On Mon, Dec 12, 2011 at 06:13:11AM +0100, Andi Kleen wrote:
> > It's ~180 bytes, so it's not really that small.
> Quite small compared to what real code uses. And also fixed
> > > is on the new stack. ISTs are not used for interrupts, only for
> > > some special exceptions.
> > IST = ???
> That's a hardware mechanism on x86-64 to switch stacks
> (Interrupt Stack Table or somesuch)
> With ISTs it would have been possible to move the the pt_regs too,
> but the software mechanism is somewhat simpler.
> > at the top of the stack frame? Is the stack unwinder walking back
> > across the interrupt stack to the previous task stack?
> Yes, the unwinder knows about all the extra stacks (interrupt
> and exception stacks) and crosses them as needed.
> BTW I suppose it wouldn't be all that hard to add more stacks and
> switch to them too, similar to what the 32bit do_IRQ does.
> Perhaps XFS could just allocate its own stack per thread
> (or maybe only if it detects some specific configuration that
> is known to need much stack)
That's possible, but rather complex, I think.
> It would need to be per thread if you could sleep inside them.
Yes, we'd need to sleep, do IO, possibly operate within a
transaction context, etc, and a workqueue handles all these cases
without having to do anything special. Splitting the stack at a
logical point is probably better, such as this patch: