On 07/31/13 17:20, Timur Tabi wrote:

On 07/31/2013 07:16 PM, Stephen Boyd wrote:

quoted

cpu_relax() is usually just a compiler barrier or an instruction hint to
the cpu that it should cool down because we're spinning in a tight loop.
It certainly shouldn't be calling into the  scheduler.

Ah yes, I remember now.  So it does seem that if we can fix the problem
of non-incrementing 'jiffies', then this macro can be used in interrupts.

That's encouraging. It looks like you introduced it to use in interrupt
context but then it got shot down[1]? I lost track in all the versions.

Of course, that assumes that spinning in interrupt context is a good
idea to begin with.  Maybe we shouldn't be encouraging it?

I read through the v5 discussion and it seems I'm about to walk through
some tall grass on the way to Cerulean City.

Andrew Morton, I choose you! Use your mind-power move to convince
everyone that having a macro for spinning on a register in interrupt
context is a good thing. At least it will be more obvious.

quoted

quoted

quoted

FYI, you might want to look at the code reviews for spin_event_timeout()
on the linuxppc-dev mailing list, back in March 2009.

Sure. Any pointers? Otherwise I'll go digging around the archives.

https://lists.ozlabs.org/pipermail/linuxppc-dev/2009-March/thread.html

[1] https://lists.ozlabs.org/pipermail/linuxppc-dev/2009-May/072521.html

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation