Re: [BUG] SPD5118 Intermittent MR11 Corruption During Suspend/Resume
From: Guenter Roeck
Date: Wed Sep 02 2026 - 11:29:40 EST
Hi Matthew,
On 9/2/26 07:00, Matthew Bettencourt wrote:
Hello Guenter,
> It would be something like
>
> sudo acpidump > acpi.log
> acpixtract acpi.log
> iasl -d *.dat
>
> You'll get a number of.dsl files with cryptic assembler-style
> code and data. Then the digging starts ... you may find something
> like "Name (SMB0, 0x0B00)" in dsdt.dsl. Then we can only hope to
> figure out if ACPI actually accesses it.
That wasn't as hard as I thought. I got everything decoding I guess, I just did a grep on all the dsl files for SMB.
$ grep -i -rn "SMB" *.dsl
That matches almost exactly the dsdt files on my main system (MSI MAG B650M
MORTAR WIFI with 9900X). Either I have the same problem (I don't usually
suspend) or we'll have to find something else to blame. I do see different
devices on the same bus, though.
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- -- -- --
10: -- -- -- -- -- 15 -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- 48 49 4a 4b -- -- -- 4f
50: UU UU UU UU -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --
[spd5118 is loaded, which explains the UU, and the system has
4 DIMMs]
Compared to yours:
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- -- -- --
10: -- -- -- -- -- 15 -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- 49 -- 4b -- -- -- --
50: -- 51 -- 53 -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- 71 -- 73 -- -- --
That does make me wonder what you have on 0x71 and 0x73,
because those are often I2C bus multiplexers. That could mess
up everything if some entity selects a different bus.
Guenter