separated helpers from micro operations

git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@204 c046a42c-6fe2-441c-8c8c-71466251a162

3 files changed, 908 insertions, 841 deletions

diff --git a/exec-i386.h b/exec-i386.h
index af82370032..d89a318cc3 100644
--- a/exec-i386.h
+++ b/exec-i386.h
@@ -190,14 +190,21 @@ register struct CPUX86State *env asm("r27");
 #endif
 
 #ifdef __alpha__
+/* the symbols are considered non exported so a br immediate is generated */
+#define __hidden __attribute__((visibility("hidden")))
+#else
+#define __hidden 
+#endif
+
+#ifdef __alpha__
 /* Suggested by Richard Henderson. This will result in code like
         ldah $0,__op_param1($29)        !gprelhigh
         lda $0,__op_param1($0)          !gprellow
    We can then conveniently change $29 to $31 and adapt the offsets to
    emit the appropriate constant.  */
-extern int __op_param1 __attribute__((visibility("hidden")));
-extern int __op_param2 __attribute__((visibility("hidden")));
-extern int __op_param3 __attribute__((visibility("hidden")));
+extern int __op_param1 __hidden;
+extern int __op_param2 __hidden;
+extern int __op_param3 __hidden;
 #define PARAM1 ({ int _r; asm("" : "=r"(_r) : "0" (&__op_param1)); _r; })
 #define PARAM2 ({ int _r; asm("" : "=r"(_r) : "0" (&__op_param2)); _r; })
 #define PARAM3 ({ int _r; asm("" : "=r"(_r) : "0" (&__op_param3)); _r; })
@@ -220,15 +227,173 @@ typedef struct CCTable {
 extern CCTable cc_table[];
 
 void load_seg(int seg_reg, int selector, unsigned cur_eip);
-void cpu_lock(void);
-void cpu_unlock(void);
+void __hidden cpu_lock(void);
+void __hidden cpu_unlock(void);
 void raise_interrupt(int intno, int is_int, int error_code, 
                      unsigned int next_eip);
 void raise_exception_err(int exception_index, int error_code);
 void raise_exception(int exception_index);
-void cpu_loop_exit(void);
+void __hidden cpu_loop_exit(void);
 void helper_fsave(uint8_t *ptr, int data32);
 void helper_frstor(uint8_t *ptr, int data32);
 
 void OPPROTO op_movl_eflags_T0(void);
 void OPPROTO op_movl_T0_eflags(void);
+void raise_interrupt(int intno, int is_int, int error_code, 
+                     unsigned int next_eip);
+void raise_exception_err(int exception_index, int error_code);
+void raise_exception(int exception_index);
+void helper_cpuid(void);
+void helper_lsl(void);
+void helper_lar(void);
+
+
+#ifdef USE_X86LDOUBLE
+/* use long double functions */
+#define lrint lrintl
+#define llrint llrintl
+#define fabs fabsl
+#define sin sinl
+#define cos cosl
+#define sqrt sqrtl
+#define pow powl
+#define log logl
+#define tan tanl
+#define atan2 atan2l
+#define floor floorl
+#define ceil ceill
+#define rint rintl
+#endif
+
+extern int lrint(CPU86_LDouble x);
+extern int64_t llrint(CPU86_LDouble x);
+extern CPU86_LDouble fabs(CPU86_LDouble x);
+extern CPU86_LDouble sin(CPU86_LDouble x);
+extern CPU86_LDouble cos(CPU86_LDouble x);
+extern CPU86_LDouble sqrt(CPU86_LDouble x);
+extern CPU86_LDouble pow(CPU86_LDouble, CPU86_LDouble);
+extern CPU86_LDouble log(CPU86_LDouble x);
+extern CPU86_LDouble tan(CPU86_LDouble x);
+extern CPU86_LDouble atan2(CPU86_LDouble, CPU86_LDouble);
+extern CPU86_LDouble floor(CPU86_LDouble x);
+extern CPU86_LDouble ceil(CPU86_LDouble x);
+extern CPU86_LDouble rint(CPU86_LDouble x);
+
+#define RC_MASK         0xc00
+#define RC_NEAR		0x000
+#define RC_DOWN		0x400
+#define RC_UP		0x800
+#define RC_CHOP		0xc00
+
+#define MAXTAN 9223372036854775808.0
+
+#ifdef USE_X86LDOUBLE
+
+/* only for x86 */
+typedef union {
+    long double d;
+    struct {
+        unsigned long long lower;
+        unsigned short upper;
+    } l;
+} CPU86_LDoubleU;
+
+/* the following deal with x86 long double-precision numbers */
+#define MAXEXPD 0x7fff
+#define EXPBIAS 16383
+#define EXPD(fp)	(fp.l.upper & 0x7fff)
+#define SIGND(fp)	((fp.l.upper) & 0x8000)
+#define MANTD(fp)       (fp.l.lower)
+#define BIASEXPONENT(fp) fp.l.upper = (fp.l.upper & ~(0x7fff)) | EXPBIAS
+
+#else
+
+typedef union {
+    double d;
+#ifndef WORDS_BIGENDIAN
+    struct {
+        uint32_t lower;
+        int32_t upper;
+    } l;
+#else
+    struct {
+        int32_t upper;
+        uint32_t lower;
+    } l;
+#endif
+    int64_t ll;
+} CPU86_LDoubleU;
+
+/* the following deal with IEEE double-precision numbers */
+#define MAXEXPD 0x7ff
+#define EXPBIAS 1023
+#define EXPD(fp)	(((fp.l.upper) >> 20) & 0x7FF)
+#define SIGND(fp)	((fp.l.upper) & 0x80000000)
+#define MANTD(fp)	(fp.ll & ((1LL << 52) - 1))
+#define BIASEXPONENT(fp) fp.l.upper = (fp.l.upper & ~(0x7ff << 20)) | (EXPBIAS << 20)
+#endif
+
+static inline void fpush(void)
+{
+    env->fpstt = (env->fpstt - 1) & 7;
+    env->fptags[env->fpstt] = 0; /* validate stack entry */
+}
+
+static inline void fpop(void)
+{
+    env->fptags[env->fpstt] = 1; /* invvalidate stack entry */
+    env->fpstt = (env->fpstt + 1) & 7;
+}
+
+#ifndef USE_X86LDOUBLE
+static inline CPU86_LDouble helper_fldt(uint8_t *ptr)
+{
+    CPU86_LDoubleU temp;
+    int upper, e;
+    /* mantissa */
+    upper = lduw(ptr + 8);
+    /* XXX: handle overflow ? */
+    e = (upper & 0x7fff) - 16383 + EXPBIAS; /* exponent */
+    e |= (upper >> 4) & 0x800; /* sign */
+    temp.ll = ((ldq(ptr) >> 11) & ((1LL << 52) - 1)) | ((uint64_t)e << 52);
+    return temp.d;
+}
+
+static inline void helper_fstt(CPU86_LDouble f, uint8_t *ptr)
+{
+    CPU86_LDoubleU temp;
+    int e;
+    temp.d = f;
+    /* mantissa */
+    stq(ptr, (MANTD(temp) << 11) | (1LL << 63));
+    /* exponent + sign */
+    e = EXPD(temp) - EXPBIAS + 16383;
+    e |= SIGND(temp) >> 16;
+    stw(ptr + 8, e);
+}
+#endif
+
+void helper_fldt_ST0_A0(void);
+void helper_fstt_ST0_A0(void);
+void helper_fbld_ST0_A0(void);
+void helper_fbst_ST0_A0(void);
+void helper_f2xm1(void);
+void helper_fyl2x(void);
+void helper_fptan(void);
+void helper_fpatan(void);
+void helper_fxtract(void);
+void helper_fprem1(void);
+void helper_fprem(void);
+void helper_fyl2xp1(void);
+void helper_fsqrt(void);
+void helper_fsincos(void);
+void helper_frndint(void);
+void helper_fscale(void);
+void helper_fsin(void);
+void helper_fcos(void);
+void helper_fxam_ST0(void);
+void helper_fstenv(uint8_t *ptr, int data32);
+void helper_fldenv(uint8_t *ptr, int data32);
+void helper_fsave(uint8_t *ptr, int data32);
+void helper_frstor(uint8_t *ptr, int data32);
+
diff --git a/helper-i386.c b/helper-i386.c
new file mode 100644
index 0000000000..5f3040bf27
--- /dev/null
+++ b/helper-i386.c
@@ -0,0 +1,735 @@
+/*
+ *  i386 helpers
+ * 
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library 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 2 of the License, or (at your option) any later version.
+ *
+ * This library 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.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include "exec-i386.h"
+
+#if 0
+/* full interrupt support (only useful for real CPU emulation, not
+   finished) - I won't do it any time soon, finish it if you want ! */
+void raise_interrupt(int intno, int is_int, int error_code, 
+                     unsigned int next_eip)
+{
+    SegmentDescriptorTable *dt;
+    uint8_t *ptr;
+    int type, dpl, cpl;
+    uint32_t e1, e2;
+    
+    dt = &env->idt;
+    if (intno * 8 + 7 > dt->limit)
+        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
+    ptr = dt->base + intno * 8;
+    e1 = ldl(ptr);
+    e2 = ldl(ptr + 4);
+    /* check gate type */
+    type = (e2 >> DESC_TYPE_SHIFT) & 0x1f;
+    switch(type) {
+    case 5: /* task gate */
+    case 6: /* 286 interrupt gate */
+    case 7: /* 286 trap gate */
+    case 14: /* 386 interrupt gate */
+    case 15: /* 386 trap gate */
+        break;
+    default:
+        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
+        break;
+    }
+    dpl = (e2 >> DESC_DPL_SHIFT) & 3;
+    cpl = env->segs[R_CS] & 3;
+    /* check privledge if software int */
+    if (is_int && dpl < cpl)
+        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
+    /* check valid bit */
+    if (!(e2 & DESC_P_MASK))
+        raise_exception_err(EXCP0B_NOSEG, intno * 8 + 2);
+}
+
+#else
+
+/*
+ * is_int is TRUE if coming from the int instruction. next_eip is the
+ * EIP value AFTER the interrupt instruction. It is only relevant if
+ * is_int is TRUE.  
+ */
+void raise_interrupt(int intno, int is_int, int error_code, 
+                     unsigned int next_eip)
+{
+    SegmentDescriptorTable *dt;
+    uint8_t *ptr;
+    int dpl, cpl;
+    uint32_t e2;
+
+    dt = &env->idt;
+    ptr = dt->base + (intno * 8);
+    e2 = ldl(ptr + 4);
+    
+    dpl = (e2 >> DESC_DPL_SHIFT) & 3;
+    cpl = 3;
+    /* check privledge if software int */
+    if (is_int && dpl < cpl)
+        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
+
+    /* Since we emulate only user space, we cannot do more than
+       exiting the emulation with the suitable exception and error
+       code */
+    if (is_int)
+        EIP = next_eip;
+    env->exception_index = intno;
+    env->error_code = error_code;
+
+    cpu_loop_exit();
+}
+
+#endif
+
+/* shortcuts to generate exceptions */
+void raise_exception_err(int exception_index, int error_code)
+{
+    raise_interrupt(exception_index, 0, error_code, 0);
+}
+
+void raise_exception(int exception_index)
+{
+    raise_interrupt(exception_index, 0, 0, 0);
+}
+
+/* We simulate a pre-MMX pentium as in valgrind */
+#define CPUID_FP87 (1 << 0)
+#define CPUID_VME  (1 << 1)
+#define CPUID_DE   (1 << 2)
+#define CPUID_PSE  (1 << 3)
+#define CPUID_TSC  (1 << 4)
+#define CPUID_MSR  (1 << 5)
+#define CPUID_PAE  (1 << 6)
+#define CPUID_MCE  (1 << 7)
+#define CPUID_CX8  (1 << 8)
+#define CPUID_APIC (1 << 9)
+#define CPUID_SEP  (1 << 11) /* sysenter/sysexit */
+#define CPUID_MTRR (1 << 12)
+#define CPUID_PGE  (1 << 13)
+#define CPUID_MCA  (1 << 14)
+#define CPUID_CMOV (1 << 15)
+/* ... */
+#define CPUID_MMX  (1 << 23)
+#define CPUID_FXSR (1 << 24)
+#define CPUID_SSE  (1 << 25)
+#define CPUID_SSE2 (1 << 26)
+
+void helper_cpuid(void)
+{
+    if (EAX == 0) {
+        EAX = 1; /* max EAX index supported */
+        EBX = 0x756e6547;
+        ECX = 0x6c65746e;
+        EDX = 0x49656e69;
+    } else if (EAX == 1) {
+        /* EAX = 1 info */
+        EAX = 0x52b;
+        EBX = 0;
+        ECX = 0;
+        EDX = CPUID_FP87 | CPUID_DE | CPUID_PSE |
+            CPUID_TSC | CPUID_MSR | CPUID_MCE |
+            CPUID_CX8;
+    }
+}
+
+/* only works if protected mode and not VM86 */
+void load_seg(int seg_reg, int selector, unsigned cur_eip)
+{
+    SegmentCache *sc;
+    SegmentDescriptorTable *dt;
+    int index;
+    uint32_t e1, e2;
+    uint8_t *ptr;
+
+    sc = &env->seg_cache[seg_reg];
+    if ((selector & 0xfffc) == 0) {
+        /* null selector case */
+        if (seg_reg == R_SS) {
+            EIP = cur_eip;
+            raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
+        } else {
+            /* XXX: each access should trigger an exception */
+            sc->base = NULL;
+            sc->limit = 0;
+            sc->seg_32bit = 1;
+        }
+    } else {
+        if (selector & 0x4)
+            dt = &env->ldt;
+        else
+            dt = &env->gdt;
+        index = selector & ~7;
+        if ((index + 7) > dt->limit) {
+            EIP = cur_eip;
+            raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
+        }
+        ptr = dt->base + index;
+        e1 = ldl(ptr);
+        e2 = ldl(ptr + 4);
+        if (!(e2 & DESC_S_MASK) ||
+            (e2 & (DESC_CS_MASK | DESC_R_MASK)) == DESC_CS_MASK) {
+            EIP = cur_eip;
+            raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
+        }
+
+        if (seg_reg == R_SS) {
+            if ((e2 & (DESC_CS_MASK | DESC_W_MASK)) == 0) {
+                EIP = cur_eip;
+                raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
+            }
+        } else {
+            if ((e2 & (DESC_CS_MASK | DESC_R_MASK)) == DESC_CS_MASK) {
+                EIP = cur_eip;
+                raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
+            }
+        }
+
+        if (!(e2 & DESC_P_MASK)) {
+            EIP = cur_eip;
+            if (seg_reg == R_SS)
+                raise_exception_err(EXCP0C_STACK, selector & 0xfffc);
+            else
+                raise_exception_err(EXCP0B_NOSEG, selector & 0xfffc);
+        }
+        
+        sc->base = (void *)((e1 >> 16) | ((e2 & 0xff) << 16) | (e2 & 0xff000000));
+        sc->limit = (e1 & 0xffff) | (e2 & 0x000f0000);
+        if (e2 & (1 << 23))
+            sc->limit = (sc->limit << 12) | 0xfff;
+        sc->seg_32bit = (e2 >> 22) & 1;
+#if 0
+        fprintf(logfile, "load_seg: sel=0x%04x base=0x%08lx limit=0x%08lx seg_32bit=%d\n", 
+                selector, (unsigned long)sc->base, sc->limit, sc->seg_32bit);
+#endif
+    }
+    env->segs[seg_reg] = selector;
+}
+
+void helper_lsl(void)
+{
+    unsigned int selector, limit;
+    SegmentDescriptorTable *dt;
+    int index;
+    uint32_t e1, e2;
+    uint8_t *ptr;
+
+    CC_SRC = cc_table[CC_OP].compute_all() & ~CC_Z;
+    selector = T0 & 0xffff;
+    if (selector & 0x4)
+        dt = &env->ldt;
+    else
+        dt = &env->gdt;
+    index = selector & ~7;
+    if ((index + 7) > dt->limit)
+        return;
+    ptr = dt->base + index;
+    e1 = ldl(ptr);
+    e2 = ldl(ptr + 4);
+    limit = (e1 & 0xffff) | (e2 & 0x000f0000);
+    if (e2 & (1 << 23))
+        limit = (limit << 12) | 0xfff;
+    T1 = limit;
+    CC_SRC |= CC_Z;
+}
+
+void helper_lar(void)
+{
+    unsigned int selector;
+    SegmentDescriptorTable *dt;
+    int index;
+    uint32_t e2;
+    uint8_t *ptr;
+
+    CC_SRC = cc_table[CC_OP].compute_all() & ~CC_Z;
+    selector = T0 & 0xffff;
+    if (selector & 0x4)
+        dt = &env->ldt;
+    else
+        dt = &env->gdt;
+    index = selector & ~7;
+    if ((index + 7) > dt->limit)
+        return;
+    ptr = dt->base + index;
+    e2 = ldl(ptr + 4);
+    T1 = e2 & 0x00f0ff00;
+    CC_SRC |= CC_Z;
+}
+
+/* FPU helpers */
+
+#ifndef USE_X86LDOUBLE
+void helper_fldt_ST0_A0(void)
+{
+    ST0 = helper_fldt((uint8_t *)A0);
+}
+
+void helper_fstt_ST0_A0(void)
+{
+    helper_fstt(ST0, (uint8_t *)A0);
+}
+#endif
+
+/* BCD ops */
+
+#define MUL10(iv) ( iv + iv + (iv << 3) )
+
+void helper_fbld_ST0_A0(void)
+{
+    uint8_t *seg;
+    CPU86_LDouble fpsrcop;
+    int m32i;
+    unsigned int v;
+
+    /* in this code, seg/m32i will be used as temporary ptr/int */
+    seg = (uint8_t *)A0 + 8;
+    v = ldub(seg--);
+    /* XXX: raise exception */
+    if (v != 0)
+        return;
+    v = ldub(seg--);
+    /* XXX: raise exception */
+    if ((v & 0xf0) != 0)
+        return;
+    m32i = v;  /* <-- d14 */
+    v = ldub(seg--);
+    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d13 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d12 */
+    v = ldub(seg--);
+    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d11 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d10 */
+    v = ldub(seg--);
+    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d9 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d8 */
+    fpsrcop = ((CPU86_LDouble)m32i) * 100000000.0;
+
+    v = ldub(seg--);
+    m32i = (v >> 4);  /* <-- d7 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d6 */
+    v = ldub(seg--);
+    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d5 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d4 */
+    v = ldub(seg--);
+    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d3 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d2 */
+    v = ldub(seg);
+    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d1 */
+    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d0 */
+    fpsrcop += ((CPU86_LDouble)m32i);
+    if ( ldub(seg+9) & 0x80 )
+        fpsrcop = -fpsrcop;
+    ST0 = fpsrcop;
+}
+
+void helper_fbst_ST0_A0(void)
+{
+    CPU86_LDouble fptemp;
+    CPU86_LDouble fpsrcop;
+    int v;
+    uint8_t *mem_ref, *mem_end;
+
+    fpsrcop = rint(ST0);
+    mem_ref = (uint8_t *)A0;
+    mem_end = mem_ref + 8;
+    if ( fpsrcop < 0.0 ) {
+        stw(mem_end, 0x8000);
+        fpsrcop = -fpsrcop;
+    } else {
+        stw(mem_end, 0x0000);
+    }
+    while (mem_ref < mem_end) {
+        if (fpsrcop == 0.0)
+            break;
+        fptemp = floor(fpsrcop/10.0);
+        v = ((int)(fpsrcop - fptemp*10.0));
+        if  (fptemp == 0.0)  { 
+            stb(mem_ref++, v); 
+            break; 
+        }
+        fpsrcop = fptemp;
+        fptemp = floor(fpsrcop/10.0);
+        v |= (((int)(fpsrcop - fptemp*10.0)) << 4);
+        stb(mem_ref++, v);
+        fpsrcop = fptemp;
+    }
+    while (mem_ref < mem_end) {
+        stb(mem_ref++, 0);
+    }
+}
+
+void helper_f2xm1(void)
+{
+    ST0 = pow(2.0,ST0) - 1.0;
+}
+
+void helper_fyl2x(void)
+{
+    CPU86_LDouble fptemp;
+    
+    fptemp = ST0;
+    if (fptemp>0.0){
+        fptemp = log(fptemp)/log(2.0);	 /* log2(ST) */
+        ST1 *= fptemp;
+        fpop();
+    } else { 
+        env->fpus &= (~0x4700);
+        env->fpus |= 0x400;
+    }
+}
+
+void helper_fptan(void)
+{
+    CPU86_LDouble fptemp;
+
+    fptemp = ST0;
+    if((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
+        env->fpus |= 0x400;
+    } else {
+        ST0 = tan(fptemp);
+        fpush();
+        ST0 = 1.0;
+        env->fpus &= (~0x400);  /* C2 <-- 0 */
+        /* the above code is for  |arg| < 2**52 only */
+    }
+}
+
+void helper_fpatan(void)
+{
+    CPU86_LDouble fptemp, fpsrcop;
+
+    fpsrcop = ST1;
+    fptemp = ST0;
+    ST1 = atan2(fpsrcop,fptemp);
+    fpop();
+}
+
+void helper_fxtract(void)
+{
+    CPU86_LDoubleU temp;
+    unsigned int expdif;
+
+    temp.d = ST0;
+    expdif = EXPD(temp) - EXPBIAS;
+    /*DP exponent bias*/
+    ST0 = expdif;
+    fpush();
+    BIASEXPONENT(temp);
+    ST0 = temp.d;
+}
+
+void helper_fprem1(void)
+{
+    CPU86_LDouble dblq, fpsrcop, fptemp;
+    CPU86_LDoubleU fpsrcop1, fptemp1;
+    int expdif;
+    int q;
+
+    fpsrcop = ST0;
+    fptemp = ST1;
+    fpsrcop1.d = fpsrcop;
+    fptemp1.d = fptemp;
+    expdif = EXPD(fpsrcop1) - EXPD(fptemp1);
+    if (expdif < 53) {
+        dblq = fpsrcop / fptemp;
+        dblq = (dblq < 0.0)? ceil(dblq): floor(dblq);
+        ST0 = fpsrcop - fptemp*dblq;
+        q = (int)dblq; /* cutting off top bits is assumed here */
+        env->fpus &= (~0x4700); /* (C3,C2,C1,C0) <-- 0000 */
+				/* (C0,C1,C3) <-- (q2,q1,q0) */
+        env->fpus |= (q&0x4) << 6; /* (C0) <-- q2 */
+        env->fpus |= (q&0x2) << 8; /* (C1) <-- q1 */
+        env->fpus |= (q&0x1) << 14; /* (C3) <-- q0 */
+    } else {
+        env->fpus |= 0x400;  /* C2 <-- 1 */
+        fptemp = pow(2.0, expdif-50);
+        fpsrcop = (ST0 / ST1) / fptemp;
+        /* fpsrcop = integer obtained by rounding to the nearest */
+        fpsrcop = (fpsrcop-floor(fpsrcop) < ceil(fpsrcop)-fpsrcop)?
+            floor(fpsrcop): ceil(fpsrcop);
+        ST0 -= (ST1 * fpsrcop * fptemp);
+    }
+}
+
+void helper_fprem(void)
+{
+    CPU86_LDouble dblq, fpsrcop, fptemp;
+    CPU86_LDoubleU fpsrcop1, fptemp1;
+    int expdif;
+    int q;
+    
+    fpsrcop = ST0;
+    fptemp = ST1;
+    fpsrcop1.d = fpsrcop;
+    fptemp1.d = fptemp;
+    expdif = EXPD(fpsrcop1) - EXPD(fptemp1);
+    if ( expdif < 53 ) {
+        dblq = fpsrcop / fptemp;
+        dblq = (dblq < 0.0)? ceil(dblq): floor(dblq);
+        ST0 = fpsrcop - fptemp*dblq;
+        q = (int)dblq; /* cutting off top bits is assumed here */
+        env->fpus &= (~0x4700); /* (C3,C2,C1,C0) <-- 0000 */
+				/* (C0,C1,C3) <-- (q2,q1,q0) */
+        env->fpus |= (q&0x4) << 6; /* (C0) <-- q2 */
+        env->fpus |= (q&0x2) << 8; /* (C1) <-- q1 */
+        env->fpus |= (q&0x1) << 14; /* (C3) <-- q0 */
+    } else {
+        env->fpus |= 0x400;  /* C2 <-- 1 */
+        fptemp = pow(2.0, expdif-50);
+        fpsrcop = (ST0 / ST1) / fptemp;
+        /* fpsrcop = integer obtained by chopping */
+        fpsrcop = (fpsrcop < 0.0)?
+            -(floor(fabs(fpsrcop))): floor(fpsrcop);
+        ST0 -= (ST1 * fpsrcop * fptemp);
+    }
+}
+
+void helper_fyl2xp1(void)
+{
+    CPU86_LDouble fptemp;
+
+    fptemp = ST0;
+    if ((fptemp+1.0)>0.0) {
+        fptemp = log(fptemp+1.0) / log(2.0); /* log2(ST+1.0) */
+        ST1 *= fptemp;
+        fpop();
+    } else { 
+        env->fpus &= (~0x4700);
+        env->fpus |= 0x400;
+    }
+}
+
+void helper_fsqrt(void)
+{
+    CPU86_LDouble fptemp;
+
+    fptemp = ST0;
+    if (fptemp<0.0) { 
+        env->fpus &= (~0x4700);  /* (C3,C2,C1,C0) <-- 0000 */
+        env->fpus |= 0x400;
+    }
+    ST0 = sqrt(fptemp);
+}
+
+void helper_fsincos(void)
+{
+    CPU86_LDouble fptemp;
+
+    fptemp = ST0;
+    if ((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
+        env->fpus |= 0x400;
+    } else {
+        ST0 = sin(fptemp);
+        fpush();
+        ST0 = cos(fptemp);
+        env->fpus &= (~0x400);  /* C2 <-- 0 */
+        /* the above code is for  |arg| < 2**63 only */
+    }
+}
+
+void helper_frndint(void)
+{
+    ST0 = rint(ST0);
+}
+
+void helper_fscale(void)
+{
+    CPU86_LDouble fpsrcop, fptemp;
+
+    fpsrcop = 2.0;
+    fptemp = pow(fpsrcop,ST1);
+    ST0 *= fptemp;
+}
+
+void helper_fsin(void)
+{
+    CPU86_LDouble fptemp;
+
+    fptemp = ST0;
+    if ((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
+        env->fpus |= 0x400;
+    } else {
+        ST0 = sin(fptemp);
+        env->fpus &= (~0x400);  /* C2 <-- 0 */
+        /* the above code is for  |arg| < 2**53 only */
+    }
+}
+
+void helper_fcos(void)
+{
+    CPU86_LDouble fptemp;
+
+    fptemp = ST0;
+    if((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
+        env->fpus |= 0x400;
+    } else {
+        ST0 = cos(fptemp);
+        env->fpus &= (~0x400);  /* C2 <-- 0 */
+        /* the above code is for  |arg5 < 2**63 only */
+    }
+}
+
+void helper_fxam_ST0(void)
+{
+    CPU86_LDoubleU temp;
+    int expdif;
+
+    temp.d = ST0;
+
+    env->fpus &= (~0x4700);  /* (C3,C2,C1,C0) <-- 0000 */
+    if (SIGND(temp))
+        env->fpus |= 0x200; /* C1 <-- 1 */
+
+    expdif = EXPD(temp);
+    if (expdif == MAXEXPD) {
+        if (MANTD(temp) == 0)
+            env->fpus |=  0x500 /*Infinity*/;
+        else
+            env->fpus |=  0x100 /*NaN*/;
+    } else if (expdif == 0) {
+        if (MANTD(temp) == 0)
+            env->fpus |=  0x4000 /*Zero*/;
+        else
+            env->fpus |= 0x4400 /*Denormal*/;
+    } else {
+        env->fpus |= 0x400;
+    }
+}
+
+void helper_fstenv(uint8_t *ptr, int data32)
+{
+    int fpus, fptag, exp, i;
+    uint64_t mant;
+    CPU86_LDoubleU tmp;
+
+    fpus = (env->fpus & ~0x3800) | (env->fpstt & 0x7) << 11;
+    fptag = 0;
+    for (i=7; i>=0; i--) {
+	fptag <<= 2;
+	if (env->fptags[i]) {
+            fptag |= 3;
+	} else {
+            tmp.d = env->fpregs[i];
+            exp = EXPD(tmp);
+            mant = MANTD(tmp);
+            if (exp == 0 && mant == 0) {
+                /* zero */
+	        fptag |= 1;
+	    } else if (exp == 0 || exp == MAXEXPD
+#ifdef USE_X86LDOUBLE
+                       || (mant & (1LL << 63)) == 0
+#endif
+                       ) {
+                /* NaNs, infinity, denormal */
+                fptag |= 2;
+            }
+        }
+    }
+    if (data32) {
+        /* 32 bit */
+        stl(ptr, env->fpuc);
+        stl(ptr + 4, fpus);
+        stl(ptr + 8, fptag);
+        stl(ptr + 12, 0);
+        stl(ptr + 16, 0);
+        stl(ptr + 20, 0);
+        stl(ptr + 24, 0);
+    } else {
+        /* 16 bit */
+        stw(ptr, env->fpuc);
+        stw(ptr + 2, fpus);
+        stw(ptr + 4, fptag);
+        stw(ptr + 6, 0);
+        stw(ptr + 8, 0);
+        stw(ptr + 10, 0);
+        stw(ptr + 12, 0);
+    }
+}
+
+void helper_fldenv(uint8_t *ptr, int data32)
+{
+    int i, fpus, fptag;
+
+    if (data32) {
+	env->fpuc = lduw(ptr);
+        fpus = lduw(ptr + 4);
+        fptag = lduw(ptr + 8);
+    }
+    else {
+	env->fpuc = lduw(ptr);
+        fpus = lduw(ptr + 2);
+        fptag = lduw(ptr + 4);
+    }
+    env->fpstt = (fpus >> 11) & 7;
+    env->fpus = fpus & ~0x3800;
+    for(i = 0;i < 7; i++) {
+        env->fptags[i] = ((fptag & 3) == 3);
+        fptag >>= 2;
+    }
+}
+
+void helper_fsave(uint8_t *ptr, int data32)
+{
+    CPU86_LDouble tmp;
+    int i;
+
+    helper_fstenv(ptr, data32);
+
+    ptr += (14 << data32);
+    for(i = 0;i < 8; i++) {
+        tmp = ST(i);
+#ifdef USE_X86LDOUBLE
+        *(long double *)ptr = tmp;
+#else
+        helper_fstt(tmp, ptr);
+#endif        
+        ptr += 10;
+    }
+
+    /* fninit */
+    env->fpus = 0;
+    env->fpstt = 0;
+    env->fpuc = 0x37f;
+    env->fptags[0] = 1;
+    env->fptags[1] = 1;
+    env->fptags[2] = 1;
+    env->fptags[3] = 1;
+    env->fptags[4] = 1;
+    env->fptags[5] = 1;
+    env->fptags[6] = 1;
+    env->fptags[7] = 1;
+}
+
+void helper_frstor(uint8_t *ptr, int data32)
+{
+    CPU86_LDouble tmp;
+    int i;
+
+    helper_fldenv(ptr, data32);
+    ptr += (14 << data32);
+
+    for(i = 0;i < 8; i++) {
+#ifdef USE_X86LDOUBLE
+        tmp = *(long double *)ptr;
+#else
+        tmp = helper_fldt(ptr);
+#endif        
+        ST(i) = tmp;
+        ptr += 10;
+    }
+}
+
diff --git a/op-i386.c b/op-i386.c
index df841cec67..4c7b3804dc 100644
--- a/op-i386.c
+++ b/op-i386.c
@@ -626,95 +626,6 @@ void OPPROTO op_jmp_im(void)
     EIP = PARAM1;
 }
 
-#if 0
-/* full interrupt support (only useful for real CPU emulation, not
-   finished) - I won't do it any time soon, finish it if you want ! */
-void raise_interrupt(int intno, int is_int, int error_code, 
-                     unsigned int next_eip)
-{
-    SegmentDescriptorTable *dt;
-    uint8_t *ptr;
-    int type, dpl, cpl;
-    uint32_t e1, e2;
-    
-    dt = &env->idt;
-    if (intno * 8 + 7 > dt->limit)
-        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
-    ptr = dt->base + intno * 8;
-    e1 = ldl(ptr);
-    e2 = ldl(ptr + 4);
-    /* check gate type */
-    type = (e2 >> DESC_TYPE_SHIFT) & 0x1f;
-    switch(type) {
-    case 5: /* task gate */
-    case 6: /* 286 interrupt gate */
-    case 7: /* 286 trap gate */
-    case 14: /* 386 interrupt gate */
-    case 15: /* 386 trap gate */
-        break;
-    default:
-        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
-        break;
-    }
-    dpl = (e2 >> DESC_DPL_SHIFT) & 3;
-    cpl = env->segs[R_CS] & 3;
-    /* check privledge if software int */
-    if (is_int && dpl < cpl)
-        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
-    /* check valid bit */
-    if (!(e2 & DESC_P_MASK))
-        raise_exception_err(EXCP0B_NOSEG, intno * 8 + 2);
-}
-
-#else
-
-/*
- * is_int is TRUE if coming from the int instruction. next_eip is the
- * EIP value AFTER the interrupt instruction. It is only relevant if
- * is_int is TRUE.  
- */
-void raise_interrupt(int intno, int is_int, int error_code, 
-                     unsigned int next_eip)
-{
-    SegmentDescriptorTable *dt;
-    uint8_t *ptr;
-    int dpl, cpl;
-    uint32_t e2;
-
-    dt = &env->idt;
-    ptr = dt->base + (intno * 8);
-    e2 = ldl(ptr + 4);
-    
-    dpl = (e2 >> DESC_DPL_SHIFT) & 3;
-    cpl = 3;
-    /* check privledge if software int */
-    if (is_int && dpl < cpl)
-        raise_exception_err(EXCP0D_GPF, intno * 8 + 2);
-
-    /* Since we emulate only user space, we cannot do more than
-       exiting the emulation with the suitable exception and error
-       code */
-    if (is_int)
-        EIP = next_eip;
-    env->exception_index = intno;
-    env->error_code = error_code;
-
-    cpu_loop_exit();
-}
-
-#endif
-
-/* shortcuts to generate exceptions */
-void raise_exception_err(int exception_index, int error_code)
-{
-    raise_interrupt(exception_index, 0, error_code, 0);
-}
-
-void raise_exception(int exception_index)
-{
-    raise_interrupt(exception_index, 0, 0, 0);
-}
-
 void OPPROTO op_raise_interrupt(void)
 {
     int intno;
@@ -833,7 +744,7 @@ label ## n:\
 #define JUMP_TB(tbparam, n, eip)\
 do {\
     static void __attribute__((unused)) *__op_label ## n = &&label ## n;\
-    goto *((TranslationBlock *)tbparam)->tb_next[n];\
+    goto *(void *)(((TranslationBlock *)tbparam)->tb_next[n]);\
 label ## n:\
     T0 = (long)(tbparam) + (n);\
     EIP = eip;\
@@ -1044,46 +955,6 @@ void OPPROTO op_rdtsc(void)
     EDX = val >> 32;
 }
 
-/* We simulate a pre-MMX pentium as in valgrind */
-#define CPUID_FP87 (1 << 0)
-#define CPUID_VME  (1 << 1)
-#define CPUID_DE   (1 << 2)
-#define CPUID_PSE  (1 << 3)
-#define CPUID_TSC  (1 << 4)
-#define CPUID_MSR  (1 << 5)
-#define CPUID_PAE  (1 << 6)
-#define CPUID_MCE  (1 << 7)
-#define CPUID_CX8  (1 << 8)
-#define CPUID_APIC (1 << 9)
-#define CPUID_SEP  (1 << 11) /* sysenter/sysexit */
-#define CPUID_MTRR (1 << 12)
-#define CPUID_PGE  (1 << 13)
-#define CPUID_MCA  (1 << 14)
-#define CPUID_CMOV (1 << 15)
-/* ... */
-#define CPUID_MMX  (1 << 23)
-#define CPUID_FXSR (1 << 24)
-#define CPUID_SSE  (1 << 25)
-#define CPUID_SSE2 (1 << 26)
-
-void helper_cpuid(void)
-{
-    if (EAX == 0) {
-        EAX = 1; /* max EAX index supported */
-        EBX = 0x756e6547;
-        ECX = 0x6c65746e;
-        EDX = 0x49656e69;
-    } else if (EAX == 1) {
-        /* EAX = 1 info */
-        EAX = 0x52b;
-        EBX = 0;
-        ECX = 0;
-        EDX = CPUID_FP87 | CPUID_DE | CPUID_PSE |
-            CPUID_TSC | CPUID_MSR | CPUID_MCE |
-            CPUID_CX8;
-    }
-}
-
 void OPPROTO op_cpuid(void)
 {
     helper_cpuid();
@@ -1221,79 +1092,6 @@ void OPPROTO op_das(void)
 
 /* segment handling */
 
-/* only works if protected mode and not VM86 */
-void load_seg(int seg_reg, int selector, unsigned cur_eip)
-{
-    SegmentCache *sc;
-    SegmentDescriptorTable *dt;
-    int index;
-    uint32_t e1, e2;
-    uint8_t *ptr;
-
-    sc = &env->seg_cache[seg_reg];
-    if ((selector & 0xfffc) == 0) {
-        /* null selector case */
-        if (seg_reg == R_SS) {
-            EIP = cur_eip;
-            raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
-        } else {
-            /* XXX: each access should trigger an exception */
-            sc->base = NULL;
-            sc->limit = 0;
-            sc->seg_32bit = 1;
-        }
-    } else {
-        if (selector & 0x4)
-            dt = &env->ldt;
-        else
-            dt = &env->gdt;
-        index = selector & ~7;
-        if ((index + 7) > dt->limit) {
-            EIP = cur_eip;
-            raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
-        }
-        ptr = dt->base + index;
-        e1 = ldl(ptr);
-        e2 = ldl(ptr + 4);
-        if (!(e2 & DESC_S_MASK) ||
-            (e2 & (DESC_CS_MASK | DESC_R_MASK)) == DESC_CS_MASK) {
-            EIP = cur_eip;
-            raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
-        }
-
-        if (seg_reg == R_SS) {
-            if ((e2 & (DESC_CS_MASK | DESC_W_MASK)) == 0) {
-                EIP = cur_eip;
-                raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
-            }
-        } else {
-            if ((e2 & (DESC_CS_MASK | DESC_R_MASK)) == DESC_CS_MASK) {
-                EIP = cur_eip;
-                raise_exception_err(EXCP0D_GPF, selector & 0xfffc);
-            }
-        }
-
-        if (!(e2 & DESC_P_MASK)) {
-            EIP = cur_eip;
-            if (seg_reg == R_SS)
-                raise_exception_err(EXCP0C_STACK, selector & 0xfffc);
-            else
-                raise_exception_err(EXCP0B_NOSEG, selector & 0xfffc);
-        }
-        
-        sc->base = (void *)((e1 >> 16) | ((e2 & 0xff) << 16) | (e2 & 0xff000000));
-        sc->limit = (e1 & 0xffff) | (e2 & 0x000f0000);
-        if (e2 & (1 << 23))
-            sc->limit = (sc->limit << 12) | 0xfff;
-        sc->seg_32bit = (e2 >> 22) & 1;
-#if 0
-        fprintf(logfile, "load_seg: sel=0x%04x base=0x%08lx limit=0x%08lx seg_32bit=%d\n", 
-                selector, (unsigned long)sc->base, sc->limit, sc->seg_32bit);
-#endif
-    }
-    env->segs[seg_reg] = selector;
-}
-
 void OPPROTO op_movl_seg_T0(void)
 {
     load_seg(PARAM1, T0 & 0xffff, PARAM2);
@@ -1326,61 +1124,11 @@ void OPPROTO op_addl_A0_seg(void)
     A0 += *(unsigned long *)((char *)env + PARAM1);
 }
 
-void helper_lsl(void)
-{
-    unsigned int selector, limit;
-    SegmentDescriptorTable *dt;
-    int index;
-    uint32_t e1, e2;
-    uint8_t *ptr;
-
-    CC_SRC = cc_table[CC_OP].compute_all() & ~CC_Z;
-    selector = T0 & 0xffff;
-    if (selector & 0x4)
-        dt = &env->ldt;
-    else
-        dt = &env->gdt;
-    index = selector & ~7;
-    if ((index + 7) > dt->limit)
-        return;
-    ptr = dt->base + index;
-    e1 = ldl(ptr);
-    e2 = ldl(ptr + 4);
-    limit = (e1 & 0xffff) | (e2 & 0x000f0000);
-    if (e2 & (1 << 23))
-        limit = (limit << 12) | 0xfff;
-    T1 = limit;
-    CC_SRC |= CC_Z;
-}
-
 void OPPROTO op_lsl(void)
 {
     helper_lsl();
 }
 
-void helper_lar(void)
-{
-    unsigned int selector;
-    SegmentDescriptorTable *dt;
-    int index;
-    uint32_t e2;
-    uint8_t *ptr;
-
-    CC_SRC = cc_table[CC_OP].compute_all() & ~CC_Z;
-    selector = T0 & 0xffff;
-    if (selector & 0x4)
-        dt = &env->ldt;
-    else
-        dt = &env->gdt;
-    index = selector & ~7;
-    if ((index + 7) > dt->limit)
-        return;
-    ptr = dt->base + index;
-    e2 = ldl(ptr + 4);
-    T1 = e2 & 0x00f0ff00;
-    CC_SRC |= CC_Z;
-}
-
 void OPPROTO op_lar(void)
 {
     helper_lar();
@@ -1678,37 +1426,6 @@ CCTable cc_table[CC_OP_NB] = {
    functions comes from the LGPL'ed x86 emulator found in the Willows
    TWIN windows emulator. */
 
-#ifdef USE_X86LDOUBLE
-/* use long double functions */
-#define lrint lrintl
-#define llrint llrintl
-#define fabs fabsl
-#define sin sinl
-#define cos cosl
-#define sqrt sqrtl
-#define pow powl
-#define log logl
-#define tan tanl
-#define atan2 atan2l
-#define floor floorl
-#define ceil ceill
-#define rint rintl
-#endif
-
-extern int lrint(CPU86_LDouble x);
-extern int64_t llrint(CPU86_LDouble x);
-extern CPU86_LDouble fabs(CPU86_LDouble x);
-extern CPU86_LDouble sin(CPU86_LDouble x);
-extern CPU86_LDouble cos(CPU86_LDouble x);
-extern CPU86_LDouble sqrt(CPU86_LDouble x);
-extern CPU86_LDouble pow(CPU86_LDouble, CPU86_LDouble);
-extern CPU86_LDouble log(CPU86_LDouble x);
-extern CPU86_LDouble tan(CPU86_LDouble x);
-extern CPU86_LDouble atan2(CPU86_LDouble, CPU86_LDouble);
-extern CPU86_LDouble floor(CPU86_LDouble x);
-extern CPU86_LDouble ceil(CPU86_LDouble x);
-extern CPU86_LDouble rint(CPU86_LDouble x);
-
 #if defined(__powerpc__)
 extern CPU86_LDouble copysign(CPU86_LDouble, CPU86_LDouble);
 
@@ -1729,60 +1446,6 @@ double qemu_rint(double x)
 #define rint qemu_rint
 #endif
 
-#define RC_MASK         0xc00
-#define RC_NEAR		0x000
-#define RC_DOWN		0x400
-#define RC_UP		0x800
-#define RC_CHOP		0xc00
-
-#define MAXTAN 9223372036854775808.0
-
-#ifdef USE_X86LDOUBLE
-
-/* only for x86 */
-typedef union {
-    long double d;
-    struct {
-        unsigned long long lower;
-        unsigned short upper;
-    } l;
-} CPU86_LDoubleU;
-
-/* the following deal with x86 long double-precision numbers */
-#define MAXEXPD 0x7fff
-#define EXPBIAS 16383
-#define EXPD(fp)	(fp.l.upper & 0x7fff)
-#define SIGND(fp)	((fp.l.upper) & 0x8000)
-#define MANTD(fp)       (fp.l.lower)
-#define BIASEXPONENT(fp) fp.l.upper = (fp.l.upper & ~(0x7fff)) | EXPBIAS
-
-#else
-
-typedef union {
-    double d;
-#ifndef WORDS_BIGENDIAN
-    struct {
-        uint32_t lower;
-        int32_t upper;
-    } l;
-#else
-    struct {
-        int32_t upper;
-        uint32_t lower;
-    } l;
-#endif
-    int64_t ll;
-} CPU86_LDoubleU;
-
-/* the following deal with IEEE double-precision numbers */
-#define MAXEXPD 0x7ff
-#define EXPBIAS 1023
-#define EXPD(fp)	(((fp.l.upper) >> 20) & 0x7FF)
-#define SIGND(fp)	((fp.l.upper) & 0x80000000)
-#define MANTD(fp)	(fp.ll & ((1LL << 52) - 1))
-#define BIASEXPONENT(fp) fp.l.upper = (fp.l.upper & ~(0x7ff << 20)) | (EXPBIAS << 20)
-#endif
-
 /* fp load FT0 */
 
 void OPPROTO op_flds_FT0_A0(void)
@@ -1899,24 +1562,6 @@ void OPPROTO op_fldt_ST0_A0(void)
     ST0 = *(long double *)A0;
 }
 #else
-static inline CPU86_LDouble helper_fldt(uint8_t *ptr)
-{
-    CPU86_LDoubleU temp;
-    int upper, e;
-    /* mantissa */
-    upper = lduw(ptr + 8);
-    /* XXX: handle overflow ? */
-    e = (upper & 0x7fff) - 16383 + EXPBIAS; /* exponent */
-    e |= (upper >> 4) & 0x800; /* sign */
-    temp.ll = ((ldq(ptr) >> 11) & ((1LL << 52) - 1)) | ((uint64_t)e << 52);
-    return temp.d;
-}
-
-void helper_fldt_ST0_A0(void)
-{
-    ST0 = helper_fldt((uint8_t *)A0);
-}
-
 void OPPROTO op_fldt_ST0_A0(void)
 {
     helper_fldt_ST0_A0();
@@ -2013,25 +1658,6 @@ void OPPROTO op_fstt_ST0_A0(void)
     *(long double *)A0 = ST0;
 }
 #else
-
-static inline void helper_fstt(CPU86_LDouble f, uint8_t *ptr)
-{
-    CPU86_LDoubleU temp;
-    int e;
-    temp.d = f;
-    /* mantissa */
-    stq(ptr, (MANTD(temp) << 11) | (1LL << 63));
-    /* exponent + sign */
-    e = EXPD(temp) - EXPBIAS + 16383;
-    e |= SIGND(temp) >> 16;
-    stw(ptr + 8, e);
-}
-
-void helper_fstt_ST0_A0(void)
-{
-    helper_fstt(ST0, (uint8_t *)A0);
-}
-
 void OPPROTO op_fstt_ST0_A0(void)
 {
     helper_fstt_ST0_A0();
@@ -2080,98 +1706,11 @@ void OPPROTO op_fistll_ST0_A0(void)
     stq((void *)A0, val);
 }
 
-/* BCD ops */
-
-#define MUL10(iv) ( iv + iv + (iv << 3) )
-
-void helper_fbld_ST0_A0(void)
-{
-    uint8_t *seg;
-    CPU86_LDouble fpsrcop;
-    int m32i;
-    unsigned int v;
-
-    /* in this code, seg/m32i will be used as temporary ptr/int */
-    seg = (uint8_t *)A0 + 8;
-    v = ldub(seg--);
-    /* XXX: raise exception */
-    if (v != 0)
-        return;
-    v = ldub(seg--);
-    /* XXX: raise exception */
-    if ((v & 0xf0) != 0)
-        return;
-    m32i = v;  /* <-- d14 */
-    v = ldub(seg--);
-    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d13 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d12 */
-    v = ldub(seg--);
-    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d11 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d10 */
-    v = ldub(seg--);
-    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d9 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d8 */
-    fpsrcop = ((CPU86_LDouble)m32i) * 100000000.0;
-
-    v = ldub(seg--);
-    m32i = (v >> 4);  /* <-- d7 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d6 */
-    v = ldub(seg--);
-    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d5 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d4 */
-    v = ldub(seg--);
-    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d3 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d2 */
-    v = ldub(seg);
-    m32i = MUL10(m32i) + (v >> 4);  /* <-- val * 10 + d1 */
-    m32i = MUL10(m32i) + (v & 0xf); /* <-- val * 10 + d0 */
-    fpsrcop += ((CPU86_LDouble)m32i);
-    if ( ldub(seg+9) & 0x80 )
-        fpsrcop = -fpsrcop;
-    ST0 = fpsrcop;
-}
-
 void OPPROTO op_fbld_ST0_A0(void)
 {
     helper_fbld_ST0_A0();
 }
 
-void helper_fbst_ST0_A0(void)
-{
-    CPU86_LDouble fptemp;
-    CPU86_LDouble fpsrcop;
-    int v;
-    uint8_t *mem_ref, *mem_end;
-
-    fpsrcop = rint(ST0);
-    mem_ref = (uint8_t *)A0;
-    mem_end = mem_ref + 8;
-    if ( fpsrcop < 0.0 ) {
-        stw(mem_end, 0x8000);
-        fpsrcop = -fpsrcop;
-    } else {
-        stw(mem_end, 0x0000);
-    }
-    while (mem_ref < mem_end) {
-        if (fpsrcop == 0.0)
-            break;
-        fptemp = floor(fpsrcop/10.0);
-        v = ((int)(fpsrcop - fptemp*10.0));
-        if  (fptemp == 0.0)  { 
-            stb(mem_ref++, v); 
-            break; 
-        }
-        fpsrcop = fptemp;
-        fptemp = floor(fpsrcop/10.0);
-        v |= (((int)(fpsrcop - fptemp*10.0)) << 4);
-        stb(mem_ref++, v);
-        fpsrcop = fptemp;
-    }
-    while (mem_ref < mem_end) {
-        stb(mem_ref++, 0);
-    }
-}
-
 void OPPROTO op_fbst_ST0_A0(void)
 {
     helper_fbst_ST0_A0();
@@ -2179,18 +1718,6 @@ void OPPROTO op_fbst_ST0_A0(void)
 
 /* FPU move */
 
-static inline void fpush(void)
-{
-    env->fpstt = (env->fpstt - 1) & 7;
-    env->fptags[env->fpstt] = 0; /* validate stack entry */
-}
-
-static inline void fpop(void)
-{
-    env->fptags[env->fpstt] = 1; /* invvalidate stack entry */
-    env->fpstt = (env->fpstt + 1) & 7;
-}
-
 void OPPROTO op_fpush(void)
 {
     fpush();
@@ -2370,33 +1897,6 @@ void OPPROTO op_fabs_ST0(void)
     ST0 = fabs(ST0);
 }
 
-void helper_fxam_ST0(void)
-{
-    CPU86_LDoubleU temp;
-    int expdif;
-
-    temp.d = ST0;
-
-    env->fpus &= (~0x4700);  /* (C3,C2,C1,C0) <-- 0000 */
-    if (SIGND(temp))
-        env->fpus |= 0x200; /* C1 <-- 1 */
-
-    expdif = EXPD(temp);
-    if (expdif == MAXEXPD) {
-        if (MANTD(temp) == 0)
-            env->fpus |=  0x500 /*Infinity*/;
-        else
-            env->fpus |=  0x100 /*NaN*/;
-    } else if (expdif == 0) {
-        if (MANTD(temp) == 0)
-            env->fpus |=  0x4000 /*Zero*/;
-        else
-            env->fpus |= 0x4400 /*Denormal*/;
-    } else {
-        env->fpus |= 0x400;
-    }
-}
-
 void OPPROTO op_fxam_ST0(void)
 {
     helper_fxam_ST0();
@@ -2442,217 +1942,6 @@ void OPPROTO op_fldz_FT0(void)
     ST0 = *(CPU86_LDouble *)&f15rk[0];
 }
 
-void helper_f2xm1(void)
-{
-    ST0 = pow(2.0,ST0) - 1.0;
-}
-
-void helper_fyl2x(void)
-{
-    CPU86_LDouble fptemp;
-    
-    fptemp = ST0;
-    if (fptemp>0.0){
-        fptemp = log(fptemp)/log(2.0);	 /* log2(ST) */
-        ST1 *= fptemp;
-        fpop();
-    } else { 
-        env->fpus &= (~0x4700);
-        env->fpus |= 0x400;
-    }
-}
-
-void helper_fptan(void)
-{
-    CPU86_LDouble fptemp;
-
-    fptemp = ST0;
-    if((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
-        env->fpus |= 0x400;
-    } else {
-        ST0 = tan(fptemp);
-        fpush();
-        ST0 = 1.0;
-        env->fpus &= (~0x400);  /* C2 <-- 0 */
-        /* the above code is for  |arg| < 2**52 only */
-    }
-}
-
-void helper_fpatan(void)
-{
-    CPU86_LDouble fptemp, fpsrcop;
-
-    fpsrcop = ST1;
-    fptemp = ST0;
-    ST1 = atan2(fpsrcop,fptemp);
-    fpop();
-}
-
-void helper_fxtract(void)
-{
-    CPU86_LDoubleU temp;
-    unsigned int expdif;
-
-    temp.d = ST0;
-    expdif = EXPD(temp) - EXPBIAS;
-    /*DP exponent bias*/
-    ST0 = expdif;
-    fpush();
-    BIASEXPONENT(temp);
-    ST0 = temp.d;
-}
-
-void helper_fprem1(void)
-{
-    CPU86_LDouble dblq, fpsrcop, fptemp;
-    CPU86_LDoubleU fpsrcop1, fptemp1;
-    int expdif;
-    int q;
-
-    fpsrcop = ST0;
-    fptemp = ST1;
-    fpsrcop1.d = fpsrcop;
-    fptemp1.d = fptemp;
-    expdif = EXPD(fpsrcop1) - EXPD(fptemp1);
-    if (expdif < 53) {
-        dblq = fpsrcop / fptemp;
-        dblq = (dblq < 0.0)? ceil(dblq): floor(dblq);
-        ST0 = fpsrcop - fptemp*dblq;
-        q = (int)dblq; /* cutting off top bits is assumed here */
-        env->fpus &= (~0x4700); /* (C3,C2,C1,C0) <-- 0000 */
-				/* (C0,C1,C3) <-- (q2,q1,q0) */
-        env->fpus |= (q&0x4) << 6; /* (C0) <-- q2 */
-        env->fpus |= (q&0x2) << 8; /* (C1) <-- q1 */
-        env->fpus |= (q&0x1) << 14; /* (C3) <-- q0 */
-    } else {
-        env->fpus |= 0x400;  /* C2 <-- 1 */
-        fptemp = pow(2.0, expdif-50);
-        fpsrcop = (ST0 / ST1) / fptemp;
-        /* fpsrcop = integer obtained by rounding to the nearest */
-        fpsrcop = (fpsrcop-floor(fpsrcop) < ceil(fpsrcop)-fpsrcop)?
-            floor(fpsrcop): ceil(fpsrcop);
-        ST0 -= (ST1 * fpsrcop * fptemp);
-    }
-}
-
-void helper_fprem(void)
-{
-    CPU86_LDouble dblq, fpsrcop, fptemp;
-    CPU86_LDoubleU fpsrcop1, fptemp1;
-    int expdif;
-    int q;
-    
-    fpsrcop = ST0;
-    fptemp = ST1;
-    fpsrcop1.d = fpsrcop;
-    fptemp1.d = fptemp;
-    expdif = EXPD(fpsrcop1) - EXPD(fptemp1);
-    if ( expdif < 53 ) {
-        dblq = fpsrcop / fptemp;
-        dblq = (dblq < 0.0)? ceil(dblq): floor(dblq);
-        ST0 = fpsrcop - fptemp*dblq;
-        q = (int)dblq; /* cutting off top bits is assumed here */
-        env->fpus &= (~0x4700); /* (C3,C2,C1,C0) <-- 0000 */
-				/* (C0,C1,C3) <-- (q2,q1,q0) */
-        env->fpus |= (q&0x4) << 6; /* (C0) <-- q2 */
-        env->fpus |= (q&0x2) << 8; /* (C1) <-- q1 */
-        env->fpus |= (q&0x1) << 14; /* (C3) <-- q0 */
-    } else {
-        env->fpus |= 0x400;  /* C2 <-- 1 */
-        fptemp = pow(2.0, expdif-50);
-        fpsrcop = (ST0 / ST1) / fptemp;
-        /* fpsrcop = integer obtained by chopping */
-        fpsrcop = (fpsrcop < 0.0)?
-            -(floor(fabs(fpsrcop))): floor(fpsrcop);
-        ST0 -= (ST1 * fpsrcop * fptemp);
-    }
-}
-
-void helper_fyl2xp1(void)
-{
-    CPU86_LDouble fptemp;
-
-    fptemp = ST0;
-    if ((fptemp+1.0)>0.0) {
-        fptemp = log(fptemp+1.0) / log(2.0); /* log2(ST+1.0) */
-        ST1 *= fptemp;
-        fpop();
-    } else { 
-        env->fpus &= (~0x4700);
-        env->fpus |= 0x400;
-    }
-}
-
-void helper_fsqrt(void)
-{
-    CPU86_LDouble fptemp;
-
-    fptemp = ST0;
-    if (fptemp<0.0) { 
-        env->fpus &= (~0x4700);  /* (C3,C2,C1,C0) <-- 0000 */
-        env->fpus |= 0x400;
-    }
-    ST0 = sqrt(fptemp);
-}
-
-void helper_fsincos(void)
-{
-    CPU86_LDouble fptemp;
-
-    fptemp = ST0;
-    if ((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
-        env->fpus |= 0x400;
-    } else {
-        ST0 = sin(fptemp);
-        fpush();
-        ST0 = cos(fptemp);
-        env->fpus &= (~0x400);  /* C2 <-- 0 */
-        /* the above code is for  |arg| < 2**63 only */
-    }
-}
-
-void helper_frndint(void)
-{
-    ST0 = rint(ST0);
-}
-
-void helper_fscale(void)
-{
-    CPU86_LDouble fpsrcop, fptemp;
-
-    fpsrcop = 2.0;
-    fptemp = pow(fpsrcop,ST1);
-    ST0 *= fptemp;
-}
-
-void helper_fsin(void)
-{
-    CPU86_LDouble fptemp;
-
-    fptemp = ST0;
-    if ((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
-        env->fpus |= 0x400;
-    } else {
-        ST0 = sin(fptemp);
-        env->fpus &= (~0x400);  /* C2 <-- 0 */
-        /* the above code is for  |arg| < 2**53 only */
-    }
-}
-
-void helper_fcos(void)
-{
-    CPU86_LDouble fptemp;
-
-    fptemp = ST0;
-    if((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
-        env->fpus |= 0x400;
-    } else {
-        ST0 = cos(fptemp);
-        env->fpus &= (~0x400);  /* C2 <-- 0 */
-        /* the above code is for  |arg5 < 2**63 only */
-    }
-}
-
 /* associated heplers to reduce generated code length and to simplify
    relocation (FP constants are usually stored in .rodata section) */
 
@@ -2789,129 +2078,6 @@ void OPPROTO op_fninit(void)
     env->fptags[7] = 1;
 }
 
-void helper_fstenv(uint8_t *ptr, int data32)
-{
-    int fpus, fptag, exp, i;
-    uint64_t mant;
-    CPU86_LDoubleU tmp;
-
-    fpus = (env->fpus & ~0x3800) | (env->fpstt & 0x7) << 11;
-    fptag = 0;
-    for (i=7; i>=0; i--) {
-	fptag <<= 2;
-	if (env->fptags[i]) {
-            fptag |= 3;
-	} else {
-            tmp.d = env->fpregs[i];
-            exp = EXPD(tmp);
-            mant = MANTD(tmp);
-            if (exp == 0 && mant == 0) {
-                /* zero */
-	        fptag |= 1;
-	    } else if (exp == 0 || exp == MAXEXPD
-#ifdef USE_X86LDOUBLE
-                       || (mant & (1LL << 63)) == 0
-#endif
-                       ) {
-                /* NaNs, infinity, denormal */
-                fptag |= 2;
-            }
-        }
-    }
-    if (data32) {
-        /* 32 bit */
-        stl(ptr, env->fpuc);
-        stl(ptr + 4, fpus);
-        stl(ptr + 8, fptag);
-        stl(ptr + 12, 0);
-        stl(ptr + 16, 0);
-        stl(ptr + 20, 0);
-        stl(ptr + 24, 0);
-    } else {
-        /* 16 bit */
-        stw(ptr, env->fpuc);
-        stw(ptr + 2, fpus);
-        stw(ptr + 4, fptag);
-        stw(ptr + 6, 0);
-        stw(ptr + 8, 0);
-        stw(ptr + 10, 0);
-        stw(ptr + 12, 0);
-    }
-}
-
-void helper_fldenv(uint8_t *ptr, int data32)
-{
-    int i, fpus, fptag;
-
-    if (data32) {
-	env->fpuc = lduw(ptr);
-        fpus = lduw(ptr + 4);
-        fptag = lduw(ptr + 8);
-    }
-    else {
-	env->fpuc = lduw(ptr);
-        fpus = lduw(ptr + 2);
-        fptag = lduw(ptr + 4);
-    }
-    env->fpstt = (fpus >> 11) & 7;
-    env->fpus = fpus & ~0x3800;
-    for(i = 0;i < 7; i++) {
-        env->fptags[i] = ((fptag & 3) == 3);
-        fptag >>= 2;
-    }
-}
-
-void helper_fsave(uint8_t *ptr, int data32)
-{
-    CPU86_LDouble tmp;
-    int i;
-
-    helper_fstenv(ptr, data32);
-
-    ptr += (14 << data32);
-    for(i = 0;i < 8; i++) {
-        tmp = ST(i);
-#ifdef USE_X86LDOUBLE
-        *(long double *)ptr = tmp;
-#else
-        helper_fstt(tmp, ptr);
-#endif        
-        ptr += 10;
-    }
-
-    /* fninit */
-    env->fpus = 0;
-    env->fpstt = 0;
-    env->fpuc = 0x37f;
-    env->fptags[0] = 1;
-    env->fptags[1] = 1;
-    env->fptags[2] = 1;
-    env->fptags[3] = 1;
-    env->fptags[4] = 1;
-    env->fptags[5] = 1;
-    env->fptags[6] = 1;
-    env->fptags[7] = 1;
-}
-
-void helper_frstor(uint8_t *ptr, int data32)
-{
-    CPU86_LDouble tmp;
-    int i;
-
-    helper_fldenv(ptr, data32);
-    ptr += (14 << data32);
-
-    for(i = 0;i < 8; i++) {
-#ifdef USE_X86LDOUBLE
-        tmp = *(long double *)ptr;
-#else
-        tmp = helper_fldt(ptr);
-#endif        
-        ST(i) = tmp;
-        ptr += 10;
-    }
-}
-
 void OPPROTO op_fnstenv_A0(void)
 {
     helper_fstenv((uint8_t *)A0, PARAM1);
@@ -2942,3 +2108,4 @@ void OPPROTO op_unlock(void)
 {
     cpu_unlock();
 }
+