Re: [PATCH v2 1/2] iopoll: Call cpu_relax() in busy loops

[Geert Uytterhoeven]

Hi David,

On Thu, May 11, 2023 at 12:49 PM David Laight <David.Laight@xxxxxxxxxx> wrote:
> > * Geert Uytterhoeven <geert+renesas@xxxxxxxxx> [230510 13:23]:
> > > It is considered good practice to call cpu_relax() in busy loops, see
> > > Documentation/process/volatile-considered-harmful.rst.  This can not
> > > only lower CPU power consumption or yield to a hyperthreaded twin
> > > processor, but also allows an architecture to mitigate hardware issues
> > > (e.g. ARM Erratum 754327 for Cortex-A9 prior to r2p0) in the
> > > architecture-specific cpu_relax() implementation.
>
> Don't you also need to call cond_resched() (at least some times).
> Otherwise the process can't be pre-empted and a RT process
> that last ran on that cpu will never be scheduled.

According to [1], cond_resched() must be called at least once per few
tens of milliseconds.

read_poll_timeout() uses usleep_range(), which calls schedule_hrtimeout_range().
read_poll_timeout_atomic() should not be used with multi-ms timeouts anyway.

So I guess we're OK?

[1] https://elixir.bootlin.com/linux/latest/source/Documentation/RCU/Design/Requirements/Requirements.rst#L2348

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds