Re: x86: A fast way to check capabilities of the current cpu

[Christoph Lameter]

On Fri, 21 Jan 2011, Tejun Heo wrote:

> On Fri, Jan 21, 2011 at 11:57:09AM -0600, Christoph Lameter wrote:
> > > Hmm?  asm/percpu.h gets included by linux/percpu.h, so it would end up
> > > in most .c files.
> >
> > The function is defined in arch/x86/include/asm/percpu.h. At least my
> > patches here are that way.
> >
> > #include "asm/percpu.h" will get you include/asm-generic/percpu.h on
> > other arches.
>
> Yeah, I was referring to generic code having visibility to the x86
> specific this_cpu op which isn't available on other archs.  I don't
> feel too strong about it so if you don't like it, just add a comment.

The code segments in generic code that use this function are in #ifdef
CONFIG_X86 sections.

Where do you want me to add a comment?

In include/x86/include/asm/percpu.h saying that the operations defined
there are x86 specific? Or in the #ifdef CONFIG_X86 sections saying that
the this_cpu_has operations are x86 specific?



Code snippet from drivers/acpi/processor_throttling.c

...

#ifdef CONFIG_X86
static int acpi_throttling_rdmsr(u64 *value)
{
        u64 msr_high, msr_low;
        u64 msr = 0;
        int ret = -1;

        if ((this_cpu_read(cpu_info.x86_vendor) != X86_VENDOR_INTEL) ||
                !this_cpu_has(X86_FEATURE_ACPI)) {
                printk(KERN_ERR PREFIX
                        "HARDWARE addr space,NOT supported yet\n");
        } else {
                msr_low = 0;
                msr_high = 0;
                rdmsr_safe(MSR_IA32_THERM_CONTROL,
                        (u32 *)&msr_low , (u32 *) &msr_high);
                msr = (msr_high << 32) | msr_low;
                *value = (u64) msr;
                ret = 0;
        }
        return ret;
}


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/