From b7e8406f6cf7f8f83fe9594d06c0861e4068241b Mon Sep 17 00:00:00 2001
From: "Yukihiro \"Matz\" Matsumoto" <matz@ruby.or.jp>
Date: Tue, 3 Nov 2020 14:58:44 +0900
Subject: Add new instructions to handle symbols/literals >255; fix #5109

New instructions:
  * OP_LOADL16
  * OP_LOADSYM16
  * OP_STRING16

Size of pools, symbols are `int16_t` but offset representation in the
bytecode was 8 bits. Size of child `irep` array is `int16_t`, too but
this change does not address it.
---
 src/codedump.c |  9 +++++++++
 src/vm.c       | 17 ++++++++++++++++-
 2 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'src')

diff --git a/src/codedump.c b/src/codedump.c
index dcc196973..9af609264 100644
--- a/src/codedump.c
+++ b/src/codedump.c
@@ -135,7 +135,10 @@ codedump(mrb_state *mrb, const mrb_irep *irep)
       printf("OP_MOVE\tR%d\tR%d\t", a, b);
       print_lv_ab(mrb, irep, a, b);
       break;
+    CASE(OP_LOADL16, BS);
+      goto op_loadl;
     CASE(OP_LOADL, BB);
+    op_loadl:
       switch (irep->pool[b].tt) {
       case IREP_TT_FLOAT:
         printf("OP_LOADL\tR%d\tL(%d)\t; %f", a, b, (double)irep->pool[b].u.f);
@@ -182,7 +185,10 @@ codedump(mrb_state *mrb, const mrb_irep *irep)
       printf("OP_LOADI_%d\tR%d\t\t", ins-(int)OP_LOADI_0, a);
       print_lv_a(mrb, irep, a);
       break;
+    CASE(OP_LOADSYM16, BS);
+      goto op_loadsym;
     CASE(OP_LOADSYM, BB);
+    op_loadsym:
       printf("OP_LOADSYM\tR%d\t:%s\t", a, mrb_sym_dump(mrb, irep->syms[b]));
       print_lv_a(mrb, irep, a);
       break;
@@ -438,7 +444,10 @@ codedump(mrb_state *mrb, const mrb_irep *irep)
       printf("OP_INTERN\tR%d", a);
       print_lv_a(mrb, irep, a);
       break;
+    CASE(OP_STRING16, BS);
+      goto op_string;
     CASE(OP_STRING, BB);
+    op_string:
       if ((irep->pool[b].tt & IREP_TT_NFLAG) == 0) {
         printf("OP_STRING\tR%d\tL(%d)\t; %s", a, b, irep->pool[b].u.str);
       }
diff --git a/src/vm.c b/src/vm.c
index bcb78c0c9..1d6f82249 100644
--- a/src/vm.c
+++ b/src/vm.c
@@ -1069,7 +1069,12 @@ RETRY_TRY_BLOCK:
       NEXT;
     }
 
+    
+    CASE(OP_LOADL16, BS) {
+      goto op_loadl;
+    }
     CASE(OP_LOADL, BB) {
+    op_loadl:
       switch (pool[b].tt) {   /* number */
       case IREP_TT_INT32:
         regs[a] = mrb_int_value(mrb, (mrb_int)pool[b].u.i32);
@@ -1138,6 +1143,11 @@ RETRY_TRY_BLOCK:
       NEXT;
     }
 
+    CASE(OP_LOADSYM16, BS) {
+      SET_SYM_VALUE(regs[a], syms[b]);
+      NEXT;
+    }
+
     CASE(OP_LOADNIL, B) {
       SET_NIL_VALUE(regs[a]);
       NEXT;
@@ -2575,8 +2585,13 @@ RETRY_TRY_BLOCK:
       NEXT;
     }
 
+    CASE(OP_STRING16, BS) {
+      goto op_string;
+    }
     CASE(OP_STRING, BB) {
-      size_t len = pool[b].tt >> 2;
+      size_t len;
+    op_string:
+      len = pool[b].tt >> 2;
       if (pool[b].tt & IREP_TT_SFLAG) {
         regs[a] = mrb_str_new_static(mrb, pool[b].u.str, len);
       }
-- 
cgit v1.2.3