On 22.09.2017 09:38, Cornelia Huck wrote: > On Thu, 21 Sep 2017 15:50:28 -0300 > Eduardo Habkost <ehabk...@redhat.com> wrote: > >> On Tue, Sep 19, 2017 at 10:55:53AM +0200, Thomas Huth wrote: >>> Historically we've marked all devices as hotpluggable by default. However, >>> most devices are not hotpluggable, and you also need a HotplugHandler to >>> support these devices. So if the user tries to "device_add" or "device_del" >>> such a non-hotpluggable device during runtime, either nothing really usable >>> happens, or QEMU even crashes/aborts unexpectedly (see for example commit >>> 84ebd3e8c7d4fe955b - "Mark diag288 watchdog as non-hotpluggable"). >>> So let's change this dangerous default behaviour and mark the devices as >>> non-hotpluggable by default. Certain parent devices classes which are known >>> as hotpluggable (e.g. PCI, USB, etc.) are marked with "hotpluggable = true", >>> so that devices that are derived from these classes continue to work as >>> expected. >> >> These seem to be missing: >> * TYPE_CPU (or at least TYPE_X86_CPU and TYPE_S390_CPU) > > I think it would be better to set it for TYPE_CPU (and have > architectures override if needed).
Hmm, no, I think TYPE_CPU should stay non-notpluggable, and we should only do this for TYPE_X86_CPU, TYPE_S390_CPU and TYPE_SPAPR_CPU_CORE. Most CPU types are not hot-pluggable, so that should be the default. Having seen all those "device_add" crashes in the past weeks, I'm afraid that we'll run into weird problems again otherwise since people will keep forgetting to add "hotpluggable = false" for new CPU types that are not hotpluggable. Thomas
