On Mon, Jun 08, 2020 at 12:11:11PM +0100, Robin Murphy wrote:

> Again, 2 cycles. The overhead of a static key alone is at least 50% of that.
> And that's not even considering whether the change in code layout caused by
> doubling up the IRQ handler might affect I-cache or branch predictor
> behaviour, where a single miss stands to more than wipe out any perceived
> saving. And all in code that has at least one obvious inefficiency left on
> the table either way.

> This thread truly epitomises Knuth's "premature optimisation" quote... ;)

In fairness the main reason this driver is so heavily tuned already (and
has lead to some really nice improvements in the core) is that there are
a number of users hitting 100% CPU utilization driving SPI devices on
some of the older RPi hardware, IIRC around IIO type applications
mostly.  I do tend to agree that this particular optimization is a bit
marginal but there has been a lot of effort put into this.