Currently, we update the SPU master run control bit (ie,
spu_enable_spu) in spufs_run_spu before we grab the context mutex. This
can result in races with other processes accessing this context's
resources.
This change moves the spu_enable_spu to after we have acquired the
context lock.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
if (mutex_lock_interruptible(&ctx->run_mutex))
return -ERESTARTSYS;
- spu_enable_spu(ctx);
ctx->event_return = 0;
ret = spu_acquire(ctx);
if (ret)
goto out_unlock;
+ spu_enable_spu(ctx);
+
spu_update_sched_info(ctx);
ret = spu_run_init(ctx, npc);