linux-kernel - Re: [PATCH] hwmon: sparx5: make it selectable for ARCH

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: <4824134d-0286-4f9c-9b62-2f969e3245d8@roeck-us.net>
Date: Thu, 25 Sep 2025 13:33:58 -0700
From: Guenter Roeck <linux@...ck-us.net>
To: Robert Marko <robert.marko@...tura.hr>
Cc: linux-hwmon@...r.kernel.org, linux-kernel@...r.kernel.org,
	daniel.machon@...rochip.com, luka.perkov@...tura.hr,
	benjamin.ryzman@...onical.com
Subject: Re: [PATCH] hwmon: sparx5: make it selectable for ARCH_MICROCHIP

On Thu, Sep 25, 2025 at 10:12:48PM +0200, Robert Marko wrote:
> On Thu, Sep 25, 2025 at 5:33 PM Guenter Roeck <linux@...ck-us.net> wrote:
> >
> > On Thu, Sep 25, 2025 at 11:15:52AM +0200, Robert Marko wrote:
> > > On Thu, Sep 25, 2025 at 12:07 AM Guenter Roeck <linux@...ck-us.net> wrote:
> > > >
> > > > On 9/24/25 14:30, Robert Marko wrote:
> > > > > LAN969x uses the same sensor and driver, so make it selectable for
> > > > > ARCH_MICROCHIP.
> > > > >
> > > > LAN969x _is_ the Ethernet switch driver for Sparx5, so this description does
> > > > not really make sense. Same as what ? Itself ?
> > > >
> > > > > Signed-off-by: Robert Marko <robert.marko@...tura.hr>
> > > > > ---
> > > > >   drivers/hwmon/Kconfig | 2 +-
> > > > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > > >
> > > > > diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
> > > > > index 840d998f850a..ba2b7b2f6c44 100644
> > > > > --- a/drivers/hwmon/Kconfig
> > > > > +++ b/drivers/hwmon/Kconfig
> > > > > @@ -632,7 +632,7 @@ config SENSORS_I5K_AMB
> > > > >
> > > > >   config SENSORS_SPARX5
> > > > >       tristate "Sparx5 SoC temperature sensor"
> > > > > -     depends on ARCH_SPARX5 || COMPILE_TEST
> > > > > +     depends on ARCH_MICROCHIP || COMPILE_TEST
> > > >
> > > > ... and silently disable it for ARCH_SPARX5 at the same time ? That is not what
> > > > the description says, and is completely unacceptable unless explained.
> > > >
> > > > >       help
> > > > >         If you say yes here you get support for temperature monitoring
> > > > >         with the Microchip Sparx5 SoC.
> > > >
> > > > ... and, on top of all that, still claim to support Sparx5 even that is
> > > > no longer the case.
> > > >
> > > > Ah, I see, this patch depends on patches in linux-next. You might want to say that.
> > > > Also, there is context missing: If the sensor is _only_ supported on Sparx5
> > > > (which everything but the dependency suggests), it does not make sense to extend
> > > > the dependencies. Why make the sensor configurable for _all_ microchip architectures
> > > > if it is only supported on Sparx5 ? Maybe there is some other series explaining
> > > > this, but this patch is all I got and it does not explain anything. It is only
> > > > confusing and does not make sense on its own.
> > >
> > > Hi Guenter,
> > > I should have extended the description a bit.
> > >
> > > LAN969x uses the same sensor as SparX-5, and they both select ARCH_MICROCHIP so
> > > SparX-5 only configs can still select this driver.
> >
> > You mean ARCH_LAN969X (in contrast to ARCH_SPARX5 and LAN969X_SWITCH) ?
> 
> Both ARCH_LAN969X (Which is now in linux-next) and ARCH_SPARX5 select
> ARCH_MICROCHIP
> which was introduced to avoid the need to keep adding new ARCH symbols
> to individual drivers.
> 
> LAN969X_SWITCH is just the switchdev driver for ARCH_LAN969X.
> 
> > That makes sense, but it really needs to be explained in the patch
> > description. Also, since ARCH_MICROCHIP does not exist for arm64 in the
> > upstream kernel, a note that this depends on other patches in linux-next
> > would have been helpful. You may know that, and others may know that, but
> > I don't, and digging through that cost me at least an hour of time which
> > I don't have.
> 
> Yes, I should have pointed out that this depends on linux-next since
> the prerequisite
> Kconfig.platforms change for ARM64 is in linux-next, but it should
> land in the regular ARM tree
> if its not already there.
> 
> If there is need for a v2, I will make sure to expand the description.

Please do.

Guenter