diff --git a/src/lj_jit.h b/src/lj_jit.h
index f52c8d33d5..b72207d6c8 100644
--- a/src/lj_jit.h
+++ b/src/lj_jit.h
@@ -223,13 +223,13 @@ static LJ_AINLINE MSize snap_nextofs(GCtrace *T, SnapShot *snap)
 /* Round-robin penalty cache for bytecodes leading to aborted traces. */
 typedef struct HotPenalty {
   MRef pc;		/* Starting bytecode PC. */
-  uint16_t val;		/* Penalty value, i.e. hotcount start. */
+  uint32_t val;		/* Penalty value, i.e. hotcount start. */
   uint16_t reason;	/* Abort reason (really TraceErr). */
 } HotPenalty;
 
 #define PENALTY_SLOTS	64	/* Penalty cache slot. Must be a power of 2. */
 #define PENALTY_MIN	(36*2)	/* Minimum penalty value. */
-#define PENALTY_MAX	60000	/* Maximum penalty value. */
+#define PENALTY_MAX	6000000	/* Maximum penalty value. */
 #define PENALTY_RNDBITS	4	/* # of random bits to add to penalty value. */
 
 /* Round-robin backpropagation cache for narrowing conversions. */
diff --git a/src/lj_record.c b/src/lj_record.c
index f4b3c0e0d5..e6e6c3be5d 100644
--- a/src/lj_record.c
+++ b/src/lj_record.c
@@ -568,7 +568,8 @@ static void rec_loop_interp(jit_State *J, const BCIns *pc, LoopEvent ev)
 /* Handle the case when an already compiled loop op is hit. */
 static void rec_loop_jit(jit_State *J, TraceNo lnk, LoopEvent ev)
 {
-  if (J->parent == 0 && J->exitno == 0) {  /* Root trace hit an inner loop. */
+  /* Root trace hit an inner loop. */
+  if (J->parent == 0 && J->exitno == 0 && !innerloopleft(J, J->startpc)) {
     /* Better let the inner loop spawn a side trace back here. */
     lj_trace_err(J, LJ_TRERR_LINNER);
   } else if (ev != LOOPEV_LEAVE) {  /* Side trace enters a compiled loop. */
diff --git a/src/lj_trace.c b/src/lj_trace.c
index c99cff1156..5be350ad17 100644
--- a/src/lj_trace.c
+++ b/src/lj_trace.c
@@ -264,6 +264,8 @@ int lj_trace_flushall(lua_State *L)
   }
   /* Clear penalty cache. */
   memset(J->penalty, 0, sizeof(J->penalty));
+  /* Reset hotcounts. */
+  lj_dispatch_init_hotcount(J2G(J));
   /* Free the whole machine code and invalidate all exit stub groups. */
   lj_mcode_free(J);
   memset(J->exitstubgroup, 0, sizeof(J->exitstubgroup));
@@ -335,7 +337,7 @@ static void penalty_pc(jit_State *J, GCproto *pt, BCIns *pc, TraceError e)
   J->penaltyslot = (J->penaltyslot + 1) & (PENALTY_SLOTS-1);
   setmref(J->penalty[i].pc, pc);
 setpenalty:
-  J->penalty[i].val = (uint16_t)val;
+  J->penalty[i].val = val;
   J->penalty[i].reason = e;
   hotcount_set(J2GG(J), pc+1, val);
 }
@@ -401,6 +403,7 @@ static void trace_stop(jit_State *J)
   TraceNo traceno = J->cur.traceno;
   GCtrace *T = J->curfinal;
   lua_State *L;
+  int i;
 
   switch (op) {
   case BC_FORL:
@@ -452,6 +455,11 @@ static void trace_stop(jit_State *J)
   J->postproc = LJ_POST_NONE;
   trace_save(J, T);
 
+  /* Clear any penalty after successful recording. */
+  for (i = 0; i < PENALTY_SLOTS; i++)
+    if (mref(J->penalty[i].pc, const BCIns) == pc)
+      J->penalty[i].val = PENALTY_MIN;
+
   L = J->L;
 }