git subrepo clone https://git.savannah.gnu.org/git/lightning.git deps/lightning

subrepo:
  subdir:   "deps/lightning"
  merged:   "b0b8eb5"
upstream:
  origin:   "https://git.savannah.gnu.org/git/lightning.git"
  branch:   "master"
  commit:   "b0b8eb5"
git-subrepo:
  version:  "0.4.1"
  origin:   "https://github.com/ingydotnet/git-subrepo.git"
  commit:   "a04d8c2"

1 files changed, 2264 insertions, 0 deletions

diff --git a/deps/lightning/lib/jit_x86.c b/deps/lightning/lib/jit_x86.c
new file mode 100644
index 0000000..c34a117
--- /dev/null
+++ b/deps/lightning/lib/jit_x86.c
@@ -0,0 +1,2264 @@
+/*
+ * Copyright (C) 2012-2019  Free Software Foundation, Inc.
+ *
+ * This file is part of GNU lightning.
+ *
+ * GNU lightning is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU Lesser General Public License as published
+ * by the Free Software Foundation; either version 3, or (at your option)
+ * any later version.
+ *
+ * GNU lightning is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public
+ * License for more details.
+ *
+ * Authors:
+ *	Paulo Cesar Pereira de Andrade
+ */
+
+#include <lightning.h>
+#include <lightning/jit_private.h>
+
+#if __X32
+#  define jit_arg_reg_p(i)		0
+#  define jit_arg_f_reg_p(i)		0
+#  define stack_framesize		20
+#  define stack_adjust			12
+#  define CVT_OFFSET			-12
+#  define REAL_WORDSIZE			4
+#  define va_gp_increment		4
+#  define va_fp_increment		8
+#else
+#  if __CYGWIN__ || _WIN32
+#    define jit_arg_reg_p(i)		((i) >= 0 && (i) < 4)
+#    define jit_arg_f_reg_p(i)		jit_arg_reg_p(i)
+#    define stack_framesize		152
+#    define va_fp_increment		8
+#  else
+#    define jit_arg_reg_p(i)		((i) >= 0 && (i) < 6)
+#    define jit_arg_f_reg_p(i)		((i) >= 0 && (i) < 8)
+#    define stack_framesize		56
+#    define first_gp_argument		rdi
+#    define first_gp_offset		offsetof(jit_va_list_t, rdi)
+#    define first_gp_from_offset(gp)	((gp) / 8)
+#    define last_gp_argument		r9
+#    define va_gp_max_offset						\
+	(offsetof(jit_va_list_t, r9) - offsetof(jit_va_list_t, rdi) + 8)
+#    define first_fp_argument		xmm0
+#    define first_fp_offset		offsetof(jit_va_list_t, xmm0)
+#    define last_fp_argument		xmm7
+#    define va_fp_max_offset						\
+	(offsetof(jit_va_list_t, xmm7) - offsetof(jit_va_list_t, rdi) + 16)
+#    define va_fp_increment		16
+#    define first_fp_from_offset(fp)	(((fp) - va_gp_max_offset) / 16)
+#  endif
+#  define va_gp_increment		8
+#  define stack_adjust			8
+#  define CVT_OFFSET			-8
+#  define REAL_WORDSIZE			8
+#endif
+
+/*
+ * Types
+ */
+#if __X32 || __CYGWIN__ || _WIN32
+typedef jit_pointer_t jit_va_list_t;
+#else
+typedef struct jit_va_list {
+    jit_int32_t		gpoff;
+    jit_int32_t		fpoff;
+    jit_pointer_t	over;
+    jit_pointer_t	save;
+    /* Declared explicitly as int64 for the x32 abi */
+    jit_int64_t		rdi;
+    jit_int64_t		rsi;
+    jit_int64_t		rdx;
+    jit_int64_t		rcx;
+    jit_int64_t		r8;
+    jit_int64_t		r9;
+    jit_float64_t	xmm0;
+    jit_float64_t	_up0;
+    jit_float64_t	xmm1;
+    jit_float64_t	_up1;
+    jit_float64_t	xmm2;
+    jit_float64_t	_up2;
+    jit_float64_t	xmm3;
+    jit_float64_t	_up3;
+    jit_float64_t	xmm4;
+    jit_float64_t	_up4;
+    jit_float64_t	xmm5;
+    jit_float64_t	_up5;
+    jit_float64_t	xmm6;
+    jit_float64_t	_up6;
+    jit_float64_t	xmm7;
+    jit_float64_t	_up7;
+} jit_va_list_t;
+#endif
+
+/*
+ * Prototypes
+ */
+#define patch(instr, node)		_patch(_jit, instr, node)
+static void _patch(jit_state_t*,jit_word_t,jit_node_t*);
+#define sse_from_x87_f(r0, r1)		_sse_from_x87_f(_jit, r0, r1)
+static void _sse_from_x87_f(jit_state_t*,jit_int32_t,jit_int32_t);
+#define sse_from_x87_d(r0, r1)		_sse_from_x87_d(_jit, r0, r1)
+static void _sse_from_x87_d(jit_state_t*,jit_int32_t,jit_int32_t);
+#define x87_from_sse_f(r0, r1)		_x87_from_sse_f(_jit, r0, r1)
+static void _x87_from_sse_f(jit_state_t*,jit_int32_t,jit_int32_t);
+#define x87_from_sse_d(r0, r1)		_x87_from_sse_d(_jit, r0, r1)
+static void _x87_from_sse_d(jit_state_t*,jit_int32_t,jit_int32_t);
+
+#define PROTO				1
+#  include "jit_x86-cpu.c"
+#  include "jit_x86-sse.c"
+#  include "jit_x86-x87.c"
+#undef PROTO
+
+/*
+ * Initialization
+ */
+jit_cpu_t		jit_cpu;
+jit_register_t		_rvs[] = {
+#if __X32
+    { rc(gpr) | rc(rg8) | 0,		"%eax" },
+    { rc(gpr) | rc(rg8) | 1,		"%ecx" },
+    { rc(gpr) | rc(rg8) | 2,		"%edx" },
+    { rc(sav) | rc(rg8) | rc(gpr) | 3,	"%ebx" },
+    { rc(sav) | rc(gpr) | 6,		"%esi" },
+    { rc(sav) | rc(gpr) | 7,		"%edi" },
+    { rc(sav) | 4,			"%esp" },
+    { rc(sav) | 5,			"%ebp" },
+    { rc(xpr) | rc(fpr) | 0,		"%xmm0" },
+    { rc(xpr) | rc(fpr) | 1,		"%xmm1" },
+    { rc(xpr) | rc(fpr) | 2,		"%xmm2" },
+    { rc(xpr) | rc(fpr) | 3,		"%xmm3" },
+    { rc(xpr) | rc(fpr) | 4,		"%xmm4" },
+    { rc(xpr) | rc(fpr) | 5,		"%xmm5" },
+    { rc(xpr) | rc(fpr) | 6,		"%xmm6" },
+    { rc(xpr) | rc(fpr) | 7,		"%xmm7" },
+    { rc(fpr) | 0,			"st(0)" },
+    { rc(fpr) | 1,			"st(1)" },
+    { rc(fpr) | 2,			"st(2)" },
+    { rc(fpr) | 3,			"st(3)" },
+    { rc(fpr) | 4,			"st(4)" },
+    { rc(fpr) | 5,			"st(5)" },
+    { rc(fpr) | 6,			"st(6)" },
+    { rc(fpr) | 7,			"st(7)" },
+#else
+#  if __CYGWIN__ || _WIN32
+    { rc(gpr) | rc(rg8) | 0,		"%rax" },
+    { rc(gpr) | rc(rg8) | rc(rg8) | 10,	"%r10" },
+    { rc(gpr) | rc(rg8) | rc(rg8) | 11,	"%r11" },
+    { rc(sav) | rc(rg8) | rc(gpr) | 3,	"%rbx" },
+    { rc(sav) | rc(gpr) | 7,		"%rdi" },
+    { rc(sav) | rc(gpr) | 6,		"%rsi" },
+    { rc(sav) | rc(gpr) | 12,		"%r12" },
+    { rc(sav) | rc(gpr) | 13,		"%r13" },
+    { rc(sav) | rc(gpr) | 14,		"%r14" },
+    { rc(sav) | rc(gpr) | 15,		"%r15" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 9,	"%r9" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 8,	"%r8" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 2,	"%rdx" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 1,	"%rcx" },
+    { rc(sav) | 4,			"%rsp" },
+    { rc(sav) | 5,			"%rbp" },
+    { rc(xpr) | rc(fpr) | 4,		"%xmm4" },
+    { rc(xpr) | rc(fpr) | 5,		"%xmm5" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 6,	"%xmm6" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 7,	"%xmm7" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 8,	"%xmm8" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 9,	"%xmm9" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 10,	"%xmm10" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 11,	"%xmm11" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 12,	"%xmm12" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 13,	"%xmm13" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 14,	"%xmm14" },
+    { rc(sav) | rc(xpr) | rc(fpr) | 15,	"%xmm15" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 3,	"%xmm3" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 2,	"%xmm2" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 1,	"%xmm1" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 0,	"%xmm0" },
+#else
+    /* %rax is a pseudo flag argument for varargs functions */
+    { rc(arg) | rc(gpr) | rc(rg8) | 0,	"%rax" },
+    { rc(gpr) | rc(rg8) | 10,		"%r10" },
+    { rc(gpr) | rc(rg8) | 11,		"%r11" },
+    { rc(sav) | rc(rg8) | rc(gpr) | 3,	"%rbx" },
+    { rc(sav) | rc(rg8) | rc(gpr) | 13,	"%r13" },
+    { rc(sav) | rc(rg8) | rc(gpr) | 14,	"%r14" },
+    { rc(sav) | rc(rg8) | rc(gpr) | 15,	"%r15" },
+    { rc(sav) | rc(gpr) | rc(rg8) | 12,	"%r12" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 9,	"%r9" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 8,	"%r8" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 1,	"%rcx" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 2,	"%rdx" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 6,	"%rsi" },
+    { rc(arg) | rc(rg8) | rc(gpr) | 7,	"%rdi" },
+    { rc(sav) | 4,			"%rsp" },
+    { rc(sav) | 5,			"%rbp" },
+    { rc(xpr) | rc(fpr) | 8,		"%xmm8" },
+    { rc(xpr) | rc(fpr) | 9,		"%xmm9" },
+    { rc(xpr) | rc(fpr) | 10,		"%xmm10" },
+    { rc(xpr) | rc(fpr) | 11,		"%xmm11" },
+    { rc(xpr) | rc(fpr) | 12,		"%xmm12" },
+    { rc(xpr) | rc(fpr) | 13,		"%xmm13" },
+    { rc(xpr) | rc(fpr) | 14,		"%xmm14" },
+    { rc(xpr) | rc(fpr) | 15,		"%xmm15" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 7,	"%xmm7" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 6,	"%xmm6" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 5,	"%xmm5" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 4,	"%xmm4" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 3,	"%xmm3" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 2,	"%xmm2" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 1,	"%xmm1" },
+    { rc(xpr) | rc(arg) | rc(fpr) | 0,	"%xmm0" },
+#  endif
+    { rc(fpr) | 0,			"st(0)" },
+    { rc(fpr) | 1,			"st(1)" },
+    { rc(fpr) | 2,			"st(2)" },
+    { rc(fpr) | 3,			"st(3)" },
+    { rc(fpr) | 4,			"st(4)" },
+    { rc(fpr) | 5,			"st(5)" },
+    { rc(fpr) | 6,			"st(6)" },
+    { rc(fpr) | 7,			"st(7)" },
+#endif
+    { _NOREG,				"<none>" },
+};
+
+/*
+ * Implementation
+ */
+void
+jit_get_cpu(void)
+{
+    union {
+	struct {
+	    jit_uint32_t sse3		: 1;
+	    jit_uint32_t pclmulqdq	: 1;
+	    jit_uint32_t dtes64		: 1;	/* amd reserved */
+	    jit_uint32_t monitor	: 1;
+	    jit_uint32_t ds_cpl		: 1;	/* amd reserved */
+	    jit_uint32_t vmx		: 1;	/* amd reserved */
+	    jit_uint32_t smx		: 1;	/* amd reserved */
+	    jit_uint32_t est		: 1;	/* amd reserved */
+	    jit_uint32_t tm2		: 1;	/* amd reserved */
+	    jit_uint32_t ssse3		: 1;
+	    jit_uint32_t cntx_id	: 1;	/* amd reserved */
+	    jit_uint32_t __reserved0	: 1;
+	    jit_uint32_t fma		: 1;
+	    jit_uint32_t cmpxchg16b	: 1;
+	    jit_uint32_t xtpr		: 1;	/* amd reserved */
+	    jit_uint32_t pdcm		: 1;	/* amd reserved */
+	    jit_uint32_t __reserved1	: 1;
+	    jit_uint32_t pcid		: 1;	/* amd reserved */
+	    jit_uint32_t dca		: 1;	/* amd reserved */
+	    jit_uint32_t sse4_1		: 1;
+	    jit_uint32_t sse4_2		: 1;
+	    jit_uint32_t x2apic		: 1;	/* amd reserved */
+	    jit_uint32_t movbe		: 1;	/* amd reserved */
+	    jit_uint32_t popcnt		: 1;
+	    jit_uint32_t tsc		: 1;	/* amd reserved */
+	    jit_uint32_t aes		: 1;
+	    jit_uint32_t xsave		: 1;
+	    jit_uint32_t osxsave	: 1;
+	    jit_uint32_t avx		: 1;
+	    jit_uint32_t __reserved2	: 1;	/* amd F16C */
+	    jit_uint32_t __reserved3	: 1;
+	    jit_uint32_t __alwayszero	: 1;	/* amd RAZ */
+	} bits;
+	jit_uword_t	cpuid;
+    } ecx;
+    union {
+	struct {
+	    jit_uint32_t fpu		: 1;
+	    jit_uint32_t vme		: 1;
+	    jit_uint32_t de		: 1;
+	    jit_uint32_t pse		: 1;
+	    jit_uint32_t tsc		: 1;
+	    jit_uint32_t msr		: 1;
+	    jit_uint32_t pae		: 1;
+	    jit_uint32_t mce		: 1;
+	    jit_uint32_t cmpxchg8b	: 1;
+	    jit_uint32_t apic		: 1;
+	    jit_uint32_t __reserved0	: 1;
+	    jit_uint32_t sep		: 1;
+	    jit_uint32_t mtrr		: 1;
+	    jit_uint32_t pge		: 1;
+	    jit_uint32_t mca		: 1;
+	    jit_uint32_t cmov		: 1;
+	    jit_uint32_t pat		: 1;
+	    jit_uint32_t pse36		: 1;
+	    jit_uint32_t psn		: 1;	/* amd reserved */
+	    jit_uint32_t clfsh		: 1;
+	    jit_uint32_t __reserved1	: 1;
+	    jit_uint32_t ds		: 1;	/* amd reserved */
+	    jit_uint32_t acpi		: 1;	/* amd reserved */
+	    jit_uint32_t mmx		: 1;
+	    jit_uint32_t fxsr		: 1;
+	    jit_uint32_t sse		: 1;
+	    jit_uint32_t sse2		: 1;
+	    jit_uint32_t ss		: 1;	/* amd reserved */
+	    jit_uint32_t htt		: 1;
+	    jit_uint32_t tm		: 1;	/* amd reserved */
+	    jit_uint32_t __reserved2	: 1;
+	    jit_uint32_t pbe		: 1;	/* amd reserved */
+	} bits;
+	jit_uword_t	cpuid;
+    } edx;
+#if __X32
+    int			ac, flags;
+#endif
+    jit_uword_t		eax, ebx;
+
+#if __X32
+    /* adapted from glibc __sysconf */
+    __asm__ volatile ("pushfl;\n\t"
+		      "popl %0;\n\t"
+		      "movl $0x240000, %1;\n\t"
+		      "xorl %0, %1;\n\t"
+		      "pushl %1;\n\t"
+		      "popfl;\n\t"
+		      "pushfl;\n\t"
+		      "popl %1;\n\t"
+		      "xorl %0, %1;\n\t"
+		      "pushl %0;\n\t"
+		      "popfl"
+		      : "=r" (flags), "=r" (ac));
+
+    /* i386 or i486 without cpuid */
+    if ((ac & (1 << 21)) == 0)
+	/* probably without x87 as well */
+	return;
+#endif
+
+    /* query %eax = 1 function */
+#if __X32 || __X64_32
+    __asm__ volatile ("xchgl %%ebx, %1; cpuid; xchgl %%ebx, %1"
+#else
+    __asm__ volatile ("xchgq %%rbx, %1; cpuid; xchgq %%rbx, %1"
+#endif
+		      : "=a" (eax), "=r" (ebx),
+		      "=c" (ecx.cpuid), "=d" (edx.cpuid)
+		      : "0" (1));
+
+    jit_cpu.fpu		= edx.bits.fpu;
+    jit_cpu.cmpxchg8b	= edx.bits.cmpxchg8b;
+    jit_cpu.cmov	= edx.bits.cmov;
+    jit_cpu.mmx		= edx.bits.mmx;
+    jit_cpu.sse		= edx.bits.sse;
+    jit_cpu.sse2	= edx.bits.sse2;
+    jit_cpu.sse3	= ecx.bits.sse3;
+    jit_cpu.pclmulqdq	= ecx.bits.pclmulqdq;
+    jit_cpu.ssse3	= ecx.bits.ssse3;
+    jit_cpu.fma		= ecx.bits.fma;
+    jit_cpu.cmpxchg16b	= ecx.bits.cmpxchg16b;
+    jit_cpu.sse4_1	= ecx.bits.sse4_1;
+    jit_cpu.sse4_2	= ecx.bits.sse4_2;
+    jit_cpu.movbe	= ecx.bits.movbe;
+    jit_cpu.popcnt	= ecx.bits.popcnt;
+    jit_cpu.aes		= ecx.bits.aes;
+    jit_cpu.avx		= ecx.bits.avx;
+
+    /* query %eax = 0x80000001 function */
+#if __X64
+#  if __X64_32
+    __asm__ volatile ("xchgl %%ebx, %1; cpuid; xchgl %%ebx, %1"
+#  else
+    __asm__ volatile ("xchgq %%rbx, %1; cpuid; xchgq %%rbx, %1"
+#  endif
+		      : "=a" (eax), "=r" (ebx),
+		      "=c" (ecx.cpuid), "=d" (edx.cpuid)
+		      : "0" (0x80000001));
+    jit_cpu.lahf	= ecx.cpuid & 1;
+#endif
+}
+
+void
+_jit_init(jit_state_t *_jit)
+{
+#if __X32
+    jit_int32_t		regno;
+    static jit_bool_t	first = 1;
+#endif
+
+    _jitc->reglen = jit_size(_rvs) - 1;
+#if __X32
+    if (first) {
+	if (!jit_cpu.sse2) {
+	    for (regno = _jitc->reglen; regno >= 0; regno--) {
+		if (_rvs[regno].spec & jit_class_xpr)
+		    _rvs[regno].spec = 0;
+	    }
+	}
+	first = 0;
+    }
+#endif
+}
+
+void
+_jit_prolog(jit_state_t *_jit)
+{
+    jit_int32_t		offset;
+
+    if (_jitc->function)
+	jit_epilog();
+    assert(jit_regset_cmp_ui(&_jitc->regarg, 0) == 0);
+    jit_regset_set_ui(&_jitc->regsav, 0);
+    offset = _jitc->functions.offset;
+    if (offset >= _jitc->functions.length) {
+	jit_realloc((jit_pointer_t *)&_jitc->functions.ptr,
+		    _jitc->functions.length * sizeof(jit_function_t),
+		    (_jitc->functions.length + 16) * sizeof(jit_function_t));
+	_jitc->functions.length += 16;
+    }
+    _jitc->function = _jitc->functions.ptr + _jitc->functions.offset++;
+    _jitc->function->self.size = stack_framesize;
+    _jitc->function->self.argi = _jitc->function->self.argf =
+	_jitc->function->self.aoff = _jitc->function->self.alen = 0;
+    /* sse/x87 conversion */
+    _jitc->function->self.aoff = CVT_OFFSET;
+    _jitc->function->self.call = jit_call_default;
+    jit_alloc((jit_pointer_t *)&_jitc->function->regoff,
+	      _jitc->reglen * sizeof(jit_int32_t));
+
+    /* _no_link here does not mean the jit_link() call can be removed
+     * by rewriting as:
+     * _jitc->function->prolog = jit_new_node(jit_code_prolog);
+     */
+    _jitc->function->prolog = jit_new_node_no_link(jit_code_prolog);
+    jit_link(_jitc->function->prolog);
+    _jitc->function->prolog->w.w = offset;
+    _jitc->function->epilog = jit_new_node_no_link(jit_code_epilog);
+    /*	u:	label value
+     *	v:	offset in blocks vector
+     *	w:	offset in functions vector
+     */
+    _jitc->function->epilog->w.w = offset;
+
+    jit_regset_new(&_jitc->function->regset);
+}
+
+jit_int32_t
+_jit_allocai(jit_state_t *_jit, jit_int32_t length)
+{
+    assert(_jitc->function);
+    switch (length) {
+	case 0:	case 1:						break;
+	case 2:		_jitc->function->self.aoff &= -2;	break;
+	case 3:	case 4:	_jitc->function->self.aoff &= -4;	break;
+	default:	_jitc->function->self.aoff &= -8;	break;
+    }
+    _jitc->function->self.aoff -= length;
+
+    /* jit_allocai() may be called from jit_x86-cpu.c, and force a function
+     * generation restart on some conditions: div/rem and qmul/qdiv, due
+     * to registers constraints.
+     * The check is to prevent an assertion of a jit_xyz() being called
+     * during code generation, and attempting to add a node to the tail
+     * of the current IR generation. */
+    if (!_jitc->realize) {
+	jit_inc_synth_ww(allocai, _jitc->function->self.aoff, length);
+	jit_dec_synth();
+    }
+
+    return (_jitc->function->self.aoff);
+}
+
+void
+_jit_allocar(jit_state_t *_jit, jit_int32_t u, jit_int32_t v)
+{
+    jit_int32_t		 reg;
+    assert(_jitc->function);
+    jit_inc_synth_ww(allocar, u, v);
+    if (!_jitc->function->allocar) {
+	_jitc->function->aoffoff = jit_allocai(sizeof(jit_int32_t));
+	_jitc->function->allocar = 1;
+    }
+    reg = jit_get_reg(jit_class_gpr);
+    jit_negr(reg, v);
+    jit_andi(reg, reg, -16);
+    jit_ldxi_i(u, JIT_FP, _jitc->function->aoffoff);
+    jit_addr(u, u, reg);
+    jit_addr(JIT_SP, JIT_SP, reg);
+    jit_stxi_i(_jitc->function->aoffoff, JIT_FP, u);
+    jit_unget_reg(reg);
+    jit_dec_synth();
+}
+
+void
+_jit_ret(jit_state_t *_jit)
+{
+    jit_node_t		*instr;
+    assert(_jitc->function);
+    jit_inc_synth(ret);
+    /* jump to epilog */
+    instr = jit_jmpi();
+    jit_patch_at(instr, _jitc->function->epilog);
+    jit_dec_synth();
+}
+
+void
+_jit_retr(jit_state_t *_jit, jit_int32_t u)
+{
+    jit_inc_synth_w(retr, u);
+    /* movr(%ret, %ret) would be optimized out */
+    if (JIT_RET != u)
+	jit_movr(JIT_RET, u);
+    /* explicitly tell it is live */
+    jit_live(JIT_RET);
+    jit_ret();
+    jit_dec_synth();
+}
+
+void
+_jit_reti(jit_state_t *_jit, jit_word_t u)
+{
+    jit_inc_synth_w(reti, u);
+    jit_movi(JIT_RET, u);
+    jit_ret();
+    jit_dec_synth();
+}
+
+void
+_jit_retr_f(jit_state_t *_jit, jit_int32_t u)
+{
+    jit_inc_synth_w(retr_f, u);
+    if (JIT_FRET != u)
+	jit_movr_f(JIT_FRET, u);
+    else
+	jit_live(JIT_FRET);
+    jit_ret();
+    jit_dec_synth();
+}
+
+void
+_jit_reti_f(jit_state_t *_jit, jit_float32_t u)
+{
+    jit_inc_synth_f(reti_f, u);
+    jit_movi_f(JIT_FRET, u);
+    jit_ret();
+    jit_dec_synth();
+}
+
+void
+_jit_retr_d(jit_state_t *_jit, jit_int32_t u)
+{
+    jit_inc_synth_w(retr_d, u);
+    if (JIT_FRET != u)
+	jit_movr_d(JIT_FRET, u);
+    else
+	jit_live(JIT_FRET);
+    jit_ret();
+    jit_dec_synth();
+}
+
+void
+_jit_reti_d(jit_state_t *_jit, jit_float64_t u)
+{
+    jit_inc_synth_d(reti_d, u);
+    jit_movi_d(JIT_FRET, u);
+    jit_ret();
+    jit_dec_synth();
+}
+
+void
+_jit_epilog(jit_state_t *_jit)
+{
+    assert(_jitc->function);
+    assert(_jitc->function->epilog->next == NULL);
+    jit_link(_jitc->function->epilog);
+    _jitc->function = NULL;
+}
+
+jit_bool_t
+_jit_arg_register_p(jit_state_t *_jit, jit_node_t *u)
+{
+    if (u->code == jit_code_arg)
+	return (jit_arg_reg_p(u->u.w));
+    assert(u->code == jit_code_arg_f || u->code == jit_code_arg_d);
+    return (jit_arg_f_reg_p(u->u.w));
+}
+
+void
+_jit_ellipsis(jit_state_t *_jit)
+{
+    jit_inc_synth(ellipsis);
+    if (_jitc->prepare) {
+	jit_link_prepare();
+	/* Remember that a varargs function call is being constructed. */
+	assert(!(_jitc->function->call.call & jit_call_varargs));
+	_jitc->function->call.call |= jit_call_varargs;
+    }
+    else {
+	jit_link_prolog();
+	/* Remember the current function is varargs. */
+	assert(!(_jitc->function->self.call & jit_call_varargs));
+	_jitc->function->self.call |= jit_call_varargs;
+
+#if __X64 && !(__CYGWIN__ || _WIN32)
+	/* Allocate va_list like object in the stack.
+	 * If applicable, with enough space to save all argument
+	 * registers, and use fixed offsets for them. */
+	_jitc->function->vaoff = jit_allocai(sizeof(jit_va_list_t));
+
+	/* Initialize gp offset in save area. */
+	if (jit_arg_reg_p(_jitc->function->self.argi))
+	    _jitc->function->vagp = _jitc->function->self.argi * 8;
+	else
+	    _jitc->function->vagp = va_gp_max_offset;
+
+	/* Initialize fp offset in save area. */
+	if (jit_arg_f_reg_p(_jitc->function->self.argf))
+	    _jitc->function->vafp = _jitc->function->self.argf * 16 +
+				    va_gp_max_offset;
+	else
+	    _jitc->function->vafp = va_fp_max_offset;
+#endif
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_va_push(jit_state_t *_jit, jit_int32_t u)
+{
+    jit_inc_synth_w(va_push, u);
+    jit_pushargr(u);
+    jit_dec_synth();
+}
+
+jit_node_t *
+_jit_arg(jit_state_t *_jit)
+{
+    jit_node_t		*node;
+    jit_int32_t		 offset;
+    assert(_jitc->function);
+    assert(!(_jitc->function->self.call & jit_call_varargs));
+#if __X64
+    if (jit_arg_reg_p(_jitc->function->self.argi)) {
+	offset = _jitc->function->self.argi++;
+#  if __CYGWIN__ || _WIN32
+	_jitc->function->self.size += sizeof(jit_word_t);
+#  endif
+    }
+    else
+#endif
+    {
+	offset = _jitc->function->self.size;
+	_jitc->function->self.size += REAL_WORDSIZE;
+    }
+    node = jit_new_node_ww(jit_code_arg, offset,
+			   ++_jitc->function->self.argn);
+    jit_link_prolog();
+    return (node);
+}
+
+jit_node_t *
+_jit_arg_f(jit_state_t *_jit)
+{
+    jit_node_t		*node;
+    jit_int32_t		 offset;
+    assert(_jitc->function);
+    assert(!(_jitc->function->self.call & jit_call_varargs));
+#if __X64
+#  if __CYGWIN__ || _WIN32
+    if (jit_arg_reg_p(_jitc->function->self.argi)) {
+	offset = _jitc->function->self.argi++;
+	_jitc->function->self.size += sizeof(jit_word_t);
+    }
+#  else
+    if (jit_arg_f_reg_p(_jitc->function->self.argf))
+	offset = _jitc->function->self.argf++;
+#  endif
+    else
+#endif
+    {
+	offset = _jitc->function->self.size;
+	_jitc->function->self.size += REAL_WORDSIZE;
+    }
+    node = jit_new_node_ww(jit_code_arg_f, offset,
+			   ++_jitc->function->self.argn);
+    jit_link_prolog();
+    return (node);
+}
+
+jit_node_t *
+_jit_arg_d(jit_state_t *_jit)
+{
+    jit_node_t		*node;
+    jit_int32_t		 offset;
+    assert(_jitc->function);
+    assert(!(_jitc->function->self.call & jit_call_varargs));
+#if __X64
+#  if __CYGWIN__ || _WIN32
+    if (jit_arg_reg_p(_jitc->function->self.argi)) {
+	offset = _jitc->function->self.argi++;
+	_jitc->function->self.size += sizeof(jit_word_t);
+    }
+#  else
+    if (jit_arg_f_reg_p(_jitc->function->self.argf))
+	offset = _jitc->function->self.argf++;
+#  endif
+    else
+#endif
+    {
+	offset = _jitc->function->self.size;
+	_jitc->function->self.size += sizeof(jit_float64_t);
+    }
+    node = jit_new_node_ww(jit_code_arg_d, offset,
+			   ++_jitc->function->self.argn);
+    jit_link_prolog();
+    return (node);
+}
+
+void
+_jit_getarg_c(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_c, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_extr_c(u, JIT_RA0 - v->u.w);
+    else
+#endif
+	jit_ldxi_c(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_uc(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_uc, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_extr_uc(u, JIT_RA0 - v->u.w);
+    else
+#endif
+	jit_ldxi_uc(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_s(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_s, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_extr_s(u, JIT_RA0 - v->u.w);
+    else
+#endif
+	jit_ldxi_s(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_us(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_us, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_extr_us(u, JIT_RA0 - v->u.w);
+    else
+#endif
+	jit_ldxi_us(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_i(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_i, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w)) {
+#  if __X64_32
+	jit_movr(u, JIT_RA0 - v->u.w);
+#  else
+	jit_extr_i(u, JIT_RA0 - v->u.w);
+#  endif
+     }
+    else
+#endif
+	jit_ldxi_i(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+#if __X64 && !__X64_32
+void
+_jit_getarg_ui(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_ui, u, v);
+    if (jit_arg_reg_p(v->u.w))
+	jit_extr_ui(u, JIT_RA0 - v->u.w);
+    else
+	jit_ldxi_ui(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_l(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(getarg_l, u, v);
+    if (jit_arg_reg_p(v->u.w))
+	jit_movr(u, JIT_RA0 - v->u.w);
+    else
+	jit_ldxi_l(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+#endif
+
+void
+_jit_putargr(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(putargr, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_movr(JIT_RA0 - v->u.w, u);
+    else
+#endif
+	jit_stxi(v->u.w, _RBP, u);
+    jit_dec_synth();
+}
+
+void
+_jit_putargi(jit_state_t *_jit, jit_word_t u, jit_node_t *v)
+{
+    jit_int32_t		regno;
+    assert(v->code == jit_code_arg);
+    jit_inc_synth_wp(putargi, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_movi(JIT_RA0 - v->u.w, u);
+    else
+#endif
+    {
+	regno = jit_get_reg(jit_class_gpr);
+	jit_movi(regno, u);
+	jit_stxi(v->u.w, _RBP, regno);
+	jit_unget_reg(regno);
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_f(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg_f);
+    jit_inc_synth_wp(getarg_f, u, v);
+#if __X64
+    if (jit_arg_f_reg_p(v->u.w))
+	jit_movr_f(u, _XMM0 - v->u.w);
+    else
+#endif
+	jit_ldxi_f(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_putargr_f(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg_f);
+    jit_inc_synth_wp(putargr_f, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_movr_f(_XMM0 - v->u.w, u);
+    else
+#endif
+	jit_stxi_f(v->u.w, _RBP, u);
+    jit_dec_synth();
+}
+
+void
+_jit_putargi_f(jit_state_t *_jit, jit_float32_t u, jit_node_t *v)
+{
+    jit_int32_t		regno;
+    assert(v->code == jit_code_arg_f);
+    jit_inc_synth_fp(putargi_f, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_movi_f(_XMM0 - v->u.w, u);
+    else
+#endif
+    {
+	regno = jit_get_reg(jit_class_gpr);
+	jit_movi_f(regno, u);
+	jit_stxi_f(v->u.w, _RBP, regno);
+	jit_unget_reg(regno);
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_getarg_d(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg_d);
+    jit_inc_synth_wp(getarg_d, u, v);
+#if __X64
+    if (jit_arg_f_reg_p(v->u.w))
+	jit_movr_d(u, _XMM0 - v->u.w);
+    else
+#endif
+	jit_ldxi_d(u, _RBP, v->u.w);
+    jit_dec_synth();
+}
+
+void
+_jit_putargr_d(jit_state_t *_jit, jit_int32_t u, jit_node_t *v)
+{
+    assert(v->code == jit_code_arg_d);
+    jit_inc_synth_wp(putargr_d, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_movr_d(_XMM0 - v->u.w, u);
+    else
+#endif
+	jit_stxi_d(v->u.w, _RBP, u);
+    jit_dec_synth();
+}
+
+void
+_jit_putargi_d(jit_state_t *_jit, jit_float64_t u, jit_node_t *v)
+{
+    jit_int32_t		regno;
+    assert(v->code == jit_code_arg_d);
+    jit_inc_synth_dp(putargi_d, u, v);
+#if __X64
+    if (jit_arg_reg_p(v->u.w))
+	jit_movi_d(_XMM0 - v->u.w, u);
+    else
+#endif
+    {
+	regno = jit_get_reg(jit_class_gpr);
+	jit_movi_d(regno, u);
+	jit_stxi_d(v->u.w, _RBP, regno);
+	jit_unget_reg(regno);
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_pushargr(jit_state_t *_jit, jit_int32_t u)
+{
+    assert(_jitc->function);
+    jit_inc_synth_w(pushargr, u);
+    jit_link_prepare();
+#if __X64
+    if (jit_arg_reg_p(_jitc->function->call.argi)) {
+	jit_movr(JIT_RA0 - _jitc->function->call.argi, u);
+	++_jitc->function->call.argi;
+#  if __CYGWIN__ || _WIN32
+	if (_jitc->function->call.call & jit_call_varargs)
+	    jit_stxi(_jitc->function->call.size, _RSP, u);
+	_jitc->function->call.size += sizeof(jit_word_t);
+#  endif
+    }
+    else
+#endif
+    {
+	jit_stxi(_jitc->function->call.size, _RSP, u);
+	_jitc->function->call.size += REAL_WORDSIZE;
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_pushargi(jit_state_t *_jit, jit_word_t u)
+{
+    jit_int32_t		 regno;
+    assert(_jitc->function);
+    jit_inc_synth_w(pushargi, u);
+    jit_link_prepare();
+#if __X64
+    if (jit_arg_reg_p(_jitc->function->call.argi)) {
+	jit_movi(JIT_RA0 - _jitc->function->call.argi, u);
+#  if __CYGWIN__ || _WIN32
+	if (_jitc->function->call.call & jit_call_varargs)
+	    jit_stxi(_jitc->function->call.size, _RSP,
+		     JIT_RA0 - _jitc->function->call.argi);
+	_jitc->function->call.size += sizeof(jit_word_t);
+#  endif
+	++_jitc->function->call.argi;
+    }
+    else
+#endif
+    {
+	regno = jit_get_reg(jit_class_gpr);
+	jit_movi(regno, u);
+	jit_stxi(_jitc->function->call.size, _RSP, regno);
+	_jitc->function->call.size += REAL_WORDSIZE;
+	jit_unget_reg(regno);
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_pushargr_f(jit_state_t *_jit, jit_int32_t u)
+{
+    assert(_jitc->function);
+    jit_inc_synth_w(pushargr_f, u);
+    jit_link_prepare();
+#if __X64
+#  if __CYGWIN__ || _WIN32
+    if (jit_arg_reg_p(_jitc->function->call.argi)) {
+	jit_movr_f(_XMM0 - _jitc->function->call.argi, u);
+	if (_jitc->function->call.call & jit_call_varargs) {
+	    jit_stxi_f(_jitc->function->call.size, _RSP,
+		       _XMM0 - _jitc->function->call.argi);
+	    jit_ldxi_i(JIT_RA0 - _jitc->function->call.argi, _RSP,
+		       _jitc->function->call.size);
+	}
+	++_jitc->function->call.argi;
+	_jitc->function->call.size += sizeof(jit_word_t);
+    }
+#  else
+    if (jit_arg_f_reg_p(_jitc->function->self.argf)) {
+	jit_movr_f(_XMM0 - _jitc->function->call.argf, u);
+	++_jitc->function->call.argf;
+    }
+#  endif
+    else
+#endif
+    {
+	jit_stxi_f(_jitc->function->call.size, _RSP, u);
+	_jitc->function->call.size += REAL_WORDSIZE;
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_pushargi_f(jit_state_t *_jit, jit_float32_t u)
+{
+    jit_int32_t		regno;
+    assert(_jitc->function);
+    jit_inc_synth_f(pushargi_f, u);
+    jit_link_prepare();
+#if __X64
+#  if __CYGWIN__ || _WIN32
+    if (jit_arg_reg_p(_jitc->function->call.argi)) {
+	jit_movi_f(_XMM0 - _jitc->function->call.argi, u);
+	if (_jitc->function->call.call & jit_call_varargs) {
+	    jit_stxi_f(_jitc->function->call.size, _RSP,
+		       _XMM0 - _jitc->function->call.argi);
+	    jit_ldxi_i(JIT_RA0 - _jitc->function->call.argi, _RSP,
+		       _jitc->function->call.size);
+	}
+	++_jitc->function->call.argi;
+	_jitc->function->call.size += sizeof(jit_word_t);
+    }
+#  else
+    if (jit_arg_f_reg_p(_jitc->function->call.argf)) {
+	jit_movi_f(_XMM0 - _jitc->function->call.argf, u);
+	++_jitc->function->call.argf;
+    }
+#  endif
+    else
+#endif
+    {
+	regno = jit_get_reg(jit_class_fpr);
+	jit_movi_f(regno, u);
+	jit_stxi_f(_jitc->function->call.size, _RSP, regno);
+	_jitc->function->call.size += REAL_WORDSIZE;
+	jit_unget_reg(regno);
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_pushargr_d(jit_state_t *_jit, jit_int32_t u)
+{
+    assert(_jitc->function);
+    jit_inc_synth_w(pushargr_d, u);
+    jit_link_prepare();
+#if __X64
+#  if __CYGWIN__ || _WIN32
+    if (jit_arg_reg_p(_jitc->function->call.argi)) {
+	jit_movr_d(_XMM0 - _jitc->function->call.argi, u);
+	if (_jitc->function->call.call & jit_call_varargs) {
+	    jit_stxi_d(_jitc->function->call.size, _RSP,
+		       _XMM0 - _jitc->function->call.argi);
+	    jit_ldxi_l(JIT_RA0 - _jitc->function->call.argi, _RSP,
+		       _jitc->function->call.size);
+	}
+	++_jitc->function->call.argi;
+	_jitc->function->call.size += sizeof(jit_word_t);
+    }
+#  else
+    if (jit_arg_f_reg_p(_jitc->function->call.argf)) {
+	jit_movr_d(_XMM0 - _jitc->function->call.argf, u);
+	++_jitc->function->call.argf;
+    }
+#  endif
+    else
+#endif
+    {
+	jit_stxi_d(_jitc->function->call.size, _RSP, u);
+	_jitc->function->call.size += sizeof(jit_float64_t);
+    }
+    jit_dec_synth();
+}
+
+void
+_jit_pushargi_d(jit_state_t *_jit, jit_float64_t u)
+{
+    jit_int32_t		 regno;
+    assert(_jitc->function);
+    jit_inc_synth_d(pushargi_d, u);
+    jit_link_prepare();
+#if __X64
+#  if __CYGWIN__ || _WIN32
+    if (jit_arg_reg_p(_jitc->function->call.argi)) {
+	jit_movi_d(_XMM0 - _jitc->function->call.argi, u);
+	if (_jitc->function->call.call & jit_call_varargs) {
+	    jit_stxi_d(_jitc->function->call.size, _RSP,
+		       _XMM0 - _jitc->function->call.argi);
+	    jit_ldxi_l(JIT_RA0 - _jitc->function->call.argi, _RSP,
+		       _jitc->function->call.size);
+	}
+	++_jitc->function->call.argi;
+	_jitc->function->call.size += sizeof(jit_word_t);
+    }
+#  else
+    if (jit_arg_f_reg_p(_jitc->function->call.argf)) {
+	jit_movi_d(_XMM0 - _jitc->function->call.argf, u);
+	++_jitc->function->call.argf;
+    }
+#  endif
+    else
+#endif
+    {
+	regno = jit_get_reg(jit_class_fpr);
+	jit_movi_d(regno, u);
+	jit_stxi_d(_jitc->function->call.size, _RSP, regno);
+	_jitc->function->call.size += sizeof(jit_float64_t);
+	jit_unget_reg(regno);
+    }
+    jit_dec_synth();
+}
+
+jit_bool_t
+_jit_regarg_p(jit_state_t *_jit, jit_node_t *node, jit_int32_t regno)
+{
+#if __X64
+    jit_int32_t		spec;
+
+    spec = jit_class(_rvs[regno].spec);
+    if (spec & jit_class_arg) {
+	if (spec & jit_class_gpr) {
+	    regno = JIT_RA0 - regno;
+	    if (regno >= 0 && regno < node->v.w)
+		return (1);
+	}
+	else if (spec & jit_class_fpr) {
+	    regno = _XMM0 - regno;
+	    if (regno >= 0 && regno < node->w.w)
+		return (1);
+	}
+    }
+#endif
+    return (0);
+}
+
+void
+_jit_finishr(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_int32_t		 reg;
+    jit_node_t		*call;
+    assert(_jitc->function);
+    reg = r0;
+    jit_inc_synth_w(finishr, r0);
+    if (_jitc->function->self.alen < _jitc->function->call.size)
+	_jitc->function->self.alen = _jitc->function->call.size;
+#if __X64
+#  if !(__CYGWIN__ || _WIN32)
+    if (_jitc->function->call.call & jit_call_varargs) {
+	if (jit_regno(reg) == _RAX) {
+	    reg = jit_get_reg(jit_class_gpr);
+	    jit_movr(reg, _RAX);
+	}
+	if (_jitc->function->call.argf)
+	    jit_movi(_RAX, _jitc->function->call.argf);
+	else
+	    jit_movi(_RAX, 0);
+	if (reg != r0)
+	    jit_unget_reg(reg);
+    }
+#  endif
+#endif
+    call = jit_callr(reg);
+    call->v.w = _jitc->function->call.argi;
+    call->w.w = _jitc->function->call.argf;
+    _jitc->function->call.argi = _jitc->function->call.argf =
+	_jitc->function->call.size = 0;
+    _jitc->prepare = 0;
+    jit_dec_synth();
+}
+
+jit_node_t *
+_jit_finishi(jit_state_t *_jit, jit_pointer_t i0)
+{
+#if __X64
+    jit_int32_t		reg;
+#endif
+    jit_node_t		*node;
+    assert(_jitc->function);
+    jit_inc_synth_w(finishi, (jit_word_t)i0);
+    if (_jitc->function->self.alen < _jitc->function->call.size)
+	_jitc->function->self.alen = _jitc->function->call.size;
+#if __X64
+    /* FIXME preventing %rax allocation is good enough, but for consistency
+     * it should automatically detect %rax is dead, in case it has run out
+     * registers, and not save/restore it, what would be wrong if using the
+     * the return value, otherwise, just a needless noop */
+    /* >> prevent %rax from being allocated as the function pointer */
+    jit_regset_setbit(&_jitc->regarg, _RAX);
+    reg = jit_get_reg(jit_class_gpr);
+    node = jit_movi(reg, (jit_word_t)i0);
+    jit_finishr(reg);
+    jit_unget_reg(reg);
+    /* << prevent %rax from being allocated as the function pointer */
+    jit_regset_clrbit(&_jitc->regarg, _RAX);
+#else
+    node = jit_calli(i0);
+    node->v.w = _jitc->function->call.argi;
+    node->w.w = _jitc->function->call.argf;
+#endif
+    _jitc->function->call.argi = _jitc->function->call.argf =
+	_jitc->function->call.size = 0;
+    _jitc->prepare = 0;
+    jit_dec_synth();
+    return (node);
+}
+
+void
+_jit_retval_c(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_c, r0);
+    jit_extr_c(r0, JIT_RET);
+    jit_dec_synth();
+}
+
+void
+_jit_retval_uc(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_uc, r0);
+    jit_extr_uc(r0, JIT_RET);
+    jit_dec_synth();
+}
+
+void
+_jit_retval_s(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_s, r0);
+    jit_extr_s(r0, JIT_RET);
+    jit_dec_synth();
+}
+
+void
+_jit_retval_us(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_us, r0);
+    jit_extr_us(r0, JIT_RET);
+    jit_dec_synth();
+}
+
+void
+_jit_retval_i(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_i, r0);
+#if __X32 || __X64_32
+    if (r0 != JIT_RET)
+	jit_movr(r0, JIT_RET);
+#else
+    jit_extr_i(r0, JIT_RET);
+#endif
+    jit_dec_synth();
+}
+
+#if __X64 && !__X64_32
+void
+_jit_retval_ui(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_ui, r0);
+    jit_extr_ui(r0, JIT_RET);
+    jit_dec_synth();
+}
+
+void
+_jit_retval_l(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_l, r0);
+    if (r0 != JIT_RET)
+	jit_movr(r0, JIT_RET);
+    jit_dec_synth();
+}
+#endif
+
+void
+_jit_retval_f(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_f, r0);
+#if __X64
+    if (r0 != JIT_FRET)
+	jit_movr_f(r0, JIT_FRET);
+#endif
+    jit_dec_synth();
+}
+
+void
+_jit_retval_d(jit_state_t *_jit, jit_int32_t r0)
+{
+    jit_inc_synth_w(retval_d, r0);
+#if __X64
+    if (r0 != JIT_FRET)
+	jit_movr_d(r0, JIT_FRET);
+#endif
+    jit_dec_synth();
+}
+
+jit_pointer_t
+_emit_code(jit_state_t *_jit)
+{
+    jit_node_t		*node;
+    jit_node_t		*temp;
+    jit_word_t		 word;
+    jit_int32_t		 value;
+    jit_int32_t		 offset;
+    struct {
+	jit_node_t	*node;
+	jit_word_t	 word;
+#if DEVEL_DISASSEMBLER
+	jit_word_t	 prevw;
+#endif
+	jit_int32_t	 patch_offset;
+    } undo;
+#if DEVEL_DISASSEMBLER
+    jit_word_t		 prevw;
+#endif
+
+    _jitc->function = NULL;
+
+    jit_reglive_setup();
+
+    undo.word = 0;
+    undo.node = NULL;
+    undo.patch_offset = 0;
+#define case_rr(name, type)						\
+	    case jit_code_##name##r##type:				\
+		name##r##type(rn(node->u.w), rn(node->v.w));		\
+		break
+#define case_rw(name, type)						\
+	    case jit_code_##name##i##type:				\
+		name##i##type(rn(node->u.w), node->v.w);		\
+		break
+#define case_rf(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->v.w))				\
+		    x87_##name##r##type(rn(node->u.w), rn(node->v.w));	\
+		else							\
+		    sse_##name##r##type(rn(node->u.w), rn(node->v.w));	\
+		break
+#define case_fr(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->u.w))				\
+		    x87_##name##r##type(rn(node->u.w), rn(node->v.w));	\
+		else							\
+		    sse_##name##r##type(rn(node->u.w), rn(node->v.w));	\
+		break
+#define case_fw(name, type)						\
+	    case jit_code_##name##i##type:				\
+		if (jit_x87_reg_p(node->u.w))				\
+		    x87_##name##i##type(rn(node->u.w), node->v.w);	\
+		else							\
+		    sse_##name##i##type(rn(node->u.w), node->v.w);	\
+		break
+#define case_wr(name, type)						\
+	    case jit_code_##name##i##type:				\
+		name##i##type(node->u.w, rn(node->v.w));		\
+		break
+#define case_wf(name, type)						\
+	    case jit_code_##name##i##type:				\
+		if (jit_x87_reg_p(node->v.w))				\
+		    x87_##name##i##type(node->u.w, rn(node->v.w));	\
+		else							\
+		    sse_##name##i##type(node->u.w, rn(node->v.w));	\
+		break
+#define case_ff(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->u.w) &&				\
+		    jit_x87_reg_p(node->v.w))				\
+		    x87_##name##r##type(rn(node->u.w), rn(node->v.w));	\
+		else							\
+		    sse_##name##r##type(rn(node->u.w), rn(node->v.w));	\
+		break;
+#define case_rrr(name, type)						\
+	    case jit_code_##name##r##type:				\
+		name##r##type(rn(node->u.w),				\
+			      rn(node->v.w), rn(node->w.w));		\
+		break
+#define case_rrrr(name, type)						\
+	    case jit_code_##name##r##type:				\
+		name##r##type(rn(node->u.q.l), rn(node->u.q.h),		\
+			      rn(node->v.w), rn(node->w.w));		\
+		break
+#define case_frr(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->u.w))				\
+		    x87_##name##r##type(rn(node->u.w),			\
+					rn(node->v.w), rn(node->w.w));	\
+		else							\
+		    sse_##name##r##type(rn(node->u.w),			\
+					rn(node->v.w), rn(node->w.w));	\
+		break
+#define case_rrf(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->w.w))				\
+		    x87_##name##r##type(rn(node->u.w),			\
+					rn(node->v.w), rn(node->w.w));	\
+		else							\
+		    sse_##name##r##type(rn(node->u.w),			\
+					rn(node->v.w), rn(node->w.w));	\
+		break
+#define case_rrw(name, type)						\
+	    case jit_code_##name##i##type:				\
+		name##i##type(rn(node->u.w), rn(node->v.w), node->w.w);	\
+		break
+#define case_rrrw(name, type)						\
+	    case jit_code_##name##i##type:				\
+		name##i##type(rn(node->u.q.l), rn(node->u.q.h),		\
+			      rn(node->v.w), node->w.w);		\
+		break
+#define case_frw(name, type)						\
+	    case jit_code_##name##i##type:				\
+		if (jit_x87_reg_p(node->u.w))				\
+		    x87_##name##i##type(rn(node->u.w),			\
+					rn(node->v.w), node->w.w);	\
+		else							\
+		    sse_##name##i##type(rn(node->u.w),			\
+					rn(node->v.w), node->w.w);	\
+		break
+#define case_wrr(name, type)						\
+	    case jit_code_##name##i##type:				\
+		name##i##type(node->u.w, rn(node->v.w), rn(node->w.w));	\
+		break
+#define case_wrf(name, type)						\
+	    case jit_code_##name##i##type:				\
+		if (jit_x87_reg_p(node->w.w))				\
+		    x87_##name##i##type(node->u.w,			\
+					rn(node->v.w), rn(node->w.w));	\
+		else							\
+		    sse_##name##i##type(node->u.w,			\
+					rn(node->v.w), rn(node->w.w));	\
+		break
+#define case_brr(name, type)						\
+	    case jit_code_##name##r##type:				\
+		temp = node->u.n;					\
+		assert(temp->code == jit_code_label ||			\
+		       temp->code == jit_code_epilog);			\
+		if (temp->flag & jit_flag_patch)			\
+		    name##r##type(temp->u.w, rn(node->v.w),		\
+				  rn(node->w.w));			\
+		else {							\
+		    word = name##r##type(_jit->pc.w,			\
+					 rn(node->v.w), rn(node->w.w));	\
+		    patch(word, node);					\
+		}							\
+		break
+#define case_brw(name, type)						\
+	    case jit_code_##name##i##type:				\
+		temp = node->u.n;					\
+		assert(temp->code == jit_code_label ||			\
+		       temp->code == jit_code_epilog);			\
+		if (temp->flag & jit_flag_patch)			\
+		    name##i##type(temp->u.w,				\
+				  rn(node->v.w), node->w.w);		\
+		else {							\
+		    word = name##i##type(_jit->pc.w,			\
+					 rn(node->v.w), node->w.w);	\
+		    patch(word, node);					\
+		}							\
+		break
+#define case_rff(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->v.w) &&				\
+		    jit_x87_reg_p(node->w.w))				\
+		    x87_##name##r##type(rn(node->u.w), rn(node->v.w),	\
+					rn(node->w.w));			\
+		else							\
+		    sse_##name##r##type(rn(node->u.w), rn(node->v.w),	\
+					rn(node->w.w));			\
+		break;
+#define case_rfw(name, type, size)					\
+	    case jit_code_##name##i##type:				\
+		assert(node->flag & jit_flag_data);			\
+		if (jit_x87_reg_p(node->v.w))				\
+		    x87_##name##i##type(rn(node->u.w), rn(node->v.w),	\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		else							\
+		    sse_##name##i##type(rn(node->u.w), rn(node->v.w),	\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		break
+#define case_fff(name, type)						\
+	    case jit_code_##name##r##type:				\
+		if (jit_x87_reg_p(node->u.w) &&				\
+		    jit_x87_reg_p(node->v.w) &&				\
+		    jit_x87_reg_p(node->w.w))				\
+		    x87_##name##r##type(rn(node->u.w),			\
+					rn(node->v.w), rn(node->w.w));	\
+		else							\
+		    sse_##name##r##type(rn(node->u.w),			\
+					rn(node->v.w), rn(node->w.w));	\
+		break
+#define case_ffw(name, type, size)					\
+	    case jit_code_##name##i##type:				\
+		assert(node->flag & jit_flag_data);			\
+		if (jit_x87_reg_p(node->u.w) &&				\
+		    jit_x87_reg_p(node->v.w))				\
+		    x87_##name##i##type(rn(node->u.w), rn(node->v.w),	\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		else							\
+		    sse_##name##i##type(rn(node->u.w), rn(node->v.w),	\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		break
+#define case_bff(name, type)						\
+	    case jit_code_b##name##r##type:				\
+		temp = node->u.n;					\
+		assert(temp->code == jit_code_label ||			\
+		       temp->code == jit_code_epilog);			\
+		if (temp->flag & jit_flag_patch) {			\
+		    if (jit_x87_reg_p(node->v.w) &&			\
+			jit_x87_reg_p(node->w.w))			\
+			x87_b##name##r##type(temp->u.w,			\
+				rn(node->v.w), rn(node->w.w));		\
+		    else						\
+			sse_b##name##r##type(temp->u.w,			\
+				rn(node->v.w), rn(node->w.w));		\
+		}							\
+		else {							\
+		    if (jit_x87_reg_p(node->v.w) &&			\
+			jit_x87_reg_p(node->w.w))			\
+			word = x87_b##name##r##type(_jit->pc.w,		\
+				rn(node->v.w), rn(node->w.w));		\
+		    else						\
+			word = sse_b##name##r##type(_jit->pc.w,		\
+				rn(node->v.w), rn(node->w.w));		\
+		    patch(word, node);					\
+		}							\
+		break
+#define case_bfw(name, type, size)					\
+	    case jit_code_b##name##i##type:				\
+		temp = node->u.n;					\
+		assert(temp->code == jit_code_label ||			\
+		       temp->code == jit_code_epilog);			\
+		if (temp->flag & jit_flag_patch) {			\
+		    if (jit_x87_reg_p(node->v.w))			\
+			x87_b##name##i##type(temp->u.w,			\
+				rn(node->v.w),				\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		    else						\
+			sse_b##name##i##type(temp->u.w,			\
+				rn(node->v.w),				\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		}							\
+		else {							\
+		    if (jit_x87_reg_p(node->v.w))			\
+			word = x87_b##name##i##type(_jit->pc.w,		\
+				rn(node->v.w),				\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		    else						\
+			word = sse_b##name##i##type(_jit->pc.w,		\
+				rn(node->v.w),				\
+				(jit_float##size##_t *)node->w.n->u.w);	\
+		    patch(word, node);					\
+		}							\
+		break
+#if DEVEL_DISASSEMBLER
+    prevw = _jit->pc.w;
+#endif
+    for (node = _jitc->head; node; node = node->next) {
+	if (_jit->pc.uc >= _jitc->code.end)
+	    return (NULL);
+
+#if DEVEL_DISASSEMBLER
+	node->offset = (jit_uword_t)_jit->pc.w - (jit_uword_t)prevw;
+	prevw = _jit->pc.w;
+#endif
+	value = jit_classify(node->code);
+	jit_regarg_set(node, value);
+	switch (node->code) {
+	    case jit_code_align:
+		assert(!(node->u.w & (node->u.w - 1)) &&
+		       node->u.w <= sizeof(jit_word_t));
+		if ((word = _jit->pc.w & (node->u.w - 1)))
+		    nop(node->u.w - word);
+		break;
+	    case jit_code_note:		case jit_code_name:
+		node->u.w = _jit->pc.w;
+		break;
+	    case jit_code_label:
+		if ((node->link || (node->flag & jit_flag_use)) &&
+		    (word = _jit->pc.w & (sizeof(jit_word_t) - 1)))
+		    nop(sizeof(jit_word_t) - word);
+		/* remember label is defined */
+		node->flag |= jit_flag_patch;
+		node->u.w = _jit->pc.w;
+		break;
+		case_rrr(add,);
+		case_rrw(add,);
+		case_rrr(addx,);
+		case_rrw(addx,);
+		case_rrr(addc,);
+		case_rrw(addc,);
+		case_rrr(sub,);
+		case_rrw(sub,);
+		case_rrr(subx,);
+		case_rrw(subx,);
+		case_rrr(subc,);
+		case_rrw(subc,);
+		case_rrw(rsb,);
+		case_rrr(mul,);
+		case_rrw(mul,);
+		case_rrrr(qmul,);
+		case_rrrw(qmul,);
+		case_rrrr(qmul, _u);
+		case_rrrw(qmul, _u);
+		case_rrr(div,);
+		case_rrw(div,);
+		case_rrr(div, _u);
+		case_rrw(div, _u);
+		case_rrrr(qdiv,);
+		case_rrrw(qdiv,);
+		case_rrrr(qdiv, _u);
+		case_rrrw(qdiv, _u);
+		case_rrr(rem,);
+		case_rrw(rem,);
+		case_rrr(rem, _u);
+		case_rrw(rem, _u);
+		case_rrr(and,);
+		case_rrw(and,);
+		case_rrr(or,);
+		case_rrw(or,);
+		case_rrr(xor,);
+		case_rrw(xor,);
+		case_rrr(lsh,);
+		case_rrw(lsh,);
+		case_rrr(rsh,);
+		case_rrw(rsh,);
+		case_rrr(rsh, _u);
+		case_rrw(rsh, _u);
+		case_rr(neg,);
+		case_rr(com,);
+		case_rrr(lt,);
+		case_rrw(lt,);
+		case_rrr(lt, _u);
+		case_rrw(lt, _u);
+		case_rrr(le,);
+		case_rrw(le,);
+		case_rrr(le, _u);
+		case_rrw(le, _u);
+		case_rrr(eq,);
+		case_rrw(eq,);
+		case_rrr(ge,);
+		case_rrw(ge,);
+		case_rrr(ge, _u);
+		case_rrw(ge, _u);
+		case_rrr(gt,);
+		case_rrw(gt,);
+		case_rrr(gt, _u);
+		case_rrw(gt, _u);
+		case_rrr(ne,);
+		case_rrw(ne,);
+		case_rr(mov,);
+	    case jit_code_movi:
+		if (node->flag & jit_flag_node) {
+		    temp = node->v.n;
+		    if (temp->code == jit_code_data ||
+			(temp->code == jit_code_label &&
+			 (temp->flag & jit_flag_patch)))
+			movi(rn(node->u.w), temp->u.w);
+		    else {
+			assert(temp->code == jit_code_label ||
+			       temp->code == jit_code_epilog);
+			word = movi_p(rn(node->u.w), node->v.w);
+			patch(word, node);
+		    }
+		}
+		else
+		    movi(rn(node->u.w), node->v.w);
+		break;
+		case_rr(hton, _us);
+		case_rr(hton, _ui);
+#if __X64 && !__X64_32
+		case_rr(hton, _ul);
+#endif
+		case_rr(ext, _c);
+		case_rr(ext, _uc);
+		case_rr(ext, _s);
+		case_rr(ext, _us);
+#if __X64 && !__X64_32
+		case_rr(ext, _i);
+		case_rr(ext, _ui);
+#endif
+		case_rf(trunc, _f_i);
+		case_rf(trunc, _d_i);
+#if __X64
+		case_rf(trunc, _f_l);
+		case_rf(trunc, _d_l);
+#endif
+		case_rr(ld, _c);
+		case_rw(ld, _c);
+		case_rr(ld, _uc);
+		case_rw(ld, _uc);
+		case_rr(ld, _s);
+		case_rw(ld, _s);
+		case_rr(ld, _us);
+		case_rw(ld, _us);
+		case_rr(ld, _i);
+		case_rw(ld, _i);
+#if __X64 && !__X64_32
+		case_rr(ld, _ui);
+		case_rw(ld, _ui);
+		case_rr(ld, _l);
+		case_rw(ld, _l);
+#endif
+		case_rrr(ldx, _c);
+		case_rrw(ldx, _c);
+		case_rrr(ldx, _uc);
+		case_rrw(ldx, _uc);
+		case_rrr(ldx, _s);
+		case_rrw(ldx, _s);
+		case_rrr(ldx, _us);
+		case_rrw(ldx, _us);
+		case_rrr(ldx, _i);
+		case_rrw(ldx, _i);
+#if __X64 && !__X64_32
+		case_rrr(ldx, _ui);
+		case_rrw(ldx, _ui);
+		case_rrr(ldx, _l);
+		case_rrw(ldx, _l);
+#endif
+		case_rr(st, _c);
+		case_wr(st, _c);
+		case_rr(st, _s);
+		case_wr(st, _s);
+		case_rr(st, _i);
+		case_wr(st, _i);
+#if __X64 && !__X64_32
+		case_rr(st, _l);
+		case_wr(st, _l);
+#endif
+		case_rrr(stx, _c);
+		case_wrr(stx, _c);
+		case_rrr(stx, _s);
+		case_wrr(stx, _s);
+		case_rrr(stx, _i);
+		case_wrr(stx, _i);
+#if __X64 && !__X64_32
+		case_rrr(stx, _l);
+		case_wrr(stx, _l);
+#endif
+		case_brr(blt,);
+		case_brw(blt,);
+		case_brr(blt, _u);
+		case_brw(blt, _u);
+		case_brr(ble,);
+		case_brw(ble,);
+		case_brr(ble, _u);
+		case_brw(ble, _u);
+		case_brr(beq,);
+		case_brw(beq,);
+		case_brr(bge,);
+		case_brw(bge,);
+		case_brr(bge, _u);
+		case_brw(bge, _u);
+		case_brr(bgt,);
+		case_brw(bgt,);
+		case_brr(bgt, _u);
+		case_brw(bgt, _u);
+		case_brr(bne,);
+		case_brw(bne,);
+		case_brr(bms,);
+		case_brw(bms,);
+		case_brr(bmc,);
+		case_brw(bmc,);
+		case_brr(boadd,);
+		case_brw(boadd,);
+		case_brr(boadd, _u);
+		case_brw(boadd, _u);
+		case_brr(bxadd,);
+		case_brw(bxadd,);
+		case_brr(bxadd, _u);
+		case_brw(bxadd, _u);
+		case_brr(bosub,);
+		case_brw(bosub,);
+		case_brr(bosub, _u);
+		case_brw(bosub, _u);
+		case_brr(bxsub,);
+		case_brw(bxsub,);
+		case_brr(bxsub, _u);
+		case_brw(bxsub, _u);
+		case_fff(add, _f);
+		case_ffw(add, _f, 32);
+		case_fff(sub, _f);
+		case_ffw(sub, _f, 32);
+		case_ffw(rsb, _f, 32);
+		case_fff(mul, _f);
+		case_ffw(mul, _f, 32);
+		case_fff(div, _f);
+		case_ffw(div, _f, 32);
+		case_ff(abs, _f);
+		case_ff(neg, _f);
+		case_ff(sqrt, _f);
+		case_fr(ext, _f);
+		case_fr(ext, _d_f);
+		case_rff(lt, _f);
+		case_rfw(lt, _f, 32);
+		case_rff(le, _f);
+		case_rfw(le, _f, 32);
+		case_rff(eq, _f);
+		case_rfw(eq, _f, 32);
+		case_rff(ge, _f);
+		case_rfw(ge, _f, 32);
+		case_rff(gt, _f);
+		case_rfw(gt, _f, 32);
+		case_rff(ne, _f);
+		case_rfw(ne, _f, 32);
+		case_rff(unlt, _f);
+		case_rfw(unlt, _f, 32);
+		case_rff(unle, _f);
+		case_rfw(unle, _f, 32);
+		case_rff(uneq, _f);
+		case_rfw(uneq, _f, 32);
+		case_rff(unge, _f);
+		case_rfw(unge, _f, 32);
+		case_rff(ungt, _f);
+		case_rfw(ungt, _f, 32);
+		case_rff(ltgt, _f);
+		case_rfw(ltgt, _f, 32);
+		case_rff(ord, _f);
+		case_rfw(ord, _f, 32);
+		case_rff(unord, _f);
+		case_rfw(unord, _f, 32);
+	    case jit_code_movr_f:
+		if (jit_x87_reg_p(node->u.w)) {
+		    if (jit_x87_reg_p(node->v.w))
+			x87_movr_f(rn(node->u.w), rn(node->v.w));
+		    else
+			x87_from_sse_f(rn(node->u.w), rn(node->v.w));
+		}
+		else {
+		    if (jit_sse_reg_p(node->v.w))
+			sse_movr_f(rn(node->u.w), rn(node->v.w));
+		    else
+			sse_from_x87_f(rn(node->u.w), rn(node->v.w));
+		}
+		break;
+	    case jit_code_movi_f:
+		assert(node->flag & jit_flag_data);
+		if (jit_x87_reg_p(node->u.w))
+		    x87_movi_f(rn(node->u.w), (jit_float32_t *)node->v.n->u.w);
+		else
+		    sse_movi_f(rn(node->u.w), (jit_float32_t *)node->v.n->u.w);
+		break;
+		case_fr(ld, _f);
+		case_fw(ld, _f);
+		case_frr(ldx, _f);
+		case_frw(ldx, _f);
+		case_rf(st, _f);
+		case_wf(st, _f);
+		case_rrf(stx, _f);
+		case_wrf(stx, _f);
+		case_bff(lt, _f);
+		case_bfw(lt, _f, 32);
+		case_bff(le, _f);
+		case_bfw(le, _f, 32);
+		case_bff(eq, _f);
+		case_bfw(eq, _f, 32);
+		case_bff(ge, _f);
+		case_bfw(ge, _f, 32);
+		case_bff(gt, _f);
+		case_bfw(gt, _f, 32);
+		case_bff(ne, _f);
+		case_bfw(ne, _f, 32);
+		case_bff(unlt, _f);
+		case_bfw(unlt, _f, 32);
+		case_bff(unle, _f);
+		case_bfw(unle, _f, 32);
+		case_bff(uneq, _f);
+		case_bfw(uneq, _f, 32);
+		case_bff(unge, _f);
+		case_bfw(unge, _f, 32);
+		case_bff(ungt, _f);
+		case_bfw(ungt, _f, 32);
+		case_bff(ltgt, _f);
+		case_bfw(ltgt, _f, 32);
+		case_bff(ord, _f);
+		case_bfw(ord, _f, 32);
+		case_bff(unord, _f);
+		case_bfw(unord, _f, 32);
+		case_fff(add, _d);
+		case_ffw(add, _d, 64);
+		case_fff(sub, _d);
+		case_ffw(sub, _d, 64);
+		case_ffw(rsb, _d, 64);
+		case_fff(mul, _d);
+		case_ffw(mul, _d, 64);
+		case_fff(div, _d);
+		case_ffw(div, _d, 64);
+		case_ff(abs, _d);
+		case_ff(neg, _d);
+		case_ff(sqrt, _d);
+		case_fr(ext, _d);
+		case_fr(ext, _f_d);
+		case_rff(lt, _d);
+		case_rfw(lt, _d, 64);
+		case_rff(le, _d);
+		case_rfw(le, _d, 64);
+		case_rff(eq, _d);
+		case_rfw(eq, _d, 64);
+		case_rff(ge, _d);
+		case_rfw(ge, _d, 64);
+		case_rff(gt, _d);
+		case_rfw(gt, _d, 64);
+		case_rff(ne, _d);
+		case_rfw(ne, _d, 64);
+		case_rff(unlt, _d);
+		case_rfw(unlt, _d, 64);
+		case_rff(unle, _d);
+		case_rfw(unle, _d, 64);
+		case_rff(uneq, _d);
+		case_rfw(uneq, _d, 64);
+		case_rff(unge, _d);
+		case_rfw(unge, _d, 64);
+		case_rff(ungt, _d);
+		case_rfw(ungt, _d, 64);
+		case_rff(ltgt, _d);
+		case_rfw(ltgt, _d, 64);
+		case_rff(ord, _d);
+		case_rfw(ord, _d, 64);
+		case_rff(unord, _d);
+		case_rfw(unord, _d, 64);
+	    case jit_code_movr_d:
+		if (jit_x87_reg_p(node->u.w)) {
+		    if (jit_x87_reg_p(node->v.w))
+			x87_movr_d(rn(node->u.w), rn(node->v.w));
+		    else
+			x87_from_sse_d(rn(node->u.w), rn(node->v.w));
+		}
+		else {
+		    if (jit_sse_reg_p(node->v.w))
+			sse_movr_d(rn(node->u.w), rn(node->v.w));
+		    else
+			sse_from_x87_d(rn(node->u.w), rn(node->v.w));
+		}
+		break;
+	    case jit_code_movi_d:
+		assert(node->flag & jit_flag_data);
+		if (jit_x87_reg_p(node->u.w))
+		    x87_movi_d(rn(node->u.w), (jit_float64_t *)node->v.n->u.w);
+		else
+		    sse_movi_d(rn(node->u.w), (jit_float64_t *)node->v.n->u.w);
+		break;
+		case_fr(ld, _d);
+		case_fw(ld, _d);
+		case_frr(ldx, _d);
+		case_frw(ldx, _d);
+		case_rf(st, _d);
+		case_wf(st, _d);
+		case_rrf(stx, _d);
+		case_wrf(stx, _d);
+		case_bff(lt, _d);
+		case_bfw(lt, _d, 64);
+		case_bff(le, _d);
+		case_bfw(le, _d, 64);
+		case_bff(eq, _d);
+		case_bfw(eq, _d, 64);
+		case_bff(ge, _d);
+		case_bfw(ge, _d, 64);
+		case_bff(gt, _d);
+		case_bfw(gt, _d, 64);
+		case_bff(ne, _d);
+		case_bfw(ne, _d, 64);
+		case_bff(unlt, _d);
+		case_bfw(unlt, _d, 64);
+		case_bff(unle, _d);
+		case_bfw(unle, _d, 64);
+		case_bff(uneq, _d);
+		case_bfw(uneq, _d, 64);
+		case_bff(unge, _d);
+		case_bfw(unge, _d, 64);
+		case_bff(ungt, _d);
+		case_bfw(ungt, _d, 64);
+		case_bff(ltgt, _d);
+		case_bfw(ltgt, _d, 64);
+		case_bff(ord, _d);
+		case_bfw(ord, _d, 64);
+		case_bff(unord, _d);
+		case_bfw(unord, _d, 64);
+	    case jit_code_jmpr:
+		jmpr(rn(node->u.w));
+		break;
+	    case jit_code_jmpi:
+		if (node->flag & jit_flag_node) {
+		    temp = node->u.n;
+		    assert(temp->code == jit_code_label ||
+			   temp->code == jit_code_epilog);
+		    if (temp->flag & jit_flag_patch)
+			jmpi(temp->u.w);
+		    else {
+			word = jmpi(_jit->pc.w);
+			patch(word, node);
+		    }
+		}
+		else
+		    jmpi(node->u.w);
+		break;
+	    case jit_code_callr:
+		callr(rn(node->u.w));
+		break;
+	    case jit_code_calli:
+		if (node->flag & jit_flag_node) {
+		    temp = node->u.n;
+		    assert(temp->code == jit_code_label ||
+			   temp->code == jit_code_epilog);
+		    word = calli(temp->u.w);
+		    if (!(temp->flag & jit_flag_patch))
+			patch(word, node);
+		}
+		else
+		    calli(node->u.w);
+		break;
+	    case jit_code_prolog:
+		_jitc->function = _jitc->functions.ptr + node->w.w;
+		undo.node = node;
+		undo.word = _jit->pc.w;
+#if DEVEL_DISASSEMBLER
+		undo.prevw = prevw;
+#endif
+		undo.patch_offset = _jitc->patches.offset;
+	    restart_function:
+		_jitc->again = 0;
+		prolog(node);
+		break;
+	    case jit_code_epilog:
+		assert(_jitc->function == _jitc->functions.ptr + node->w.w);
+		if (_jitc->again) {
+		    for (temp = undo.node->next;
+			 temp != node; temp = temp->next) {
+			if (temp->code == jit_code_label ||
+			    temp->code == jit_code_epilog)
+			    temp->flag &= ~jit_flag_patch;
+		    }
+		    temp->flag &= ~jit_flag_patch;
+		    node = undo.node;
+		    _jit->pc.w = undo.word;
+#if DEVEL_DISASSEMBLER
+		    prevw = undo.prevw;
+#endif
+		    _jitc->patches.offset = undo.patch_offset;
+		    goto restart_function;
+		}
+		if (node->link &&
+		    (word = _jit->pc.w & (sizeof(jit_word_t) - 1)))
+		    nop(sizeof(jit_word_t) - word);
+		/* remember label is defined */
+		node->flag |= jit_flag_patch;
+		node->u.w = _jit->pc.w;
+		epilog(node);
+		_jitc->function = NULL;
+		break;
+	    case jit_code_va_start:
+		vastart(rn(node->u.w));
+		break;
+	    case jit_code_va_arg:
+		vaarg(rn(node->u.w), rn(node->v.w));
+		break;
+	    case jit_code_va_arg_d:
+		vaarg_d(rn(node->u.w), rn(node->v.w), jit_x87_reg_p(node->u.w));
+		break;
+	    case jit_code_live:			case jit_code_ellipsis:
+	    case jit_code_va_push:
+	    case jit_code_allocai:		case jit_code_allocar:
+	    case jit_code_arg:
+	    case jit_code_arg_f:		case jit_code_arg_d:
+	    case jit_code_va_end:
+	    case jit_code_ret:
+	    case jit_code_retr:			case jit_code_reti:
+	    case jit_code_retr_f:		case jit_code_reti_f:
+	    case jit_code_retr_d:		case jit_code_reti_d:
+	    case jit_code_getarg_c:		case jit_code_getarg_uc:
+	    case jit_code_getarg_s:		case jit_code_getarg_us:
+	    case jit_code_getarg_i:
+#if __X64 && !__X64_32
+	    case jit_code_getarg_ui:		case jit_code_getarg_l:
+#endif
+	    case jit_code_getarg_f:		case jit_code_getarg_d:
+	    case jit_code_putargr:		case jit_code_putargi:
+	    case jit_code_putargr_f:		case jit_code_putargi_f:
+	    case jit_code_putargr_d:		case jit_code_putargi_d:
+	    case jit_code_pushargr:		case jit_code_pushargi:
+	    case jit_code_pushargr_f:		case jit_code_pushargi_f:
+	    case jit_code_pushargr_d:		case jit_code_pushargi_d:
+	    case jit_code_retval_c:		case jit_code_retval_uc:
+	    case jit_code_retval_s:		case jit_code_retval_us:
+	    case jit_code_retval_i:
+#if __X64 && !__X32
+	    case jit_code_retval_ui:		case jit_code_retval_l:
+#endif
+	    case jit_code_prepare:
+	    case jit_code_finishr:		case jit_code_finishi:
+		break;
+	    case jit_code_retval_f:
+#if __X32
+		if (jit_sse_reg_p(node->u.w)) {
+		    fstpr(_ST1_REGNO);
+		    sse_from_x87_f(rn(node->u.w), _ST0_REGNO);
+		}
+		else
+		    fstpr(rn(node->u.w) + 1);
+#endif
+		break;
+	    case jit_code_retval_d:
+#if __X32
+		if (jit_sse_reg_p(node->u.w)) {
+		    fstpr(_ST1_REGNO);
+		    sse_from_x87_d(rn(node->u.w), _ST0_REGNO);
+		}
+		else
+		    fstpr(rn(node->u.w) + 1);
+#endif
+		break;
+	    default:
+		abort();
+	}
+	jit_regarg_clr(node, value);
+	assert(_jitc->regarg == 0 && _jitc->synth == 0);
+	/* update register live state */
+	jit_reglive(node);
+    }
+#undef case_bfw
+#undef case_bff
+#undef case_ffw
+#undef case_rfw
+#undef case_rff
+#undef case_brw
+#undef case_brr
+#undef case_wrf
+#undef case_wrr
+#undef case_frw
+#undef case_rrf
+#undef case_rrw
+#undef case_frr
+#undef case_rrr
+#undef case_wf
+#undef case_fw
+#undef case_fr
+#undef case_rr
+
+    for (offset = 0; offset < _jitc->patches.offset; offset++) {
+	node = _jitc->patches.ptr[offset].node;
+	word = node->code == jit_code_movi ? node->v.n->u.w : node->u.n->u.w;
+	patch_at(node, _jitc->patches.ptr[offset].inst, word);
+    }
+
+    jit_flush(_jit->code.ptr, _jit->pc.uc);
+
+    return (_jit->code.ptr);
+}
+
+#define CODE				1
+#  include "jit_x86-cpu.c"
+#  include "jit_x86-sse.c"
+#  include "jit_x86-x87.c"
+#undef CODE
+
+void
+jit_flush(void *fptr, void *tptr)
+{
+}
+
+void
+_emit_ldxi(jit_state_t *_jit, jit_gpr_t r0, jit_gpr_t r1, jit_word_t i0)
+{
+    ldxi(rn(r0), rn(r1), i0);
+}
+
+void
+_emit_stxi(jit_state_t *_jit, jit_word_t i0, jit_gpr_t r0, jit_gpr_t r1)
+{
+    stxi(i0, rn(r0), rn(r1));
+}
+
+void
+_emit_ldxi_d(jit_state_t *_jit, jit_fpr_t r0, jit_gpr_t r1, jit_word_t i0)
+{
+    if (jit_x87_reg_p(r0))
+	x87_ldxi_d(rn(r0), rn(r1), i0);
+    else
+	sse_ldxi_d(rn(r0), rn(r1), i0);
+}
+
+void
+_emit_stxi_d(jit_state_t *_jit, jit_word_t i0, jit_gpr_t r0, jit_fpr_t r1)
+{
+    if (jit_x87_reg_p(r1))
+	x87_stxi_d(i0, rn(r0), rn(r1));
+    else
+	sse_stxi_d(i0, rn(r0), rn(r1));
+}
+
+static void
+_patch(jit_state_t *_jit, jit_word_t instr, jit_node_t *node)
+{
+    jit_int32_t		flag;
+
+    assert(node->flag & jit_flag_node);
+    if (node->code == jit_code_movi)
+	flag = node->v.n->flag;
+    else
+	flag = node->u.n->flag;
+    assert(!(flag & jit_flag_patch));
+    if (_jitc->patches.offset >= _jitc->patches.length) {
+	jit_realloc((jit_pointer_t *)&_jitc->patches.ptr,
+		    _jitc->patches.length * sizeof(jit_patch_t),
+		    (_jitc->patches.length + 1024) * sizeof(jit_patch_t));
+	_jitc->patches.length += 1024;
+    }
+    _jitc->patches.ptr[_jitc->patches.offset].inst = instr;
+    _jitc->patches.ptr[_jitc->patches.offset].node = node;
+    ++_jitc->patches.offset;
+}
+
+static void
+_sse_from_x87_f(jit_state_t *_jit, jit_int32_t r0, jit_int32_t r1)
+{
+    x87_stxi_f(CVT_OFFSET, _RBP_REGNO, r1);
+    sse_ldxi_f(r0, _RBP_REGNO, CVT_OFFSET);
+}
+
+static void
+_sse_from_x87_d(jit_state_t *_jit, jit_int32_t r0, jit_int32_t r1)
+{
+    x87_stxi_d(CVT_OFFSET, _RBP_REGNO, r1);
+    sse_ldxi_d(r0, _RBP_REGNO, CVT_OFFSET);
+}
+
+static void
+_x87_from_sse_f(jit_state_t *_jit, jit_int32_t r0, jit_int32_t r1)
+{
+    sse_stxi_f(CVT_OFFSET, _RBP_REGNO, r1);
+    x87_ldxi_f(r0, _RBP_REGNO, CVT_OFFSET);
+}
+
+static void
+_x87_from_sse_d(jit_state_t *_jit, jit_int32_t r0, jit_int32_t r1)
+{
+    sse_stxi_d(CVT_OFFSET, _RBP_REGNO, r1);
+    x87_ldxi_d(r0, _RBP_REGNO, CVT_OFFSET);
+}