mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Matthew Bettencourt <matthew.bettencourt@gmail.com>
To: Guenter Roeck <linux@roeck-us.net>
Cc: linux-hwmon@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [BUG] SPD5118 Intermittent MR11 Corruption During Suspend/Resume
Date: Wed, 2 Sep 2026 12:37:32 -0500	[thread overview]
Message-ID: <bb74ebc1-9ea6-4ab0-9443-d25f2f971e4e@gmail.com> (raw)
In-Reply-To: <56826d45-e18a-4f9e-94b0-3e6f335c4ee7@roeck-us.net>

Hello Guenter

 > I2C bus multiplexers only return a single data byte reflecting the
 > currently
 > selected bus. On writes they also accept a single byte, where a set bit
 > indicates
 > which of its channel(s) should be enabled. So i2cdump does not really 
work.
 > Actually, using i2cdump might have messed it up because the writes 
involved
 > might change the active bus... sorry, I should have mentioned before.
 > In general it is not safe to use i2cdump on an unknown device. Worst case
 > it can turn the hardware into a brick (seriously - I have seen that
 > happen).
 >
 > You could try "i2cget 12 0x71" and "i2cget 12 0x73". That should return
 > the active channel(s) if those are really I2C bus multiplexers. If they
 > are,
 > only a single bit should be set (again, i2cdump might have messed it up,
 > so you might want to reboot).

No worries, the system did not brick itself haha. I rebooted and 
executed the i2cget commands

# i2cget 12 0x71
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will read from device file /dev/i2c-12, chip address 0x71, current data
address, using read byte.
Continue? [Y/n] y
0x1f
# i2cget 12 0x73
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will read from device file /dev/i2c-12, chip address 0x73, current data
address, using read byte.
Continue? [Y/n] y
0x1f

 > so I don't have any of the AMDI0010 devices. Maybe that is the built-in
 > GPU ?

I thought I had disabled the igpu in the BIOS, but I might have reset 
the BIOS to factory settings at some point so it might have been become 
enabled again. I can check the BIOS settings when I get home.


 >> If it is worth while I have access to second AMD machine, PRO B850M-A
 >> WIFI/7600x3d/Teamgroup 2x16GB - FLBD532G6000HC30DC01, which I can swap
 >> parts from/to. Even though its a completely different system it might
 >> worth it to swap parts out and see then what comes up on either system
 >> to help identify what might be going on with the i2c bus devices.
 >>
 >
 > It might be worthwhile if that system does not suffer from the same
 > problem.
 >
 > Bus-wise I can see that you seem to have different I2C busses in your
 > system. I have

Yes I can do that. I will swap the RAM kits on both systems and see what 
happens, checking for the sleep/resume error message with the piix4 
transaction log.

Thanks,
Matthew

On 9/2/26 12:21 PM, Guenter Roeck wrote:
> Hi Matthew,
> 
> On 9/2/26 08:26, Matthew Bettencourt wrote:
>> Hello Guenter,
>>
>> Here is the i2cdump for those two 0x71 and 0x73 devices
>>
>> # i2cdump -y -f 12 0x73 b
>>       0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
>> 00: ff ff ff ff XX ff 12 ff 12 ff ff ff XX XX XX XX    ....X.?.?...XXXX
>> 10: XX 0f XX XX XX XX XX XX XX XX XX XX XX XX XX XX    X?XXXXXXXXXXXXXX
>> 20: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 30: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 40: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 50: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 60: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 70: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 80: 00 e1 f9 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> 90: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> a0: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> b0: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> c0: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> d0: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> e0: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> f0: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> # i2cdump -y -f 12 0x71 b
>>       0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
>> 00: ff ff ff ff XX ff 57 ff 57 ff ff ff XX XX XX XX    ....X.W.W...XXXX
>> 10: XX 0f XX XX XX XX XX XX XX XX XX XX XX XX XX XX    X?XXXXXXXXXXXXXX
>> 20: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 30: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 40: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 50: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 60: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 70: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX    XXXXXXXXXXXXXXXX
>> 80: 00 0b e0 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> 90: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> a0: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> b0: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> c0: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> d0: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>> e0: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f    .???????????????
>> f0: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f    ????????????????
>>
> 
> I2C bus multiplexers only return a single data byte reflecting the 
> currently
> selected bus. On writes they also accept a single byte, where a set bit 
> indicates
> which of its channel(s) should be enabled. So i2cdump does not really work.
> Actually, using i2cdump might have messed it up because the writes involved
> might change the active bus... sorry, I should have mentioned before.
> In general it is not safe to use i2cdump on an unknown device. Worst case
> it can turn the hardware into a brick (seriously - I have seen that 
> happen).
> 
> You could try "i2cget 12 0x71" and "i2cget 12 0x73". That should return
> the active channel(s) if those are really I2C bus multiplexers. If they 
> are,
> only a single bit should be set (again, i2cdump might have messed it up,
> so you might want to reboot).
> 
>> # ls -l /sys/bus/i2c/devices/
>> total 0
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 12-0051 -> ../../../devices/ 
>> pci0000:00/0000:00:14.0/i2c-12/12-0051
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 12-0053 -> ../../../devices/ 
>> pci0000:00/0000:00:14.0/i2c-12/12-0053
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-0 -> ../../../devices/ 
>> platform/AMDI0010:00/i2c-0
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-1 -> ../../../devices/ 
>> platform/AMDI0010:01/i2c-1
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-10 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/drm/ 
>> card1/card1-DP-2/i2c-10
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-11 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/drm/ 
>> card1/card1-DP-3/i2c-11
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-12 -> ../../../devices/ 
>> pci0000:00/0000:00:14.0/i2c-12
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-13 -> ../../../devices/ 
>> pci0000:00/0000:00:14.0/i2c-13
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-14 -> ../../../devices/ 
>> pci0000:00/0000:00:14.0/i2c-14
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-2 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-2
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-3 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-3
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-4 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-4
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-5 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-5
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-6 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-6
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-7 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-7
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-8 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-8
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-9 -> ../../../devices/ 
>> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/drm/ 
>> card1/card1-DP-1/i2c-9
>> lrwxrwxrwx. 1 root root 0 Sep  2 10:08 i2c-MSFT8000:00 -> ../../../ 
>> devices/platform/AMDI0010:01/i2c-1/i2c-MSFT8000:00
>>
>>
>> If it is worth while I have access to second AMD machine, PRO B850M-A 
>> WIFI/7600x3d/Teamgroup 2x16GB - FLBD532G6000HC30DC01, which I can swap 
>> parts from/to. Even though its a completely different system it might 
>> worth it to swap parts out and see then what comes up on either system 
>> to help identify what might be going on with the i2c bus devices.
>>
> 
> It might be worthwhile if that system does not suffer from the same 
> problem.
> 
> Bus-wise I can see that you seem to have different I2C busses in your 
> system. I have
> 
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 0-0050 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-0/0-0050
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 0-0051 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-0/0-0051
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 0-0052 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-0/0-0052
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 0-0053 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-0/0-0053
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-0 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-0
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-1 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-1
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-2 -> ../../../devices/ 
> pci0000:00/0000:00:14.0/i2c-2
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-3 -> ../../../devices/ 
> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-3
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-4 -> ../../../devices/ 
> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-4
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-5 -> ../../../devices/ 
> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-5
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-6 -> ../../../devices/ 
> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/i2c-6
> lrwxrwxrwx 1 root root 0 Aug 29 16:26 i2c-7 -> ../../../devices/ 
> pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.0/drm/ 
> card1/card1-DP-1/i2c-7
> 
> so I don't have any of the AMDI0010 devices. Maybe that is the built-in 
> GPU ?
> 
> Thanks,
> Guenter
> 


  reply	other threads:[~2026-09-02 17:37 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-30 23:21 Matthew Bettencourt
2026-08-31 15:12 ` Guenter Roeck
2026-08-31 17:28   ` Guenter Roeck
2026-08-31 23:36     ` Matthew Bettencourt
     [not found]     ` <37132c40-aa82-4df1-b5ce-337601f3c085@gmail.com>
     [not found]       ` <3c4e9f59-5272-48cf-a4a0-eeefcba955a9@roeck-us.net>
2026-09-01  0:16         ` Matthew Bettencourt
2026-09-01  3:10           ` Guenter Roeck
2026-09-01 16:31             ` Matthew Bettencourt
2026-09-01 17:04               ` Guenter Roeck
2026-09-01 20:53                 ` Matthew Bettencourt
2026-09-01 22:36                   ` Guenter Roeck
2026-09-02  0:57                     ` Matthew Bettencourt
2026-09-02  1:32                       ` Guenter Roeck
2026-09-02  2:14                         ` Matthew Bettencourt
2026-09-02  3:19                           ` Guenter Roeck
2026-09-02 14:00                             ` Matthew Bettencourt
2026-09-02 14:52                               ` Guenter Roeck
2026-09-02 15:26                                 ` Matthew Bettencourt
2026-09-02 17:21                                   ` Guenter Roeck
2026-09-02 17:37                                     ` Matthew Bettencourt [this message]
2026-09-02 16:57                               ` Armin Wolf
2026-09-02 17:08                                 ` Matthew Bettencourt
2026-09-02 17:22                                   ` Armin Wolf
2026-09-02 17:28                                     ` Matthew Bettencourt
2026-09-02 19:19                                       ` Armin Wolf
2026-09-02 19:25                                         ` Matthew Bettencourt
2026-09-02 19:31                                           ` Armin Wolf
2026-09-02 19:40                                             ` Matthew Bettencourt
2026-09-02 19:51                                               ` Armin Wolf
2026-09-03 18:01                                                 ` Matthew Bettencourt
2026-09-03 18:16                                                   ` Guenter Roeck
2026-09-03 18:25                                                     ` Matthew Bettencourt
2026-09-03 21:05                                                       ` Guenter Roeck
2026-09-03 22:47                                                         ` Armin Wolf
2026-09-04  0:24                                                           ` Matthew Bettencourt
2026-09-04 17:58                                                             ` Guenter Roeck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bb74ebc1-9ea6-4ab0-9443-d25f2f971e4e@gmail.com \
    --to=matthew.bettencourt@gmail.com \
    --cc=linux-hwmon@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@roeck-us.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®