> There are tons of Sparc responsibilities I need to attend to which are
> long overdue, and people are assuming I'm the Intel/SMP maintainer
> now, I am not.
True. Thats me. I've now got a 2.1.24 set that I can actually run on
the SMP box (it does some real stuff so it has to vaguely stay up). There's
a giant diff heading out soon.
Ok, I want to suggest though that we organize people to begin
multi-threading various subsystems of the kernel. This was of course
the main motiviation behind my enormous changes.
On the Intel side of things the first thing someone needs to do is
implement traditional spinlocks in asm-i386/smp_lock.h Check out
asm-sparc/smp_lock.h for the interfaces needed. Should be simple.
---------------------------------------------////
Yow! 11.26 MB/s remote host TCP bandwidth & ////
199 usec remote TCP latency over 100Mb/s ////
ethernet. Beat that! ////
-----------------------------------------////__________ o
David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><