[Cbe-oss-dev] [PATCH 2/6] spufs: fix timing dependent false return from spufs_run_spu

[Luke Browning]

On Thu, 2008-02-07 at 07:05 +0100, Christoph Hellwig wrote:
> On Wed, Feb 06, 2008 at 03:43:29PM -0200, Luke Browning wrote:
> > Eliminate timing dependent false return from spufs_run_spu().
> > 
> > Stop bits are only valid when the running bit is not set.  Status bits carry 
> > over from one invocation of spufs_run_spu() to another, so the RUNNING bit 
> > gets added to the previous state of the register which may have been a remote
> > library call.  In this case, it looks like another library routine should be
> > invoked, but the spe is actually running.  This solves a problem with a 
> > testcase that exercises the scheduler.
> 
> This looks like a fix for sched_simple indeed.  I can't get at my blade
> currently, but I'll test it later today.
> 

Yes, I developed these changes to fix the sched_simple problem.

Luke