On Fri, Aug 12, 2022 at 2:47 AM Tony Lindgren <tony@xxxxxxxxxxx> wrote:
>
> Hi,
>
> * Saravana Kannan <saravanak@xxxxxxxxxx> [220810 05:54]:
> > The driver core now:
> > - Has the parent device of a supplier pick up the consumers if the
> > supplier never has a device created for it.
> > - Ignores a supplier if the supplier has no parent device and will never
> > be probed by a driver
> >
> > And already prevents creating a device link with the consumer as a
> > supplier of a parent.
> >
> > So, we no longer need to find the "compatible" node of the supplier or
> > do any other checks in of_link_to_phandle(). We simply need to make sure
> > that the supplier is available in DT.
>
> This patch fixes booting for me, so it should be applied as a fix and
> tagged with:
>
> Fixes: 5a46079a9645 ("PM: domains: Delete usage of driver_deferred_probe_check_state()")
>
> If there are dependencies to the other patches in this series, it might
> make sense to revert commit 5a46079a9645 instead.
Yes, there are dependencies on the rest of the patches in this series.
For linux-next, I think we should pick up this series once we get more
Tested-bys.
So if 5a46079a9645 is causing any regression in stable branches, we
should pick up the revert series [1] instead of this series we are
replying to.
[1] - https://lore.kernel.org/all/20220727185012.3255200-1-saravanak@xxxxxxxxxx/
> Anyways, thanks for fixing the issue, for this patch:
>
> Reviewed-by: Tony Lindgren <tony@xxxxxxxxxxx>
> Tested-by: Tony Lindgren <tony@xxxxxxxxxxx>
Thanks!
> For the process, looks like the earlier series got merged despite the
> issues reported.
If I'm not mistaken, the issues were reported after the series got
picked up. And the series got some tested-by s before it was picked
up. And once it's in git history, we obviously can't drop it.
> And we had non-booting Linux next for at least some SoCs
> for weeks. And now we are about to have a non-booting -rc1 unless things
> get fixed fast. Annoying glitches, sigh..
Sorry for breaking some boards -- so mean "creative" corner cases :)
This rewrite is way more flexible (removes a lot of limitations in
fw_devlink) and hopefully this handles all the corner cases. We'll
see.
-Saravana
