Re: [PATCH] Lazy interrupt disabling for 64-bit machines

On Tue, 2006-09-26 at 14:55 +1000, Paul Mackerras wrote:

This implements a lazy strategy for disabling interrupts.  This means
that local_irq_disable() et al. just clear the 'interrupts are
enabled' flag in the paca.  If an interrupt comes along, the interrupt
entry code notices that interrupts are supposed to be disabled, and
clears the EE bit in SRR1, clears the 'interrupts are hard-enabled'
flag in the paca, and returns.  This means that interrupts only
actually get disabled in the processor when an interrupt comes along.

I saw this once testing kdump, although I haven't hit it again
(3 retries).

Badness in _local_bh_enable at /home/michael/src/auto/git/kernel/softirq.c:120
Call Trace:
[C000000009F83B60] [C00000000200F810] .show_stack+0x6c/0x1a0 (unreliable)
[C000000009F83C00] [C0000000020219F0] .program_check_exception+0x19c/0x5d0
[C000000009F83CA0] [C0000000020043D0] program_check_common+0xd0/0x100

--- Exception: 700 at ._local_bh_enable+0x34/0x78
    LR = .call_do_softirq+0x14/0x24

[C0000000028FF8B0] [C00000000200C4A8] .do_softirq+0x6c/0xb0
[C0000000028FF940] [C00000000204E550] .irq_exit+0x4c/0x64
[C0000000028FF9C0] [C00000000201FC94] .timer_interrupt+0x3c0/0x3fc
[C0000000028FFA90] [C00000000200354C] decrementer_common+0xcc/0x100

--- Exception: 901 at .__delay+0x8/0x24
    LR = .__cpu_up+0x118/0x214

[C0000000028FFD80] [C000000002024C80] .__cpu_up+0xfc/0x214 (unreliable)
[C0000000028FFE10] [C00000000206987C] .cpu_up+0xd4/0x164
[C0000000028FFEC0] [C0000000020093A0] .init+0xd4/0x39c
[C0000000028FFF90] [C000000002022F0C] .kernel_thread+0x4c/0x68


Looks like it's hitting the in_irq() warning:

void _local_bh_enable(void)
{
        WARN_ON_ONCE(in_irq());
        WARN_ON_ONCE(!irqs_disabled());
	...

cheers

-- 
Michael Ellerman
OzLabs, IBM Australia Development Lab

wwweb: http://michael.ellerman.id.au
phone: +61 2 6212 1183 (tie line 70 21183)

We do not inherit the earth from our ancestors,
we borrow it from our children. - S.M.A.R.T Person