and file-as-dir/metas/whatever) 
> will be able to do everything xattrs did, only much better, faster, 
> and more powerfully.
>
> So, it's more worth their time to work on other things, like:
> - kernel acceptance
> - repacker
> - crypto/compression
> - security ("views" or ACLs)
>
> Of course, if someone wanted to pay them for xattrs, and they didn't 
> have another contract going, they'd probably support xattrs.  But for 
> now, it's probably up to us to add xattr support.
>
> Once it's added, these should do the same thing:
>
> openat ("some_attribute")
> open ("metas/some_attribute")
> // or maybe
> open ("metas/xattrs/some_attribute")
>
> Well, there's more to it than that, but the most important thing is, 
> the xattrs should be a subset of the metas namespace.
>
It seems to me that getting in the Kernel is the most important thing. 
Then getting it compatible with existing standards. Then you build up 
the user based (addicts) and then you add the innovative stuff. 
Otherwise people are going to use Ext3 because it has ACLs and Reiser4 
doesn't. That's where I'm at. If it doesn't do ACLs it doesn't work. I 
need acls.

The way I see it - and I may well be missing something - s that the 
current API is just a way of talking to the security layer. So why not 
support multiple ways to talk to it?

Re: Reiser4 and ACLs

[michael chang]

On 8/15/05, Marc Perkel <marc@perkel.com> wrote:
> It seems to me that getting in the Kernel is the most important thing.

This is because nothing else can happen because no one will know it
exists.  Everything else will happen *as soon as* Reiser4 gets in. 
Otherwise, it's practically useless.

> Then getting it compatible with existing standards. Then you build up

But then you're basically rewriting ext2 or 3.  There's no point in that.

> the user based (addicts) and then you add the innovative stuff.

You do this first, because it's what sets you appart.  Otherwise,
people won't convert away from EXT2/3.  Priorities first.

> Otherwise people are going to use Ext3 because it has ACLs and Reiser4
> doesn't. That's where I'm at. If it doesn't do ACLs it doesn't work. I
> need acls.

So why fix something that isn't broken?  Just because one person who
uses ACLs doesn't use Reiser4 right away doesn't mean he won't use it
later.  Compare the number of people who use Ubuntu Linux (who won't
support Reiser4 until it's vanilla) and other Vanilla kernels, to the
number of ACL users.  The numbers matter.

> The way I see it - and I may well be missing something - s that the
> current API is just a way of talking to the security layer. So why not
> support multiple ways to talk to it?

There is nothing wrong with this idea.  It's just time consuming. 
It's not done, yet (for Reiser4).  If someone is willing to pay the
time/sweat or the money to get it done, it will be done.  Otherwise,
it won't.  Simple as that.

-- 
~Mike
 - Just my two cents
 - No man is an island, and no man is unable.

Re: Reiser4 and ACLs

[David Masover]

michael chang wrote:
> On 8/15/05, Marc Perkel <marc@perkel.com> wrote:
> 
>>It seems to me that getting in the Kernel is the most important thing.
> 
> 
> This is because nothing else can happen because no one will know it
> exists.  Everything else will happen *as soon as* Reiser4 gets in. 
> Otherwise, it's practically useless.
> 
> 
>>Then getting it compatible with existing standards. Then you build up
> 
> 
> But then you're basically rewriting ext2 or 3.  There's no point in that.
> 
> 
>>the user based (addicts) and then you add the innovative stuff.
> 
> 
> You do this first, because it's what sets you appart.  Otherwise,
> people won't convert away from EXT2/3.  Priorities first.
> 
> 
>>Otherwise people are going to use Ext3 because it has ACLs and Reiser4
>>doesn't. That's where I'm at. If it doesn't do ACLs it doesn't work. I
>>need acls.
> 
> 
> So why fix something that isn't broken?  Just because one person who
> uses ACLs doesn't use Reiser4 right away doesn't mean he won't use it
> later.  Compare the number of people who use Ubuntu Linux (who won't
> support Reiser4 until it's vanilla) and other Vanilla kernels, to the
> number of ACL users.  The numbers matter.
> 
> 
>>The way I see it - and I may well be missing something - s that the
>>current API is just a way of talking to the security layer. So why not
>>support multiple ways to talk to it?
> 
> 
> There is nothing wrong with this idea.  It's just time consuming. 
> It's not done, yet (for Reiser4).  If someone is willing to pay the
> time/sweat or the money to get it done, it will be done.  Otherwise,
> it won't.  Simple as that.

And, just to clarify, people are willing to pay the money for the 
time/sweat that it takes to implement the nice, clean, new ways of doing 
things like ACLs and xattrs.  I don't think anyone's willing to spend 
money (yet) to get the old things working -- no point.

I'm betting that at some point, there's going to be a question of the 
amount of time it'd take to rewrite an existing project vs. adding the 
xattrs compatibility layer, and that layer will be easier to do, because 
someone has a rediculously un-modular system.  But not soon, and not 
before it's in the vanilla kernel.



Completely offtopic, inane comment follows:

"Vanilla" should really be called "Essence of Madagascar Orchid", or 
some such.  It is, in fact, one of the most exotic tastes on the planet. 
  Read up at:  http://en.wikipedia.org/wiki/Vanilla_%28orchid%29

So, the Vanilla kernel should never be seen as something boring ;)

Re: Reiser4 and ACLs

[Hubert Chan]

On Mon, 15 Aug 2005 16:30:32 -0700, Marc Perkel <marc@perkel.com> said:

> It seems to me that getting in the Kernel is the most important
> thing. Then getting it compatible with existing standards. Then you
> build up the user based (addicts) and then you add the innovative
> stuff. Otherwise people are going to use Ext3 because it has ACLs and
> Reiser4 doesn't. That's where I'm at. If it doesn't do ACLs it doesn't
> work. I need acls.

AFAIK, ACLs are separate from xattr.  It's just that Ext3 uses xattr to
store their ACLs.  So it should be possible to implement ACLs (following
the same standards) via sys_reiser4.  Perhaps someone more knowledgeable
can confirm (or deny).

-- 
Hubert Chan <hubert@uhoreg.ca> - http://www.uhoreg.ca/
PGP/GnuPG key: 1024D/124B61FA
Fingerprint: 96C5 012F 5F74 A5F7 1FF7  5291 AF29 C719 124B 61FA
Key available at wwwkeys.pgp.net.   Encrypted e-mail preferred.

Re: Reiser4 and ACLs

[michael chang]

On 8/15/05, Hubert Chan <hubert@uhoreg.ca> wrote:
> On Mon, 15 Aug 2005 16:30:32 -0700, Marc Perkel <marc@perkel.com> said:
> 
> > It seems to me that getting in the Kernel is the most important
> > thing. Then getting it compatible with existing standards. Then you
> > build up the user based (addicts) and then you add the innovative
> > stuff. Otherwise people are going to use Ext3 because it has ACLs and
> > Reiser4 doesn't. That's where I'm at. If it doesn't do ACLs it doesn't
> > work. I need acls.
> 
> AFAIK, ACLs are separate from xattr.  It's just that Ext3 uses xattr to
> store their ACLs.  So it should be possible to implement ACLs (following
> the same standards) via sys_reiser4.

It is.  That's what reiser4 is doing.  It's just that older xattr
tools don't know how to access the new ACLs access system, (a.k.a.
access ACLs *NOT* via xattrs) and it's not written yet.

> Perhaps someone more knowledgeable
> can confirm (or deny).

-- 
~Mike
 - Just my two cents
 - No man is an island, and no man is unable.

Re: Reiser4 and ACLs

[David Masover]

Jonathan Briggs wrote:
> On Tue, 2005-08-16 at 00:19 +0400, Nikita Danilov wrote:
> 
>>Hubert Chan writes:
>> > On Sun, 14 Aug 2005 17:24:17 +0400, Nikita Danilov <nikita@clusterfs.com> said:
>> > 
>> > > Not exactly. As a matter of fact, ACL and EA support was already
>> > > implemented in reiser4. But it used standard xattrs API to interface
>> > > to the user-land, and it was decided that reiser4 should go
>> > > sys_reiser4() route instead. So, it was reaped.
>> > 
>> > Does this mean that file-as-dir has also been abandoned?
>>
>>Sorry, it seems I was too vague. It's exactly the opposite: standard
>>xattr API was abandoned in favour of accessing EAs and ACLs through
>>pseudo files. The latter method is not implemented yet, and I don't know
>>how stable sys_reiser4() API is currently. Hans is the proper person to
>>ask.
> 
> 
> There is no reason not to support the standard xattr calls,

Except time, money, and more important things to do.

Actually, Hans probably doesn't like the idea of supporting them at all, 
but even Hans has to accept that everyone won't rewrite apps for 
pseudo-files at once.

> whether it
> is done in-kernel or in the C library.  The pseudo files would also be
> good, but several programs already use the current xattr support.

Several programs already use Windows.  We do have Wine, but would you 
rather all the kernel hackers started hacking on making Wine more 
compatible instead of making Linux better?

> Namesys may not want to add the support themselves but I hope they do
> not reject contributed code for it.

Probably not, just make sure you fully understand what pseudo files are 
and what they do, and how you're going to fit the two together.  If the 
patch makes xattrs a completely separate namespace, you're probably 
going to get a "try again".

Re: Reiser4 and ACLs

[Marc Perkel]

Nikita Danilov wrote:

>Sorry, it seems I was too vague. It's exactly the opposite: standard
>xattr API was abandoned in favour of accessing EAs and ACLs through
>pseudo files. The latter method is not implemented yet, and I don't know
>how stable sys_reiser4() API is currently. Hans is the proper person to
>ask.
>
> 
>
Why not have more than one interface if you need that? Is there any 
reason for not supporting the standard that is in the kernel now? Reiser 
3 supports it.

Re: Reiser4 and ACLs

[David Masover]

Marc Perkel wrote:
> 
> 
> Nikita Danilov wrote:
> 
>> Sorry, it seems I was too vague. It's exactly the opposite: standard
>> xattr API was abandoned in favour of accessing EAs and ACLs through
>> pseudo files. The latter method is not implemented yet, and I don't know
>> how stable sys_reiser4() API is currently. Hans is the proper person to
>> ask.
>>
>>
>>
> Why not have more than one interface if you need that? Is there any 
> reason for not supporting the standard that is in the kernel now? Reiser 
> 3 supports it.

With third-party patches, yes.  And because there was no reiser4 and no 
metas.

Re: Reiser4 and ACLs

[David Masover]

Marc Perkel wrote:
> 
> 
> Mark Nipper wrote:
> 
>>On 13 Aug 2005, Marc Perkel wrote:
>>  
>>
>>>So - how do you get Reiser 4 to work with ACLs? Can't find a mount 
>>>option in any docs.
>>>    
>>>
>>
>>	They are not supported yet under Reiser4.  I'm sure this
>>is keeping just a few people from early adoption at the moment.
>>:)
>>
>>  
>>

> I had been reading that the very reason you want to use Reiser 4 is 
> because that's where the ACLs were going to be. It is after all built on 
> a database

No, it's not.  It's just easy to build a database on top of, without 
having to redesign the storage system.  As I said in my last message, go 
read the whitepaper.

> How can there be a claim that Reiser4 is ready to be merged into the 
> Kernel without ACL support?

Because things don't have to have every feature imaginable to be merged. 
  In fact, quite the opposite -- they have to be sane and stable.