Bug#1012440: linux-image-amd64: Long time to load ACPI module battery.ko
Package: src:linux
Version: 5.10.113-1
Followup-For: Bug #1012440
X-Debbugs-Cc: macgyver0@gmail.com
Wow, thanks for the very fast answer!
> This sounds like a BIOS problem to me. Have you checked whether a new one is available?
The first thing I did after buying the laptop was updating the BIOS to the last version, just before wiping the original Windows installation and I was lucky since from what I found so far seems that Lenovo does not give other possibilities for BIOS update for my model! (I don't know how will I do for the next BIOS version that will be released in the future)
I forgot to mention that apart the long time to load, then the battery ACPI support seems to work properly, reporting all the informations (charging/dischargin rate, last full cacpcity, desing capacity etc..)
> To see/verify whether it may be a kernel issue, it's useful to test other kernel versions. If it works properly with another kernel version, then it's mostly likely a kernel issue, but if it's consistently slow across kernel versions, then a HW/BIOS problem seems most likely.
I just did the tests you suggested measuring precisely with /usr/bin/time (with the time function built in bash I was not able to redirect the output) and the results are very similar:
binary deb packages:
5.10.0-10-amd64 0.00user 0.00system 0:07.57elapsed 0%CPU
5.10.0-14-amd64 ~8.55s (was builtin, not module)
5.16.0-0.bpo.4-amd64 0.00user 0.00system 0:07.74elapsed 0%CPU
kernel source built by me:
5.18.2-mrf0 0.00user 0.00system 0:07.48elapsed 0%CPU
If you need other tests I can carry on.
Don't know if it's related, but I get lots of error lines during kernel boot like these:
...
[ 0.248050] ACPI BIOS Error (bug): Failure creating named object [\_SB.PCI0.T
XHC.RHUB.SS01._UPC], AE_ALREADY_EXISTS (20210730/dswload2-327)
[ 0.248057] ACPI Error: AE_ALREADY_EXISTS, During name lookup/catalog (202107
30/psobject-221)
[ 0.248060] ACPI: Skipping parse of AML opcode: OpcodeName unavailable (0x001
4)
...
I tried various acpi_osi=... without any improvement. From what I understand it should not be a dramatic problem and seems not to affect the system so far.
Regards,
Luca
Reply to: