Re: benh 2.4.6 kernel

[gtgj@pacbell.net (Gregorio Gervasio Jr.)]

>>>>> On Tue, 10 Jul 2001 18:37:41 -0500, Adam Lazur <adam@lazur.org> said:

[ TiPB sleep problem with 2.4.6 ]

a> cpu 0: vector: 300 at pc = c0007894, lr = c0007894, msr = b032, sp = c76d9d90 [c76d9ce0]
a> dar = 358, dsisr = 40000000
a> current = c76d8000, pid = 283, comm = pmud
a> 0:mon>

(c007894 is in "msr_vec_debug".)

        I get the same trace.  As Olaf Hering suggested, it's failing
when "giveup_altivec" is called.  It looks like a processor version
check was recently added to "giveup_altivec" (in
arch/ppc/kernel/head.S) but it seems to be incorrect:

	.globl	giveup_altivec
giveup_altivec:
	mfpvr	r24			/* check if we are on a G4 */
	srwi	r24,r24,16
	cmpwi	r24,PVR_7400@h
	beq	3f			/* continue */
	mflr	r24
	bl	msr_vec_debug		/* debug thingy in process.c */
	mtlr	r24
	blr
3:
...

I think PVR_7400 is the unshifted value?  (Also, the TiPB uses a 7410,
so what is the right comparison?)  Anyway, without this check, sleep
seems to be fine in both console and X11 mode.

Gregorio Gervasio, Jr.
gtgj@pacbell.net