Re: 2.6.17-rc4-mm2

From: Martin J. Bligh
Date: Sun May 21 2006 - 18:28:58 EST

Andrew Morton wrote:
"Martin J. Bligh" <mbligh@xxxxxxxxxx> wrote:

Panic on boot on 2-way PPC64

Bad page state in process 'idle'
page:c0000000010c3100 flags:0x0003300000000000 mapping:0000000000000000 mapcount:0 count:0
Trying to fix it up, but a reboot is needed
Call Trace:
[C0000000004CBB70] [C00000000000EEE8] .show_stack+0x74/0x1b4 (unreliable)
[C0000000004CBC20] [C000000000098D04] .bad_page+0x80/0x134
[C0000000004CBCB0] [C000000000099F28] .__free_pages_ok+0x134/0x280
[C0000000004CBD70] [C00000000039C4F8] .free_all_bootmem_core+0x15c/0x320
[C0000000004CBE50] [C0000000003923AC] .mem_init+0xc0/0x294
[C0000000004CBEF0] [C000000000385700] .start_kernel+0x208/0x300
[C0000000004CBF90] [C000000000008594] .start_here_common+0x88/0x8c

I dunno, Martin. I obviously need to resurrect the
takes-seven-minutes-to-boot pseries machine here, but I need to get -mm3

Don't kill yourself over it - the blade seems to do it, but the p650
(8x) doesn't, so it may not help.

out without Mel's patches to see how much that fixes. So if you could test
-mm3 and if that fixes it then we have a likely culprit.

Sure, no problem. I suppose I should really make this thing do automated
bisection chopsearch ... I had that half-working when at IBM, but it'll
have to wait for the next generation now ... shouldn't take all that long.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at