2007-02-05 19:29:50

by Lu, Yinghai

[permalink] [raw]
Subject: RE: [PATCH 2/2] x86_64 irq: Handle irqs pending in IRR during irq migration.

Eric,

How about let apic_hangle_pending_vector take the irq too?
I wonder if there any chance that you have two IRQ_MOVE_PENDING?


BYW, Andi, Do you want to try "reuse vector when do the irq-balance
between CPUS"?

If So, I will update the patch that you dropped some months ago.

YH



2007-02-05 20:00:39

by Eric W. Biederman

[permalink] [raw]
Subject: Re: [PATCH 2/2] x86_64 irq: Handle irqs pending in IRR during irq migration.

"Lu, Yinghai" <[email protected]> writes:

> Eric,
>
> How about let apic_hangle_pending_vector take the irq too?

We can't compute the vector by reading the hardware registers after
we have acknowledged the irq.

I hope that was the answer you were looking for I'm not quite certain
what you mean by take.

> I wonder if there any chance that you have two IRQ_MOVE_PENDING?

I'm not quite certain what you are asking on the face of it there
is one bit per irq so it is impossible.

I don't think we can get into any trouble we don't let a vector be
reused on a cpu until we know it isn't in irr. So even if things
come in highly out of order the should still function.

I also have seen irqs migrate on nearly every irq received.

Eric