py: Compress load-int, load-fast, store-fast, unop, binop bytecodes.

There is a lot potential in compress bytecodes and make more use of the
coding space.  This patch introduces "multi" bytecodes which have their
argument included in the bytecode (by addition).

UNARY_OP and BINARY_OP now no longer take a 1 byte argument for the
opcode.  Rather, the opcode is included in the first byte itself.

LOAD_FAST_[0,1,2] and STORE_FAST_[0,1,2] are removed in favour of their
multi versions, which can take an argument between 0 and 15 inclusive.
The majority of LOAD_FAST/STORE_FAST codes fit in this range and so this
saves a byte for each of these.

LOAD_CONST_SMALL_INT_MULTI is used to load small ints between -16 and 47
inclusive.  Such ints are quite common and now only need 1 byte to
store, and now have much faster decoding.

In all this patch saves about 2% RAM for typically bytecode (1.8% on
64-bit test, 2.5% on pyboard test).  It also reduces the binary size
(because bytecodes are simplified) and doesn't harm performance.

1 files changed, 16 insertions, 37 deletions

diff --git a/py/showbc.c b/py/showbc.c
index 13d257d30..93fe2c378 100644
--- a/py/showbc.c
+++ b/py/showbc.c
@@ -190,18 +190,6 @@ void mp_bytecode_print2(const byte *ip, mp_uint_t len) {
                 printf("LOAD_NULL");
                 break;
 
-            case MP_BC_LOAD_FAST_0:
-                printf("LOAD_FAST_0");
-                break;
-
-            case MP_BC_LOAD_FAST_1:
-                printf("LOAD_FAST_1");
-                break;
-
-            case MP_BC_LOAD_FAST_2:
-                printf("LOAD_FAST_2");
-                break;
-
             case MP_BC_LOAD_FAST_N:
                 DECODE_UINT;
                 printf("LOAD_FAST_N " UINT_FMT, unum);
@@ -240,18 +228,6 @@ void mp_bytecode_print2(const byte *ip, mp_uint_t len) {
                 printf("LOAD_SUBSCR");
                 break;
 
-            case MP_BC_STORE_FAST_0:
-                printf("STORE_FAST_0");
-                break;
-
-            case MP_BC_STORE_FAST_1:
-                printf("STORE_FAST_1");
-                break;
-
-            case MP_BC_STORE_FAST_2:
-                printf("STORE_FAST_2");
-                break;
-
             case MP_BC_STORE_FAST_N:
                 DECODE_UINT;
                 printf("STORE_FAST_N " UINT_FMT, unum);
@@ -397,16 +373,6 @@ void mp_bytecode_print2(const byte *ip, mp_uint_t len) {
                 printf("NOT");
                 break;
 
-            case MP_BC_UNARY_OP:
-                unum = *ip++;
-                printf("UNARY_OP " UINT_FMT, unum);
-                break;
-
-            case MP_BC_BINARY_OP:
-                unum = *ip++;
-                printf("BINARY_OP " UINT_FMT, unum);
-                break;
-
             case MP_BC_BUILD_TUPLE:
                 DECODE_UINT;
                 printf("BUILD_TUPLE " UINT_FMT, unum);
@@ -534,9 +500,22 @@ void mp_bytecode_print2(const byte *ip, mp_uint_t len) {
                 break;
 
             default:
-                printf("code %p, byte code 0x%02x not implemented\n", ip, ip[-1]);
-                assert(0);
-                return;
+                if (ip[-1] < MP_BC_LOAD_CONST_SMALL_INT_MULTI + 64) {
+                    printf("LOAD_CONST_SMALL_INT " INT_FMT, (mp_int_t)ip[-1] - MP_BC_LOAD_CONST_SMALL_INT_MULTI - 16);
+                } else if (ip[-1] < MP_BC_LOAD_FAST_MULTI + 16) {
+                    printf("LOAD_FAST " UINT_FMT, (mp_uint_t)ip[-1] - MP_BC_LOAD_FAST_MULTI);
+                } else if (ip[-1] < MP_BC_STORE_FAST_MULTI + 16) {
+                    printf("STORE_FAST " UINT_FMT, (mp_uint_t)ip[-1] - MP_BC_STORE_FAST_MULTI);
+                } else if (ip[-1] < MP_BC_UNARY_OP_MULTI + 5) {
+                    printf("UNARY_OP " UINT_FMT, (mp_uint_t)ip[-1] - MP_BC_UNARY_OP_MULTI);
+                } else if (ip[-1] < MP_BC_BINARY_OP_MULTI + 35) {
+                    printf("BINARY_OP " UINT_FMT, (mp_uint_t)ip[-1] - MP_BC_BINARY_OP_MULTI);
+                } else {
+                    printf("code %p, byte code 0x%02x not implemented\n", ip, ip[-1]);
+                    assert(0);
+                    return;
+                }
+                break;
         }
         printf("\n");
     }