Re: [BISECTED][REGRESSION] 5.10.56 longterm kernel breakage on m68k/aranym

[Geert Uytterhoeven]

CC Mike

On Mon, Aug 9, 2021 at 3:32 PM Mikael Pettersson <mikpelinux@xxxxxxxxx> wrote:
> On Mon, Aug 9, 2021 at 3:59 AM Finn Thain <fthain@xxxxxxxxxxxxxx> wrote:
> > On Sun, 8 Aug 2021, Mikael Pettersson wrote:
> > > On Sun, Aug 8, 2021 at 1:20 AM Finn Thain <fthain@xxxxxxxxxxxxxx> wrote:
> > > > On Sat, 7 Aug 2021, Mikael Pettersson wrote:
> > > > > I updated the 5.10 longterm kernel on one of my m68k/aranym VMs from
> > > > > 5.10.47 to 5.10.56, and the new kernel failed to boot:
> > > > >
> > > > > ARAnyM 1.1.0
> > > > > Using config file: 'aranym1.headless.config'
> > > > > Could not open joystick 0
> > > > > ARAnyM RTC Timer: /dev/rtc: Permission denied
> > > > > ARAnyM LILO: Error loading ramdisk 'root.bin'
> > > > > Blitter tried to read byte from register ff8a00 at 0077ee
> > > > >
> > > > > At this point it kept running, but produced no output to the console,
> > > > > and would never get to the point of starting user-space. Attaching gdb
> > > > > to aranym showed nothing interesting, i.e. it seemed to be executing
> > > > > normally.
> > > > >
> > > > > A git bisect identified the following commit between 5.10.52 and
> > > > > 5.10.53 as the culprit:
> > > > > # first bad commit: [9e1cf2d1ed37c934c9935f2c0b2f8b15d9355654]
> > > > > mm/userfaultfd: fix uffd-wp special cases for fork()
> > > > >
> > > >
> > > > That commit appeared in mainline between v5.13 and v5.14-rc1. Is mainline
> > > > also affected? e.g. v5.14-rc4.
> > >
> > > 5.14-rc4 boots fine. I suspect the commit has some dependency that
> > > hasn't been backported to 5.10 stable.
> > >
> >
> > On mainline, 9e1cf2d1ed3 is known as commit 8f34f1eac382 ("mm/userfaultfd:
> > fix uffd-wp special cases for fork()").
> >
> > There are differences between the two commits that may be relevant. I
> > don't know.
> >
> > If you checkout 8f34f1eac382 and if that works, it would indicate either
> > missing dependencies in -stable, or those differences are important.
> >
> > OTOH, if 8f34f1eac382 fails in the same way as linux-5.10.y, it would
> > indicate that -stable is missing a fix that's present in v5.14-rc4.
>
> My initial bisect was wrong. I tried reverting 8f34f1eac382 from
> 5.10.57 but that made no difference, so I re-ran the git bisect with
> all known good points pre-marked. This landed on:
> # first bad commit: [ce6ee46e0f39ed97e23ebf7b5a565e0266a8a1a3]
> mm/page_alloc: fix memory map initialization for descending nodes
>
> Reverting _that_ from 5.10.57 does unbreak that kernel.
>
> Sorry about the confusion.