Add minimize memory to PE connection resource algorithm

• Added min_mem_to_pe resource allocation/assignment algorithm.
• Fixed bug where two PEs/memories with the same name broke the architecture rendering by enforcing unique names.
• Fixed comments from !507 (merged)

Closes: #347 (closed)

b_asic/architecture.py

@@ -615,13 +615,23 @@ of :class:`~b_asic.architecture.ProcessingElement`
         direct_interconnects: ProcessCollection | None = None,
     ):
         super().__init__(entity_name)
+
+        pe_names = [pe._entity_name for pe in processing_elements]
+        if len(pe_names) != len(set(pe_names)):
+            raise ValueError("Entity names of processing elements needs to be unique.")
         self._processing_elements = (
             [processing_elements]
             if isinstance(processing_elements, ProcessingElement)
             else list(processing_elements)
         )
+
+        mem_names = [mem._entity_name for mem in memories]
+        if len(mem_names) != len(set(mem_names)):
+            raise ValueError("Entity names of memories needs to be unique.")
         self._memories = [memories] if isinstance(memories, Memory) else list(memories)
+
         self._direct_interconnects = direct_interconnects
+
         self._variable_input_port_to_resource: defaultdict[
             InputPort, set[tuple[Resource, int]]
         ] = defaultdict(set)
@@ -929,7 +939,7 @@ of :class:`~b_asic.architecture.ProcessingElement`
         fmt: str | None = None,
         branch_node: bool = True,
         cluster: bool = True,
-        splines: str = "spline",
+        splines: Literal["spline", "line", "ortho", "polyline", "curved"] = "spline",
         io_cluster: bool = True,
         multiplexers: bool = True,
         colored: bool = True,
@@ -975,7 +985,7 @@ of :class:`~b_asic.architecture.ProcessingElement`
         self,
         branch_node: bool = True,
         cluster: bool = True,
-        splines: str = "spline",
+        splines: Literal["spline", "line", "ortho", "polyline", "curved"] = "spline",
         io_cluster: bool = True,
         multiplexers: bool = True,
         colored: bool = True,

[Oscar Gustafsson]

If you call entity_name here instead, you will not have to check and raise an error since that will do it for you.

[Simon Bjurek]

Hmm, dont really see how that would help.

There is a check in set_entity_name but nothing in entity_name.

[Oscar Gustafsson]

Ahh, I mixed it up with https://gitlab.liu.se/da/B-ASIC/-/blob/master/b_asic/architecture.py?ref_type=heads#L83

But do you actually need to use the entity names here? It seems like the wrong place to check it.

[Simon Bjurek]

Don't know really, my thinking was that a resource can have a missing entity name as long as its not part of an architecture.

Otherwise, why is the name optional?

[Simon Bjurek]

But yeah, I'm up for alternatives.

[Oscar Gustafsson]

It is optional because you probably do not want to come up with a name every time just to play around. One can also imagine some sort of automatic naming (as is the case for the top architecture or use something like GraphID). But, especially, since we do not have code generation yet, it seems inefficient to strictly require an entity name.

I think that the use case of evaluating different algorithms without having to specify names is a good enough reason at the moment to not require a name.

We may reevaluate it later if it becomes a problem, but there are things to consider that are not really clear yet.

[Simon Bjurek]

Sure, I guess that automatic naming is the best option then?

[Simon Bjurek]

Alternatively, fix the architecture rendering for undefined entity names.

[Oscar Gustafsson]

And here.