Re: m68k boot failure in -next bisected to 'xarray: Replace exceptional entries'

[Guenter Roeck]

On 06/23/2018 08:26 PM, Matthew Wilcox wrote:
> On Sat, Jun 23, 2018 at 07:52:35PM -0700, Guenter Roeck wrote:
> > On 06/23/2018 06:17 PM, Matthew Wilcox wrote:
> > > On Fri, Jun 22, 2018 at 03:33:35PM -0700, Guenter Roeck wrote:
> > > > It may be odd that fs/quota/netlink.c:quota_genl_family is not word
> > > > aligned, but on the other side I don't think there is a rule that
> > > > the function parameter to genl_register_family() - or the second
> > > > parameter of idr_alloc() - must be word aligned. Am I missing
> > > > something ? After all, it could be a pointer to the nth element
> > > > of a string, or the caller could on purpose allocate IDRs for
> > > > (ptr), (ptr + 1), and so on.
> > >
> > > This is probably going to cure the problem for you (apply on top
> > > of the existing radix tree / IDR changes).
> > >
> > > I realised when I was reviewing the patch that it's no good; if you
> > > call idr_replace() with a radix_tree_is_internal_node(), then you'll
> > > still get a crash.  I'm going to have to disable the optimisation of
> > > shrinking the tree all the way into the head for the IDR.  But this
> > > will probably get m68k working again.
> >
> > It doesn't crash anymore, but there is still a backtrace.
>
> Oof.  Thank you!  Updated patch (still doesn't fix idr_replace, but does
> fix the idr_for_each() problem, at least according to the tests I added
> to the test-suite.

Yes, that is much better:

Build reference: next-20180622-1-g023ad92e94da

Building mcf5208evb:m5208:m5208evb_defconfig:initrd ... running .... passed
Building q800:m68040:mac_defconfig:initrd ... running .... passed
Building q800:m68040:mac_defconfig:rootfs ... running .... passed

Guenter