Subject: Re: Prototype kernel continuation-passing for NetBSD
To: Matt Thomas <matt@3am-software.com>
From: Martin Husemann <martin@duskware.de>
List: tech-kern
Date: 03/26/2004 11:04:07
On Thu, Mar 25, 2004 at 03:19:36PM -0800, Matt Thomas wrote:
> At this point, I would instead make generic softintr required fuctionality,
> extend their capability, and kill kconts.  Or kill generic softintr and
> use kconts instead.  One or other but not both.

While there I'd like to note that some archs already support
__GENERIC_SOFT_INTERRUPTS_ALL_LEVELS (i.e. allow arbitrary IPLs for
soft interrupts). I'm not sure if there are any archs that have reasons
not to support this feature, my impression was that the restrictions on
IPLs happened to make the generic soft interrupt introduction less
intrusive.


Martin