On Thu, 01 Jan 2009, John David Anglin wrote:
> I was think of something like this (untested):
While trying to test this, I hit the following with Kyle's 2.6.28-rc7
parisc tree:
Debian GNU/Linux 5.0 mx3210 ttyS1
mx3210 login: Backtrace:
[<000000004014240c>] dequeue_entity+0x34/0x190
[<0000000040144154>] __set_se_shares+0x74/0xb8
[<0000000040146674>] tg_shares_up+0x214/0x2c0
[<000000004013e254>] walk_tg_tree+0xac/0x110
[<0000000040148ea8>] update_shares+0x78/0x90
[<0000000040149198>] try_to_wake_up+0x2d8/0x2e0
[<0000000040149234>] wake_up_process+0x24/0x38
[<000000004018962c>] softlockup_tick+0x234/0x250
[<000000004015fe68>] run_local_timers+0x38/0x48
[<000000004015ff48>] update_process_times+0x50/0xc8
[<000000004011d208>] timer_interrupt+0x208/0x210
[<0000000040189f30>] handle_IRQ_event+0x70/0xd0
[<000000004018a05c>] __do_IRQ+0xcc/0x280
[<000000004011d8d0>] do_cpu_irq_mask+0x1a8/0x298
[<0000000040104074>] intr_return+0x0/0x4
[<000000004011d208>] timer_interrupt+0x208/0x210
Bad Address (null pointer deref?): Code=15 regs=000000007f54dc30 (Addr=c000000053c33ffd)
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00001000000001100000000000001110 Tainted: G W
r00-03 000000000806000e 000000007f54daf0 000000004014240c 000000007f54db90
r04-07 00000000405b86a0 e840d00053c33f8d e840d00053c33f8d e840d00053c33f8d
r08-11 0000000000000010 0000000000000002 000000007f54da20 000000007ed15178
r12-15 0000000000001000 0000000000000400 0000000000000400 0000000000000001
r16-19 000000004276c1d0 00000000405baea0 0000000000001001 000000007fb1f740
r20-23 0000000000000000 0000000080080000 0000000000000000 0000000000000000
r24-27 0000000000000000 000000004013e254 e840d00053c33f8d 00000000405b86a0
r28-31 00000000081b0244 000000007f54db60 000000007f54dc30 000000007fb1f740
sr00-03 0000000003840000 0000000000000000 0000000000000000 0000000003840000
sr04-07 0000000000000000 0000000000000000 0000000000000000 0000000000000000
IASQ: 0000000000000000 0000000000000000 IAOQ: 000000004013e550 000000004013e554
IIR: 535c00e0 ISR: 000000002840d000 IOR: c000000053c33ffd
CPU: 2 CR30: 000000007f524000 CR31: 0000000040554000
ORIG_R28: d84156c5635688c0
IAOQ[0]: update_curr+0x28/0x150
IAOQ[1]: update_curr+0x2c/0x150
RP(r2): dequeue_entity+0x34/0x190
000000004013e528 <update_curr>:
4013e528: 08 03 02 41 copy r3,r1
4013e52c: 0f c2 12 c1 std rp,-10(sp)
4013e530: 08 1e 02 43 copy sp,r3
4013e534: 73 c1 01 48 std,ma r1,a0(sp)
4013e538: 70 66 00 30 std r6,18(r3)
4013e53c: 08 1a 02 46 copy r26,r6
4013e540: 0c 68 12 d0 std r8,8(r3)
4013e544: 70 67 00 20 std r7,10(r3)
4013e548: 70 65 00 40 std r5,20(r3)
4013e54c: 70 64 00 50 std r4,28(r3)
4013e550: 53 5c 00 e0 ldd 70(r26),ret0
Not sure what's going on with r26.
Config attached.
Dave
--
J. David Anglin dave.anglin@xxxxxxxxxxxxxx
National Research Council of Canada (613) 990-0752 (FAX: 952-6602)
Attachment:
config-mx3210.gz
Description: application/gunzip
