On Mon, Dec 30, 2024 at 01:10:09PM +0100, Borislav Petkov wrote: > On Mon, Dec 30, 2024 at 07:40:18PM +0800, Feng Tang wrote: > > Still we may need to honor the user setting, say if user specifically set > > "panic=XXX" in the cmdline, we should detect that case and skip overwritting > > the panic_timeout? > > So the user has set panic timeout to something, machine encounters a hw error > which you want to log/report earlier but user's panic setting prevents you > from doing that. > > So what do you do? > > What has higher prio? > > I guess we're something like this: > > https://youtu.be/gLFQystE8vU?t=71 Hah! As per kernel config, most ARCH has 'panic_timeout' as 0 by default, so need to set the kcmdline. For the case in my commit log, where user had clear requirement for not-reboot and wait, the manually set 'panic=0' should take priority here? Thanks, Feng > > :-P > > -- > Regards/Gruss, > Boris. > > https://people.kernel.org/tglx/notes-about-netiquette
