RE: [PATCH v9 14/14] powerpc: rewrite local_t using soft_irq

From: Nicholas Piggin

Sent: 04 August 2017 10:04
On Fri, 04 Aug 2017 11:40:43 +1000
Benjamin Herrenschmidt [off-list ref] wrote:
=20

quoted

On Fri, 2017-08-04 at 03:50 +1000, Nicholas Piggin wrote:

quoted

Hey, so... why are any of these implemented in asm? We should
just do them all in C, right? I looked a bit harder at code gen
and a couple of them are still emitting larx/stcx.

As long as we can guarantee that the C compiler won't play games
moving stuff around. But yes, I tend to agree.

=20
=20
I believe so. I mean we already depend on the same pattern for any
other sequence of local_irq_disable(); c code; local_irq_enable();
so we'd have other problems if we couldn't.

I'd guess that a "memory" clobber on the irq_disable/enable would be enough=
.
It could be restricted to the memory area being updated.

	David