Re: [PATCH v3] hwmon: Add driver for fsp-3y PSUs and PDUs

From: "Václav Kubernát" <kubernat@cesnet.cz>
To: Guenter Roeck <linux@roeck-us.net>
Cc: Jean Delvare <jdelvare@suse.com>,
	Jonathan Corbet <corbet@lwn.net>,
	linux-hwmon@vger.kernel.org, linux-doc@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3] hwmon: Add driver for fsp-3y PSUs and PDUs
Date: Tue, 13 Apr 2021 12:54:05 +0200	[thread overview]
Message-ID: <CABKa3nqhbnDTsSQc4xgo0gvGR3xeiznJPt_nK-wA_U9aoYXg+w@mail.gmail.com> (raw)
In-Reply-To: <78016097-21df-8321-8b8b-33d15a6e6ff2@roeck-us.net>

út 13. 4. 2021 v 7:36 odesílatel Guenter Roeck <linux@roeck-us.net> napsal:
>
> On 4/12/21 9:31 PM, Václav Kubernát wrote:
> [ ... ]
> > Okay, I have made some additional testing. Some of the registers
> > return 0xFFFF, some don't. These are the ones that pmbus_core queries
> > when the driver is loading (with values I got from
> > i2c_smbus_read_word_data):
> > PMBUS_POUT_MAX 0xFFFF
> > PMBUS_FAN_COMMAND_1 0x0
> > PMBUS_VOUT_OV_FAULT_LIMIT 0xFFFF
> > PMBUS_VOUT_OV_WARN_LIMIT 0xFFFF
> > PMBUS_VOUT_UV_WARN_LIMIT 0xFFFF
> > PMBUS_VOUT_UV_FAULT_LIMIT 0xFFFF
> > PMBUS_IOUT_OC_FAULT_LIMIT 0xFFFF
> > PMBUS_IOUT_OC_WARN_LIMIT 0x10
> > PMBUS_IOUT_UC_FAULT_LIMIT 0xFFFF
> > PMBUS_OT_FAULT_LIMIT 0xFFFF
> > PMBUS_OT_WARN_LIMIT 0xFFFFFFB6
> > PMBUS_UT_WARN_LIMIT 0xFFFF
> > PMBUS_UT_FAULT_LIMIT 0xFFFF
> > PMBUS_VIN_OV_FAULT_LIMIT 0xFFFF
> > PMBUS_VIN_OV_WARN_LIMIT 0xFFFF
> > PMBUS_VIN_UV_WARN_LIMIT 0xFFFF
> > PMBUS_VIN_UV_FAULT_LIMIT 0xFFFF
> > PMBUS_IIN_OC_FAULT_LIMIT 0xFFFF
> > PMBUS_IIN_OC_WARN_LIMIT 0x0
> > PMBUS_POUT_OP_FAULT_LIMIT 0xFFFF
> > PMBUS_POUT_OP_WARN_LIMIT 0xFFFFFFB6
> > PMBUS_PIN_OP_WARN_LIMIT 0x0
> > PMBUS_READ_FAN_SPEED_2 0x0
> > PMBUS_MFR_VIN_MIN 0x5A
> > PMBUS_MFR_VIN_MAX 0x108
> > PMBUS_MFR_IIN_MAX 0x3
> > PMBUS_MFR_PIN_MAX 0xC8
> > PMBUS_MFR_VOUT_MIN 0x1748
> > PMBUS_MFR_VOUT_MAX 0x18B0
> > PMBUS_MFR_IOUT_MAX 0xF
> > PMBUS_MFR_POUT_MAX 0x96
> > PMBUS_MFR_MAX_TEMP_1 0xFFFF
> > PMBUS_MFR_MAX_TEMP_2 0xFFFF
> >
>
> The question is really what the status register reports for those.
> I bet that PMBUS_STATUS_CML will set the "invalid command" bit
> for many if not all registers returning 0xffff. Those will
> be filtered out by the PMBus core automatically. That leaves the
> ones returning some data. Of course it would be desirable
> to have those supported, but I can understand if you don't know
> the encoding. The reason for masking those needs to be explained.
>

I thought I sent the v4 patch before, but it seems that I only sent it
to myself. Also, I realized I've sent some of the emails only to you.
That wasn't my intention. Sorry. I'm bad at emailing.

Anyway, I have posted the v4 patch right now and it includes a comment
about the masking.

Václav

> Thanks,
> Guenter