Board: Decomp view (progress tiers, function inspector, next up per band, deferred) and RE tools in Health; Verified tier from the emulator

harbor/queue.py

@@ -129,14 +129,14 @@ class FunctionQueue:
         return addr in library or any(lo <= addr < hi for lo, hi in ranges)
 
     # ------------------------------------------------------------------ queue
-    def score(self, addr, facts, named, rejected, similar_readable=frozenset()):
+    def score(self, addr, facts, named, rejected, similar_readable=frozenset(), with_convention=True):
         d = facts[addr]
         size = int(d.get("size") or 0)
         calls = [int(c, 16) if isinstance(c, str) else c for c in d.get("calls", [])]
         named_share = (sum(1 for c in calls if c in named) / len(calls)) if calls else 1.0
         score = 1.0 / (1.0 + size / 250.0)                        # small first
         score *= 0.5 + 0.5 * named_share                          # callees with names make readable code possible
-        conv = self.r.convention(addr) if size <= 4000 else None
+        conv = self.r.convention(addr) if with_convention and size <= 4000 else None
         if conv and conv["summary"].startswith("custom"):
             score *= 0.75                                         # register arguments: prototypes.h work too
         if self._category(addr) not in ("game", "unsorted"):
@@ -186,9 +186,12 @@ class FunctionQueue:
             head = self.r.read(addr, 10) if int(facts[addr].get("size") or 0) == 10 else b""
             if head[:1] and 0xB8 <= head[0] <= 0xBF and head[5:6] == b"\xe9":
                 continue        # mov reg, imm32; jmp: an at-exit destructor thunk or an SEH handler stub (compiler's)
-            score, reasons = self.score(addr, facts, named, rejected, similar_readable)
+            score, reasons = self.score(addr, facts, named, rejected, similar_readable, with_convention=False)
             ranked.append((score, addr, reasons))
         ranked.sort(reverse=True)
+        # the convention read from the bytes only lowers a score, and it is the slow part: read it for the head only
+        head = [(*self.score(a, facts, named, rejected, similar_readable), a) for _, a, _ in ranked[:max(4 * n, 40)]]
+        ranked = sorted(((s, a, why) for s, why, a in head), reverse=True)
         return [{"address": f"0x{a:X}", "name": self.r.name_of(a), "score": round(s, 4), "why": ", ".join(why),
                  "file": statuses.get(a, (None, None))[1]} for s, a, why in ranked[:n]]
 
@@ -207,7 +210,9 @@ class FunctionQueue:
         facts, statuses, deferred = self.r._facts(), self.statuses(), self.deferrals()
         exact = {a for a, row in self._results("match_results").items()
                  if str(row.get("status")) in ("match", "exact", "exact-relaxed")}
-        verified = {a for a, row in self._results("diffrun_results").items() if str(row.get("verdict")) == "same"}
+        # behaves the same in the emulator, with at least one trial that ran to the end in both (not 'weak')
+        verified = {a for a, row in self._results("diffrun_results").items()
+                    if str(row.get("verdict")) == "same" and not row.get("weak")}
         library = set(self._results("library_results"))
         ranges = self.library_ranges()
         tiers = {t: {"functions": 0, "bytes": 0} for t in TIER_ORDER + ("library", "engine")}