Skip to contentSkip to conversation

" Change size of node ids from u8 to u16 fixup" causes early panic in __build_all_zonelists

DISCUSSION Andi Kleen 3 participants 3 messages Jan 19, 2008

lore
Share & tools
lore

Andi Kleen

author
#0 lore raw

One of my test systems didn't boot with latest git-x86. I bisected it down to f1321f875910172bcc3e1f302fe145a9e4d3bdf7

With later patches the fault seemed to happen even earlier before other initialization messages.

Config is available at http://halobates.de/config

-Andi

commit f1321f875910172bcc3e1f302fe145a9e4d3bdf7

Author: [email protected] <[email protected]>
Date:   Fri Jan 18 23:05:33 2008 +0100

x86: Change size of node ids from u8 to u16 fixup

Change the size of node ids for X86_64 from 8 bits to 16 bits to accomodate more than 256 nodes.

Introduce a "numanode_t" type for x86-generic usage.
...

swsusp: Registered nosave memory region: 00000000ff700000 - 0000000100000000 Allocating PCI resources starting at e2000000 (gap: e0000000:1ec00000)
SMP: Allowing 4 CPUs, 0 hotplug CPUs
PERCPU: Allocating 34912 bytes of per cpu data
PANIC: early exception 0e rip 10:ffffffff802602cd error 0 cr2 6b0
Pid: 0, comm: swapper Not tainted 2.6.24-rc8-g3c2d7552 #27

Call Trace:
 [<ffffffff802602cd>] __build_all_zonelists+0x2a9/0x40a
 [<ffffffff80260110>] __build_all_zonelists+0xec/0x40a
 [<ffffffff802605ce>] build_all_zonelists+0x1a0/0x244
 [<ffffffff80811950>] start_kernel+0x110/0x2bd
 [<ffffffff808111c6>] _sinittext+0x1c6/0x1cd

RIP 0x10

#1 lore raw
Quoted 7 lines · Andi Kleen

One of my test systems didn't boot with latest git-x86. I bisected it down to f1321f875910172bcc3e1f302fe145a9e4d3bdf7

With later patches the fault seemed to happen even earlier before other initialization messages.

Config is available at http://halobates.de/config

i've reverted that patch for now, but we need to figure this crash out - the patch looks a trivial extension so it's probably some bad assumption elsewhere, which we want to fix.

Ingo

#2 lore raw
Quoted 8 lines · Ingo Molnar
Quoted 6 lines · Andi Kleen

One of my test systems didn't boot with latest git-x86. I bisected it down to f1321f875910172bcc3e1f302fe145a9e4d3bdf7

With later patches the fault seemed to happen even earlier before other initialization messages.

Quoted 7 lines
Quoted 1 line

Config is available at http://halobates.de/config

i've reverted that patch for now, but we need to figure this crash out - the patch looks a trivial extension so it's probably some bad assumption elsewhere, which we want to fix.

Ingo

I'm testing the new version of the patch using:

git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git mm

The straight mm version doesn't like your .config file.