On 08.03.2017 15:50, Eduardo Habkost wrote: > On Wed, Mar 08, 2017 at 03:31:01PM +0100, Peter Maydell wrote: >> On 3 March 2017 at 15:58, Eduardo Habkost <ehabk...@redhat.com> wrote: >>> On Tue, Jan 31, 2017 at 02:11:59PM +0100, Thomas Huth wrote: >>>> When running with KVM on POWER, we register some CPU types during >>>> the initialization function of the ppc64 KVM code (which unfortunately >>>> also can not be done via a type_init() like it is done on x86). >>> >>> Can you elaborate why it can't be done via type_init()? If the >>> QOM type hierarchy depends on any runtime data unavailable at >>> type_init(), we should fix that. >> >> On ARM we (currently) have the KVM-only 'host' CPU type be >> added to the type hierarchy only at runtime in kvm_init(), >> but we deal with the '-help' problem by having arm_cpu_list() do >> >> #ifdef CONFIG_KVM >> /* The 'host' CPU type is dynamically registered only if KVM is >> * enabled, so we have to special-case it here: >> */ >> (*cpu_fprintf)(f, " host (only available in KVM mode)\n"); >> #endif >> > > We already have similar code at ppc_cpu_list(): > > #ifdef CONFIG_KVM > cpu_fprintf(f, "\n"); > cpu_fprintf(f, "PowerPC %-16s\n", "host"); > #endif > > If I understood it correctly, the current problem is just > inaccurate alias information being printed because the alias > table is modified by the accelerator.
Yes. > My current suggestion is to avoid printing anything that depends > on the current machine/accelerator at the help output. I'll have a look at that... Thomas
