]> pilppa.com Git - linux-2.6-omap-h63xx.git/commit
[POWERPC] spusched: Dynamic timeslicing for SCHED_OTHER
authorChristoph Hellwig <hch@lst.de>
Fri, 29 Jun 2007 00:57:52 +0000 (10:57 +1000)
committerPaul Mackerras <paulus@samba.org>
Tue, 3 Jul 2007 05:24:44 +0000 (15:24 +1000)
commitfe443ef2ac421c9c652e251e8733e2479d8e411a
tree60914cfaf29232b9c3d4f555eac3b19b85a84254
parent379018022071489a7dffee74db2a267465dab561
[POWERPC] spusched: Dynamic timeslicing for SCHED_OTHER

Enable preemptive scheduling for non-RT contexts.

We use the same algorithms as the CPU scheduler to calculate the time
slice length, and for now we also use the same timeslice length as the
CPU scheduler. This might be not enough for good performance and can be
changed after some benchmarking.

Note that currently we do not boost the priority for contexts waiting
on the runqueue for a long time, so contexts with a higher nice value
could starve ones with less priority.  This could easily be fixed once
the rework of the spu lists that Luke and I discussed is done.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
arch/powerpc/platforms/cell/spufs/context.c
arch/powerpc/platforms/cell/spufs/sched.c
arch/powerpc/platforms/cell/spufs/spufs.h