You are not logged in.
I've been having a weird issue with a couple of HDDs, where there is an exception on the ATA link and the SATA link being hard reset as the result - downgrading the connection from 6.0Gb/s to 3.0Gb/s and eventually to 1.5Gb/s.
The thing is, no such behavior on the same cable, same slot, same HDD on windows. I've found some threads, for example this one (that's the errors I was seeing too, just on a different port), some people stating it was fixed for them, others saying it's still an issue.
For me setting libata.noacpi=1 gets rid off the problem. The drives also don't seem to suffer from any side-effects, running at 6.0Gb/s, files are being written and read fine (checksums match for about ~500GB r/w) and no more exceptions in dmesg.
So what is this option doing exactly, and what are the implications? All I was able to find so far was
libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
when set.
Format: <int>
(from here), but that's not really telling me much.
[ Arch x86_64 | linux | Framework 13 | AMD Ryzen™ 5 7640U | 32GB RAM | KDE Plasma Wayland ]
Offline