With 4.8.0-rc1 my machine always crashes trying to handle an interrupt on shutdown. Please reassign the bug to an appropriate context. Last known good: 4.7.0-ARCH-13902-gd491e80-dirty.
Created attachment 229281 [details] screenshot of the backtrace Please also have a look at bug 153361; similar and also related to 4.8.0-rc1.
Ok, let's see if reverting a certain commit fixes your issue and if it does let's try narrowing down where it's started your regression.
Created attachment 229501 [details] Revert Commit
Created attachment 229511 [details] dmesg of 4.8.0-rc1+revert-patch including backtrace Unfortunately the revert patch did not resolve the issue. When booting with the patch applied I had also discovered a backtrace on bootup as well as the usual backtrace on shutdown (see for the attachement). By the time I have already started bisecting the issue. However that is going to take a long time so that I will post intermediate results here. Please tell me when I can stop bisecting as that endeavour costs a lot of time and CPU power while I do already have other issues about kernel 4.8.0 or before which need bisecting (specifically something about various Wifi-drivers).
good: 4.7.0-ARCH-13902-gd491e80 bad: 4.8.0-rc1 bad: [g]1c88e19
Bisect this issue first completely and send me the back commit id. Otherwise a incomplete git bisect is not that useful, sorry. In addition if your also having issues with wifi drivers open new bug reports and bisect them too if possible or send me the dmesgs first as those need to be fixed too.
Can you try applying 65ea11ec6a82b1d44aba62b59e9eb20247e57c6e [x86/hweight: Don't clobber %rdi]. Not a perfect match but is close enough it deserves a try.
You mean check it out + apply the Revert Commit? (also bad: f7816ad[0f878dac...])
(In reply to Elmar Stellnberger from comment #8) > You mean check it out + apply the Revert Commit? (also bad: > f7816ad[0f878dac...]) You need commit 65ea11ec6a82b1d44aba62b59e9eb20247e57c6e, so either apply it to what you are building or if you can build the latest mainline, try that.
Mark, today we all have reason to celebrate! This bug as well as kernel bugs 153361, 117581 and freedesktop bug 93887 (https://bugs.freedesktop.org/show_bug.cgi?id=93887) have in deed been resolved with commit 65ea11ec6a82b1d44aba62b59e9eb20247e57c6e (Ville Syrjälä - x86/hweight: Don't clobber %rdi) - and these were just the issues reported by me - not to count all the others you have been talking about.
Good to hear this is fixed. Please open bug reports for your wi-fi drivers so I can take a look at the other regressions you are currently facing on your hardware.