linux-kernel - Re: ARM iMX6sx board fails to boot with kernel 6.17

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAPDyKFooV3zH+0egCQhefRxM_twG+MCkQdqWB=KaSkMXYNcD0Q@mail.gmail.com>
Date: Tue, 7 Oct 2025 17:19:55 +0200
From: Ulf Hansson <ulf.hansson@...aro.org>
To: Sergey Organov <sorganov@...il.com>
Cc: Fabio Estevam <festevam@...il.com>, linux-kernel@...r.kernel.org, 
	Shawn Guo <shawnguo@...nel.org>, "Rob Herring (Arm)" <robh@...nel.org>, Anson Huang <Anson.Huang@....com>
Subject: Re: ARM iMX6sx board fails to boot with kernel 6.17

On Tue, 7 Oct 2025 at 13:35, Sergey Organov <sorganov@...il.com> wrote:
>
> Fabio Estevam <festevam@...il.com> writes:
>
> > On Mon, Oct 6, 2025 at 6:22 PM Sergey Organov <sorganov@...il.com> wrote:
> >
> >> It is built form multiple .dtsi, so I figure I attach the one after
> >> pre-processing stage. Please let me know if you'd prefer I rather
> >> manually insert all the custom .dtsi into a single .dts, and send that
> >> one instead.
> >
> > This format is hard to follow.
> >
> > To make things easier for debugging, you could create a minimal board
> > dts file with only UART and eMMC nodes to reproduce the problem.
> >
> > It's not clear to me the relationship between the ANATOP regulators
> > and the eMMC power on your board.
>
> Thanks, Fabio! OK, I'll do, re-check, and get back to you.

FYI, we have had some strange issues cropping up in regards to the
so-called ->sync_state() support in genpd, that was added in
v6.17-rc1. Not sure if this could be related.

Most of the problems pointed to commit: 0e789b491ba0 "pmdomain: core:
Leave powered-on genpds on until sync_state".

Although, it's not possible to make a plain revert of the above
commit, one can opt-out from the new behavior in genpd, by setting
GENPD_FLAG_NO_STAY_ON for the corresponding genpds that are registered
with pm_genpd_init().

Kind regards
Uffe