Re: [PATCH] speed up on find_first_bit for i386 (let compiler do the work)

[Richard Kennedy <richard@rsk.demon.co.uk>]

Hi,
FWIW the following routine is consistently slightly faster using
Steven's test harness , with a big win when no bit set.

static inline int new_find_first_bit(const unsigned long *b, unsigned
size)
{
	int x = 0;
	do {
		unsigned long v = *b++;
	  	if (v)
			return __ffs(v) + x;
		if (x >= size)
			break;
		x += 32;
	} while (1);
	return x;
}

Tested on P III M 933MHz / gcc 4.0.1

clock speed = 00000000:17c56980 398813568 ticks per second

no bit set
ffb=320  my=320 new=320
generic ffb: 00000000:02fd6660
time: 0.125776182us
my ffb: 00000000:03c314e9
time: 0.158260714us
new ffb : 00000000:02d9190b
time: 0.119810758us

last bit set
ffb=319  my=319 new=319 
generic ffb: 00000000:04e5900c
time: 0.205994717us
my ffb: 00000000:0327475d
time: 0.132658024us
new ffb: 00000000:02c86938
time: 0.117068655us

middle bit set
ffb=159  my=159 new=159
generic ffb: 00000000:03c2bc56
time: 0.158203865us
my ffb: 00000000:01356b8b
time: 0.050846204us
new ffb: 00000000:0115f133
time: 0.045673521us

first bit set
ffb=0  my=0 new=0 
generic ffb: 00000000:02d07460
time: 0.118390436us
my ffb: 00000000:005d3079
time: 0.015313564us
new ffb: 00000000:005cca07
time: 0.015247804us

Cheers
Richard
Not subscribed please CC -- thanks.


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