Re: [PATCH 2/5] bitops: compile time optimization for hweight_long(CONSTANT)

["H. Peter Anvin" <hpa@zytor.com>]

On 02/12/2010 09:06 AM, Borislav Petkov wrote:
> On Thu, Feb 11, 2010 at 09:33:49AM -0800, H. Peter Anvin wrote:
>> You don't do the push/pop inline -- if you're going to take the hit of
>> pushing this into the caller, it's better to list them as explicit
>> clobbers and let the compiler figure out how to do it.  The point of
>> doing an explicit push/pop is that it can be pushed into the out-of-line
>> subroutine.
> 
> Doh! Of course, this was wrong. See below.
> 
> diff --git a/arch/x86/lib/hweight.c b/arch/x86/lib/hweight.c
> new file mode 100644
> index 0000000..9b4bfa5
> --- /dev/null
> +++ b/arch/x86/lib/hweight.c
> @@ -0,0 +1,72 @@
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/bitops.h>
> +
> +#ifdef CONFIG_64BIT
> +/* popcnt %rdi, %rax */
> +#define POPCNT ".byte 0xf3\n\t.byte 0x48\n\t.byte 0x0f\n\t.byte 0xb8\n\t.byte 0xc7"
> +#define REG_IN "D"
> +#define REG_OUT "a"
> +#else
> +/* popcnt %eax, %eax */
> +#define POPCNT ".byte 0xf3\n\t.byte 0x0f\n\t.byte 0xb8\n\t.byte 0xc0"
> +#define REG_IN "a"
> +#define REG_OUT "a"
> +#endif
> +
> +/*
> + * Those are called out-of-line in the alternative below and are added here only
> + * so that gcc is able to figure out which registers have been clobbered by
> + * __sw_hweightXX so that it could restore their values before returning from
> + * the __arch_hweightXX versions. See also <arch/x86/lib/Makefile>.
> + */
> +unsigned int _sw_hweight32(unsigned int w)
> +{
> +	return __sw_hweight32(w);
> +}
> +
> +unsigned long _sw_hweight64(__u64 w)
> +{
> +	return __sw_hweight64(w);
> +}
> +
> +unsigned int __arch_hweight32(unsigned int w)
> +{
> +	unsigned int res = 0;
> +
> +	asm (ALTERNATIVE("call _sw_hweight32", POPCNT, X86_FEATURE_POPCNT)
> +		     : "="REG_OUT (res)
> +		     : REG_IN (w));
> +

Now you have no clobbers and no assembly wrapper.  That really doesn't work.

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.