On Fri, Jun 21, 2002 at 11:48:23AM +0200, Wolfgang Grandegger wrote: > Hello, > > I jump into the discussion as well. > > On 06/21/2002 12:10 AM, Tom Rini wrote: > > >Well, I looked at it (after I sent but before your mail) and it wasn't a > >patch but some cut'n paste bits. And to possibly Dan's dismay I don't > >always agree with his opinions now :) > > > > I first wanted to present my ideas ... before I prepare and send a > patch.
For future reference, unified diffs are the preferred way to present ideas which you've already implemented. :) > >Well, having never used RTAI (and by looking at the 'patch' at > >http://lists.linuxppc.org/linuxppc-embedded/200110/txt00002.txt) I don't > >get it.. > > RTAI and RTLinux actually use the "irq_desc[].handler" functions to > handle the interrupts and with the previous cpm_* functions it was > non-trivial to access CPM interrupts within RTAI because they are > handled in a separate interface. Ah, which is why you just expanded things the way you did.. Which works for that context but still leaves the larger problem of request_irq() panic()'ing on unknown interrupt numbers. [snip] > I understand DAN's argument, that the usage of an offset can be quite > messy especially during driver development but I cannot see a simple > way around it without breaking the (PC) "request_irq" function. Well maybe Dan will get encouraged enough to try and properly fix request_irq() and break some of the legacy PC drivers in 2.5 again. :) -- Tom Rini (TR1265) http://gate.crashing.org/~trini/ ** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
