Re: FreeBSD-15 kernel panic when the amdtemp device is in the kernel
Date: Sun, 03 Sep 2023 13:17:36 UTC
On Sat, 02 Sep 2023 18:02:03 +0200, Gary Jennejohn wrote: > > On Sat, 02 Sep 2023 15:36:36 +0200 > "Herbert J. Skuhra" <herbert@gojira.at> wrote: > > > On Fri, 01 Sep 2023 18:05:34 +0200, Gary Jennejohn wrote: > > > > > > On Fri, 1 Sep 2023 14:43:21 +0000 > > > Gary Jennejohn <garyj@gmx.de> wrote: > > > > > > > A git-bisect is probably required. > > > > > > > > > > I did a bisect and the result was commit > > > 9a7add6d01f3c5f7eba811e794cf860d2bce131d. > > > > > > However, that can't be correct because this commit was made on > > > Mon Jul 17 19:29:20 2023 and my FBSD-14 kernel from August 13th > > > boots successfully :( > > > > Commit date is August 19th, 2023(!): > > > > commit 9a7add6d01f3c5f7eba811e794cf860d2bce131d > > Author: Colin Percival > > AuthorDate: Mon Jul 17 19:29:20 2023 -0700 > > Commit: Colin Percival > > CommitDate: Sat Aug 19 22:04:56 2023 -0700 > > > > > > Reverting this commit seems to resolve the issue for me: > > > > FreeBSD 15.0-CURRENT amd64 1500000 #0 main-n265137-2ad756a6bbb3 > > > > $ git status > > On branch main > > Your branch is up to date with 'freebsd/main'. > > > > You are currently reverting commit 9a7add6d01f3. > > (all conflicts fixed: run "git revert --continue") > > (use "git revert --skip" to skip this patch) > > (use "git revert --abort" to cancel the revert operation) > > > > Changes to be committed: > > (use "git restore --staged <file>..." to unstage) > > modified: sys/kern/init_main.c > > > > # dmesg |egrep "(amdsmn|amdtemp)" > > amdsmn0: <AMD Family 17h System Management Network> on hostb0 > > amdtemp0: <AMD CPU On-Die Thermal Sensors> on hostb0 > > > > $ sysctl kern.conftxt |grep amdt > > device amdtemp > > > > Really? I did a git log and July 17 is what pops out for this commit. > > Ah, I see that git log doesn't show the commit date. > > So I guess that the git bisect really did find the commit which caused > all our problems. > > If reverting it fixes things then this requires some action from Colin > Percival. > > This would also explain why my FBSD-14 kernel from August 13 was > OK. Probably best to file a PR: https://bugs.freebsd.org/bugzilla/ -- Herbert