| Message ID | 1418956884-24022-6-git-send-email-ehabkost@redhat.com |
|---|---|
| State | New |
| Headers | show |
On 19/12/2014 03:41, Eduardo Habkost wrote: > + object_property_set_int(OBJECT(cpu), CPU(cpu)->cpu_index, "apic-id", > + &error); > + if (error) { > + goto out; > + } > + Should this use &error_abort? Paolo
On Fri, Dec 19, 2014 at 12:22:30PM +0100, Paolo Bonzini wrote: > On 19/12/2014 03:41, Eduardo Habkost wrote: > > + object_property_set_int(OBJECT(cpu), CPU(cpu)->cpu_index, "apic-id", > > + &error); > > + if (error) { > > + goto out; > > + } > > + > > Should this use &error_abort? I normally use &error_abort only if there is no way to indicate errors to the caller at all. In this case, we print the error message and return NULL. However, I was assuming that all callers of cpu_init() check for NULL properly, and cpu_copy() at linux-user/main.c doesn't. As I don't want to touch linux-user code or change the cpu_init() signature, I will use &error_abort in the next version.
diff --git a/target-i386/cpu.c b/target-i386/cpu.c index bb9525d..4b0e0a5 100644 --- a/target-i386/cpu.c +++ b/target-i386/cpu.c @@ -2161,6 +2161,12 @@ CPUX86State *cpu_x86_init_user(const char *cpu_model) goto out; } + object_property_set_int(OBJECT(cpu), CPU(cpu)->cpu_index, "apic-id", + &error); + if (error) { + goto out; + } + object_property_set_bool(OBJECT(cpu), true, "realized", &error); out:
The PC CPU initialization code already sets apic-id based on the CPU topology, and CONFIG_USER doesn't need the topology-based APIC ID calculation code. Make CONFIG_USER set apic-id before realizing the CPU (just like PC already does), so we can simplify x86_cpu_initfn later. As there is no CPU topology configuration in CONFIG_USER, just use cpu_index as the APIC ID. Signed-off-by: Eduardo Habkost <ehabkost@redhat.com> Cc: Gu Zheng <guz.fnst@cn.fujitsu.com> --- target-i386/cpu.c | 6 ++++++ 1 file changed, 6 insertions(+)