linux-kernel - Re: [PATCH] hwmon: (pmbus/lm25066) Swap low/high current coefficients for LM5066(i)

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: <CAFRv3wzNoMeYymxTVw8w9-rWgHKGJUBZNXZWhWc8=uXK4D3_Jw@mail.gmail.com>
Date:   Mon, 27 Nov 2017 16:31:40 -0800
From:   Rob Lippert <rlippert@...gle.com>
To:     Guenter Roeck <linux@...ck-us.net>
Cc:     Robert Lippert <roblip@...il.com>, linux-hwmon@...r.kernel.org,
        jdelvare@...e.com, linux-kernel@...r.kernel.org,
        Xo Wang <xow@...gle.com>
Subject: Re: [PATCH] hwmon: (pmbus/lm25066) Swap low/high current coefficients
 for LM5066(i)

On Mon, Nov 27, 2017 at 2:08 PM, Guenter Roeck <linux@...ck-us.net> wrote:
> On Mon, Nov 27, 2017 at 01:51:35PM -0800, Rob Lippert wrote:
>> On Wed, Nov 22, 2017 at 5:17 PM, Guenter Roeck <linux@...ck-us.net> wrote:
>> > Hi Rob,
>> >
>> > On 11/22/2017 03:39 PM, Rob Lippert wrote:
>> >>
>> >> On Wed, Nov 22, 2017 at 3:28 PM, Guenter Roeck <linux@...ck-us.net> wrote:
>> >>>
>> >>>
>> >>> On Wed, Nov 22, 2017 at 02:07:28PM -0800, Robert Lippert wrote:
>> >>>>
>> >>>> The _L low-current mode coefficient values should reference the
>> >>>> datasheet rows with CL=VDD but it seems were mistakenly pulled from
>> >>>> the rows with CL=GND.
>> >>>>
>> >>>> This causes the current/power to be reported as approximately double
>> >>>> the actual value when CL=GND and half the actual value when CL=VDD.
>> >>>>
>> >>>
>> >>> This would affect all chips supported by this driver. Hmm, and I was sure
>> >>> I tested this. I'll have to dig out my hardware and confirm.
>> >>
>> >>
>> >> I'm still not 100% convinced this commit is correct as I haven't been
>> >> able to validate the measurements against an external probe yet (and
>> >> my test board uses a non-standard sense resistor which means it needs
>> >> additional massaging of the data anyhow).
>> >>
>> >>>
>> >>>
>> >>> The code currently only uses bit 4 of the DEVICE_SETUP (D9h) command
>> >>> to determine which current limit setting to use. Looking into the
>> >>> datasheet, it looks like it also has to evaluate bit 2, and I wonder
>> >>> if there is a means to determine CL if bit 2 = 0. Any idea ?
>> >>
>> >>
>> >> On my test board CL=floating (equivalent to GND) and the value of
>> >> register 0xD9 is all zeroes.
>> >>
>> > Are you sure that floating is equivalent to GND ? I didn't check the
>> > datasheet, but it is more common for chips to have an internal pull-up.
>> >
>> >>>
>> >>> Does bit 4 report the CL pin value if bit 2 = 0 ?
>> >>
>> >>
>> >> I can't tell by reading the datasheet that 0xD9 bit4 will ever report
>> >> the pin value as the language is difficult to parse :)
>> >
>> >
>> > Same here.
>> >
>> >> But I don't have any hardware setup with CL=VDD to test...
>> >>
>> >
>> > I do have various evaluation boards, so I should be able to do some testing.
>> > I hope I'll get to it over the weekend.
>>
>> Actually turns out my board does tie CL=VDD as recommended by the
>> LM5066i datasheet to improve the current/power reporting accuracy.
>> But the value in 0xD9 is always zero so it seems there is no way to
>> read this CL pin setting from the device.
>>
>> I think my commit is technically correct but it seems will likely
>> break the readings for most boards that follow the datasheet typical
>> circuit which recommends CL=VDD.
>>
>> Would it make sense to remove the code trying to read the CL setting
>> and default the coefficient values to the "low current limit" CL=VDD
>> setting?  (and maybe something in devtree or module param to pick the
>> other coefficients?)
>>
>
> The code should also check bit 2. If bit 2 is not set, it is ok to assume
> CL=VDD. We can add a devicetree property if/when needed. Question though
> is still how to handle the problem for the other chips supported by the
> driver; if we change the code we should fix the problem for all chips,
> not just for one. Wonder if we can just swap the defines to minimize code
> changes.

OK sent patch "hwmon: (pmbus/lm25066) Default coefficients for low
current limit" that I think does what was discussed.

I have not checked any of the other chip variant datasheet tables or
recommended default designs to see if they all match up though...

-Rob