https://bugzilla.kernel.org/show_bug.cgi?id=221587
--- Comment #9 from Dmytro Skarbinets ([email protected]) --- (In reply to Artem S. Tashkinov from comment #6) > Try to boot only with this flag > > idle=poll > > and see how your device behaves. IT WORKS! So, what it does is make the CPU never enter low-power state and instead always loop infintely until it gets a real task? Now that i think about it, when i just flashed 318 bios, i installed Fedora and noticed that the laptop is a lot quieter and colder than it was in windows when under no load. Might that be that windows somehow finds this issue and also decides to disable low-power states? Though, shouldn't cpuidle.off=1 do the same thing? Either way, looks like we got our solution. Now i guess i can see how is this a hardware issue, faulty/low-quality MOSFETs or something maybe? Maybe i should take a multimeter and poke around... Still doesn't explain why did it start working normally after flashing 318... Maybe it was a giant fluke? What if this is also one? Though, when i was reinstalling windows after flashing 318, i noticed that the laptop got noticeably hotter than it typically would during a reinstall... Also funnily enough, with idle=poll i stopped getting BadTLP fixable errors on nvidia card as often as i was, they were almost constant before, pcie_aspm=off fixed that, but not the crashes. Though i've gotten one barrage of these errors during boot, but that's it. Either way, now it works. I've seen many computers work with no load and crumble under load, but never one to crumble under no load and work under load... Guess there's first time for everything. Thanks a lot for your help :) -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug. _______________________________________________ acpi-bugzilla mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla
