linux-kernel - Re: [PATCH v3 1/4] gpio-f7188x: Add GPIO support for Nuvoton NCT6116

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]

Date:   Mon, 22 Aug 2022 18:01:10 +0200
From:   Henning Schild <henning.schild@...mens.com>
To:     simon.guinot@...uanux.org
Cc:     Linus Walleij <linus.walleij@...aro.org>,
        Bartosz Golaszewski <brgl@...ev.pl>,
        Pavel Machek <pavel@....cz>,
        Hans de Goede <hdegoede@...hat.com>,
        Mark Gross <markgross@...nel.org>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Lee Jones <lee@...nel.org>, linux-gpio@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-leds@...r.kernel.org,
        platform-driver-x86@...r.kernel.org,
        Sheng-Yuan Huang <syhuang3@...oton.com>,
        Tasanakorn Phaipool <tasanakorn@...il.com>
Subject: Re: [PATCH v3 1/4] gpio-f7188x: Add GPIO support for Nuvoton
 NCT6116

Am Fri, 12 Aug 2022 10:43:03 +0200
schrieb simon.guinot@...uanux.org:

> On Thu, Aug 11, 2022 at 05:39:05PM +0200, Henning Schild wrote:
> > Add GPIO support for Nuvoton NCT6116 chip. Nuvoton SuperIO chips are
> > very similar to the ones from Fintek. In other subsystems they also
> > share drivers and are called a family of drivers.
> > 
> > For the GPIO subsystem the only difference is that the direction
> > bit is reversed and that there is only one data bit per pin. On the
> > SuperIO level the logical device is another one.
> > 
> > Signed-off-by: Henning Schild <henning.schild@...mens.com>
> > ---
> >  drivers/gpio/gpio-f7188x.c | 71
> > +++++++++++++++++++++++++++----------- 1 file changed, 51
> > insertions(+), 20 deletions(-)
> > 
> > diff --git a/drivers/gpio/gpio-f7188x.c b/drivers/gpio/gpio-f7188x.c
> > index 18a3147f5a42..7b05ecc611e9 100644
> > --- a/drivers/gpio/gpio-f7188x.c
> > +++ b/drivers/gpio/gpio-f7188x.c
> > @@ -1,6 +1,7 @@
> >  // SPDX-License-Identifier: GPL-2.0-or-later
> >  /*
> >   * GPIO driver for Fintek Super-I/O F71869, F71869A, F71882,
> > F71889 and F81866
> > + * and Nuvoton Super-I/O NCT6116D
> >   *
> >   * Copyright (C) 2010-2013 LaCie
> >   *
> > @@ -22,13 +23,12 @@
> >  #define SIO_LDSEL		0x07	/* Logical device
> > select */ #define SIO_DEVID		0x20	/* Device ID
> > (2 bytes) */ #define SIO_DEVREV		0x22	/*
> > Device revision */ -#define SIO_MANID		0x23	/*
> > Fintek ID (2 bytes) */ 
> > -#define SIO_LD_GPIO		0x06	/* GPIO logical
> > device */ #define SIO_UNLOCK_KEY		0x87	/* Key
> > to enable Super-I/O */ #define SIO_LOCK_KEY
> > 0xAA	/* Key to disable Super-I/O */ 
> > -#define SIO_FINTEK_ID		0x1934	/* Manufacturer
> > ID */ +#define SIO_LD_GPIO_FINTEK	0x06	/* GPIO
> > logical device */ +#define SIO_LD_GPIO_NUVOTON	0x07
> > /* GPIO logical device */  
> 
> Please indulge me and add a new line here.
> 
> >  #define SIO_F71869_ID		0x0814	/* F71869
> > chipset ID */ #define SIO_F71869A_ID		0x1007
> > /* F71869A chipset ID */ #define SIO_F71882_ID
> > 0x0541	/* F71882 chipset ID */ @@ -37,7 +37,7 @@
> >  #define SIO_F81866_ID		0x1010	/* F81866
> > chipset ID */ #define SIO_F81804_ID		0x1502  /*
> > F81804 chipset ID, same for f81966 */ #define SIO_F81865_ID
> > 	0x0704	/* F81865 chipset ID */ -
> > +#define SIO_NCT6116D_ID		0xD283  /* NCT6116D chipset
> > ID */ 
> 
> ... snip ...
> 
> > @@ -485,12 +516,8 @@ static int __init f7188x_find(int addr, struct
> > f7188x_sio *sio) return err;
> >  
> >  	err = -ENODEV;
> > -	devid = superio_inw(addr, SIO_MANID);
> > -	if (devid != SIO_FINTEK_ID) {
> > -		pr_debug(DRVNAME ": Not a Fintek device at
> > 0x%08x\n", addr);
> > -		goto err;
> > -	}  
> 
> Sorry for missing that at my first review. You can't remove this block
> of code. This driver is poking around on the I2C bus, which is not
> great. So we want to make sure as much as possible that we are
> speaking to the right device.

Unfortunately the Nuvoton Super IOs do not have a global manufacturer
ID, just that chip ID in their global control registers.

So i think we should really just drop the checking of the manufacturer
ID all together, like proposed here. Or we could check manid+chipid for
fintek and only chipid for nuvoton (like i.e. all the wdt and hwmon
drivers for nuvoton do already).

In fact i will implement the best checking we can do, so match
manufacturer and chip where possible and drop to chip only where not.

Henning

> Simon