Hi,
> > We made some investigations for this issue and found commit number which
> > breaks JFFS2 power loss recoverability.
> > It is: [JFFS2] Improve getdents vs. f_pos handling on NOR flash.
> > http://git.infradead.org/mtd-2.6.git?a=commitdiff;h=15953580e79b58caefb107e77f218e009b9992e6
> >
> > Currently we don't know why it breaks JFFS2 and how it could be fixed
> > except roll back.
>
> Does everything work when you unmerge this one commit?
>
Correct, commit roll back fixes the problem.
Thanks
Alexey
On Mon, 2008-06-02 at 12:41 +0100, Alexey Korolev wrote:
> Hi,
> > > We made some investigations for this issue and found commit number which
> > > breaks JFFS2 power loss recoverability.
> > > It is: [JFFS2] Improve getdents vs. f_pos handling on NOR flash.
> > > http://git.infradead.org/mtd-2.6.git?a=commitdiff;h=15953580e79b58caefb107e77f218e009b9992e6
> > >
> > > Currently we don't know why it breaks JFFS2 and how it could be fixed
> > > except roll back.
> >
> > Does everything work when you unmerge this one commit?
> >
> Correct, commit roll back fixes the problem.
But leaves another one. I was away last week, and I'm home now -- I'll
try to reproduce and understand what's going on.
--
dwmw2
Hi David,
> On Mon, 2008-06-02 at 12:41 +0100, Alexey Korolev wrote:
> > Hi,
> > > > We made some investigations for this issue and found commit number which
> > > > breaks JFFS2 power loss recoverability.
> > > > It is: [JFFS2] Improve getdents vs. f_pos handling on NOR flash.
> > > > http://git.infradead.org/mtd-2.6.git?a=commitdiff;h=15953580e79b58caefb107e77f218e009b9992e6
> > > >
> > > > Currently we don't know why it breaks JFFS2 and how it could be fixed
> > > > except roll back.
> > >
> > > Does everything work when you unmerge this one commit?
> > >
> > Correct, commit roll back fixes the problem.
>
> But leaves another one. I was away last week, and I'm home now -- I'll
> try to reproduce and understand what's going on.
>
>
We made more investigations of this problem. Joakim has assisted us to understan the patch. This patch can't break PLR feature on NAND. The only way it affects on tests is timing. So commit roll back will not fix the problem, but hides.
Now we are collecting more information about the problem. We will let you know as
soon as we find something interesting about it.
Thanks,
Alexey