* [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
@ 2002-04-11 16:05 Jason R Thorpe
0 siblings, 0 replies; 8+ messages in thread
From: Jason R Thorpe @ 2002-04-11 16:05 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 1204 bytes --]
Per my query on gdb@ about cross-debugging of core files, I'd like to move
handling of i386 register supply/fill from i387-nat.c to i387-tdep.c. This
renders i387-nat.c empty.
OK to commit?
* Makefile.in (ALLDEPFILES): Remove i387-nat.c.
(i387-nat.o): Delete rule.
* i387-nat.c: Delete file, moving contents to...
* i387-tdep.c: ...here.
* i387-nat.h: Rename...
* i387-tdep.h: ...to this.
* go32-nat.c: Include i387-tdep.h instead of i387-nat.h.
* i386-linux-nat.c: Likewise.
* i386bsd-nat.c: Likewise.
* i386gnu-nat.c: Likewise.
* i386nbsd-nat.c: Likewise.
* i386v4-nat.c: Likewise.
* x86-64-linux-nat.c: Likewise.
* config/i386/fbsd.mh (NATDEPFILES): Remove i387-nat.o.
* config/i386/go32.mh (NATDEPFILES): Likewise.
* config/i386/i386gnu.mh (NATDEPFILES): Likewise.
* config/i386/i386sol2.mh (NATDEPFILES): Likewise.
* config/i386/i386v42mp.mh (NATDEPFILES): Likewise.
* config/i386/linux.mh (NATDEPFILES): Likewise.
* config/i386/nbsd.mh (NATDEPFILES): Likewise.
* config/i386/nbsdelf.mh (NATDEPFILES): Likewise.
* config/i386/obsd.mh (NATDEPFILES): Likewise.
* config/i386/x86-64linux.mh (NATDEPFILES): Likewise.
--
-- Jason R. Thorpe <thorpej@wasabisystems.com>
[-- Attachment #2: i387-tdep.diff --]
[-- Type: text/plain, Size: 39107 bytes --]
Index: Makefile.in
===================================================================
RCS file: /cvs/src/src/gdb/Makefile.in,v
retrieving revision 1.169
diff -c -r1.169 Makefile.in
*** Makefile.in 9 Apr 2002 03:06:13 -0000 1.169
--- Makefile.in 11 Apr 2002 22:43:31 -0000
***************
*** 1184,1190 ****
i386-tdep.c i386b-nat.c i386mach-nat.c i386v-nat.c i386-linux-nat.c \
i386aix-nat.c i386m3-nat.c i386v4-nat.c i386ly-tdep.c \
i386bsd-nat.c i386bsd-tdep.c i386fbsd-nat.c \
! i387-tdep.c i387-nat.c \
i386-linux-tdep.c i386-nat.c \
i960-tdep.c \
ia64-linux-nat.c ia64-linux-tdep.c ia64-tdep.c \
--- 1184,1190 ----
i386-tdep.c i386b-nat.c i386mach-nat.c i386v-nat.c i386-linux-nat.c \
i386aix-nat.c i386m3-nat.c i386v4-nat.c i386ly-tdep.c \
i386bsd-nat.c i386bsd-tdep.c i386fbsd-nat.c \
! i387-tdep.c \
i386-linux-tdep.c i386-nat.c \
i960-tdep.c \
ia64-linux-nat.c ia64-linux-tdep.c ia64-tdep.c \
***************
*** 1611,1619 ****
i387-tdep.o: i387-tdep.c $(floatformat_h) $(defs_h) $(gdbcore_h) \
$(inferior_h) $(language_h) $(regcache_h) $(doublest_h) i386-tdep.h
-
- i387-nat.o: i387-nat.c $(defs_h) $(inferior_h) $(value_h) $(regcache_h) \
- i387-nat.h
i960-tdep.o: i960-tdep.c $(floatformat_h) $(defs_h) $(expression_h) \
$(frame_h) $(gdbtypes_h) $(symtab_h) $(value_h) $(gdbcore_h) \
--- 1611,1616 ----
Index: go32-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/go32-nat.c,v
retrieving revision 1.31
diff -c -r1.31 go32-nat.c
*** go32-nat.c 27 Mar 2002 21:35:35 -0000 1.31
--- go32-nat.c 11 Apr 2002 22:43:33 -0000
***************
*** 29,35 ****
#include "gdbcmd.h"
#include "floatformat.h"
#include "buildsym.h"
! #include "i387-nat.h"
#include "i386-tdep.h"
#include "value.h"
#include "regcache.h"
--- 29,35 ----
#include "gdbcmd.h"
#include "floatformat.h"
#include "buildsym.h"
! #include "i387-tdep.h"
#include "i386-tdep.h"
#include "value.h"
#include "regcache.h"
Index: i386-linux-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386-linux-nat.c,v
retrieving revision 1.35
diff -c -r1.35 i386-linux-nat.c
*** i386-linux-nat.c 24 Feb 2002 22:14:33 -0000 1.35
--- i386-linux-nat.c 11 Apr 2002 22:43:34 -0000
***************
*** 57,63 ****
#include "gregset.h"
/* Prototypes for i387_supply_fsave etc. */
! #include "i387-nat.h"
/* Defines for XMM0_REGNUM etc. */
#include "i386-tdep.h"
--- 57,63 ----
#include "gregset.h"
/* Prototypes for i387_supply_fsave etc. */
! #include "i387-tdep.h"
/* Defines for XMM0_REGNUM etc. */
#include "i386-tdep.h"
Index: i386bsd-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386bsd-nat.c,v
retrieving revision 1.11
diff -c -r1.11 i386bsd-nat.c
*** i386bsd-nat.c 5 Jan 2002 18:36:32 -0000 1.11
--- i386bsd-nat.c 11 Apr 2002 22:43:34 -0000
***************
*** 148,154 ****
regcache_collect (i, REG_ADDR (gregsetp, i));
}
! #include "i387-nat.h"
/* Fill GDB's register array with the floating-point register values
in *FPREGSETP. */
--- 148,154 ----
regcache_collect (i, REG_ADDR (gregsetp, i));
}
! #include "i387-tdep.h"
/* Fill GDB's register array with the floating-point register values
in *FPREGSETP. */
Index: i386gnu-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386gnu-nat.c,v
retrieving revision 1.8
diff -c -r1.8 i386gnu-nat.c
*** i386gnu-nat.c 19 Jan 2002 12:51:04 -0000 1.8
--- i386gnu-nat.c 11 Apr 2002 22:43:34 -0000
***************
*** 34,42 ****
#include <mach/exception.h>
#include "i386-tdep.h"
#include "gnu-nat.h"
- #include "i387-nat.h"
/* Offset to the thread_state_t location where REG is stored. */
--- 34,42 ----
#include <mach/exception.h>
#include "i386-tdep.h"
+ #include "i387-tdep.h"
#include "gnu-nat.h"
/* Offset to the thread_state_t location where REG is stored. */
Index: i386nbsd-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386nbsd-nat.c,v
retrieving revision 1.10
diff -c -r1.10 i386nbsd-nat.c
*** i386nbsd-nat.c 5 Jan 2002 18:36:32 -0000 1.10
--- i386nbsd-nat.c 11 Apr 2002 22:43:34 -0000
***************
*** 39,45 ****
#include "gregset.h"
/* Prototypes for i387_supply_fsave etc. */
! #include "i387-nat.h"
struct md_core
{
--- 39,45 ----
#include "gregset.h"
/* Prototypes for i387_supply_fsave etc. */
! #include "i387-tdep.h"
struct md_core
{
Index: i386v4-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386v4-nat.c,v
retrieving revision 1.9
diff -c -r1.9 i386v4-nat.c
*** i386v4-nat.c 13 Feb 2002 08:10:30 -0000 1.9
--- i386v4-nat.c 11 Apr 2002 22:43:34 -0000
***************
*** 27,33 ****
#ifdef HAVE_SYS_REG_H
#include <sys/reg.h>
#endif
! #include "i387-nat.h"
#ifdef HAVE_SYS_PROCFS_H
--- 27,33 ----
#ifdef HAVE_SYS_REG_H
#include <sys/reg.h>
#endif
! #include "i387-tdep.h"
#ifdef HAVE_SYS_PROCFS_H
Index: i387-nat.c
===================================================================
RCS file: i387-nat.c
diff -N i387-nat.c
*** i387-nat.c 27 Dec 2001 15:22:27 -0000 1.9
--- /dev/null 1 Jan 1970 00:00:00 -0000
***************
*** 1,346 ****
- /* Native-dependent code for the i387.
- Copyright 2000, 2001 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 "inferior.h"
- #include "value.h"
- #include "regcache.h"
-
- #include "i387-nat.h"
- #include "i386-tdep.h"
-
- /* FIXME: kettenis/2000-05-21: Right now more than a few i386 targets
- define their own routines to manage the floating-point registers in
- GDB's register array. Most (if not all) of these targets use the
- format used by the "fsave" instruction in their communication with
- the OS. They should all be converted to use the routines below. */
-
- /* 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 * FPU_REG_RAW_SIZE, /* FP0_REGNUM through ... */
- 28 + 1 * FPU_REG_RAW_SIZE,
- 28 + 2 * FPU_REG_RAW_SIZE,
- 28 + 3 * FPU_REG_RAW_SIZE,
- 28 + 4 * FPU_REG_RAW_SIZE,
- 28 + 5 * FPU_REG_RAW_SIZE,
- 28 + 6 * FPU_REG_RAW_SIZE,
- 28 + 7 * FPU_REG_RAW_SIZE, /* ... FP7_REGNUM. */
- 0, /* FCTRL_REGNUM (16 bits). */
- 4, /* FSTAT_REGNUM (16 bits). */
- 8, /* FTAG_REGNUM (16 bits). */
- 16, /* FISEG_REGNUM (16 bits). */
- 12, /* FIOFF_REGNUM. */
- 24, /* FOSEG_REGNUM. */
- 20, /* FOOFF_REGNUM. */
- 18 /* FOP_REGNUM (bottom 11 bits). */
- };
-
- #define FSAVE_ADDR(fsave, regnum) (fsave + fsave_offset[regnum - FP0_REGNUM])
- \f
-
- /* Fill register REGNUM in GDB's register array with the appropriate
- value from *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
- void
- i387_supply_register (int regnum, char *fsave)
- {
- /* Most of the FPU control registers occupy only 16 bits in
- the fsave area. Give those a special treatment. */
- if (regnum >= FPC_REGNUM
- && regnum != FIOFF_REGNUM && regnum != FOOFF_REGNUM)
- {
- unsigned int val = *(unsigned short *) (FSAVE_ADDR (fsave, regnum));
-
- if (regnum == FOP_REGNUM)
- {
- val &= ((1 << 11) - 1);
- supply_register (regnum, (char *) &val);
- }
- else
- supply_register (regnum, (char *) &val);
- }
- else
- supply_register (regnum, FSAVE_ADDR (fsave, regnum));
- }
-
- /* Fill GDB's register array with the floating-point register values
- in *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
- void
- i387_supply_fsave (char *fsave)
- {
- int i;
-
- for (i = FP0_REGNUM; i < XMM0_REGNUM; i++)
- i387_supply_register (i, fsave);
- }
-
- /* Fill register REGNUM (if it is a floating-point register) in *FSAVE
- with the value in GDB's register array. 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 (char *fsave, int regnum)
- {
- int i;
-
- for (i = FP0_REGNUM; i < 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 >= FPC_REGNUM
- && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
- {
- if (i == FOP_REGNUM)
- {
- unsigned short oldval, newval;
-
- /* The opcode occupies only 11 bits. */
- oldval = (*(unsigned short *) (FSAVE_ADDR (fsave, i)));
- newval = *(unsigned short *) ®isters[REGISTER_BYTE (i)];
- newval &= ((1 << 11) - 1);
- newval |= oldval & ~((1 << 11) - 1);
- memcpy (FSAVE_ADDR (fsave, i), &newval, 2);
- }
- else
- memcpy (FSAVE_ADDR (fsave, i), ®isters[REGISTER_BYTE (i)], 2);
- }
- else
- memcpy (FSAVE_ADDR (fsave, i), ®isters[REGISTER_BYTE (i)],
- REGISTER_RAW_SIZE (i));
- }
- }
- \f
-
- /* 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, /* FP0_REGNUM through ... */
- 48,
- 64,
- 80,
- 96,
- 112,
- 128,
- 144, /* ... FP7_REGNUM (80 bits each). */
- 0, /* FCTRL_REGNUM (16 bits). */
- 2, /* FSTAT_REGNUM (16 bits). */
- 4, /* FTAG_REGNUM (16 bits). */
- 12, /* FISEG_REGNUM (16 bits). */
- 8, /* FIOFF_REGNUM. */
- 20, /* FOSEG_REGNUM (16 bits). */
- 16, /* FOOFF_REGNUM. */
- 6, /* FOP_REGNUM (bottom 11 bits). */
- 160, /* XMM0_REGNUM through ... */
- 176,
- 192,
- 208,
- 224,
- 240,
- 256,
- 272, /* ... XMM7_REGNUM (128 bits each). */
- 24, /* MXCSR_REGNUM. */
- };
-
- #define FXSAVE_ADDR(fxsave, regnum) \
- (fxsave + fxsave_offset[regnum - FP0_REGNUM])
-
- static int i387_tag (unsigned char *raw);
- \f
-
- /* Fill GDB's register array with the floating-point and SSE register
- values in *FXSAVE. This function masks off any of the reserved
- bits in *FXSAVE. */
-
- void
- i387_supply_fxsave (char *fxsave)
- {
- int i;
-
- for (i = FP0_REGNUM; i <= MXCSR_REGNUM; i++)
- {
- /* Most of the FPU control registers occupy only 16 bits in
- the fxsave area. Give those a special treatment. */
- if (i >= FPC_REGNUM && i < XMM0_REGNUM
- && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
- {
- unsigned long val = *(unsigned short *) (FXSAVE_ADDR (fxsave, i));
-
- if (i == FOP_REGNUM)
- {
- val &= ((1 << 11) - 1);
- supply_register (i, (char *) &val);
- }
- else if (i== 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;
- unsigned long fstat;
- int fpreg;
- int top;
-
- fstat = *(unsigned short *) (FXSAVE_ADDR (fxsave, FSTAT_REGNUM));
- top = ((fstat >> 11) & 0x7);
-
- for (fpreg = 7; fpreg >= 0; fpreg--)
- {
- int tag;
-
- if (val & (1 << fpreg))
- {
- int regnum = (fpreg + 8 - top) % 8 + FP0_REGNUM;
- tag = i387_tag (FXSAVE_ADDR (fxsave, regnum));
- }
- else
- tag = 3; /* Empty */
-
- ftag |= tag << (2 * fpreg);
- }
- supply_register (i, (char *) &ftag);
- }
- else
- supply_register (i, (char *) &val);
- }
- else
- supply_register (i, FXSAVE_ADDR (fxsave, i));
- }
- }
-
- /* Fill register REGNUM (if it is a floating-point or SSE register) in
- *FXSAVE with the value in GDB's register array. 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 (char *fxsave, int regnum)
- {
- int i;
-
- for (i = FP0_REGNUM; i <= 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 >= FPC_REGNUM && i < XMM0_REGNUM
- && i != FIOFF_REGNUM && i != FDOFF_REGNUM)
- {
- if (i == FOP_REGNUM)
- {
- unsigned short oldval, newval;
-
- /* The opcode occupies only 11 bits. */
- oldval = (*(unsigned short *) (FXSAVE_ADDR (fxsave, i)));
- newval = *(unsigned short *) ®isters[REGISTER_BYTE (i)];
- newval &= ((1 << 11) - 1);
- newval |= oldval & ~((1 << 11) - 1);
- memcpy (FXSAVE_ADDR (fxsave, i), &newval, 2);
- }
- else if (i == FTAG_REGNUM)
- {
- /* Converting back is much easier. */
-
- unsigned short val = 0;
- unsigned short ftag;
- int fpreg;
-
- ftag = *(unsigned short *) ®isters[REGISTER_BYTE (i)];
-
- for (fpreg = 7; fpreg >= 0; fpreg--)
- {
- int tag = (ftag >> (fpreg * 2)) & 3;
-
- if (tag != 3)
- val |= (1 << fpreg);
- }
-
- memcpy (FXSAVE_ADDR (fxsave, i), &val, 2);
- }
- else
- memcpy (FXSAVE_ADDR (fxsave, i),
- ®isters[REGISTER_BYTE (i)], 2);
- }
- else
- memcpy (FXSAVE_ADDR (fxsave, i), ®isters[REGISTER_BYTE (i)],
- REGISTER_RAW_SIZE (i));
- }
- }
-
- /* Recreate the FTW (tag word) valid bits from the 80-bit FP data in
- *RAW. */
-
- static int
- i387_tag (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);
- }
- }
- }
--- 0 ----
Index: i387-nat.h
===================================================================
RCS file: i387-nat.h
diff -N i387-nat.h
*** i387-nat.h 17 Mar 2001 23:02:10 -0000 1.4
--- /dev/null 1 Jan 1970 00:00:00 -0000
***************
*** 1,56 ****
- /* Native-dependent code for the i387.
- Copyright 2000, 2001 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. */
-
- #ifndef I387_NAT_H
- #define I387_NAT_H
-
- /* Fill register REGNUM in GDB's register array with the appropriate
- value from *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
- extern void i387_supply_register (int regnum, char *fsave);
-
- /* Fill GDB's register array with the floating-point register values
- in *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
- extern void i387_supply_fsave (char *fsave);
-
- /* Fill register REGNUM (if it is a floating-point register) in *FSAVE
- with the value in GDB's register array. If REGNUM is -1, do this
- for all registers. This function doesn't touch any of the reserved
- bits in *FSAVE. */
-
- extern void i387_fill_fsave (char *fsave, int regnum);
-
- /* Fill GDB's register array with the floating-point and SSE register
- values in *FXSAVE. This function masks off any of the reserved
- bits in *FXSAVE. */
-
- extern void i387_supply_fxsave (char *fxsave);
-
- /* Fill register REGNUM (if it is a floating-point or SSE register) in
- *FXSAVE with the value in GDB's register array. If REGNUM is -1, do
- this for all registers. This function doesn't touch any of the
- reserved bits in *FXSAVE. */
-
- extern void i387_fill_fxsave (char *fxsave, int regnum);
-
- #endif /* i387-nat.h */
--- 0 ----
Index: i387-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/i387-tdep.c,v
retrieving revision 1.16
diff -c -r1.16 i387-tdep.c
*** i387-tdep.c 12 Nov 2001 22:27:35 -0000 1.16
--- i387-tdep.c 11 Apr 2002 22:43:34 -0000
***************
*** 388,390 ****
--- 388,709 ----
printf_filtered ("Opcode: %s\n",
local_hex_string_custom (fop ? (fop | 0xd800) : 0, "04"));
}
+
+ /* FIXME: kettenis/2000-05-21: Right now more than a few i386 targets
+ define their own routines to manage the floating-point registers in
+ GDB's register array. Most (if not all) of these targets use the
+ format used by the "fsave" instruction in their communication with
+ the OS. They should all be converted to use the routines below. */
+
+ /* 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 * FPU_REG_RAW_SIZE, /* FP0_REGNUM through ... */
+ 28 + 1 * FPU_REG_RAW_SIZE,
+ 28 + 2 * FPU_REG_RAW_SIZE,
+ 28 + 3 * FPU_REG_RAW_SIZE,
+ 28 + 4 * FPU_REG_RAW_SIZE,
+ 28 + 5 * FPU_REG_RAW_SIZE,
+ 28 + 6 * FPU_REG_RAW_SIZE,
+ 28 + 7 * FPU_REG_RAW_SIZE, /* ... FP7_REGNUM. */
+ 0, /* FCTRL_REGNUM (16 bits). */
+ 4, /* FSTAT_REGNUM (16 bits). */
+ 8, /* FTAG_REGNUM (16 bits). */
+ 16, /* FISEG_REGNUM (16 bits). */
+ 12, /* FIOFF_REGNUM. */
+ 24, /* FOSEG_REGNUM. */
+ 20, /* FOOFF_REGNUM. */
+ 18 /* FOP_REGNUM (bottom 11 bits). */
+ };
+
+ #define FSAVE_ADDR(fsave, regnum) (fsave + fsave_offset[regnum - FP0_REGNUM])
+ \f
+
+ /* Fill register REGNUM in GDB's register array with the appropriate
+ value from *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+ void
+ i387_supply_register (int regnum, char *fsave)
+ {
+ /* Most of the FPU control registers occupy only 16 bits in
+ the fsave area. Give those a special treatment. */
+ if (regnum >= FPC_REGNUM
+ && regnum != FIOFF_REGNUM && regnum != FOOFF_REGNUM)
+ {
+ unsigned int val = *(unsigned short *) (FSAVE_ADDR (fsave, regnum));
+
+ if (regnum == FOP_REGNUM)
+ {
+ val &= ((1 << 11) - 1);
+ supply_register (regnum, (char *) &val);
+ }
+ else
+ supply_register (regnum, (char *) &val);
+ }
+ else
+ supply_register (regnum, FSAVE_ADDR (fsave, regnum));
+ }
+
+ /* Fill GDB's register array with the floating-point register values
+ in *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+ void
+ i387_supply_fsave (char *fsave)
+ {
+ int i;
+
+ for (i = FP0_REGNUM; i < XMM0_REGNUM; i++)
+ i387_supply_register (i, fsave);
+ }
+
+ /* Fill register REGNUM (if it is a floating-point register) in *FSAVE
+ with the value in GDB's register array. 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 (char *fsave, int regnum)
+ {
+ int i;
+
+ for (i = FP0_REGNUM; i < 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 >= FPC_REGNUM
+ && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
+ {
+ if (i == FOP_REGNUM)
+ {
+ unsigned short oldval, newval;
+
+ /* The opcode occupies only 11 bits. */
+ oldval = (*(unsigned short *) (FSAVE_ADDR (fsave, i)));
+ newval = *(unsigned short *) ®isters[REGISTER_BYTE (i)];
+ newval &= ((1 << 11) - 1);
+ newval |= oldval & ~((1 << 11) - 1);
+ memcpy (FSAVE_ADDR (fsave, i), &newval, 2);
+ }
+ else
+ memcpy (FSAVE_ADDR (fsave, i), ®isters[REGISTER_BYTE (i)], 2);
+ }
+ else
+ memcpy (FSAVE_ADDR (fsave, i), ®isters[REGISTER_BYTE (i)],
+ REGISTER_RAW_SIZE (i));
+ }
+ }
+ \f
+
+ /* 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, /* FP0_REGNUM through ... */
+ 48,
+ 64,
+ 80,
+ 96,
+ 112,
+ 128,
+ 144, /* ... FP7_REGNUM (80 bits each). */
+ 0, /* FCTRL_REGNUM (16 bits). */
+ 2, /* FSTAT_REGNUM (16 bits). */
+ 4, /* FTAG_REGNUM (16 bits). */
+ 12, /* FISEG_REGNUM (16 bits). */
+ 8, /* FIOFF_REGNUM. */
+ 20, /* FOSEG_REGNUM (16 bits). */
+ 16, /* FOOFF_REGNUM. */
+ 6, /* FOP_REGNUM (bottom 11 bits). */
+ 160, /* XMM0_REGNUM through ... */
+ 176,
+ 192,
+ 208,
+ 224,
+ 240,
+ 256,
+ 272, /* ... XMM7_REGNUM (128 bits each). */
+ 24, /* MXCSR_REGNUM. */
+ };
+
+ #define FXSAVE_ADDR(fxsave, regnum) \
+ (fxsave + fxsave_offset[regnum - FP0_REGNUM])
+
+ static int i387_tag (unsigned char *raw);
+ \f
+
+ /* Fill GDB's register array with the floating-point and SSE register
+ values in *FXSAVE. This function masks off any of the reserved
+ bits in *FXSAVE. */
+
+ void
+ i387_supply_fxsave (char *fxsave)
+ {
+ int i;
+
+ for (i = FP0_REGNUM; i <= MXCSR_REGNUM; i++)
+ {
+ /* Most of the FPU control registers occupy only 16 bits in
+ the fxsave area. Give those a special treatment. */
+ if (i >= FPC_REGNUM && i < XMM0_REGNUM
+ && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
+ {
+ unsigned long val = *(unsigned short *) (FXSAVE_ADDR (fxsave, i));
+
+ if (i == FOP_REGNUM)
+ {
+ val &= ((1 << 11) - 1);
+ supply_register (i, (char *) &val);
+ }
+ else if (i== 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;
+ unsigned long fstat;
+ int fpreg;
+ int top;
+
+ fstat = *(unsigned short *) (FXSAVE_ADDR (fxsave, FSTAT_REGNUM));
+ top = ((fstat >> 11) & 0x7);
+
+ for (fpreg = 7; fpreg >= 0; fpreg--)
+ {
+ int tag;
+
+ if (val & (1 << fpreg))
+ {
+ int regnum = (fpreg + 8 - top) % 8 + FP0_REGNUM;
+ tag = i387_tag (FXSAVE_ADDR (fxsave, regnum));
+ }
+ else
+ tag = 3; /* Empty */
+
+ ftag |= tag << (2 * fpreg);
+ }
+ supply_register (i, (char *) &ftag);
+ }
+ else
+ supply_register (i, (char *) &val);
+ }
+ else
+ supply_register (i, FXSAVE_ADDR (fxsave, i));
+ }
+ }
+
+ /* Fill register REGNUM (if it is a floating-point or SSE register) in
+ *FXSAVE with the value in GDB's register array. 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 (char *fxsave, int regnum)
+ {
+ int i;
+
+ for (i = FP0_REGNUM; i <= 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 >= FPC_REGNUM && i < XMM0_REGNUM
+ && i != FIOFF_REGNUM && i != FDOFF_REGNUM)
+ {
+ if (i == FOP_REGNUM)
+ {
+ unsigned short oldval, newval;
+
+ /* The opcode occupies only 11 bits. */
+ oldval = (*(unsigned short *) (FXSAVE_ADDR (fxsave, i)));
+ newval = *(unsigned short *) ®isters[REGISTER_BYTE (i)];
+ newval &= ((1 << 11) - 1);
+ newval |= oldval & ~((1 << 11) - 1);
+ memcpy (FXSAVE_ADDR (fxsave, i), &newval, 2);
+ }
+ else if (i == FTAG_REGNUM)
+ {
+ /* Converting back is much easier. */
+
+ unsigned short val = 0;
+ unsigned short ftag;
+ int fpreg;
+
+ ftag = *(unsigned short *) ®isters[REGISTER_BYTE (i)];
+
+ for (fpreg = 7; fpreg >= 0; fpreg--)
+ {
+ int tag = (ftag >> (fpreg * 2)) & 3;
+
+ if (tag != 3)
+ val |= (1 << fpreg);
+ }
+
+ memcpy (FXSAVE_ADDR (fxsave, i), &val, 2);
+ }
+ else
+ memcpy (FXSAVE_ADDR (fxsave, i),
+ ®isters[REGISTER_BYTE (i)], 2);
+ }
+ else
+ memcpy (FXSAVE_ADDR (fxsave, i), ®isters[REGISTER_BYTE (i)],
+ REGISTER_RAW_SIZE (i));
+ }
+ }
+
+ /* Recreate the FTW (tag word) valid bits from the 80-bit FP data in
+ *RAW. */
+
+ static int
+ i387_tag (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);
+ }
+ }
+ }
Index: i387-tdep.h
===================================================================
RCS file: i387-tdep.h
diff -N i387-tdep.h
*** /dev/null 1 Jan 1970 00:00:00 -0000
--- i387-tdep.h 11 Apr 2002 22:43:34 -0000
***************
*** 0 ****
--- 1,56 ----
+ /* Target-dependent code for the i387.
+ Copyright 2000, 2001 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. */
+
+ #ifndef I387_TDEP_H
+ #define I387_TDEP_H
+
+ /* Fill register REGNUM in GDB's register array with the appropriate
+ value from *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+ extern void i387_supply_register (int regnum, char *fsave);
+
+ /* Fill GDB's register array with the floating-point register values
+ in *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+ extern void i387_supply_fsave (char *fsave);
+
+ /* Fill register REGNUM (if it is a floating-point register) in *FSAVE
+ with the value in GDB's register array. If REGNUM is -1, do this
+ for all registers. This function doesn't touch any of the reserved
+ bits in *FSAVE. */
+
+ extern void i387_fill_fsave (char *fsave, int regnum);
+
+ /* Fill GDB's register array with the floating-point and SSE register
+ values in *FXSAVE. This function masks off any of the reserved
+ bits in *FXSAVE. */
+
+ extern void i387_supply_fxsave (char *fxsave);
+
+ /* Fill register REGNUM (if it is a floating-point or SSE register) in
+ *FXSAVE with the value in GDB's register array. If REGNUM is -1, do
+ this for all registers. This function doesn't touch any of the
+ reserved bits in *FXSAVE. */
+
+ extern void i387_fill_fxsave (char *fxsave, int regnum);
+
+ #endif /* i387-tdep.h */
Index: x86-64-linux-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/x86-64-linux-nat.c,v
retrieving revision 1.9
diff -c -r1.9 x86-64-linux-nat.c
*** x86-64-linux-nat.c 10 Apr 2002 12:12:33 -0000 1.9
--- x86-64-linux-nat.c 11 Apr 2002 22:43:35 -0000
***************
*** 25,31 ****
#include "inferior.h"
#include "gdbcore.h"
#include "regcache.h"
! #include "i387-nat.h"
#include "gdb_assert.h"
#include "x86-64-tdep.h"
--- 25,31 ----
#include "inferior.h"
#include "gdbcore.h"
#include "regcache.h"
! #include "i387-tdep.h"
#include "gdb_assert.h"
#include "x86-64-tdep.h"
Index: config/i386/fbsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/fbsd.mh,v
retrieving revision 1.11
diff -c -r1.11 fbsd.mh
*** config/i386/fbsd.mh 31 Mar 2002 17:47:17 -0000 1.11
--- config/i386/fbsd.mh 11 Apr 2002 22:43:36 -0000
***************
*** 4,7 ****
NAT_FILE= nm-fbsd.h
# NOTE: Do not spread NATDEPFILES over several lines - it hurts BSD make.
! NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-svr4.o solib-legacy.o corelow.o core-aout.o core-regset.o i386-nat.o i387-nat.o i386bsd-nat.o i386fbsd-nat.o gcore.o fbsd-proc.o
--- 4,7 ----
NAT_FILE= nm-fbsd.h
# NOTE: Do not spread NATDEPFILES over several lines - it hurts BSD make.
! NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-svr4.o solib-legacy.o corelow.o core-aout.o core-regset.o i386-nat.o i386bsd-nat.o i386fbsd-nat.o gcore.o fbsd-proc.o
Index: config/i386/go32.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/go32.mh,v
retrieving revision 1.5
diff -c -r1.5 go32.mh
*** config/i386/go32.mh 18 Jan 2002 04:50:59 -0000 1.5
--- config/i386/go32.mh 11 Apr 2002 22:43:36 -0000
***************
*** 7,13 ****
XM_FILE= xm-go32.h
NAT_FILE= nm-go32.h
! NATDEPFILES= go32-nat.o i386-nat.o i387-nat.o
TERMCAP=
HOST_IPC=
--- 7,13 ----
XM_FILE= xm-go32.h
NAT_FILE= nm-go32.h
! NATDEPFILES= go32-nat.o i386-nat.o
TERMCAP=
HOST_IPC=
Index: config/i386/i386gnu.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/i386gnu.mh,v
retrieving revision 1.6
diff -c -r1.6 i386gnu.mh
*** config/i386/i386gnu.mh 18 Jan 2002 04:50:59 -0000 1.6
--- config/i386/i386gnu.mh 11 Apr 2002 22:43:36 -0000
***************
*** 1,5 ****
# Host: Intel 386 running the GNU Hurd
! NATDEPFILES= i386gnu-nat.o i387-nat.o gnu-nat.o fork-child.o solib.o solib-svr4.o solib-legacy.o corelow.o notify_S.o process_reply_S.o msg_reply_S.o msg_U.o exc_request_U.o exc_request_S.o
XM_FILE= xm-i386gnu.h
NAT_FILE= nm-gnu.h
MH_CFLAGS = -D_GNU_SOURCE
--- 1,5 ----
# Host: Intel 386 running the GNU Hurd
! NATDEPFILES= i386gnu-nat.o gnu-nat.o fork-child.o solib.o solib-svr4.o solib-legacy.o corelow.o notify_S.o process_reply_S.o msg_reply_S.o msg_U.o exc_request_U.o exc_request_S.o
XM_FILE= xm-i386gnu.h
NAT_FILE= nm-gnu.h
MH_CFLAGS = -D_GNU_SOURCE
Index: config/i386/i386sol2.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/i386sol2.mh,v
retrieving revision 1.4
diff -c -r1.4 i386sol2.mh
*** config/i386/i386sol2.mh 14 Feb 2002 08:13:33 -0000 1.4
--- config/i386/i386sol2.mh 11 Apr 2002 22:43:36 -0000
***************
*** 4,8 ****
XM_CLIBS= -lsocket -lnsl
NAT_FILE= nm-i386sol2.h
! NATDEPFILES= core-regset.o fork-child.o i386v4-nat.o i387-nat.o corelow.o \
procfs.o proc-api.o proc-events.o proc-flags.o proc-why.o gcore.o
--- 4,8 ----
XM_CLIBS= -lsocket -lnsl
NAT_FILE= nm-i386sol2.h
! NATDEPFILES= core-regset.o fork-child.o i386v4-nat.o corelow.o \
procfs.o proc-api.o proc-events.o proc-flags.o proc-why.o gcore.o
Index: config/i386/i386v42mp.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/i386v42mp.mh,v
retrieving revision 1.7
diff -c -r1.7 i386v42mp.mh
*** config/i386/i386v42mp.mh 11 Apr 2002 18:13:21 -0000 1.7
--- config/i386/i386v42mp.mh 11 Apr 2002 22:43:36 -0000
***************
*** 14,20 ****
# continuation character (backslash) to extend a commented line. As a
# consequence, make considers subsequent tab-indented lines to be
# some sort of error.
! NATDEPFILES= corelow.o core-regset.o fork-child.o i386v4-nat.o i387-nat.o \
gcore.o solib.o solib-svr4.o solib-legacy.o procfs.o proc-api.o \
proc-events.o proc-flags.o proc-why.o uw-thread.o
--- 14,20 ----
# continuation character (backslash) to extend a commented line. As a
# consequence, make considers subsequent tab-indented lines to be
# some sort of error.
! NATDEPFILES= corelow.o core-regset.o fork-child.o i386v4-nat.o \
gcore.o solib.o solib-svr4.o solib-legacy.o procfs.o proc-api.o \
proc-events.o proc-flags.o proc-why.o uw-thread.o
Index: config/i386/linux.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/linux.mh,v
retrieving revision 1.11
diff -c -r1.11 linux.mh
*** config/i386/linux.mh 18 Jan 2002 04:50:59 -0000 1.11
--- config/i386/linux.mh 11 Apr 2002 22:43:36 -0000
***************
*** 4,10 ****
NAT_FILE= nm-linux.h
NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o linux-proc.o \
! core-aout.o i386-nat.o i386-linux-nat.o i387-nat.o \
proc-service.o thread-db.o lin-lwp.o linux-proc.o gcore.o
# The dynamically loaded libthread_db needs access to symbols in the
--- 4,10 ----
NAT_FILE= nm-linux.h
NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o linux-proc.o \
! core-aout.o i386-nat.o i386-linux-nat.o \
proc-service.o thread-db.o lin-lwp.o linux-proc.o gcore.o
# The dynamically loaded libthread_db needs access to symbols in the
Index: config/i386/nbsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/nbsd.mh,v
retrieving revision 1.12
diff -c -r1.12 nbsd.mh
*** config/i386/nbsd.mh 18 Jan 2002 04:50:59 -0000 1.12
--- config/i386/nbsd.mh 11 Apr 2002 22:43:36 -0000
***************
*** 6,12 ****
# continuation character (backslash) to extend a commented line. As a
# consequence, BSD make considers subsequent tab-indented lines to be
# "unassociated shell commands".
! NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i387-nat.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-sunos.o
XM_FILE= xm-nbsd.h
NAT_FILE= nm-nbsd.h
--- 6,12 ----
# continuation character (backslash) to extend a commented line. As a
# consequence, BSD make considers subsequent tab-indented lines to be
# "unassociated shell commands".
! NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-sunos.o
XM_FILE= xm-nbsd.h
NAT_FILE= nm-nbsd.h
Index: config/i386/nbsdelf.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/nbsdelf.mh,v
retrieving revision 1.8
diff -c -r1.8 nbsdelf.mh
*** config/i386/nbsdelf.mh 18 Jan 2002 04:50:59 -0000 1.8
--- config/i386/nbsdelf.mh 11 Apr 2002 22:43:36 -0000
***************
*** 1,4 ****
# Host: Intel 386 running NetBSD
! NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i387-nat.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-svr4.o solib-legacy.o
XM_FILE= xm-nbsd.h
NAT_FILE= nm-nbsdelf.h
--- 1,4 ----
# Host: Intel 386 running NetBSD
! NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-svr4.o solib-legacy.o
XM_FILE= xm-nbsd.h
NAT_FILE= nm-nbsdelf.h
Index: config/i386/obsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/obsd.mh,v
retrieving revision 1.2
diff -c -r1.2 obsd.mh
*** config/i386/obsd.mh 18 Jan 2002 04:50:59 -0000 1.2
--- config/i386/obsd.mh 11 Apr 2002 22:43:36 -0000
***************
*** 4,7 ****
NAT_FILE= nm-obsd.h
# NOTE: Do not spread NATDEPFILES over several lines - it hurts BSD make.
! NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-sunos.o corelow.o core-aout.o i386-nat.o i387-nat.o i386bsd-nat.o
--- 4,7 ----
NAT_FILE= nm-obsd.h
# NOTE: Do not spread NATDEPFILES over several lines - it hurts BSD make.
! NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-sunos.o corelow.o core-aout.o i386-nat.o i386bsd-nat.o
Index: config/i386/x86-64linux.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/x86-64linux.mh,v
retrieving revision 1.5
diff -c -r1.5 x86-64linux.mh
*** config/i386/x86-64linux.mh 21 Feb 2002 12:48:36 -0000 1.5
--- config/i386/x86-64linux.mh 11 Apr 2002 22:43:36 -0000
***************
*** 5,11 ****
NAT_FILE= nm-x86-64.h
NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o \
core-aout.o i386-nat.o x86-64-linux-nat.o \
! i387-nat.o proc-service.o thread-db.o lin-lwp.o \
linux-proc.o gcore.o
LOADLIBES = -ldl -rdynamic
--- 5,11 ----
NAT_FILE= nm-x86-64.h
NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o \
core-aout.o i386-nat.o x86-64-linux-nat.o \
! proc-service.o thread-db.o lin-lwp.o \
linux-proc.o gcore.o
LOADLIBES = -ldl -rdynamic
^ permalink raw reply [flat|nested] 8+ messages in thread* [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
@ 2002-05-09 16:45 Jason R Thorpe
2002-05-09 17:12 ` Daniel Jacobowitz
2002-05-10 0:26 ` Eli Zaretskii
0 siblings, 2 replies; 8+ messages in thread
From: Jason R Thorpe @ 2002-05-09 16:45 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 1507 bytes --]
This is a re-submit of an earlier patch, which takes into account
Mark's recent changes to i387-nat.c.
Basically, this moves all of the register supply/fill stuff for i387
from i387-nat.c to i387-tdep.c, to facilitate cross-debugging of core
files. This renders i387-nat.c empty, and so it is removed in this
patch.
OK to commit?
* Makefile.in (ALLDEPFILES): Remove i387-nat.c.
(i387-nat.o): Delete rule.
* i387-nat.c: Delete file, moving contents to...
* i387-tdep.c: ...here.
* i387-nat.h: Rename...
* i387-tdep.h: ...to this.
* go32-nat.c: Include i387-tdep.h instead of i387-nat.h.
* i386-linux-nat.c: Likewise.
* i386bsd-nat.c: Likewise.
* i386gnu-nat.c: Likewise.
* i386nbsd-nat.c: Likewise.
* i386v4-nat.c: Likewise.
* x86-64-linux-nat.c: Likewise.
* config/i386/fbsd.mh (NATDEPFILES): Remove i387-nat.o.
* config/i386/go32.mh (NATDEPFILES): Likewise.
* config/i386/i386gnu.mh (NATDEPFILES): Likewise.
* config/i386/i386sol2.mh (NATDEPFILES): Likewise.
* config/i386/i386v42mp.mh (NATDEPFILES): Likewise.
* config/i386/linux.mh (NATDEPFILES): Likewise.
* config/i386/nbsd.mh (NATDEPFILES): Likewise.
* config/i386/nbsdelf.mh (NATDEPFILES): Likewise.
* config/i386/obsd.mh (NATDEPFILES): Likewise.
* config/i386/x86-64linux.mh (NATDEPFILES): Likewise.
--
-- Jason R. Thorpe <thorpej@wasabisystems.com>
[-- Attachment #2: i387-patch --]
[-- Type: text/plain, Size: 33399 bytes --]
Index: go32-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/go32-nat.c,v
retrieving revision 1.32
diff -u -r1.32 go32-nat.c
--- go32-nat.c 19 Apr 2002 19:28:48 -0000 1.32
+++ go32-nat.c 9 May 2002 23:33:58 -0000
@@ -29,7 +29,7 @@
#include "gdbcmd.h"
#include "floatformat.h"
#include "buildsym.h"
-#include "i387-nat.h"
+#include "i387-tdep.h"
#include "i386-tdep.h"
#include "value.h"
#include "regcache.h"
Index: i386-linux-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386-linux-nat.c,v
retrieving revision 1.35
diff -u -r1.35 i386-linux-nat.c
--- i386-linux-nat.c 24 Feb 2002 22:14:33 -0000 1.35
+++ i386-linux-nat.c 9 May 2002 23:33:58 -0000
@@ -57,7 +57,7 @@
#include "gregset.h"
/* Prototypes for i387_supply_fsave etc. */
-#include "i387-nat.h"
+#include "i387-tdep.h"
/* Defines for XMM0_REGNUM etc. */
#include "i386-tdep.h"
Index: i386bsd-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386bsd-nat.c,v
retrieving revision 1.11
diff -u -r1.11 i386bsd-nat.c
--- i386bsd-nat.c 5 Jan 2002 18:36:32 -0000 1.11
+++ i386bsd-nat.c 9 May 2002 23:33:58 -0000
@@ -148,7 +148,7 @@
regcache_collect (i, REG_ADDR (gregsetp, i));
}
-#include "i387-nat.h"
+#include "i387-tdep.h"
/* Fill GDB's register array with the floating-point register values
in *FPREGSETP. */
Index: i386gnu-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386gnu-nat.c,v
retrieving revision 1.9
diff -u -r1.9 i386gnu-nat.c
--- i386gnu-nat.c 24 Apr 2002 21:44:46 -0000 1.9
+++ i386gnu-nat.c 9 May 2002 23:33:59 -0000
@@ -34,9 +34,9 @@
#include <mach/exception.h>
#include "i386-tdep.h"
+#include "i387-tdep.h"
#include "gnu-nat.h"
-#include "i387-nat.h"
#ifdef HAVE_SYS_PROCFS_H
# include <sys/procfs.h>
Index: i386nbsd-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386nbsd-nat.c,v
retrieving revision 1.10
diff -u -r1.10 i386nbsd-nat.c
--- i386nbsd-nat.c 5 Jan 2002 18:36:32 -0000 1.10
+++ i386nbsd-nat.c 9 May 2002 23:33:59 -0000
@@ -39,7 +39,7 @@
#include "gregset.h"
/* Prototypes for i387_supply_fsave etc. */
-#include "i387-nat.h"
+#include "i387-tdep.h"
struct md_core
{
Index: i386v4-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/i386v4-nat.c,v
retrieving revision 1.9
diff -u -r1.9 i386v4-nat.c
--- i386v4-nat.c 13 Feb 2002 08:10:30 -0000 1.9
+++ i386v4-nat.c 9 May 2002 23:33:59 -0000
@@ -27,7 +27,7 @@
#ifdef HAVE_SYS_REG_H
#include <sys/reg.h>
#endif
-#include "i387-nat.h"
+#include "i387-tdep.h"
#ifdef HAVE_SYS_PROCFS_H
Index: i387-nat.c
===================================================================
RCS file: i387-nat.c
diff -N i387-nat.c
--- i387-nat.c 6 May 2002 11:39:51 -0000 1.10
+++ /dev/null 1 Jan 1970 00:00:00 -0000
@@ -1,351 +0,0 @@
-/* Native-dependent code for the i387.
- Copyright 2000, 2001 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 "inferior.h"
-#include "value.h"
-#include "regcache.h"
-
-#include "i387-nat.h"
-#include "i386-tdep.h"
-
-/* FIXME: kettenis/2000-05-21: Right now more than a few i386 targets
- define their own routines to manage the floating-point registers in
- GDB's register array. Most (if not all) of these targets use the
- format used by the "fsave" instruction in their communication with
- the OS. They should all be converted to use the routines below. */
-
-/* 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 * FPU_REG_RAW_SIZE, /* FP0_REGNUM through ... */
- 28 + 1 * FPU_REG_RAW_SIZE,
- 28 + 2 * FPU_REG_RAW_SIZE,
- 28 + 3 * FPU_REG_RAW_SIZE,
- 28 + 4 * FPU_REG_RAW_SIZE,
- 28 + 5 * FPU_REG_RAW_SIZE,
- 28 + 6 * FPU_REG_RAW_SIZE,
- 28 + 7 * FPU_REG_RAW_SIZE, /* ... FP7_REGNUM. */
- 0, /* FCTRL_REGNUM (16 bits). */
- 4, /* FSTAT_REGNUM (16 bits). */
- 8, /* FTAG_REGNUM (16 bits). */
- 16, /* FISEG_REGNUM (16 bits). */
- 12, /* FIOFF_REGNUM. */
- 24, /* FOSEG_REGNUM. */
- 20, /* FOOFF_REGNUM. */
- 18 /* FOP_REGNUM (bottom 11 bits). */
-};
-
-#define FSAVE_ADDR(fsave, regnum) (fsave + fsave_offset[regnum - FP0_REGNUM])
-\f
-
-/* Fill register REGNUM in GDB's register array with the appropriate
- value from *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
-void
-i387_supply_register (int regnum, char *fsave)
-{
- /* Most of the FPU control registers occupy only 16 bits in
- the fsave area. Give those a special treatment. */
- if (regnum >= FPC_REGNUM
- && regnum != FIOFF_REGNUM && regnum != FOOFF_REGNUM)
- {
- unsigned int val = *(unsigned short *) (FSAVE_ADDR (fsave, regnum));
-
- if (regnum == FOP_REGNUM)
- {
- val &= ((1 << 11) - 1);
- supply_register (regnum, (char *) &val);
- }
- else
- supply_register (regnum, (char *) &val);
- }
- else
- supply_register (regnum, FSAVE_ADDR (fsave, regnum));
-}
-
-/* Fill GDB's register array with the floating-point register values
- in *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
-void
-i387_supply_fsave (char *fsave)
-{
- int i;
-
- for (i = FP0_REGNUM; i < XMM0_REGNUM; i++)
- i387_supply_register (i, fsave);
-}
-
-/* Fill register REGNUM (if it is a floating-point register) in *FSAVE
- with the value in GDB's register array. 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 (char *fsave, int regnum)
-{
- int i;
-
- for (i = FP0_REGNUM; i < 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 >= FPC_REGNUM
- && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
- {
- char buf[4];
-
- regcache_collect (i, buf);
-
- if (i == FOP_REGNUM)
- {
- unsigned short oldval, newval;
-
- /* The opcode occupies only 11 bits. */
- oldval = (*(unsigned short *) (FSAVE_ADDR (fsave, i)));
- newval = *(unsigned short *) buf;
- newval &= ((1 << 11) - 1);
- newval |= oldval & ~((1 << 11) - 1);
- memcpy (FSAVE_ADDR (fsave, i), &newval, 2);
- }
- else
- memcpy (FSAVE_ADDR (fsave, i), buf, 2);
- }
- else
- regcache_collect (i, FSAVE_ADDR (fsave, i));
- }
-}
-\f
-
-/* 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, /* FP0_REGNUM through ... */
- 48,
- 64,
- 80,
- 96,
- 112,
- 128,
- 144, /* ... FP7_REGNUM (80 bits each). */
- 0, /* FCTRL_REGNUM (16 bits). */
- 2, /* FSTAT_REGNUM (16 bits). */
- 4, /* FTAG_REGNUM (16 bits). */
- 12, /* FISEG_REGNUM (16 bits). */
- 8, /* FIOFF_REGNUM. */
- 20, /* FOSEG_REGNUM (16 bits). */
- 16, /* FOOFF_REGNUM. */
- 6, /* FOP_REGNUM (bottom 11 bits). */
- 160, /* XMM0_REGNUM through ... */
- 176,
- 192,
- 208,
- 224,
- 240,
- 256,
- 272, /* ... XMM7_REGNUM (128 bits each). */
- 24, /* MXCSR_REGNUM. */
-};
-
-#define FXSAVE_ADDR(fxsave, regnum) \
- (fxsave + fxsave_offset[regnum - FP0_REGNUM])
-
-static int i387_tag (unsigned char *raw);
-\f
-
-/* Fill GDB's register array with the floating-point and SSE register
- values in *FXSAVE. This function masks off any of the reserved
- bits in *FXSAVE. */
-
-void
-i387_supply_fxsave (char *fxsave)
-{
- int i;
-
- for (i = FP0_REGNUM; i <= MXCSR_REGNUM; i++)
- {
- /* Most of the FPU control registers occupy only 16 bits in
- the fxsave area. Give those a special treatment. */
- if (i >= FPC_REGNUM && i < XMM0_REGNUM
- && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
- {
- unsigned long val = *(unsigned short *) (FXSAVE_ADDR (fxsave, i));
-
- if (i == FOP_REGNUM)
- {
- val &= ((1 << 11) - 1);
- supply_register (i, (char *) &val);
- }
- else if (i== 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;
- unsigned long fstat;
- int fpreg;
- int top;
-
- fstat = *(unsigned short *) (FXSAVE_ADDR (fxsave, FSTAT_REGNUM));
- top = ((fstat >> 11) & 0x7);
-
- for (fpreg = 7; fpreg >= 0; fpreg--)
- {
- int tag;
-
- if (val & (1 << fpreg))
- {
- int regnum = (fpreg + 8 - top) % 8 + FP0_REGNUM;
- tag = i387_tag (FXSAVE_ADDR (fxsave, regnum));
- }
- else
- tag = 3; /* Empty */
-
- ftag |= tag << (2 * fpreg);
- }
- supply_register (i, (char *) &ftag);
- }
- else
- supply_register (i, (char *) &val);
- }
- else
- supply_register (i, FXSAVE_ADDR (fxsave, i));
- }
-}
-
-/* Fill register REGNUM (if it is a floating-point or SSE register) in
- *FXSAVE with the value in GDB's register array. 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 (char *fxsave, int regnum)
-{
- int i;
-
- for (i = FP0_REGNUM; i <= 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 >= FPC_REGNUM && i < XMM0_REGNUM
- && i != FIOFF_REGNUM && i != FDOFF_REGNUM)
- {
- char buf[4];
-
- regcache_collect (i, buf);
-
- if (i == FOP_REGNUM)
- {
- unsigned short oldval, newval;
-
- /* The opcode occupies only 11 bits. */
- oldval = (*(unsigned short *) (FXSAVE_ADDR (fxsave, i)));
- newval = *(unsigned short *) buf;
- newval &= ((1 << 11) - 1);
- newval |= oldval & ~((1 << 11) - 1);
- memcpy (FXSAVE_ADDR (fxsave, i), &newval, 2);
- }
- else if (i == FTAG_REGNUM)
- {
- /* Converting back is much easier. */
-
- unsigned short val = 0;
- unsigned short ftag;
- int fpreg;
-
- ftag = *(unsigned short *) buf;
-
- for (fpreg = 7; fpreg >= 0; fpreg--)
- {
- int tag = (ftag >> (fpreg * 2)) & 3;
-
- if (tag != 3)
- val |= (1 << fpreg);
- }
-
- memcpy (FXSAVE_ADDR (fxsave, i), &val, 2);
- }
- else
- memcpy (FXSAVE_ADDR (fxsave, i), buf, 2);
- }
- else
- regcache_collect (i, FXSAVE_ADDR (fxsave, i));
- }
-}
-
-/* Recreate the FTW (tag word) valid bits from the 80-bit FP data in
- *RAW. */
-
-static int
-i387_tag (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);
- }
- }
-}
Index: i387-nat.h
===================================================================
RCS file: i387-nat.h
diff -N i387-nat.h
--- i387-nat.h 17 Mar 2001 23:02:10 -0000 1.4
+++ /dev/null 1 Jan 1970 00:00:00 -0000
@@ -1,56 +0,0 @@
-/* Native-dependent code for the i387.
- Copyright 2000, 2001 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. */
-
-#ifndef I387_NAT_H
-#define I387_NAT_H
-
-/* Fill register REGNUM in GDB's register array with the appropriate
- value from *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
-extern void i387_supply_register (int regnum, char *fsave);
-
-/* Fill GDB's register array with the floating-point register values
- in *FSAVE. This function masks off any of the reserved
- bits in *FSAVE. */
-
-extern void i387_supply_fsave (char *fsave);
-
-/* Fill register REGNUM (if it is a floating-point register) in *FSAVE
- with the value in GDB's register array. If REGNUM is -1, do this
- for all registers. This function doesn't touch any of the reserved
- bits in *FSAVE. */
-
-extern void i387_fill_fsave (char *fsave, int regnum);
-
-/* Fill GDB's register array with the floating-point and SSE register
- values in *FXSAVE. This function masks off any of the reserved
- bits in *FXSAVE. */
-
-extern void i387_supply_fxsave (char *fxsave);
-
-/* Fill register REGNUM (if it is a floating-point or SSE register) in
- *FXSAVE with the value in GDB's register array. If REGNUM is -1, do
- this for all registers. This function doesn't touch any of the
- reserved bits in *FXSAVE. */
-
-extern void i387_fill_fxsave (char *fxsave, int regnum);
-
-#endif /* i387-nat.h */
Index: i387-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/i387-tdep.c,v
retrieving revision 1.16
diff -u -r1.16 i387-tdep.c
--- i387-tdep.c 12 Nov 2001 22:27:35 -0000 1.16
+++ i387-tdep.c 9 May 2002 23:33:59 -0000
@@ -388,3 +388,327 @@
printf_filtered ("Opcode: %s\n",
local_hex_string_custom (fop ? (fop | 0xd800) : 0, "04"));
}
+
+/* FIXME: kettenis/2000-05-21: Right now more than a few i386 targets
+ define their own routines to manage the floating-point registers in
+ GDB's register array. Most (if not all) of these targets use the
+ format used by the "fsave" instruction in their communication with
+ the OS. They should all be converted to use the routines below. */
+
+/* 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 * FPU_REG_RAW_SIZE, /* FP0_REGNUM through ... */
+ 28 + 1 * FPU_REG_RAW_SIZE,
+ 28 + 2 * FPU_REG_RAW_SIZE,
+ 28 + 3 * FPU_REG_RAW_SIZE,
+ 28 + 4 * FPU_REG_RAW_SIZE,
+ 28 + 5 * FPU_REG_RAW_SIZE,
+ 28 + 6 * FPU_REG_RAW_SIZE,
+ 28 + 7 * FPU_REG_RAW_SIZE, /* ... FP7_REGNUM. */
+ 0, /* FCTRL_REGNUM (16 bits). */
+ 4, /* FSTAT_REGNUM (16 bits). */
+ 8, /* FTAG_REGNUM (16 bits). */
+ 16, /* FISEG_REGNUM (16 bits). */
+ 12, /* FIOFF_REGNUM. */
+ 24, /* FOSEG_REGNUM. */
+ 20, /* FOOFF_REGNUM. */
+ 18 /* FOP_REGNUM (bottom 11 bits). */
+};
+
+#define FSAVE_ADDR(fsave, regnum) (fsave + fsave_offset[regnum - FP0_REGNUM])
+\f
+
+/* Fill register REGNUM in GDB's register array with the appropriate
+ value from *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+void
+i387_supply_register (int regnum, char *fsave)
+{
+ /* Most of the FPU control registers occupy only 16 bits in
+ the fsave area. Give those a special treatment. */
+ if (regnum >= FPC_REGNUM
+ && regnum != FIOFF_REGNUM && regnum != FOOFF_REGNUM)
+ {
+ unsigned int val = *(unsigned short *) (FSAVE_ADDR (fsave, regnum));
+
+ if (regnum == FOP_REGNUM)
+ {
+ val &= ((1 << 11) - 1);
+ supply_register (regnum, (char *) &val);
+ }
+ else
+ supply_register (regnum, (char *) &val);
+ }
+ else
+ supply_register (regnum, FSAVE_ADDR (fsave, regnum));
+}
+
+/* Fill GDB's register array with the floating-point register values
+ in *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+void
+i387_supply_fsave (char *fsave)
+{
+ int i;
+
+ for (i = FP0_REGNUM; i < XMM0_REGNUM; i++)
+ i387_supply_register (i, fsave);
+}
+
+/* Fill register REGNUM (if it is a floating-point register) in *FSAVE
+ with the value in GDB's register array. 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 (char *fsave, int regnum)
+{
+ int i;
+
+ for (i = FP0_REGNUM; i < 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 >= FPC_REGNUM
+ && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
+ {
+ char buf[4];
+
+ regcache_collect (i, buf);
+
+ if (i == FOP_REGNUM)
+ {
+ unsigned short oldval, newval;
+
+ /* The opcode occupies only 11 bits. */
+ oldval = (*(unsigned short *) (FSAVE_ADDR (fsave, i)));
+ newval = *(unsigned short *) buf;
+ newval &= ((1 << 11) - 1);
+ newval |= oldval & ~((1 << 11) - 1);
+ memcpy (FSAVE_ADDR (fsave, i), &newval, 2);
+ }
+ else
+ memcpy (FSAVE_ADDR (fsave, i), buf, 2);
+ }
+ else
+ regcache_collect (i, FSAVE_ADDR (fsave, i));
+ }
+}
+\f
+
+/* 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, /* FP0_REGNUM through ... */
+ 48,
+ 64,
+ 80,
+ 96,
+ 112,
+ 128,
+ 144, /* ... FP7_REGNUM (80 bits each). */
+ 0, /* FCTRL_REGNUM (16 bits). */
+ 2, /* FSTAT_REGNUM (16 bits). */
+ 4, /* FTAG_REGNUM (16 bits). */
+ 12, /* FISEG_REGNUM (16 bits). */
+ 8, /* FIOFF_REGNUM. */
+ 20, /* FOSEG_REGNUM (16 bits). */
+ 16, /* FOOFF_REGNUM. */
+ 6, /* FOP_REGNUM (bottom 11 bits). */
+ 160, /* XMM0_REGNUM through ... */
+ 176,
+ 192,
+ 208,
+ 224,
+ 240,
+ 256,
+ 272, /* ... XMM7_REGNUM (128 bits each). */
+ 24, /* MXCSR_REGNUM. */
+};
+
+#define FXSAVE_ADDR(fxsave, regnum) \
+ (fxsave + fxsave_offset[regnum - FP0_REGNUM])
+
+static int i387_tag (unsigned char *raw);
+\f
+
+/* Fill GDB's register array with the floating-point and SSE register
+ values in *FXSAVE. This function masks off any of the reserved
+ bits in *FXSAVE. */
+
+void
+i387_supply_fxsave (char *fxsave)
+{
+ int i;
+
+ for (i = FP0_REGNUM; i <= MXCSR_REGNUM; i++)
+ {
+ /* Most of the FPU control registers occupy only 16 bits in
+ the fxsave area. Give those a special treatment. */
+ if (i >= FPC_REGNUM && i < XMM0_REGNUM
+ && i != FIOFF_REGNUM && i != FOOFF_REGNUM)
+ {
+ unsigned long val = *(unsigned short *) (FXSAVE_ADDR (fxsave, i));
+
+ if (i == FOP_REGNUM)
+ {
+ val &= ((1 << 11) - 1);
+ supply_register (i, (char *) &val);
+ }
+ else if (i== 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;
+ unsigned long fstat;
+ int fpreg;
+ int top;
+
+ fstat = *(unsigned short *) (FXSAVE_ADDR (fxsave, FSTAT_REGNUM));
+ top = ((fstat >> 11) & 0x7);
+
+ for (fpreg = 7; fpreg >= 0; fpreg--)
+ {
+ int tag;
+
+ if (val & (1 << fpreg))
+ {
+ int regnum = (fpreg + 8 - top) % 8 + FP0_REGNUM;
+ tag = i387_tag (FXSAVE_ADDR (fxsave, regnum));
+ }
+ else
+ tag = 3; /* Empty */
+
+ ftag |= tag << (2 * fpreg);
+ }
+ supply_register (i, (char *) &ftag);
+ }
+ else
+ supply_register (i, (char *) &val);
+ }
+ else
+ supply_register (i, FXSAVE_ADDR (fxsave, i));
+ }
+}
+
+/* Fill register REGNUM (if it is a floating-point or SSE register) in
+ *FXSAVE with the value in GDB's register array. 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 (char *fxsave, int regnum)
+{
+ int i;
+
+ for (i = FP0_REGNUM; i <= 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 >= FPC_REGNUM && i < XMM0_REGNUM
+ && i != FIOFF_REGNUM && i != FDOFF_REGNUM)
+ {
+ char buf[4];
+
+ regcache_collect (i, buf);
+
+ if (i == FOP_REGNUM)
+ {
+ unsigned short oldval, newval;
+
+ /* The opcode occupies only 11 bits. */
+ oldval = (*(unsigned short *) (FXSAVE_ADDR (fxsave, i)));
+ newval = *(unsigned short *) buf;
+ newval &= ((1 << 11) - 1);
+ newval |= oldval & ~((1 << 11) - 1);
+ memcpy (FXSAVE_ADDR (fxsave, i), &newval, 2);
+ }
+ else if (i == FTAG_REGNUM)
+ {
+ /* Converting back is much easier. */
+
+ unsigned short val = 0;
+ unsigned short ftag;
+ int fpreg;
+
+ ftag = *(unsigned short *) buf;
+
+ for (fpreg = 7; fpreg >= 0; fpreg--)
+ {
+ int tag = (ftag >> (fpreg * 2)) & 3;
+
+ if (tag != 3)
+ val |= (1 << fpreg);
+ }
+
+ memcpy (FXSAVE_ADDR (fxsave, i), &val, 2);
+ }
+ else
+ memcpy (FXSAVE_ADDR (fxsave, i), buf, 2);
+ }
+ else
+ regcache_collect (i, FXSAVE_ADDR (fxsave, i));
+ }
+}
+
+/* Recreate the FTW (tag word) valid bits from the 80-bit FP data in
+ *RAW. */
+
+static int
+i387_tag (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);
+ }
+ }
+}
Index: i387-tdep.h
===================================================================
RCS file: i387-tdep.h
diff -N i387-tdep.h
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ i387-tdep.h 9 May 2002 23:33:59 -0000
@@ -0,0 +1,56 @@
+/* Target-dependent code for the i387.
+ Copyright 2000, 2001 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. */
+
+#ifndef I387_TDEP_H
+#define I387_TDEP_H
+
+/* Fill register REGNUM in GDB's register array with the appropriate
+ value from *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+extern void i387_supply_register (int regnum, char *fsave);
+
+/* Fill GDB's register array with the floating-point register values
+ in *FSAVE. This function masks off any of the reserved
+ bits in *FSAVE. */
+
+extern void i387_supply_fsave (char *fsave);
+
+/* Fill register REGNUM (if it is a floating-point register) in *FSAVE
+ with the value in GDB's register array. If REGNUM is -1, do this
+ for all registers. This function doesn't touch any of the reserved
+ bits in *FSAVE. */
+
+extern void i387_fill_fsave (char *fsave, int regnum);
+
+/* Fill GDB's register array with the floating-point and SSE register
+ values in *FXSAVE. This function masks off any of the reserved
+ bits in *FXSAVE. */
+
+extern void i387_supply_fxsave (char *fxsave);
+
+/* Fill register REGNUM (if it is a floating-point or SSE register) in
+ *FXSAVE with the value in GDB's register array. If REGNUM is -1, do
+ this for all registers. This function doesn't touch any of the
+ reserved bits in *FXSAVE. */
+
+extern void i387_fill_fxsave (char *fxsave, int regnum);
+
+#endif /* i387-tdep.h */
Index: x86-64-linux-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/x86-64-linux-nat.c,v
retrieving revision 1.11
diff -u -r1.11 x86-64-linux-nat.c
--- x86-64-linux-nat.c 7 May 2002 11:22:54 -0000 1.11
+++ x86-64-linux-nat.c 9 May 2002 23:34:00 -0000
@@ -25,7 +25,7 @@
#include "inferior.h"
#include "gdbcore.h"
#include "regcache.h"
-#include "i387-nat.h"
+#include "i387-tdep.h"
#include "gdb_assert.h"
#include "x86-64-tdep.h"
Index: config/i386/fbsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/fbsd.mh,v
retrieving revision 1.11
diff -u -r1.11 fbsd.mh
--- config/i386/fbsd.mh 31 Mar 2002 17:47:17 -0000 1.11
+++ config/i386/fbsd.mh 9 May 2002 23:34:00 -0000
@@ -4,4 +4,4 @@
NAT_FILE= nm-fbsd.h
# NOTE: Do not spread NATDEPFILES over several lines - it hurts BSD make.
-NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-svr4.o solib-legacy.o corelow.o core-aout.o core-regset.o i386-nat.o i387-nat.o i386bsd-nat.o i386fbsd-nat.o gcore.o fbsd-proc.o
+NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-svr4.o solib-legacy.o corelow.o core-aout.o core-regset.o i386-nat.o i386bsd-nat.o i386fbsd-nat.o gcore.o fbsd-proc.o
Index: config/i386/go32.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/go32.mh,v
retrieving revision 1.5
diff -u -r1.5 go32.mh
--- config/i386/go32.mh 18 Jan 2002 04:50:59 -0000 1.5
+++ config/i386/go32.mh 9 May 2002 23:34:01 -0000
@@ -7,7 +7,7 @@
XM_FILE= xm-go32.h
NAT_FILE= nm-go32.h
-NATDEPFILES= go32-nat.o i386-nat.o i387-nat.o
+NATDEPFILES= go32-nat.o i386-nat.o
TERMCAP=
HOST_IPC=
Index: config/i386/i386gnu.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/i386gnu.mh,v
retrieving revision 1.7
diff -u -r1.7 i386gnu.mh
--- config/i386/i386gnu.mh 24 Apr 2002 21:44:47 -0000 1.7
+++ config/i386/i386gnu.mh 9 May 2002 23:34:01 -0000
@@ -1,5 +1,5 @@
# Host: Intel 386 running the GNU Hurd
-NATDEPFILES= i386gnu-nat.o i387-nat.o gnu-nat.o corelow.o core-regset.o \
+NATDEPFILES= i386gnu-nat.o gnu-nat.o corelow.o core-regset.o \
fork-child.o solib.o solib-svr4.o solib-legacy.o \
notify_S.o process_reply_S.o msg_reply_S.o \
msg_U.o exc_request_U.o exc_request_S.o
Index: config/i386/i386sol2.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/i386sol2.mh,v
retrieving revision 1.4
diff -u -r1.4 i386sol2.mh
--- config/i386/i386sol2.mh 14 Feb 2002 08:13:33 -0000 1.4
+++ config/i386/i386sol2.mh 9 May 2002 23:34:01 -0000
@@ -4,5 +4,5 @@
XM_CLIBS= -lsocket -lnsl
NAT_FILE= nm-i386sol2.h
-NATDEPFILES= core-regset.o fork-child.o i386v4-nat.o i387-nat.o corelow.o \
+NATDEPFILES= core-regset.o fork-child.o i386v4-nat.o corelow.o \
procfs.o proc-api.o proc-events.o proc-flags.o proc-why.o gcore.o
Index: config/i386/i386v42mp.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/i386v42mp.mh,v
retrieving revision 1.7
diff -u -r1.7 i386v42mp.mh
--- config/i386/i386v42mp.mh 11 Apr 2002 18:13:21 -0000 1.7
+++ config/i386/i386v42mp.mh 9 May 2002 23:34:01 -0000
@@ -14,7 +14,7 @@
# continuation character (backslash) to extend a commented line. As a
# consequence, make considers subsequent tab-indented lines to be
# some sort of error.
-NATDEPFILES= corelow.o core-regset.o fork-child.o i386v4-nat.o i387-nat.o \
+NATDEPFILES= corelow.o core-regset.o fork-child.o i386v4-nat.o \
gcore.o solib.o solib-svr4.o solib-legacy.o procfs.o proc-api.o \
proc-events.o proc-flags.o proc-why.o uw-thread.o
Index: config/i386/linux.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/linux.mh,v
retrieving revision 1.11
diff -u -r1.11 linux.mh
--- config/i386/linux.mh 18 Jan 2002 04:50:59 -0000 1.11
+++ config/i386/linux.mh 9 May 2002 23:34:01 -0000
@@ -4,7 +4,7 @@
NAT_FILE= nm-linux.h
NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o linux-proc.o \
- core-aout.o i386-nat.o i386-linux-nat.o i387-nat.o \
+ core-aout.o i386-nat.o i386-linux-nat.o \
proc-service.o thread-db.o lin-lwp.o linux-proc.o gcore.o
# The dynamically loaded libthread_db needs access to symbols in the
Index: config/i386/nbsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/nbsd.mh,v
retrieving revision 1.12
diff -u -r1.12 nbsd.mh
--- config/i386/nbsd.mh 18 Jan 2002 04:50:59 -0000 1.12
+++ config/i386/nbsd.mh 9 May 2002 23:34:01 -0000
@@ -6,7 +6,7 @@
# continuation character (backslash) to extend a commented line. As a
# consequence, BSD make considers subsequent tab-indented lines to be
# "unassociated shell commands".
-NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i387-nat.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-sunos.o
+NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-sunos.o
XM_FILE= xm-nbsd.h
NAT_FILE= nm-nbsd.h
Index: config/i386/nbsdelf.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/nbsdelf.mh,v
retrieving revision 1.8
diff -u -r1.8 nbsdelf.mh
--- config/i386/nbsdelf.mh 18 Jan 2002 04:50:59 -0000 1.8
+++ config/i386/nbsdelf.mh 9 May 2002 23:34:01 -0000
@@ -1,4 +1,4 @@
# Host: Intel 386 running NetBSD
-NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i387-nat.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-svr4.o solib-legacy.o
+NATDEPFILES= fork-child.o infptrace.o inftarg.o corelow.o i386bsd-nat.o i386nbsd-nat.o solib.o solib-svr4.o solib-legacy.o
XM_FILE= xm-nbsd.h
NAT_FILE= nm-nbsdelf.h
Index: config/i386/obsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/obsd.mh,v
retrieving revision 1.2
diff -u -r1.2 obsd.mh
--- config/i386/obsd.mh 18 Jan 2002 04:50:59 -0000 1.2
+++ config/i386/obsd.mh 9 May 2002 23:34:01 -0000
@@ -4,4 +4,4 @@
NAT_FILE= nm-obsd.h
# NOTE: Do not spread NATDEPFILES over several lines - it hurts BSD make.
-NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-sunos.o corelow.o core-aout.o i386-nat.o i387-nat.o i386bsd-nat.o
+NATDEPFILES= fork-child.o infptrace.o inftarg.o solib.o solib-sunos.o corelow.o core-aout.o i386-nat.o i386bsd-nat.o
Index: config/i386/x86-64linux.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/x86-64linux.mh,v
retrieving revision 1.5
diff -u -r1.5 x86-64linux.mh
--- config/i386/x86-64linux.mh 21 Feb 2002 12:48:36 -0000 1.5
+++ config/i386/x86-64linux.mh 9 May 2002 23:34:01 -0000
@@ -5,7 +5,7 @@
NAT_FILE= nm-x86-64.h
NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o \
core-aout.o i386-nat.o x86-64-linux-nat.o \
- i387-nat.o proc-service.o thread-db.o lin-lwp.o \
+ proc-service.o thread-db.o lin-lwp.o \
linux-proc.o gcore.o
LOADLIBES = -ldl -rdynamic
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
2002-05-09 16:45 Jason R Thorpe
@ 2002-05-09 17:12 ` Daniel Jacobowitz
2002-05-09 17:23 ` Jason R Thorpe
2002-05-10 0:26 ` Eli Zaretskii
1 sibling, 1 reply; 8+ messages in thread
From: Daniel Jacobowitz @ 2002-05-09 17:12 UTC (permalink / raw)
To: gdb-patches; +Cc: Jason R Thorpe
On Thu, May 09, 2002 at 04:45:46PM -0700, Jason R Thorpe wrote:
> This is a re-submit of an earlier patch, which takes into account
> Mark's recent changes to i387-nat.c.
>
> Basically, this moves all of the register supply/fill stuff for i387
> from i387-nat.c to i387-tdep.c, to facilitate cross-debugging of core
> files. This renders i387-nat.c empty, and so it is removed in this
> patch.
Thanks! This is the biggest chunk of i386 cross core debugging. The
other nasty one is ARM's evil registers...
[disclaimer: not approving anything. I support this patch, I've been
using something almost identical for some time.]
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
2002-05-09 17:12 ` Daniel Jacobowitz
@ 2002-05-09 17:23 ` Jason R Thorpe
2002-05-09 18:28 ` Daniel Jacobowitz
2002-05-10 3:54 ` Mark Kettenis
0 siblings, 2 replies; 8+ messages in thread
From: Jason R Thorpe @ 2002-05-09 17:23 UTC (permalink / raw)
To: gdb-patches
On Thu, May 09, 2002 at 08:12:05PM -0400, Daniel Jacobowitz wrote:
> Thanks! This is the biggest chunk of i386 cross core debugging. The
> other nasty one is ARM's evil registers...
I have changes for some of the ARM stuff, too, but I need to clean up
some stuff and then discuss it with Richard.
Note, that I think there might be byte-order snafus in the i387
supply/fill goo, but it'd be nice to move it first, and then fix
the bugs in it :-)
--
-- Jason R. Thorpe <thorpej@wasabisystems.com>
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
2002-05-09 17:23 ` Jason R Thorpe
@ 2002-05-09 18:28 ` Daniel Jacobowitz
2002-05-10 3:54 ` Mark Kettenis
1 sibling, 0 replies; 8+ messages in thread
From: Daniel Jacobowitz @ 2002-05-09 18:28 UTC (permalink / raw)
To: gdb-patches; +Cc: Jason R Thorpe
On Thu, May 09, 2002 at 05:23:24PM -0700, Jason R Thorpe wrote:
> On Thu, May 09, 2002 at 08:12:05PM -0400, Daniel Jacobowitz wrote:
>
> > Thanks! This is the biggest chunk of i386 cross core debugging. The
> > other nasty one is ARM's evil registers...
>
> I have changes for some of the ARM stuff, too, but I need to clean up
> some stuff and then discuss it with Richard.
>
> Note, that I think there might be byte-order snafus in the i387
> supply/fill goo, but it'd be nice to move it first, and then fix
> the bugs in it :-)
Maybe, but I've used it from Solaris->x86 crosses...
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
2002-05-09 17:23 ` Jason R Thorpe
2002-05-09 18:28 ` Daniel Jacobowitz
@ 2002-05-10 3:54 ` Mark Kettenis
2002-05-10 8:10 ` Jason R Thorpe
1 sibling, 1 reply; 8+ messages in thread
From: Mark Kettenis @ 2002-05-10 3:54 UTC (permalink / raw)
To: thorpej; +Cc: gdb-patches
Jason R Thorpe <thorpej@wasabisystems.com> writes:
> Note, that I think there might be byte-order snafus in the i387
> supply/fill goo, but it'd be nice to move it first, and then fix
> the bugs in it :-)
Not only byte-order snafus, but also assumptions about the size of
certain types (basically that a short is two bytes).
Thos problems should be solved before moving the code into
i387-tdep.c. Sorry.
Mark
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
2002-05-10 3:54 ` Mark Kettenis
@ 2002-05-10 8:10 ` Jason R Thorpe
0 siblings, 0 replies; 8+ messages in thread
From: Jason R Thorpe @ 2002-05-10 8:10 UTC (permalink / raw)
To: Mark Kettenis; +Cc: gdb-patches
On Fri, May 10, 2002 at 12:54:39PM +0200, Mark Kettenis wrote:
> Not only byte-order snafus, but also assumptions about the size of
> certain types (basically that a short is two bytes).
>
> Thos problems should be solved before moving the code into
> i387-tdep.c. Sorry.
No problem, I'll submit another patch shortly.
--
-- Jason R. Thorpe <thorpej@wasabisystems.com>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c
2002-05-09 16:45 Jason R Thorpe
2002-05-09 17:12 ` Daniel Jacobowitz
@ 2002-05-10 0:26 ` Eli Zaretskii
1 sibling, 0 replies; 8+ messages in thread
From: Eli Zaretskii @ 2002-05-10 0:26 UTC (permalink / raw)
To: thorpej; +Cc: gdb-patches
> Date: Thu, 9 May 2002 16:45:46 -0700
> From: Jason R Thorpe <thorpej@wasabisystems.com>
>
> This is a re-submit of an earlier patch, which takes into account
> Mark's recent changes to i387-nat.c.
>
> Basically, this moves all of the register supply/fill stuff for i387
> from i387-nat.c to i387-tdep.c, to facilitate cross-debugging of core
> files. This renders i387-nat.c empty, and so it is removed in this
> patch.
>
> OK to commit?
Okay with me, as far as go32 stuff is concerned.
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2002-05-10 15:10 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-04-11 16:05 [PATCH/RFA] Move i387 register supply/fill to i387-tdep.c Jason R Thorpe
2002-05-09 16:45 Jason R Thorpe
2002-05-09 17:12 ` Daniel Jacobowitz
2002-05-09 17:23 ` Jason R Thorpe
2002-05-09 18:28 ` Daniel Jacobowitz
2002-05-10 3:54 ` Mark Kettenis
2002-05-10 8:10 ` Jason R Thorpe
2002-05-10 0:26 ` Eli Zaretskii
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox