On Thu, Mar 5, 2015 at 12:27 PM, Catalin Marinas <catalin.marinas@xxxxxxx> wrote: > On Thu, Mar 05, 2015 at 04:03:21PM +0800, Hanjun Guo wrote: >> On 2015/3/5 6:46, Rafael J. Wysocki wrote: >> > IMO, you really need to define phys_cpuid_t in a common place or people will >> > forget that it may be 64-bit, because they'll only be looking at their arch. >> >> Since x86 and ARM64 are using different types for phys_cpuid_t, we need to >> introduce something like following if define it in common place: >> >> in linux/acpi.h, >> >> #if defined(CONFIG_X86) || defined(CONFIG_IA64) >> typedef u32 phys_cpuid_t; >> #define PHYS_CPUID_INVALID (phys_cpuid_t)(-1) >> #else if defined(CONFIG_ARM64) >> typedef u64 phys_cpuid_t; >> #define PHYS_CPUID_INVALID INVALID_HWID >> #endif >> >> I think it's awful, did I miss something? Well, you can define the type and PHYS_CPUID_INVALID in the arch code and then do this in a common header: #ifndef PHYS_CPUID_INVALID typedef u32 phys_cpuid_t; #define PHYS_CPUID_INVALID (phys_cpuid_t)(-1) #endif That would allow you to avoid the need to duplicate the definitions where it is not necessary. > I also think that's awful. I'm rather in favour of a per-arch > phys_cpuid_t. OK, so what about the above? Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
