2.2.18pre17 again

[Gabriel Paubert]

On Tue, 31 Oct 2000, Benjamin Herrenschmidt wrote:

>
> >- ICTC support in the sysctl interface:
>
> Won't this require recalibrating the loops_per_jiffy for udelay() ?

Why isn't udelay based on the timebase then ? Easy to scale
microseconds to timebase.

There is a multiplier for mulhwu somewhere in a global variable, I should
remember the name since I wrote the code, tb_to_us if I'm not messing up.

It needs some care for 601 because of the different registers (RTC) that
has the bad habit of jumping from 1e9 to 0. It is trivial to implement as
a subroutine and check for CPU inside the subroutine. After all it's a
delay, so performance should not be critical.

udelay:
	lis	r4,tb_to_us at ha
	mfpvr	r5
	lwz	r4,tb_to_us at l(r4)
	srwi	r5,r5,16
	mulhwu	r4,r3,r4
	cmpwi	cr0,r5,1
	beq	2f
	mftbl	r5
1:	mftbl	r3
	sub	r3,r3,r5
	cmplw	cr0,r3,r4	# works for delays of up to almost 2^32 tb ticks
	blt	cr0,1b
	blr

2:	mfrtcl	r5
3:	mfrtcl	r3
	sub.	r3,r3,r5	# if wraparound has occurred
	bnl+	4f		# bump elapsed time by one second
	addis	r3,r3,1000000000 at ha
	addi	r3,r3,1000000000 at l
4:	cmplw	r3,r3,r4	# works for delays of up to almost 1 second
	blt	cr0,3b
	blr

Caution, untested... BTW, that's the kind of routine in which I'd like to
tell the compiler that it only uses r3/r4/r5/cr0 so that it can keep more
variables in the registers of the caller.

	Gabriel.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/