On 3/3/23 16:30, Thorsten Leemhuis wrote:
On 03.03.23 08:10, Linux regression tracking (Thorsten Leemhuis) wrote:
Hi, this is your Linux kernel regression tracker.
I noticed a regression report in bugzilla.kernel.org that apparently affects 6.2 and later as well as 6.1.13 and later, as it was already backported there.
As many (most?) kernel developer don't keep an eye on bugzilla, I decided to forward the report by mail. Quoting from https://bugzilla.kernel.org/show_bug.cgi?id=217114 :
emmi@emmixis.net 2023-03-02 11:25:00 UTC
As per kernel problem found in https://bbs.archlinux.org/viewtopic.php?id=283906 ,
Commit 104ff59af73aba524e57ae0fef70121643ff270e
[FWIW: That's "ata: ahci: Add Tiger Lake UP{3,4} AHCI controller" from Simon Gaiser]
BTW, there is one thing I wondered after sending above mail: was it really wise to merge this to mainline two days before 6.2 was released? Yes, the change subject's makes it sounds like this is a hardware enablement, but the `Mark the Tiger Lake UP{3,4} AHCI controller as "low_power"` at the beginning of the change description shines a different light on it.
Yes, I made the decision to send this patch as a "fix" rather than a change, and that was rc8. In retrospect, maybe not the best decision. But the patch was fixing issues for Simon, so...
Anyway, will follow this. I requested more information on Bugzilla. The issue here is that it may be due to the device having a bad LPM support (there are many) rather than the controller itself. Need to sort this out.