Hi Christoph,
I'm trying to find information about "read-only bind mount" patches
posted to LKML last year by Dave Hansen. As far as I can tell there were
a number of postings and several patches in the set were included in
2.6.19 as preparation for more of the patch set.
I've sent mail to Dave but he must be to busy, on leave or not keen on
getting these patches included any more as I've not had a response.
It looks like you had some interest in the patches so I was wondering if
you know who, if anyone, is working on them or where I could locate a
coherent, most up to date set of the remaining patches?
The reason I'm chasing this is that recent changes to the NFS client
cause all mount flags to be ignored on all but the first mount which is
causing a fair bit of unrest for some users. Trond seemed sure that the
VFS read-only bind mount updates would resolve this.
Any help, information or input would be much appreciated.
Ian
On Mon, 2007-05-14 at 21:22 +0800, Ian Kent wrote:
> I've sent mail to Dave but he must be to busy, on leave or not keen on
> getting these patches included any more as I've not had a response.
I'm just now catching up on email. I was out for a couple of weeks.
I'm planning on refreshing the patches and giving them another run at
mainline in the next month or so.
-- Dave
On Mon, 2007-05-14 at 08:10 -0700, Dave Hansen wrote:
> On Mon, 2007-05-14 at 21:22 +0800, Ian Kent wrote:
> > I've sent mail to Dave but he must be to busy, on leave or not keen on
> > getting these patches included any more as I've not had a response.
>
> I'm just now catching up on email. I was out for a couple of weeks.
>
> I'm planning on refreshing the patches and giving them another run at
> mainline in the next month or so.
Anything I can do to help?
If so maybe we could reduce the time to posting a bit.
Ian
On Mon, 2007-05-14 at 23:55 +0800, Ian Kent wrote:
>
> Anything I can do to help?
> If so maybe we could reduce the time to posting a bit.
Probably nothing to actually speed up the development, but I'd really
appreciate some testing once I do post them. How about I send you an
advance copy, or cc you on the RFC?
-- Dave
On Mon, 2007-05-14 at 09:01 -0700, Dave Hansen wrote:
> On Mon, 2007-05-14 at 23:55 +0800, Ian Kent wrote:
> >
> > Anything I can do to help?
> > If so maybe we could reduce the time to posting a bit.
>
> Probably nothing to actually speed up the development, but I'd really
> appreciate some testing once I do post them. How about I send you an
> advance copy, or cc you on the RFC?
Sure.
My personal test environment is limited.
I can test them with the autofs Connectathon suite and do sanity
checking and the like.
In my other life I can spin a RHEL5 kernel in a private branch and get
that to customers that are complaining, hopefully they will be willing
to test this stuff and persevere with any needed debugging (ya right).
I'm happy to screen reports before passing them on and hopefully provide
you with sensible feedback. But it would be best if we could at least
get this into the mm kernel. That should get them a fair bit of
exposure.
The other question for me is I don't know yet if this will resolve the
problem that lead me to chase this. That's the problem where all NFS
client mounts to the same exported filesystem always use the mount
options of the first mount.
Are you sure I can't help with development, even just re-spinning them
against the current kernel may help (of course you would know best).
Having read the patch set in the archives I see that I'm familiar with
the code in most of the areas of the VFS they touch (but maybe not to
the depth that I would like).
Ian