INTR_FAST

[Ian Lepore]

On Thu, 2020-08-27 at 21:13 +0530, Kamal R. Prasad wrote:
> because i hsve 3 spi slave devices attached to the pl022 controller and the
> drain of rx queue after i initiare spi transfer is showing some
> inconsistency.
> 

I finally had a chance to do a quick test today... it looks like a
filter handler will get control 6-8 microseconds sooner on average than
a threaded handler would, on a 1ghz imx6 (32-bit cortex-a9) cpu. 
That's a bit longer than I expected (I was expecting a difference in
the 1-3us range).

-- Ian

> 
> On Thu, Aug 27, 2020 at 9:02 PM Ian Lepore <ian at freebsd.org> wrote:
> 
> > On Thu, 2020-08-27 at 19:20 +0530, Kamal R. Prasad wrote:
> > > Hello,
> > > 
> > >  This page
> > > 
> > 
> > 
https://www.freebsd.org/doc/en_US.ISO8859-1/books/arch-handbook/smp-design.html
> > > 
> > > says that
> > > "These interrupt handlers are currently misnamed “fast” interrupt
> > 
> > handlers
> > > since the INTR_FAST flag used in earlier versions of the kernel
> > > is used
> > 
> > to
> > > mark these handlers. The only interrupts which currently use
> > > these types
> > 
> > of
> > > interrupt handlers are clock interrupts and serial I/O device
> > 
> > interrupts."
> > > 
> > > I want to use INTR_FAST for my spi controller, but i find that it
> > > is not
> > > defined on freebsd stable11. Is INTR_EXCL related in any way to
> > 
> > INTR_FAST?
> > > if not, how do I setup a fast interrupt handler?
> > > 
> > > thanks
> > > -kamal
> > > 
> > 
> > An important question here is:  Why do you want to use a filter
> > handler
> > (formerly called INTR_FAST) instead of a normal threaded
> > handler?  For
> > something as slow as a SPI bus, I can't imagine you're going to see
> > any
> > noticible performance difference.
> > 
> > -- Ian
> > 
> > _______________________________________________
> > freebsd-arm at freebsd.org mailing list
> > https://lists.freebsd.org/mailman/listinfo/freebsd-arm
> > To unsubscribe, send any mail to "
> > freebsd-arm-unsubscribe at freebsd.org"
> >