Re: [PATCH 1/3] mull'ify multiplication with HZ in __const_udelay() [Was: Re: Too much error in __const_udelay() ?]

[john stultz <johnstul@us.ibm.com>]

On Mon, 2004-06-07 at 14:20, Dominik Brodowski wrote:
> Move the multiplication of (loops_per_jiffy * xloops) with HZ into
> the "mull" asm operation. This increases the accuracy of the delay functions
> largely:
> 
[snip]
> diff -ruN linux-original/arch/i386/lib/delay.c linux/arch/i386/lib/delay.c
> --- linux-original/arch/i386/lib/delay.c	2004-06-07 22:01:46.608351088 +0200
> +++ linux/arch/i386/lib/delay.c	2004-06-07 22:05:03.299449496 +0200
> @@ -33,8 +33,8 @@
>  	int d0;
>  	__asm__("mull %0"
>  		:"=d" (xloops), "=&a" (d0)
> -		:"1" (xloops),"0" (current_cpu_data.loops_per_jiffy));
> -        __delay(xloops * HZ);
> +		:"1" (xloops),"0" (current_cpu_data.loops_per_jiffy * HZ));
> +        __delay(xloops);
>  }

Kurt Garloff brought up a good point that loops_per_jiffy*HZ is only
good up to 4Ghz time sources. The workaround he suggested was to
multiply xloops by 4 first and divide HZ by 4. This will allow for
frequencies up to 16Ghz. 

So something like:

	xloops *= 4;
	__asm__("mull %0"
		:"=d" (xloops), "=&a" (d0)
		:"1" (xloops),"0" (LPJ*(HZ/4)));

-john