Am 23.12.2013 13:59, schrieb Peter Maydell:
> On 23 December 2013 12:50, Paolo Bonzini <[email protected]> wrote:
>> Il 23/12/2013 13:37, Peter Maydell ha scritto:
>>> At a minimum, if we take this approach we should add TODO comments
>>> to the effect that the NULL terminator and the if() can be removed
>>> when the first real AArch64 CPU is added.
>>>
>>> I think I'd rather put the if (!info->name) continue into the function
>>> which is doing the looping over the array.
>> Or just change the termination condition from a check on the array size
>> to one on info->name.
> That would take it out of line with the equivalent 32 bit ARM code
> (and also moxie and openrisc for what little that's worth) and be
> fractionally more tedious to revert later.
>
> thanks
> -- PMM
What about adding a dummy CPU (which can be removed later)?
I also got a warning here when I used cgcc / smatch recently, but did
not send a patch because there are too many possible fixes and none of
them seemed to be elegant :-)
Nevertheless, here is one of them:
--- a/target-arm/cpu64.c
+++ b/target-arm/cpu64.c
@@ -46,6 +46,11 @@ static void aarch64_any_initfn(Object *obj)
set_feature(&cpu->env, ARM_FEATURE_V7MP);
set_feature(&cpu->env, ARM_FEATURE_AARCH64);
}
+#else
+static void aarch64_dummy_initfn(Object *obj)
+{
+ hw_error("Dummy CPU not supported");
+}
#endif
typedef struct ARMCPUInfo {
@@ -57,6 +62,9 @@ typedef struct ARMCPUInfo {
static const ARMCPUInfo aarch64_cpus[] = {
#ifdef CONFIG_USER_ONLY
{ .name = "any", .initfn = aarch64_any_initfn },
+#else
+ /* TODO: Dummy CPU to avoid empty array. Fix when a real CPU is
added. */
+ { .name = "dummy", .initfn = aarch64_dummy_initfn },
#endif
};
Of course, any other fix is also okay.
Cheers and Merry Christmas
Stefan
