til: introduce til_frame_plan_t and .cpu_affinity

modules/checkers w/fill_module=$module requires a consistent
mapping of cpu to fragnum since it creates a per-cpu
til_module_context_t for the fill_module.

The existing implementation for threaded rendering maximizes
performance by letting *any* scheduled to run thread advance
fragnum atomically and render the acquired fragnum
indiscriminately.  A side effect of this is any given frame, even
rendered by the same module, will have a random mapping of
cpus/threads to fragnums.

With this change, the simple til_module_t.prepare_frame() API of
returning a bare fragmenter function is changed to instead return
a "frame plan" in til_frame_plan_t.  Right now til_frame_plan_t
just contains the same fragmenter as before, but also has a
.cpu_affinity member for setting if the frame requires a stable
relationship of cpu/thread to fragnum.

Setting .cpu_affinity should be avoided if unnecessary, and that
is the default if you don't mention .cpu_affinity at all when
initializing the plan in the ergonomic manner w/designated
initializers.  This is because the way .cpu_affinity is
implemented will leave threads spinning while they poll for
*their* next fragnum using atomic intrinsics.  There's probably
some room for improvement here, but this is good enough for now
to get things working and correct.

1 files changed, 8 insertions, 8 deletions

diff --git a/src/til.c b/src/til.c
index 951a398..81a300e 100644
--- a/src/til.c
+++ b/src/til.c
@@ -108,9 +108,9 @@ void til_shutdown(void)
 }
 
 
-static void _blank_prepare_frame(til_module_context_t *context, unsigned ticks, til_fb_fragment_t *fragment, til_fragmenter_t *res_fragmenter)
+static void _blank_prepare_frame(til_module_context_t *context, unsigned ticks, til_fb_fragment_t *fragment, til_frame_plan_t *res_frame_plan)
 {
-	*res_fragmenter = til_fragmenter_slice_per_cpu;
+	*res_frame_plan = (til_frame_plan_t){ .fragmenter = til_fragmenter_slice_per_cpu };
 }
 
 
@@ -182,22 +182,22 @@ static void module_render_fragment(til_module_context_t *context, til_threads_t
 	module = context->module;
 
 	if (context->n_cpus > 1 && module->prepare_frame) {
-		til_fragmenter_t	fragmenter;
+		til_frame_plan_t	frame_plan = {};
 
-		module->prepare_frame(context, ticks, fragment, &fragmenter);
+		module->prepare_frame(context, ticks, fragment, &frame_plan);
 
 		if (module->render_fragment) {
-			til_threads_frame_submit(threads, fragment, fragmenter, module->render_fragment, context, ticks);
+			til_threads_frame_submit(threads, fragment, &frame_plan, module->render_fragment, context, ticks);
 			til_threads_wait_idle(threads);
 		}
 	} else if (module->prepare_frame) {
-		til_fragmenter_t	fragmenter;
+		til_frame_plan_t	frame_plan = {};
 		unsigned		fragnum = 0;
 		til_fb_fragment_t	frag;
 
-		module->prepare_frame(context, ticks, fragment, &fragmenter);
+		module->prepare_frame(context, ticks, fragment, &frame_plan);
 
-		while (fragmenter(context, fragment, fragnum++, &frag))
+		while (frame_plan.fragmenter(context, fragment, fragnum++, &frag))
 			module->render_fragment(context, ticks, 0, &frag);
 	} else if (module->render_fragment)
 		module->render_fragment(context, ticks, 0, fragment);