Hi Salvatore, Greg,
On 6/6/21 2:34 PM, Salvatore Bonaccorso wrote:
Instead of doing a specific backport, maybe it is enough to pick
a46393c02c76 ("ACPI: probe ECDT before loading AML tables regardless
of module-level code flag") frst on 4.19.y and then the mentioned fix
b1c0330823fe ("ACPI: EC: Look for ECDT EC after calling
acpi_load_tables()").
Many thanks for looking into this. I cherry-picked
d737f333b211361b6e239fc753b84c3be2634aaa and
b1c0330823fe842dbb34641f1410f0afa51c29d3 on linux-4.19.y, they indeed
apply cleanly as Salvatore wrote. I also compiled and tested the kernel,
the battery is correctly displayed as discharging when unplugged, with a
7.5 hours remaining as expected.
Does it still make sense for me to resubmit my patch with the "commit
<full SHA1 hash> upstream." comment fixed? I would trust upstream more
than me making one commit look reasonable while missing the other commit
it was based on. Greg, what would you prefer?
Best regards,
Laurențiu
