linux-kernel - Re: [RFC PATCH] soc: imx: Add AHB to IP Bridge configuration driver

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] [day] [month] [year] [list]

Message-ID: <CAEnQRZAG4Piw6NfQC7Yyc5-d2ymfq3YeL=FpV5_hNt26AzRhjQ@mail.gmail.com>
Date: Tue, 19 Nov 2024 17:26:35 +0200
From: Daniel Baluta <daniel.baluta@...il.com>
To: Frank Li <Frank.li@....com>
Cc: Daniel Baluta <daniel.baluta@....com>, conor+dt@...nel.org, shawnguo@...nel.org, 
	s.hauer@...gutronix.de, festevam@...il.com, imx@...ts.linux.dev, 
	robh@...nel.org, krzk+dt@...nel.org, kernel@...gutronix.de, 
	devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org, 
	linux-kernel@...r.kernel.org, aisheng.dong@....com, peng.fan@....com, 
	iuliana.prodan@....com, laurentiu.mihalcea@....com, shengjiu.wang@....com
Subject: Re: [RFC PATCH] soc: imx: Add AHB to IP Bridge configuration driver

On Tue, Nov 19, 2024 at 5:00 PM Frank Li <Frank.li@....com> wrote:
>
> On Tue, Nov 19, 2024 at 03:07:26PM +0200, Daniel Baluta wrote:
> > This is a RFC patch in order to allow AHB to IP Bridge (AIPSTZ)
> > configuration to be restored after a suspend/resume operation.
> >
> > This is particularly useful for aips5 bus where the aipstz
> > configuration is lost at suspend.
> >
> > In order to configure aipstz bridge we register a platform driver
> > that will set default configuration permission at probe.
> >
> > Because AIPS configuration is lost at suspend/resume time we register
> > a power notifier callback that will take care of re-initializing the
> > configuration at resume.
>
> why not use suspend/resume callback? what's happen if devices under AIPS
> suspend/resume before AIPS resume function.

We are attaching to AIPS power domain. So any device that will resume
will have first to power up the domain which in turn will configure the bridge.

So with this it doesn't really matter the suspend/resume order
becasuse the notifier
that will configure the bridge is already register.

But I wonder if there could be a problem if a device goes
suspend/resume before the bridge
probe function gets a chance to be called.

> supposed
>         bridge@...f0000 {
>                 ...
>                 some child node {
>                 }
>
>                 so child node's devies suspend/resume will after bidged's
> suspend resume.
>
>         }