Support Ruby3.0 keyword arguments.

The Difference

Since Ruby1.9, the keyword arguments were emulated by Ruby using the hash
object at the bottom of the arguments. But we have gradually moved toward
keyword arguments separated from normal (positinal) arguments.

At the same time, we value compatibility, so that Ruby3.0 keyword
arguments are somewhat compromise. Basically, keyword arguments are
separated from positional arguments, except when the method does not
take any formal keyword arguments, given keyword arguments (packed
in the hash object) are considered as the last argument.

And we also allow non symbol keys in the keyword arguments. In that
case, those keys are just passed in the `**` hash (or raise
`ArgumentError` for unknown keys).

The Instruction Changes

We have changed `OP_SEND` instruction. `OP_SEND` instruction used to
take 3 operands, the register, the symbol, the number of (positional)
arguments. The meaning of the third operand has been changed. It is now
considered as `n|(nk<<4)`, where `n` is the number of positional
arguments, and `nk` is the number of keyword arguments, both occupies
4 bits in the operand.

The number `15` in both `n` and `nk` means variable sized arguments are
packed in the object. Positional arguments will be packed in the array,
and keyword arguments will be packed in the hash object. That means
arguments more than 14 values are always packed in the object.

Arguments information for other instructions (`OP_SENDB` and `OP_SUPER`)
are also changed. It works as the third operand of `OP_SEND`. the
difference between `OP_SEND` and `OP_SENDB` is just trivial. It assigns
`nil` to the block hidden arguments (right after arguments).

The instruction `OP_SENDV` and `OP_SENDVB` are removed. Those
instructions are replaced by `OP_SEND` and `OP_SENDB` respectively with
the `15` (variable sized) argument information.

Calling Convention

When calling a method, the stack elements shall be in the order of the
receiver of the method, positional arguments, keyword arguments and the
block argument. If the number of positional or keyword arugument (`n` or
`nk`) is zero, corresponding arguments will be empty. So when `n=0` and
`nk=0` the stack layout (from bottom to top) will be:

+-----------------------+
| recv | block (or nil) |
+-----------------------+

The last elements `block` should be explicitly filled before `OP_SEND`
or assigned to `nil` by `OP_SENDB` internally. In other words, the
following have exactly same behavior:

OP_SENDB clears `block` implicitly:

```
OP_SENDB reg sym 0
```

OP_SEND clears `block` implicitly:

```
OP_LOADNIL  R2
OP_SEND     R2 sym 0
```

When calling a method with only positional arguments (n=0..14) without
keyword arguments, the stack layout will be like following:

+--------------------------------------------+
| recv | arg1 | ... | arg_n | block (or nil) |
+--------------------------------------------+

When calling a method with arguments packed in the array (n=15) which
means argument splat (*) is used in the actual arguments, or more than
14 arguments are passed the stack layout will be like following:

+-------------------------------+
| recv | array | block (or nil) |
+-------------------------------+

The number of the actual arguments is determined by the length of the
argument array.

When keyword arguments are given (nk>0), keyword arguments are passed
between positional arguments and the block argument. For example, when
we pass one positional argument `1` and one keyword argument `a: 2`,
the stack layout will be like:

+------------------------------------+
| recv | 1 | :a | 2 | block (or nil) |
+------------------------------------+

Note that keyword arguments consume `2*nk` elements in the stack when
`nk=0..14` (unpacked).

When calling a method with keyword arguments packed in the hash object
(nk=15) which means keyword argument splat (**) is used or more than
14 keyword arguments in the actual arguments, the stack layout will
be like:

+------------------------------+
| recv | hash | block (or nil) |
+------------------------------+

Note for mruby/c

When mruby/c authors try to support new keyword arguments, they need
to handle the new meaning of the argument information operand. If they
choose not to support keyword arguments in mruby/c, it just raise
error when `nk` (taken by `(c>>4)&0xf`) is not zero. And combine
`OP_SENDV` behavior with `OP_SEND` when `n` is `15`.

If they want to support keyword arguments seriously, contact me at
<[email protected]> or `@yukihiro_matz`. I can help you.

3 files changed, 72 insertions, 35 deletions

diff --git a/include/mruby.h b/include/mruby.h
index d82cff581..f80971543 100644
--- a/include/mruby.h
+++ b/include/mruby.h
@@ -153,9 +153,10 @@ typedef void* (*mrb_allocf) (struct mrb_state *mrb, void*, size_t, void *ud);
 #endif
 
 typedef struct {
-  mrb_sym mid;
+  uint8_t n:4;                  /* (15=*) c=n|nk<<4 */
+  uint8_t nk:4;                 /* (15=*) */
   uint8_t cci;                  /* called from C function */
-  int16_t argc;
+  mrb_sym mid;
   const struct RProc *proc;
   mrb_value *stack;
   const mrb_code *pc;           /* current address on iseq of this proc */
diff --git a/include/mruby/boxing_nan.h b/include/mruby/boxing_nan.h
index 90866ab55..cbf7953ed 100644
--- a/include/mruby/boxing_nan.h
+++ b/include/mruby/boxing_nan.h
@@ -24,13 +24,22 @@
 #define MRB_FIXNUM_MIN INT32_MIN
 #define MRB_FIXNUM_MAX INT32_MAX
 
+enum mrb_nanbox_tt_inline {
+  MRB_NANBOX_TT_POINTER = 1,
+  MRB_NANBOX_TT_INTEGER,
+  MRB_NANBOX_TT_SYMBOL,
+  MRB_NANBOX_TT_MISC,
+#ifndef MRB_NO_FLOAT
+  MRB_NANBOX_TT_FLOAT,
+#endif
+};
+
 /* value representation by nan-boxing:
  *   float : FFFFFFFFFFFFFFFF FFFFFFFFFFFFFFFF FFFFFFFFFFFFFFFF FFFFFFFFFFFFFFFF
- *   object: 111111111111TTTT TTPPPPPPPPPPPPPP PPPPPPPPPPPPPPPP PPPPPPPPPPPPPPPP
- *   int   : 1111111111110001 0000000000000000 IIIIIIIIIIIIIIII IIIIIIIIIIIIIIII
- *   sym   : 1111111111110001 0100000000000000 SSSSSSSSSSSSSSSS SSSSSSSSSSSSSSSS
- * In order to get enough bit size to save TT, all pointers are shifted 2 bits
- * in the right direction. Also, TTTTTT is the mrb_vtype + 1;
+ *   object: 1111111111110001 PPPPPPPPPPPPPPPP PPPPPPPPPPPPPPPP PPPPPPPPPPPPPPPP
+ *   int   : 1111111111110010 0000000000000000 IIIIIIIIIIIIIIII IIIIIIIIIIIIIIII
+ *   sym   : 1111111111110011 0100000000000000 SSSSSSSSSSSSSSSS SSSSSSSSSSSSSSSS
+ *   misc  : 1111111111110100 0100000000000000 0000000000000000 TTTTTT000000MMMM
  */
 typedef struct mrb_value {
   uint64_t u;
@@ -41,16 +50,7 @@ union mrb_value_ {
   uint64_t u;
 #ifdef MRB_64BIT
   void *p;
-# define NANBOX_IMMEDIATE_VALUE uint32_t i
-#else
-# define NANBOX_IMMEDIATE_VALUE union { uint32_t i; void *p; }
 #endif
-  struct {
-    MRB_ENDIAN_LOHI(
-      uint32_t ttt;
-      ,NANBOX_IMMEDIATE_VALUE;
-    )
-  };
   mrb_value value;
 };
 
@@ -60,26 +60,59 @@ static inline union mrb_value_
 mrb_val_union(mrb_value v)
 {
   union mrb_value_ x;
-  x.value = v;
+  x.u = v.u;
   return x;
 }
 
-#define mrb_tt(o)       ((enum mrb_vtype)((mrb_val_union(o).ttt & 0xfc000)>>14)-1)
-#define mrb_type(o)     (enum mrb_vtype)((uint32_t)0xfff00000 < mrb_val_union(o).ttt ? mrb_tt(o) : MRB_TT_FLOAT)
-#define mrb_float(o)    mrb_val_union(o).f
-#define mrb_fixnum(o)   ((mrb_int)mrb_val_union(o).i)
+static inline mrb_float
+mrb_float(mrb_value v)
+{
+  union {
+    mrb_float f;
+    uint64_t u;
+  } x;
+  x.u = v.u;
+  return x.f;
+}
+
+#define mrb_tt_(o)       ((enum mrb_nanbox_tt_inline)((o).u >> 48)&0xf)
+
+MRB_INLINE enum mrb_vtype
+mrb_type(mrb_value o)
+{
+  switch (mrb_tt_(o)) {
+  case MRB_NANBOX_TT_POINTER:
+    return RBASIC(o)->tt;
+  case MRB_NANBOX_TT_INTEGER:
+    return MRB_TT_INTEGER;
+  case MRB_NANBOX_TT_SYMBOL:
+    return MRB_TT_SYMBOL;
+  case MRB_NANBOX_TT_MISC:
+    return (enum mrb_vtype)(o.u >> 10) & 0x3f;
+#ifndef MRB_NO_FLOAT
+  default:
+    return MRB_TT_FLOAT;
+#endif
+  }
+  return MRB_TT_UNDEF;
+}
+
+#define mrb_symbol(o)   ((mrb_sym)((o).u & 0x3fffffff))
+
+#ifdef MRB_INT64
+#define mrb_fixnum(o)   ((mrb_int)((o).u & 0xffffffffffffL))
+#define mrb_integer(o)  ((mrb_tt(o)==MRB_NANBOX_TT_POINTER)?(((struct RInteger*)mrb_ptr(o))->i):mrb_fixnum(o)))
+#else  /* MRB_INT32 */
+#define mrb_fixnum(o)   ((mrb_int)((o).u & 0xffffffff))
 #define mrb_integer(o)  mrb_fixnum(o)
-#define mrb_symbol(o)   ((mrb_sym)mrb_val_union(o).i)
+#endif
 
 #ifdef MRB_64BIT
-#define mrb_ptr(o)      ((void*)((((uintptr_t)0x3fffffffffff)&((uintptr_t)(mrb_val_union(o).p)))<<2))
-#define mrb_cptr(o)     (((struct RCptr*)mrb_ptr(o))->p)
-#define NANBOX_SHIFT_LONG_POINTER(v) (((uintptr_t)(v)>>34)&0x3fff)
+#define mrb_ptr(o)      ((void*)(((uintptr_t)(o).u) & 0xffffffffffff))
 #else
-#define mrb_ptr(o)      ((void*)mrb_val_union(o).i)
-#define mrb_cptr(o)     mrb_ptr(o)
-#define NANBOX_SHIFT_LONG_POINTER(v) 0
+#define mrb_ptr(o)      ((void*)(((uintptr_t)(o).u) & 0xffffffff))
 #endif
+#define mrb_cptr(o)     mrb_ptr(o)
 
 #define NANBOX_SET_VALUE(o, tt, attr, v) do { \
   union mrb_value_ mrb_value_union_variable; \
@@ -102,8 +135,7 @@ mrb_val_union(mrb_value v)
 #define SET_FLOAT_VALUE(mrb,r,v) do { \
   union mrb_value_ mrb_value_union_variable; \
   if ((v) != (v)) { /* NaN */ \
-    mrb_value_union_variable.ttt = 0x7ff80000; \
-    mrb_value_union_variable.i = 0; \
+    mrb_value_union_variable.u = 0x7ff8000000000000UL; \
   } \
   else { \
     mrb_value_union_variable.f = (v); \
@@ -111,6 +143,13 @@ mrb_val_union(mrb_value v)
   r = mrb_value_union_variable.value; \
 } while(0)
 
+#define NANBOX_SET_MISC_VALUE(o, tt, attr, v) do { \
+  union mrb_value_ mrb_value_union_variable; \
+  mrb_value_union_variable.attr = (v);\
+  mrb_value_union_variable.ttt = 0xfff00000 | (((tt)+1)<<14);\
+  o = mrb_value_union_variable.value;\
+} while (0)
+
 #define SET_NIL_VALUE(r) NANBOX_SET_VALUE(r, MRB_TT_FALSE, i, 0)
 #define SET_FALSE_VALUE(r) NANBOX_SET_VALUE(r, MRB_TT_FALSE, i, 1)
 #define SET_TRUE_VALUE(r) NANBOX_SET_VALUE(r, MRB_TT_TRUE, i, 1)
diff --git a/include/mruby/ops.h b/include/mruby/ops.h
index 7b5ea40ca..af7051833 100644
--- a/include/mruby/ops.h
+++ b/include/mruby/ops.h
@@ -57,11 +57,8 @@ OPCODE(JMPUW,      S)        /* unwind_and_jump_to(a) */
 OPCODE(EXCEPT,     B)        /* R(a) = exc */
 OPCODE(RESCUE,     BB)       /* R(b) = R(a).isa?(R(b)) */
 OPCODE(RAISEIF,    B)        /* raise(R(a)) if R(a) */
-OPCODE(SENDV,      BB)       /* R(a) = call(R(a),Syms(b),*R(a+1)) */
-OPCODE(SENDVB,     BB)       /* R(a) = call(R(a),Syms(b),*R(a+1),&R(a+2)) */
-OPCODE(SEND,       BBB)      /* R(a) = call(R(a),Syms(b),R(a+1),...,R(a+c)) */
-OPCODE(SENDB,      BBB)      /* R(a) = call(R(a),Syms(b),R(a+1),...,R(a+c),&R(a+c+1)) */
-OPCODE(SENDVK,     BB)       /* R(a) = call(R(a),Syms(b),*R(a+1),**(a+2),&R(a+3)) # todo */
+OPCODE(SEND,       BBB)      /* R(a) = call(R(a),Syms(b),R(a+1)..,R(a+n+1):R(a+n+2)..) (c=n|k<<4) */
+OPCODE(SENDB,      BBB)      /* R(a) = call(R(a),Syms(b),R(a+1)..,R(a+n+1):R(a+n+2)..,&R(a+n+2k+1)) */
 OPCODE(CALL,       Z)        /* R(0) = self.call(frame.argc, frame.argv) */
 OPCODE(SUPER,      BB)       /* R(a) = super(R(a+1),... ,R(a+b+1)) */
 OPCODE(ARGARY,     BS)       /* R(a) = argument array (16=m5:r1:m5:d1:lv4) */