All of lore.kernel.org
 help / color / mirror / Atom feed
* mks.jffs2 uid/gid for root inode
@ 2010-09-03 12:12 Arno Steffen
  2010-09-07  8:02 ` Artem Bityutskiy
  0 siblings, 1 reply; 10+ messages in thread
From: Arno Steffen @ 2010-09-03 12:12 UTC (permalink / raw)
  To: linux-mtd

I use mkfs.jffs2 to generate a partition out of a directory tree (not
the hole root-filesystem).
It seem the owner of the root point of this directory tree is always
root:root, althouh it might have other owners.

Example

mkfs.jffs2 -lnp -e 0x20000 -r /opt/rootfs/opt -o /mnt/hgfs/share/fs_opt.jffs2

In my NFS-rootfs /opt/rootfs/opt is owned by user but flashing the
created jffs file will be mounted as owner root.
>From my point of view this is a bug, as it doesn't care about the real
owner of this part of filetree.
Or did I misused the -r option?

I hope you can give me a comment or some feedback, how to change this behaviour.

Best regards
Arno

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-03 12:12 mks.jffs2 uid/gid for root inode Arno Steffen
@ 2010-09-07  8:02 ` Artem Bityutskiy
  2010-09-09  6:17   ` Arno Steffen
  0 siblings, 1 reply; 10+ messages in thread
From: Artem Bityutskiy @ 2010-09-07  8:02 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-mtd

On Fri, 2010-09-03 at 14:12 +0200, Arno Steffen wrote:
> I use mkfs.jffs2 to generate a partition out of a directory tree (not
> the hole root-filesystem).
> It seem the owner of the root point of this directory tree is always
> root:root, althouh it might have other owners.
> 
> Example
> 
> mkfs.jffs2 -lnp -e 0x20000 -r /opt/rootfs/opt -o /mnt/hgfs/share/fs_opt.jffs2
> 
> In my NFS-rootfs /opt/rootfs/opt is owned by user but flashing the
> created jffs file will be mounted as owner root.
> From my point of view this is a bug, as it doesn't care about the real
> owner of this part of filetree.
> Or did I misused the -r option?
> 
> I hope you can give me a comment or some feedback, how to change this behaviour.

Sounds like a bug. I'll take a look at this as soon as I have some time.
Or you can send a patch.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-07  8:02 ` Artem Bityutskiy
@ 2010-09-09  6:17   ` Arno Steffen
  2010-09-09  8:09     ` Artem Bityutskiy
  0 siblings, 1 reply; 10+ messages in thread
From: Arno Steffen @ 2010-09-09  6:17 UTC (permalink / raw)
  To: dedekind1; +Cc: linux-mtd

2010/9/7 Artem Bityutskiy <dedekind1@gmail.com>:
> On Fri, 2010-09-03 at 14:12 +0200, Arno Steffen wrote:
>> I use mkfs.jffs2 to generate a partition out of a directory tree (not
>> the hole root-filesystem).
>> It seem the owner of the root point of this directory tree is always
>> root:root, althouh it might have other owners.
>>
>> Example
>>
>> mkfs.jffs2 -lnp -e 0x20000 -r /opt/rootfs/opt -o /mnt/hgfs/share/fs_opt.jffs2
>>
>> In my NFS-rootfs /opt/rootfs/opt is owned by user but flashing the
>> created jffs file will be mounted as owner root.
>> From my point of view this is a bug, as it doesn't care about the real
>> owner of this part of filetree.
>> Or did I misused the -r option?
>>
>> I hope you can give me a comment or some feedback, how to change this behaviour.
>
> Sounds like a bug. I'll take a look at this as soon as I have some time.
> Or you can send a patch.
>
> --
> Best Regards,
> Artem Bityutskiy (Артём Битюцкий)

Thanks, thought already nobody is interested in this. Regarding a
patch: I did look at the code, but my knowledge about the details is
to less.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-09  6:17   ` Arno Steffen
@ 2010-09-09  8:09     ` Artem Bityutskiy
       [not found]       ` <AANLkTi=SGuBghcHnFoZfe6j3jVxQbKxzSU7t4z+vwPxc@mail.gmail.com>
  0 siblings, 1 reply; 10+ messages in thread
From: Artem Bityutskiy @ 2010-09-09  8:09 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-mtd

On Thu, 2010-09-09 at 08:17 +0200, Arno Steffen wrote:
> Thanks, thought already nobody is interested in this. Regarding a
> patch: I did look at the code, but my knowledge about the details is
> to less.

I've sent you the patch. Please, test v2 of the patch and let me know if
it works for you. Thanks!

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
       [not found]         ` <1284030609.27765.4.camel@localhost>
@ 2010-09-09 11:54           ` Arno Steffen
  2010-09-09 12:05             ` Artem Bityutskiy
  0 siblings, 1 reply; 10+ messages in thread
From: Arno Steffen @ 2010-09-09 11:54 UTC (permalink / raw)
  To: dedekind1, linux-mtd

2010/9/9 Artem Bityutskiy <dedekind1@gmail.com>:
> On Thu, 2010-09-09 at 11:57 +0200, Arno Steffen wrote:
>> 2010/9/9 Artem Bityutskiy <dedekind1@gmail.com>:
>> > On Thu, 2010-09-09 at 08:17 +0200, Arno Steffen wrote:
>> >> Thanks, thought already nobody is interested in this. Regarding a
>> >> patch: I did look at the code, but my knowledge about the details is
>> >> to less.
>> >
>> > I've sent you the patch. Please, test v2 of the patch and let me know if
>> > it works for you. Thanks!
>> >
>> > --
>> > Best Regards,
>> > Artem Bityutskiy (Артём Битюцкий)
>>
>> Havn't got e mail with patch ?!?
>
> Attached. It was sent to the MTD ML and you was in CC.
>
> --
> Best Regards,
> Artem Bityutskiy (Артём Битюцкий)
>

Thanks for the patch. But this is for ubifs, while I detected this bug
in jffs2 - this is what I am using.
So I can't test this.
Best regards
Arno

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-09 11:54           ` Arno Steffen
@ 2010-09-09 12:05             ` Artem Bityutskiy
  2010-09-24  9:52               ` Arno Steffen
  0 siblings, 1 reply; 10+ messages in thread
From: Artem Bityutskiy @ 2010-09-09 12:05 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-mtd

On Thu, 2010-09-09 at 13:54 +0200, Arno Steffen wrote:
> Thanks for the patch. But this is for ubifs, while I detected this bug
> in jffs2 - this is what I am using.

Oh my god, I really need to start wearing glasses. Sorry. But the same
bug exists in mkfs.ubifs, anyway. Since I'm not interested in JFFS2
anymore, I'm not going to fix it. You can do analogous thing for
mkfs.jffs2, if you wish.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-09 12:05             ` Artem Bityutskiy
@ 2010-09-24  9:52               ` Arno Steffen
  2010-09-24  9:56                 ` Artem Bityutskiy
  0 siblings, 1 reply; 10+ messages in thread
From: Arno Steffen @ 2010-09-24  9:52 UTC (permalink / raw)
  To: dedekind1, David.Woodhouse; +Cc: linux-mtd

2010/9/9 Artem Bityutskiy <dedekind1@gmail.com>:
> On Thu, 2010-09-09 at 13:54 +0200, Arno Steffen wrote:
>> Thanks for the patch. But this is for ubifs, while I detected this bug
>> in jffs2 - this is what I am using.
>
> Oh my god, I really need to start wearing glasses. Sorry. But the same
> bug exists in mkfs.ubifs, anyway. Since I'm not interested in JFFS2
> anymore, I'm not going to fix it. You can do analogous thing for
> mkfs.jffs2, if you wish.
>
> --
> Best Regards,
> Artem Bityutskiy (Артём Битюцкий)

If it's mkfs.jffs2 is not your fokus, Artem, maybe David can have a look at it.
It make sense to keep the same functionality for all mkfs tools.

Best regards
Arno

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-24  9:52               ` Arno Steffen
@ 2010-09-24  9:56                 ` Artem Bityutskiy
  2010-09-24 10:07                   ` Arno Steffen
  0 siblings, 1 reply; 10+ messages in thread
From: Artem Bityutskiy @ 2010-09-24  9:56 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-mtd, David.Woodhouse

On Fri, 2010-09-24 at 11:52 +0200, Arno Steffen wrote:
> 2010/9/9 Artem Bityutskiy <dedekind1@gmail.com>:
> > On Thu, 2010-09-09 at 13:54 +0200, Arno Steffen wrote:
> >> Thanks for the patch. But this is for ubifs, while I detected this bug
> >> in jffs2 - this is what I am using.
> >
> > Oh my god, I really need to start wearing glasses. Sorry. But the same
> > bug exists in mkfs.ubifs, anyway. Since I'm not interested in JFFS2
> > anymore, I'm not going to fix it. You can do analogous thing for
> > mkfs.jffs2, if you wish.
> >
> > --
> > Best Regards,
> > Artem Bityutskiy (Артём Битюцкий)
> 
> If it's mkfs.jffs2 is not your fokus, Artem, maybe David can have a look at it.
> It make sense to keep the same functionality for all mkfs tools.

May be, unsurprisingly, I have no jurisdiction over him, he is free to
do or not to do something for you.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-24  9:56                 ` Artem Bityutskiy
@ 2010-09-24 10:07                   ` Arno Steffen
  2011-01-17 10:53                     ` Arno Steffen
  0 siblings, 1 reply; 10+ messages in thread
From: Arno Steffen @ 2010-09-24 10:07 UTC (permalink / raw)
  To: dedekind1; +Cc: linux-mtd, David.Woodhouse

2010/9/24 Artem Bityutskiy <dedekind1@gmail.com>:
> On Fri, 2010-09-24 at 11:52 +0200, Arno Steffen wrote:
>> 2010/9/9 Artem Bityutskiy <dedekind1@gmail.com>:
>> > On Thu, 2010-09-09 at 13:54 +0200, Arno Steffen wrote:
>> >> Thanks for the patch. But this is for ubifs, while I detected this bug
>> >> in jffs2 - this is what I am using.
>> >
>> > Oh my god, I really need to start wearing glasses. Sorry. But the same
>> > bug exists in mkfs.ubifs, anyway. Since I'm not interested in JFFS2
>> > anymore, I'm not going to fix it. You can do analogous thing for
>> > mkfs.jffs2, if you wish.
>> >
>> > --
>> > Best Regards,
>> > Artem Bityutskiy (Артём Битюцкий)
>>
>> If it's mkfs.jffs2 is not your fokus, Artem, maybe David can have a look at it.
>> It make sense to keep the same functionality for all mkfs tools.
>
> May be, unsurprisingly, I have no jurisdiction over him, he is free to
> do or not to do something for you.
>
> --
> Best Regards,
> Artem Bityutskiy (Артём Битюцкий)
>
>

Hasn't expect an order. Just a broad hint for David ;) (or someone
else who has the ability).
If It would be just for me, I wouldn't push. I know that my only
chance is to argue.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: mks.jffs2 uid/gid for root inode
  2010-09-24 10:07                   ` Arno Steffen
@ 2011-01-17 10:53                     ` Arno Steffen
  0 siblings, 0 replies; 10+ messages in thread
From: Arno Steffen @ 2011-01-17 10:53 UTC (permalink / raw)
  To: linux-mtd; +Cc: David.Woodhouse, dedekind1

Unfortunatly, this bug isn't fixed until now. (I just compiled the head)
The infomation about the owner for the mount point seems not to be part of jffs2

To make this situation more clean: My flash has 2 partitions
/ and /opt

And I also create two jffs2 files, one for each partition: root.jffs2
and opt.jffs2

If I mount /opt - the ownership of /opt is root, although it has been
"user" in the tree I generated the jffs2 files from.

Best regards
Arno



2010/9/24 Arno Steffen <arno.steffen@googlemail.com>:
> 2010/9/24 Artem Bityutskiy <dedekind1@gmail.com>:
>> On Fri, 2010-09-24 at 11:52 +0200, Arno Steffen wrote:
>>> 2010/9/9 Artem Bityutskiy <dedekind1@gmail.com>:
>>> > On Thu, 2010-09-09 at 13:54 +0200, Arno Steffen wrote:
>>> >> Thanks for the patch. But this is for ubifs, while I detected this bug
>>> >> in jffs2 - this is what I am using.
>>> >
>>> > Oh my god, I really need to start wearing glasses. Sorry. But the same
>>> > bug exists in mkfs.ubifs, anyway. Since I'm not interested in JFFS2
>>> > anymore, I'm not going to fix it. You can do analogous thing for
>>> > mkfs.jffs2, if you wish.
>>> >
>>> > --
>>> > Best Regards,
>>> > Artem Bityutskiy (Артём Битюцкий)
>>>
>>> If it's mkfs.jffs2 is not your fokus, Artem, maybe David can have a look at it.
>>> It make sense to keep the same functionality for all mkfs tools.
>>
>> May be, unsurprisingly, I have no jurisdiction over him, he is free to
>> do or not to do something for you.
>>
>> --
>> Best Regards,
>> Artem Bityutskiy (Артём Битюцкий)
>>
>>
>
> Hasn't expect an order. Just a broad hint for David ;) (or someone
> else who has the ability).
> If It would be just for me, I wouldn't push. I know that my only
> chance is to argue.
>

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2011-01-17 10:53 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-09-03 12:12 mks.jffs2 uid/gid for root inode Arno Steffen
2010-09-07  8:02 ` Artem Bityutskiy
2010-09-09  6:17   ` Arno Steffen
2010-09-09  8:09     ` Artem Bityutskiy
     [not found]       ` <AANLkTi=SGuBghcHnFoZfe6j3jVxQbKxzSU7t4z+vwPxc@mail.gmail.com>
     [not found]         ` <1284030609.27765.4.camel@localhost>
2010-09-09 11:54           ` Arno Steffen
2010-09-09 12:05             ` Artem Bityutskiy
2010-09-24  9:52               ` Arno Steffen
2010-09-24  9:56                 ` Artem Bityutskiy
2010-09-24 10:07                   ` Arno Steffen
2011-01-17 10:53                     ` Arno Steffen

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.