binutils-gdb/gdb/i387-tdep.c
2004-06-26 23:20:30 +00:00

784 lines
22 KiB
C
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

/* Intel 387 floating point stuff.
Copyright 1988, 1989, 1991, 1992, 1993, 1994, 1998, 1999, 2000,
2001, 2002, 2003, 2004 Free Software Foundation, Inc.
This file is part of GDB.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program 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 General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA. */
#include "defs.h"
#include "doublest.h"
#include "floatformat.h"
#include "frame.h"
#include "gdbcore.h"
#include "inferior.h"
#include "language.h"
#include "regcache.h"
#include "value.h"
#include "gdb_assert.h"
#include "gdb_string.h"
#include "i386-tdep.h"
#include "i387-tdep.h"
/* Implement the `info float' layout based on the register definitions
in `tm-i386.h'. */
/* Print the floating point number specified by RAW. */
static void
print_i387_value (char *raw, struct ui_file *file)
{
DOUBLEST value;
/* Using extract_typed_floating here might affect the representation
of certain numbers such as NaNs, even if GDB is running natively.
This is fine since our caller already detects such special
numbers and we print the hexadecimal representation anyway. */
value = extract_typed_floating (raw, builtin_type_i387_ext);
/* We try to print 19 digits. The last digit may or may not contain
garbage, but we'd better print one too many. We need enough room
to print the value, 1 position for the sign, 1 for the decimal
point, 19 for the digits and 6 for the exponent adds up to 27. */
#ifdef PRINTF_HAS_LONG_DOUBLE
fprintf_filtered (file, " %-+27.19Lg", (long double) value);
#else
fprintf_filtered (file, " %-+27.19g", (double) value);
#endif
}
/* Print the classification for the register contents RAW. */
static void
print_i387_ext (unsigned char *raw, struct ui_file *file)
{
int sign;
int integer;
unsigned int exponent;
unsigned long fraction[2];
sign = raw[9] & 0x80;
integer = raw[7] & 0x80;
exponent = (((raw[9] & 0x7f) << 8) | raw[8]);
fraction[0] = ((raw[3] << 24) | (raw[2] << 16) | (raw[1] << 8) | raw[0]);
fraction[1] = (((raw[7] & 0x7f) << 24) | (raw[6] << 16)
| (raw[5] << 8) | raw[4]);
if (exponent == 0x7fff && integer)
{
if (fraction[0] == 0x00000000 && fraction[1] == 0x00000000)
/* Infinity. */
fprintf_filtered (file, " %cInf", (sign ? '-' : '+'));
else if (sign && fraction[0] == 0x00000000 && fraction[1] == 0x40000000)
/* Real Indefinite (QNaN). */
fputs_unfiltered (" Real Indefinite (QNaN)", file);
else if (fraction[1] & 0x40000000)
/* QNaN. */
fputs_filtered (" QNaN", file);
else
/* SNaN. */
fputs_filtered (" SNaN", file);
}
else if (exponent < 0x7fff && exponent > 0x0000 && integer)
/* Normal. */
print_i387_value (raw, file);
else if (exponent == 0x0000)
{
/* Denormal or zero. */
print_i387_value (raw, file);
if (integer)
/* Pseudo-denormal. */
fputs_filtered (" Pseudo-denormal", file);
else if (fraction[0] || fraction[1])
/* Denormal. */
fputs_filtered (" Denormal", file);
}
else
/* Unsupported. */
fputs_filtered (" Unsupported", file);
}
/* Print the status word STATUS. */
static void
print_i387_status_word (unsigned int status, struct ui_file *file)
{
fprintf_filtered (file, "Status Word: %s",
local_hex_string_custom (status, "04"));
fputs_filtered (" ", file);
fprintf_filtered (file, " %s", (status & 0x0001) ? "IE" : " ");
fprintf_filtered (file, " %s", (status & 0x0002) ? "DE" : " ");
fprintf_filtered (file, " %s", (status & 0x0004) ? "ZE" : " ");
fprintf_filtered (file, " %s", (status & 0x0008) ? "OE" : " ");
fprintf_filtered (file, " %s", (status & 0x0010) ? "UE" : " ");
fprintf_filtered (file, " %s", (status & 0x0020) ? "PE" : " ");
fputs_filtered (" ", file);
fprintf_filtered (file, " %s", (status & 0x0080) ? "ES" : " ");
fputs_filtered (" ", file);
fprintf_filtered (file, " %s", (status & 0x0040) ? "SF" : " ");
fputs_filtered (" ", file);
fprintf_filtered (file, " %s", (status & 0x0100) ? "C0" : " ");
fprintf_filtered (file, " %s", (status & 0x0200) ? "C1" : " ");
fprintf_filtered (file, " %s", (status & 0x0400) ? "C2" : " ");
fprintf_filtered (file, " %s", (status & 0x4000) ? "C3" : " ");
fputs_filtered ("\n", file);
fprintf_filtered (file,
" TOP: %d\n", ((status >> 11) & 7));
}
/* Print the control word CONTROL. */
static void
print_i387_control_word (unsigned int control, struct ui_file *file)
{
fprintf_filtered (file, "Control Word: %s",
local_hex_string_custom (control, "04"));
fputs_filtered (" ", file);
fprintf_filtered (file, " %s", (control & 0x0001) ? "IM" : " ");
fprintf_filtered (file, " %s", (control & 0x0002) ? "DM" : " ");
fprintf_filtered (file, " %s", (control & 0x0004) ? "ZM" : " ");
fprintf_filtered (file, " %s", (control & 0x0008) ? "OM" : " ");
fprintf_filtered (file, " %s", (control & 0x0010) ? "UM" : " ");
fprintf_filtered (file, " %s", (control & 0x0020) ? "PM" : " ");
fputs_filtered ("\n", file);
fputs_filtered (" PC: ", file);
switch ((control >> 8) & 3)
{
case 0:
fputs_filtered ("Single Precision (24-bits)\n", file);
break;
case 1:
fputs_filtered ("Reserved\n", file);
break;
case 2:
fputs_filtered ("Double Precision (53-bits)\n", file);
break;
case 3:
fputs_filtered ("Extended Precision (64-bits)\n", file);
break;
}
fputs_filtered (" RC: ", file);
switch ((control >> 10) & 3)
{
case 0:
fputs_filtered ("Round to nearest\n", file);
break;
case 1:
fputs_filtered ("Round down\n", file);
break;
case 2:
fputs_filtered ("Round up\n", file);
break;
case 3:
fputs_filtered ("Round toward zero\n", file);
break;
}
}
/* Print out the i387 floating point state. Note that we ignore FRAME
in the code below. That's OK since floating-point registers are
never saved on the stack. */
void
i387_print_float_info (struct gdbarch *gdbarch, struct ui_file *file,
struct frame_info *frame, const char *args)
{
struct gdbarch_tdep *tdep = gdbarch_tdep (get_frame_arch (frame));
char buf[4];
ULONGEST fctrl;
ULONGEST fstat;
ULONGEST ftag;
ULONGEST fiseg;
ULONGEST fioff;
ULONGEST foseg;
ULONGEST fooff;
ULONGEST fop;
int fpreg;
int top;
gdb_assert (gdbarch == get_frame_arch (frame));
/* Define I387_ST0_REGNUM such that we use the proper definitions
for FRAME's architecture. */
#define I387_ST0_REGNUM tdep->st0_regnum
fctrl = get_frame_register_unsigned (frame, I387_FCTRL_REGNUM);
fstat = get_frame_register_unsigned (frame, I387_FSTAT_REGNUM);
ftag = get_frame_register_unsigned (frame, I387_FTAG_REGNUM);
fiseg = get_frame_register_unsigned (frame, I387_FISEG_REGNUM);
fioff = get_frame_register_unsigned (frame, I387_FIOFF_REGNUM);
foseg = get_frame_register_unsigned (frame, I387_FOSEG_REGNUM);
fooff = get_frame_register_unsigned (frame, I387_FOOFF_REGNUM);
fop = get_frame_register_unsigned (frame, I387_FOP_REGNUM);
top = ((fstat >> 11) & 7);
for (fpreg = 7; fpreg >= 0; fpreg--)
{
unsigned char raw[I386_MAX_REGISTER_SIZE];
int tag = (ftag >> (fpreg * 2)) & 3;
int i;
fprintf_filtered (file, "%sR%d: ", fpreg == top ? "=>" : " ", fpreg);
switch (tag)
{
case 0:
fputs_filtered ("Valid ", file);
break;
case 1:
fputs_filtered ("Zero ", file);
break;
case 2:
fputs_filtered ("Special ", file);
break;
case 3:
fputs_filtered ("Empty ", file);
break;
}
get_frame_register (frame, (fpreg + 8 - top) % 8 + I387_ST0_REGNUM, raw);
fputs_filtered ("0x", file);
for (i = 9; i >= 0; i--)
fprintf_filtered (file, "%02x", raw[i]);
if (tag != 3)
print_i387_ext (raw, file);
fputs_filtered ("\n", file);
}
fputs_filtered ("\n", file);
print_i387_status_word (fstat, file);
print_i387_control_word (fctrl, file);
fprintf_filtered (file, "Tag Word: %s\n",
local_hex_string_custom (ftag, "04"));
fprintf_filtered (file, "Instruction Pointer: %s:",
local_hex_string_custom (fiseg, "02"));
fprintf_filtered (file, "%s\n", local_hex_string_custom (fioff, "08"));
fprintf_filtered (file, "Operand Pointer: %s:",
local_hex_string_custom (foseg, "02"));
fprintf_filtered (file, "%s\n", local_hex_string_custom (fooff, "08"));
fprintf_filtered (file, "Opcode: %s\n",
local_hex_string_custom (fop ? (fop | 0xd800) : 0, "04"));
#undef I387_ST0_REGNUM
}
/* Read a value of type TYPE from register REGNUM in frame FRAME, and
return its contents in TO. */
void
i387_register_to_value (struct frame_info *frame, int regnum,
struct type *type, void *to)
{
char from[I386_MAX_REGISTER_SIZE];
gdb_assert (i386_fp_regnum_p (regnum));
/* We only support floating-point values. */
if (TYPE_CODE (type) != TYPE_CODE_FLT)
{
warning ("Cannot convert floating-point register value "
"to non-floating-point type.");
return;
}
/* Convert to TYPE. This should be a no-op if TYPE is equivalent to
the extended floating-point format used by the FPU. */
get_frame_register (frame, regnum, from);
convert_typed_floating (from, builtin_type_i387_ext, to, type);
}
/* Write the contents FROM of a value of type TYPE into register
REGNUM in frame FRAME. */
void
i387_value_to_register (struct frame_info *frame, int regnum,
struct type *type, const void *from)
{
char to[I386_MAX_REGISTER_SIZE];
gdb_assert (i386_fp_regnum_p (regnum));
/* We only support floating-point values. */
if (TYPE_CODE (type) != TYPE_CODE_FLT)
{
warning ("Cannot convert non-floating-point type "
"to floating-point register value.");
return;
}
/* Convert from TYPE. This should be a no-op if TYPE is equivalent
to the extended floating-point format used by the FPU. */
convert_typed_floating (from, type, to, builtin_type_i387_ext);
put_frame_register (frame, regnum, to);
}
/* Handle FSAVE and FXSAVE formats. */
/* FIXME: kettenis/20030927: The functions below should accept a
`regcache' argument, but I don't want to change the function
signature just yet. There's some band-aid in the functions below
in the form of the `regcache' local variables. This will ease the
transition later on. */
/* At fsave_offset[REGNUM] you'll find the offset to the location in
the data structure used by the "fsave" instruction where GDB
register REGNUM is stored. */
static int fsave_offset[] =
{
28 + 0 * 10, /* %st(0) ... */
28 + 1 * 10,
28 + 2 * 10,
28 + 3 * 10,
28 + 4 * 10,
28 + 5 * 10,
28 + 6 * 10,
28 + 7 * 10, /* ... %st(7). */
0, /* `fctrl' (16 bits). */
4, /* `fstat' (16 bits). */
8, /* `ftag' (16 bits). */
16, /* `fiseg' (16 bits). */
12, /* `fioff'. */
24, /* `foseg' (16 bits). */
20, /* `fooff'. */
18 /* `fop' (bottom 11 bits). */
};
#define FSAVE_ADDR(fsave, regnum) \
(fsave + fsave_offset[regnum - I387_ST0_REGNUM])
/* Fill register REGNUM in REGCACHE with the appropriate value from
*FSAVE. This function masks off any of the reserved bits in
*FSAVE. */
void
i387_supply_fsave (struct regcache *regcache, int regnum, const void *fsave)
{
struct gdbarch_tdep *tdep = gdbarch_tdep (get_regcache_arch (regcache));
const char *regs = fsave;
int i;
gdb_assert (tdep->st0_regnum >= I386_ST0_REGNUM);
/* Define I387_ST0_REGNUM such that we use the proper definitions
for REGCACHE's architecture. */
#define I387_ST0_REGNUM tdep->st0_regnum
for (i = I387_ST0_REGNUM; i < I387_XMM0_REGNUM; i++)
if (regnum == -1 || regnum == i)
{
if (fsave == NULL)
{
regcache_raw_supply (regcache, i, NULL);
continue;
}
/* Most of the FPU control registers occupy only 16 bits in the
fsave area. Give those a special treatment. */
if (i >= I387_FCTRL_REGNUM
&& i != I387_FIOFF_REGNUM && i != I387_FOOFF_REGNUM)
{
unsigned char val[4];
memcpy (val, FSAVE_ADDR (regs, i), 2);
val[2] = val[3] = 0;
if (i == I387_FOP_REGNUM)
val[1] &= ((1 << 3) - 1);
regcache_raw_supply (regcache, i, val);
}
else
regcache_raw_supply (regcache, i, FSAVE_ADDR (regs, i));
}
#undef I387_ST0_REGNUM
}
/* Fill register REGNUM (if it is a floating-point register) in *FSAVE
with the value from REGCACHE. If REGNUM is -1, do this for all
registers. This function doesn't touch any of the reserved bits in
*FSAVE. */
void
i387_collect_fsave (const struct regcache *regcache, int regnum, void *fsave)
{
struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch);
char *regs = fsave;
int i;
gdb_assert (tdep->st0_regnum >= I386_ST0_REGNUM);
/* Define I387_ST0_REGNUM such that we use the proper definitions
for REGCACHE's architecture. */
#define I387_ST0_REGNUM tdep->st0_regnum
for (i = I387_ST0_REGNUM; i < I387_XMM0_REGNUM; i++)
if (regnum == -1 || regnum == i)
{
/* Most of the FPU control registers occupy only 16 bits in
the fsave area. Give those a special treatment. */
if (i >= I387_FCTRL_REGNUM
&& i != I387_FIOFF_REGNUM && i != I387_FOOFF_REGNUM)
{
unsigned char buf[4];
regcache_raw_collect (regcache, i, buf);
if (i == I387_FOP_REGNUM)
{
/* The opcode occupies only 11 bits. Make sure we
don't touch the other bits. */
buf[1] &= ((1 << 3) - 1);
buf[1] |= ((FSAVE_ADDR (regs, i))[1] & ~((1 << 3) - 1));
}
memcpy (FSAVE_ADDR (regs, i), buf, 2);
}
else
regcache_raw_collect (regcache, i, FSAVE_ADDR (regs, i));
}
#undef I387_ST0_REGNUM
}
/* Fill register REGNUM (if it is a floating-point register) in *FSAVE
with the value in GDB's register cache. If REGNUM is -1, do this
for all registers. This function doesn't touch any of the reserved
bits in *FSAVE. */
void
i387_fill_fsave (void *fsave, int regnum)
{
i387_collect_fsave (current_regcache, regnum, fsave);
}
/* At fxsave_offset[REGNUM] you'll find the offset to the location in
the data structure used by the "fxsave" instruction where GDB
register REGNUM is stored. */
static int fxsave_offset[] =
{
32, /* %st(0) through ... */
48,
64,
80,
96,
112,
128,
144, /* ... %st(7) (80 bits each). */
0, /* `fctrl' (16 bits). */
2, /* `fstat' (16 bits). */
4, /* `ftag' (16 bits). */
12, /* `fiseg' (16 bits). */
8, /* `fioff'. */
20, /* `foseg' (16 bits). */
16, /* `fooff'. */
6, /* `fop' (bottom 11 bits). */
160 + 0 * 16, /* %xmm0 through ... */
160 + 1 * 16,
160 + 2 * 16,
160 + 3 * 16,
160 + 4 * 16,
160 + 5 * 16,
160 + 6 * 16,
160 + 7 * 16,
160 + 8 * 16,
160 + 9 * 16,
160 + 10 * 16,
160 + 11 * 16,
160 + 12 * 16,
160 + 13 * 16,
160 + 14 * 16,
160 + 15 * 16, /* ... %xmm15 (128 bits each). */
};
#define FXSAVE_ADDR(fxsave, regnum) \
(fxsave + fxsave_offset[regnum - I387_ST0_REGNUM])
/* We made an unfortunate choice in putting %mxcsr after the SSE
registers %xmm0-%xmm7 instead of before, since it makes supporting
the registers %xmm8-%xmm15 on AMD64 a bit involved. Therefore we
don't include the offset for %mxcsr here above. */
#define FXSAVE_MXCSR_ADDR(fxsave) (fxsave + 24)
static int i387_tag (const unsigned char *raw);
/* Fill register REGNUM in REGCACHE with the appropriate
floating-point or SSE register value from *FXSAVE. This function
masks off any of the reserved bits in *FXSAVE. */
void
i387_supply_fxsave (struct regcache *regcache, int regnum, const void *fxsave)
{
struct gdbarch_tdep *tdep = gdbarch_tdep (get_regcache_arch (regcache));
const char *regs = fxsave;
int i;
gdb_assert (tdep->st0_regnum >= I386_ST0_REGNUM);
gdb_assert (tdep->num_xmm_regs > 0);
/* Define I387_ST0_REGNUM and I387_NUM_XMM_REGS such that we use the
proper definitions for REGCACHE's architecture. */
#define I387_ST0_REGNUM tdep->st0_regnum
#define I387_NUM_XMM_REGS tdep->num_xmm_regs
for (i = I387_ST0_REGNUM; i < I387_MXCSR_REGNUM; i++)
if (regnum == -1 || regnum == i)
{
if (regs == NULL)
{
regcache_raw_supply (regcache, i, NULL);
continue;
}
/* Most of the FPU control registers occupy only 16 bits in
the fxsave area. Give those a special treatment. */
if (i >= I387_FCTRL_REGNUM && i < I387_XMM0_REGNUM
&& i != I387_FIOFF_REGNUM && i != I387_FOOFF_REGNUM)
{
unsigned char val[4];
memcpy (val, FXSAVE_ADDR (regs, i), 2);
val[2] = val[3] = 0;
if (i == I387_FOP_REGNUM)
val[1] &= ((1 << 3) - 1);
else if (i== I387_FTAG_REGNUM)
{
/* The fxsave area contains a simplified version of
the tag word. We have to look at the actual 80-bit
FP data to recreate the traditional i387 tag word. */
unsigned long ftag = 0;
int fpreg;
int top;
top = ((FXSAVE_ADDR (regs, I387_FSTAT_REGNUM))[1] >> 3);
top &= 0x7;
for (fpreg = 7; fpreg >= 0; fpreg--)
{
int tag;
if (val[0] & (1 << fpreg))
{
int regnum = (fpreg + 8 - top) % 8 + I387_ST0_REGNUM;
tag = i387_tag (FXSAVE_ADDR (regs, regnum));
}
else
tag = 3; /* Empty */
ftag |= tag << (2 * fpreg);
}
val[0] = ftag & 0xff;
val[1] = (ftag >> 8) & 0xff;
}
regcache_raw_supply (regcache, i, val);
}
else
regcache_raw_supply (regcache, i, FXSAVE_ADDR (regs, i));
}
if (regnum == I387_MXCSR_REGNUM || regnum == -1)
{
if (regs == NULL)
regcache_raw_supply (regcache, I387_MXCSR_REGNUM, NULL);
else
regcache_raw_supply (regcache, I387_MXCSR_REGNUM,
FXSAVE_MXCSR_ADDR (regs));
}
#undef I387_ST0_REGNUM
#undef I387_NUM_XMM_REGS
}
/* Fill register REGNUM (if it is a floating-point or SSE register) in
*FXSAVE with the value from REGCACHE. If REGNUM is -1, do this for
all registers. This function doesn't touch any of the reserved
bits in *FXSAVE. */
void
i387_collect_fxsave (const struct regcache *regcache, int regnum, void *fxsave)
{
struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch);
char *regs = fxsave;
int i;
gdb_assert (tdep->st0_regnum >= I386_ST0_REGNUM);
gdb_assert (tdep->num_xmm_regs > 0);
/* Define I387_ST0_REGNUM and I387_NUM_XMM_REGS such that we use the
proper definitions for REGCACHE's architecture. */
#define I387_ST0_REGNUM tdep->st0_regnum
#define I387_NUM_XMM_REGS tdep->num_xmm_regs
for (i = I387_ST0_REGNUM; i < I387_MXCSR_REGNUM; i++)
if (regnum == -1 || regnum == i)
{
/* Most of the FPU control registers occupy only 16 bits in
the fxsave area. Give those a special treatment. */
if (i >= I387_FCTRL_REGNUM && i < I387_XMM0_REGNUM
&& i != I387_FIOFF_REGNUM && i != I387_FOOFF_REGNUM)
{
unsigned char buf[4];
regcache_raw_collect (regcache, i, buf);
if (i == I387_FOP_REGNUM)
{
/* The opcode occupies only 11 bits. Make sure we
don't touch the other bits. */
buf[1] &= ((1 << 3) - 1);
buf[1] |= ((FXSAVE_ADDR (regs, i))[1] & ~((1 << 3) - 1));
}
else if (i == I387_FTAG_REGNUM)
{
/* Converting back is much easier. */
unsigned short ftag;
int fpreg;
ftag = (buf[1] << 8) | buf[0];
buf[0] = 0;
buf[1] = 0;
for (fpreg = 7; fpreg >= 0; fpreg--)
{
int tag = (ftag >> (fpreg * 2)) & 3;
if (tag != 3)
buf[0] |= (1 << fpreg);
}
}
memcpy (FXSAVE_ADDR (regs, i), buf, 2);
}
else
regcache_raw_collect (regcache, i, FXSAVE_ADDR (regs, i));
}
if (regnum == I387_MXCSR_REGNUM || regnum == -1)
regcache_raw_collect (regcache, I387_MXCSR_REGNUM,
FXSAVE_MXCSR_ADDR (regs));
#undef I387_ST0_REGNUM
#undef I387_NUM_XMM_REGS
}
/* Fill register REGNUM (if it is a floating-point or SSE register) in
*FXSAVE with the value in GDB's register cache. If REGNUM is -1, do
this for all registers. This function doesn't touch any of the
reserved bits in *FXSAVE. */
void
i387_fill_fxsave (void *fxsave, int regnum)
{
i387_collect_fxsave (current_regcache, regnum, fxsave);
}
/* Recreate the FTW (tag word) valid bits from the 80-bit FP data in
*RAW. */
static int
i387_tag (const unsigned char *raw)
{
int integer;
unsigned int exponent;
unsigned long fraction[2];
integer = raw[7] & 0x80;
exponent = (((raw[9] & 0x7f) << 8) | raw[8]);
fraction[0] = ((raw[3] << 24) | (raw[2] << 16) | (raw[1] << 8) | raw[0]);
fraction[1] = (((raw[7] & 0x7f) << 24) | (raw[6] << 16)
| (raw[5] << 8) | raw[4]);
if (exponent == 0x7fff)
{
/* Special. */
return (2);
}
else if (exponent == 0x0000)
{
if (fraction[0] == 0x0000 && fraction[1] == 0x0000 && !integer)
{
/* Zero. */
return (1);
}
else
{
/* Special. */
return (2);
}
}
else
{
if (integer)
{
/* Valid. */
return (0);
}
else
{
/* Special. */
return (2);
}
}
}
/* Prepare the FPU stack in REGCACHE for a function return. */
void
i387_return_value (struct gdbarch *gdbarch, struct regcache *regcache)
{
struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
ULONGEST fstat;
/* Define I387_ST0_REGNUM such that we use the proper
definitions for the architecture. */
#define I387_ST0_REGNUM tdep->st0_regnum
/* Set the top of the floating-point register stack to 7. The
actual value doesn't really matter, but 7 is what a normal
function return would end up with if the program started out with
a freshly initialized FPU. */
regcache_raw_read_unsigned (regcache, I387_FSTAT_REGNUM, &fstat);
fstat |= (7 << 11);
regcache_raw_write_unsigned (regcache, I387_FSTAT_REGNUM, fstat);
/* Mark %st(1) through %st(7) as empty. Since we set the top of the
floating-point register stack to 7, the appropriate value for the
tag word is 0x3fff. */
regcache_raw_write_unsigned (regcache, I387_FTAG_REGNUM, 0x3fff);
#undef I387_ST0_REGNUM
}