Message-ID: <SA1PR09MB7440DE6CE6882186ECDE3401E7BE9@SA1PR09MB7440.namprd09.prod.outlook.com>
Date:   Thu, 21 Oct 2021 05:37:18 +0000
From:   "Wilson, David T. (GSFC-5870)" <david.wilson@...a.gov>
To:     Guenter Roeck <linux@...ck-us.net>
CC:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-hwmon@...r.kernel.org" <linux-hwmon@...r.kernel.org>
Subject: Re: [EXTERNAL] Re: Potential issue with smb word operations for
 tmp461 device in tmp401 driver

Guenter,

I updated several files and applied the lm90.c and tmp401.c patches. It should be noted that my target platform only supports v5.4, so 
I had to update additional files (i.e. hwmon.c/.h) in order to successfully compile and apply the patch to lm90.c, which could suggest
that some of my findings may have been due to poor updating of files.

In regards to testing, I got temp1_input and temp2_input values that look good from a glance and that steadily increased after booting
my target platform. There were three areas of possible concern.
1. The lm90.c patch includes  "+#define TMP461_REG_DFC 0x23". If this define is referring to the address of the digital filter control, 
    then the value should be 0x24, since 0x23 is the n-Factor Correction register
2. The values I recorded had a resolution of 0.125 deg resolution (e.g., 30625, 25250, etc...). However, the datasheet for the tmp461
    notes that the resolution is 0.0625 deg for local and remote channels, so I was wondering if that would also be supported in lm90.c
3. For temp1_min and temp2_min, I used "cat temp1_min" to get the value -128000. Assuming that is -128.0 deg, I don't believe the
    tmp461 should be returning a value underneath -64 deg based on the temp data format if I'm understanding the datasheet correctly.

Thanks,
David

From: Guenter Roeck <groeck7@...il.com> on behalf of Guenter Roeck <linux@...ck-us.net>
Sent: Tuesday, October 19, 2021 10:18 AM
To: Wilson, David T. (GSFC-5870) <david.wilson@...a.gov>
Cc: linux-kernel@...r.kernel.org <linux-kernel@...r.kernel.org>; linux-hwmon@...r.kernel.org <linux-hwmon@...r.kernel.org>
Subject: Re: [EXTERNAL] Re: Potential issue with smb word operations for tmp461 device in tmp401 driver 
 
David,

On 10/18/21 10:42 PM, Wilson, David T. (GSFC-5870) wrote:
> Guenter,
> 
> I've tried testing the patch by checking out the most recent commit's version for tmp401.c and applying the patch. However, the temperature values seem to be to low <1000 for each read.
> 
> If I'm understanding the driver code correctly, I believe there's two places of interest that might explain the low values
> 1. In tmp401_update_device_reg16, the final assignment to data->temp[j][i] overrides the patch's assignment to data->temp[j][i] in the tmp461 branch
> 2. In SENSOR_DEVICE_ATTR_2_RW, the temp2_offset is at (6, 1), but the newly added LSB array does not contain the LSB address at (6, 1)
> 
Thanks a lot for the information.

> Regarding your most recent email, I will try the lm90 by changing the driver tree entry to "ti,tmp451" and I'll let you know if that works for me.
> 
I just sent two patches, one to remove tmp461 support from the tmp401 driver, and one to add
tmp461 support to the lm90 driver. Turns out tmp461 is almost identical to tmp451, which was
already supported by the lm90 driver. Adding support for tmp461 to the lm90 driver makes
much more sense than trying to fix the tmp401 driver.

I module tested the lm90 driver for both tmp451 and tmp461 (I was able to find
register dumps for both chips), so I am reasonably sure that it works.

Thanks,
Guenter

> Thanks,
> David
> 
> From: Guenter Roeck <groeck7@...il.com> on behalf of Guenter Roeck <linux@...ck-us.net>
> Sent: Tuesday, October 19, 2021 12:56 AM
> To: Wilson, David T. (GSFC-5870) <david.wilson@...a.gov>
> Cc: linux-kernel@...r.kernel.org <linux-kernel@...r.kernel.org>; linux-hwmon@...r.kernel.org <linux-hwmon@...r.kernel.org>
> Subject: [EXTERNAL] Re: Potential issue with smb word operations for tmp461 device in tmp401 driver
>   
> David,
> 
> On 10/15/21 10:43 AM, Wilson, David T. (GSFC-5870) wrote:
>> Hi,
>>
>> I am reporting what I believe is a potential issue in the tmp401 driver for the tmp461 device specifically. I am new to reporting issues, so I apologize in advance if I've provided insufficient information for an issue report.
>>
>> The problem I'm encountering is that when I use the tmp401 linux driver to read temperature values from the tmp461, all of the read temperature values end with 996 (e.g. 33996, 38996, etc...).
>>
>> Looking further into the tmp401 commit messages, I see that the driver was changed to use smb word operations instead of separate byte operations. Although the other supported devices (i.e. tmp432, etc...) are noted to support 16-bit read operations in their respective datasheets, I see no indications of 16-bit read support in the tmp461 datasheet, which is supported by my inquiry in the TI forums (https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fe2e.ti.com%2Fsupport%2Fsensors-group%2Fsensors%2Ff%2Fsensors-forum%2F1044935%2Ftmp461-linux-driver-support-and-16-bit-temperature-register-reads&amp;data=04%7C01%7Cdavid.wilson%40nasa.gov%7Ce801cf6c7ac146b0e90908d9930b4794%7C7005d45845be48ae8140d43da96dd17b%7C0%7C0%7C637702498923858681%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=gGK9KbA0TWUJ4r3We%2BG9f5LOY7m%2BM5FkYuxhnEZFifY%3D&amp;reserved=0).
>>
>> Reverting the driver to the commit before the smb word change, I am then able to read temperature values that do not end only with 996. As a result, I believe that the tmp461 support may be partially broken by the switch to smb word operations.
>>
> 
> can you try to instantiate the lm90 driver (instead of the tmp401 driver)
> and let me know if it works for you ? If your system uses devicetree,
> you might have to select "ti,tmp451" instead of "ti,tmp461".
> 
> Thanks,
> Guenter
> 

Powered by blists - more mailing lists