powerpc/spufs: fix missed stop-and-signal event

There is a delay in the transition to the stopped state for class 2
interrupts. In some cases, the controlling thread detects the state of
the spu as running, and goes back to sleep resulting in a hung
application as the event is missed.

This change detects the stop condition and re-generates the wakeup event
after a context save.
Signed-off-by: NLuke Browning <lukebrowning@us.ibm.com>
Signed-off-by: NJeremy Kerr <jk@ozlabs.org>

arch/powerpc/platforms/cell/spufs/sched.c

@@ -407,6 +407,8 @@ static int has_affinity(struct spu_context *ctx)
  */
 static void spu_unbind_context(struct spu *spu, struct spu_context *ctx)
 {
+	u32 status;
+
 	spu_context_trace(spu_unbind_context__enter, ctx, spu);
 
 	spuctx_switch_state(ctx, SPU_UTIL_SYSTEM);
@@ -452,6 +454,9 @@ static void spu_unbind_context(struct spu *spu, struct spu_context *ctx)
 	/* This maps the underlying spu state to idle */
 	spuctx_switch_state(ctx, SPU_UTIL_IDLE_LOADED);
 	ctx->spu = NULL;
+
+	if (spu_stopped(ctx, &status))
+		wake_up_all(&ctx->stop_wq);
 }
 
 /**