On Wed, 29 May 2019, Peter Zijlstra wrote: > > > I verified that it succesfully makes it to the point where restore_image() > > > is called from swsusp_arch_resume() (and verified that only BSP is alive > > > at that time), but the old kernel never comes back and triplefault-like > > > reboot happens. > > which means that even without nosmt all 'other' CPUs are offline. And > when I look at resume_target_kernel() I see it call > hibernate_resume_nonboot_cpu_disable(). > > So how is the SMT offline different from that offline? afaict they all > get into play_dead()->native_play_dead()->mwait_play_dead().
There is no way those other CPUs have been offlined before to the native_play_dead() state, as this is way before any userspace was alive to initiate any kind of hotplug. So they are guaranteed to have been all online, and then offlined properly to resume_play_dead(). 'nosmt' is the only exception there, as it's the only kind of offlining that has already happened at this point. Let's continue in the other thread. Thanks, -- Jiri Kosina SUSE Labs
