Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
Sender: Guenter Roeck <groeck7@gmail.com>
Date:   Wed, 27 Apr 2022 04:53:33 -0700
From:   Guenter Roeck <linux@roeck-us.net>
To:     Zev Weiss <zev@bewilderbeest.net>
Cc:     Jean Delvare <jdelvare@suse.com>, linux-hwmon@vger.kernel.org,
        linux-kernel@vger.kernel.org, openbmc@lists.ozlabs.org
Subject: Re: [PATCH] hwmon: (pmbus) delta-ahe50dc-fan: work around hardware
 quirk
Message-ID: <20220427115333.GA2686708@roeck-us.net>
References: <20220427035109.3819-1-zev@bewilderbeest.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220427035109.3819-1-zev@bewilderbeest.net>
Precedence: bulk

On Tue, Apr 26, 2022 at 08:51:09PM -0700, Zev Weiss wrote:
> CLEAR_FAULTS commands can apparently sometimes trigger catastrophic
> power output glitches on the ahe-50dc, so block them from being sent
> at all.
> 
> Signed-off-by: Zev Weiss <zev@bewilderbeest.net>
> Cc: stable@vger.kernel.org

Applied.

Thanks,
Guenter

> ---
>  drivers/hwmon/pmbus/delta-ahe50dc-fan.c | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
> 
> diff --git a/drivers/hwmon/pmbus/delta-ahe50dc-fan.c b/drivers/hwmon/pmbus/delta-ahe50dc-fan.c
> index 40dffd9c4cbf..f546f0c12497 100644
> --- a/drivers/hwmon/pmbus/delta-ahe50dc-fan.c
> +++ b/drivers/hwmon/pmbus/delta-ahe50dc-fan.c
> @@ -14,6 +14,21 @@
>  
>  #define AHE50DC_PMBUS_READ_TEMP4 0xd0
>  
> +static int ahe50dc_fan_write_byte(struct i2c_client *client, int page, u8 value)
> +{
> +	/*
> +	 * The CLEAR_FAULTS operation seems to sometimes (unpredictably, perhaps
> +	 * 5% of the time or so) trigger a problematic phenomenon in which the
> +	 * fan speeds surge momentarily and at least some (perhaps all?) of the
> +	 * system's power outputs experience a glitch.
> +	 *
> +	 * However, according to Delta it should be OK to simply not send any
> +	 * CLEAR_FAULTS commands (the device doesn't seem to be capable of
> +	 * reporting any faults anyway), so just blackhole them unconditionally.
> +	 */
> +	return value == PMBUS_CLEAR_FAULTS ? -EOPNOTSUPP : -ENODATA;
> +}
> +
>  static int ahe50dc_fan_read_word_data(struct i2c_client *client, int page, int phase, int reg)
>  {
>  	/* temp1 in (virtual) page 1 is remapped to mfr-specific temp4 */
> @@ -68,6 +83,7 @@ static struct pmbus_driver_info ahe50dc_fan_info = {
>  		PMBUS_HAVE_VIN | PMBUS_HAVE_FAN12 | PMBUS_HAVE_FAN34 |
>  		PMBUS_HAVE_STATUS_FAN12 | PMBUS_HAVE_STATUS_FAN34 | PMBUS_PAGE_VIRTUAL,
>  	.func[1] = PMBUS_HAVE_TEMP | PMBUS_PAGE_VIRTUAL,
> +	.write_byte = ahe50dc_fan_write_byte,
>  	.read_word_data = ahe50dc_fan_read_word_data,
>  };
>