On Sat, 10 Aug 2002, Robert Davies wrote:
> Hardware trap vectors are common, it's one of the advantages of APICs, you
> have fewer shared interrupt chains.
We're talking about different things ;-) SPARC defines 5 trap levels with
separated GPRs - at least the more recent chips do according to the docs.
> If you can remember how we got into this it was due to 'the less interrupts
> the better', and the disabling of interrupts and going into poll mode under
> busy conditions certainly counts under that.
My argument is that "going in to polling" should be the kind of hack one
uses to lower latency of response - it's not the same as a "polling"
driver by any means. I define a polling driver as one which always works
the same way and not the hybrid situation you described before.
Jon.
--------------------------------------------------------------------
http://www.lug.org.uk http://www.linuxportal.co.uk
http://www.linuxjob.co.uk http://www.linuxshop.co.uk
--------------------------------------------------------------------
This archive was generated by hypermail 2.1.3 : Sat 10 Aug 2002 - 18:21:35 BST