Skip to content

R
riscv-gcc-1
Commit 83774849 by Richard Henderson Committed by Richard Henderson
Options

i386.c (dbx_register_map, [...]): New. 

        * i386.c (dbx_register_map, svr4_dbx_register_map): New.
        * i386.h (DBX_REGISTER_NUMBER): Use them.
        * i386/beos-elf.h, i386/freebsd-elf.h, i386/i386elf.h: Likewise.
        * i386/linux.h, i386/osfrose.h, i386/ptx4-i.h: Likewise.
        * i386/rtemself.h, i386/sco5.h, i386/sysv4.h: Likewise.
        * i386/sequent.h: Kill incorrect comment.

From-SVN: r31575
parent d416576b
Showing with 102 additions and 558 deletions
gcc/ChangeLog
2000-01-23 Richard Henderson <rth@cygnus.com>
* i386.c (dbx_register_map, svr4_dbx_register_map): New.
* i386.h (DBX_REGISTER_NUMBER): Use them.
* i386/beos-elf.h, i386/freebsd-elf.h, i386/i386elf.h: Likewise.
* i386/linux.h, i386/osfrose.h, i386/ptx4-i.h: Likewise.
* i386/rtemself.h, i386/sco5.h, i386/sysv4.h: Likewise.
* i386/sequent.h: Kill incorrect comment.
2000-01-23 Mark Mitchell <mark@codesourcery.com> 2000-01-23 Mark Mitchell <mark@codesourcery.com>
* ggc-page.c (struct page_entry): Make `context_depth' an * ggc-page.c (struct page_entry): Make `context_depth' an
......
gcc/config/i386/beos-elf.h
...@@ -48,73 +48,8 @@ Boston, MA 02111-1307, USA. */ ...@@ -48,73 +48,8 @@ Boston, MA 02111-1307, USA. */
necessary when compiling PIC code. */ necessary when compiling PIC code. */
#define JUMP_TABLES_IN_TEXT_SECTION (flag_pic) #define JUMP_TABLES_IN_TEXT_SECTION (flag_pic)
/* Copy this from the svr4 specifications... */
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* Output assembler code to FILE to increment profiler label # LABELNO /* Output assembler code to FILE to increment profiler label # LABELNO
for profiling a function entry. */ for profiling a function entry. */
......
gcc/config/i386/freebsd-elf.h
...@@ -49,73 +49,8 @@ Boston, MA 02111-1307, USA. */ ...@@ -49,73 +49,8 @@ Boston, MA 02111-1307, USA. */
necessary when compiling PIC code. */ necessary when compiling PIC code. */
#define JUMP_TABLES_IN_TEXT_SECTION (flag_pic) #define JUMP_TABLES_IN_TEXT_SECTION (flag_pic)
/* Copy this from the svr4 specifications... */
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* Tell final.c that we don't need a label passed to mcount. */ /* Tell final.c that we don't need a label passed to mcount. */
......
gcc/config/i386/i386.c
...@@ -233,11 +233,83 @@ enum reg_class const regclass_map[FIRST_PSEUDO_REGISTER] = ...@@ -233,11 +233,83 @@ enum reg_class const regclass_map[FIRST_PSEUDO_REGISTER] =
FP_TOP_REG, FP_SECOND_REG, FLOAT_REGS, FLOAT_REGS, FP_TOP_REG, FP_SECOND_REG, FLOAT_REGS, FLOAT_REGS,
FLOAT_REGS, FLOAT_REGS, FLOAT_REGS, FLOAT_REGS, FLOAT_REGS, FLOAT_REGS, FLOAT_REGS, FLOAT_REGS,
/* arg pointer */ /* arg pointer */
INDEX_REGS, NON_Q_REGS,
/* flags, fpsr */ /* flags, fpsr, dirflag */
NO_REGS, NO_REGS NO_REGS, NO_REGS, NO_REGS
}; };
/* The "default" register map. */
int const dbx_register_map[FIRST_PSEUDO_REGISTER] =
{
0, 2, 1, 3, 6, 7, 4, 5, /* general regs */
12, 13, 14, 15, 16, 17, 18, 19, /* fp regs */
-1, -1, -1, -1, /* arg, flags, fpsr, dir */
};
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gcc regno = 0)
1 for %ecx (gcc regno = 2)
2 for %edx (gcc regno = 1)
3 for %ebx (gcc regno = 3)
4 for %esp (gcc regno = 7)
5 for %ebp (gcc regno = 6)
6 for %esi (gcc regno = 4)
7 for %edi (gcc regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gcc equivalent)
9 for %eflags (gcc regno = 17)
10 for %trapno (no gcc equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gcc regno = 8)
12 for %st(1) (gcc regno = 9)
13 for %st(2) (gcc regno = 10)
14 for %st(3) (gcc regno = 11)
15 for %st(4) (gcc regno = 12)
16 for %st(5) (gcc regno = 13)
17 for %st(6) (gcc regno = 14)
18 for %st(7) (gcc regno = 15)
*/
int const svr4_dbx_register_map[FIRST_PSEUDO_REGISTER] =
{
0, 2, 1, 3, 6, 7, 5, 4, /* general regs */
11, 12, 13, 14, 15, 16, 17, 18, /* fp regs */
-1, 9, -1, -1, /* arg, flags, fpsr, dir */
};
/* Test and compare insns in i386.md store the information needed to /* Test and compare insns in i386.md store the information needed to
generate branch and scc insns here. */ generate branch and scc insns here. */
......
gcc/config/i386/i386.h
...@@ -2185,17 +2185,10 @@ number as al, and ax. ...@@ -2185,17 +2185,10 @@ number as al, and ax.
/* How to renumber registers for dbx and gdb. */ /* How to renumber registers for dbx and gdb. */
/* {0,2,1,3,6,7,4,5,12,13,14,15,16,17} */ #define DBX_REGISTER_NUMBER(n) dbx_register_map[n]
#define DBX_REGISTER_NUMBER(n) \
((n) == 0 ? 0 : \ extern int const dbx_register_map[FIRST_PSEUDO_REGISTER];
(n) == 1 ? 2 : \ extern int const svr4_dbx_register_map[FIRST_PSEUDO_REGISTER];
(n) == 2 ? 1 : \
(n) == 3 ? 3 : \
(n) == 4 ? 6 : \
(n) == 5 ? 7 : \
(n) == 6 ? 4 : \
(n) == 7 ? 5 : \
(n) + 4)
/* Before the prologue, RA is at 0(%esp). */ /* Before the prologue, RA is at 0(%esp). */
#define INCOMING_RETURN_ADDR_RTX \ #define INCOMING_RETURN_ADDR_RTX \
......
gcc/config/i386/i386elf.h
...@@ -115,82 +115,8 @@ do { long value[3]; \ ...@@ -115,82 +115,8 @@ do { long value[3]; \
fprintf (FILE, "\t.version\t\"01.01\"\n"); \ fprintf (FILE, "\t.version\t\"01.01\"\n"); \
} while (0) } while (0)
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The verison of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* The routine used to output sequences of byte values. We use a special /* The routine used to output sequences of byte values. We use a special
version of this for most svr4 targets because doing so makes the version of this for most svr4 targets because doing so makes the
......
gcc/config/i386/linux.h
...@@ -51,73 +51,8 @@ Boston, MA 02111-1307, USA. */ ...@@ -51,73 +51,8 @@ Boston, MA 02111-1307, USA. */
necessary when compiling PIC code. */ necessary when compiling PIC code. */
#define JUMP_TABLES_IN_TEXT_SECTION (flag_pic) #define JUMP_TABLES_IN_TEXT_SECTION (flag_pic)
/* Copy this from the svr4 specifications... */
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* Output assembler code to FILE to increment profiler label # LABELNO /* Output assembler code to FILE to increment profiler label # LABELNO
for profiling a function entry. */ for profiling a function entry. */
......
gcc/config/i386/osfrose.h
...@@ -877,34 +877,10 @@ while (0) ...@@ -877,34 +877,10 @@ while (0)
we want to retain compatibility with older gcc versions. */ we want to retain compatibility with older gcc versions. */
#define DEFAULT_PCC_STRUCT_RETURN 0 #define DEFAULT_PCC_STRUCT_RETURN 0
/* Map i386 registers to the numbers dwarf expects. Of course this is different /* Map i386 registers to the numbers dwarf expects. Of course this is
from what stabs expects. */ different from what stabs expects. */
#define DWARF_DBX_REGISTER_NUMBER(n) \
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* Now what stabs expects in the register. */
#define STABS_DBX_REGISTER_NUMBER(n) \
((n) == 0 ? 0 : \
(n) == 1 ? 2 : \
(n) == 2 ? 1 : \
(n) == 3 ? 3 : \
(n) == 4 ? 6 : \
(n) == 5 ? 7 : \
(n) == 6 ? 4 : \
(n) == 7 ? 5 : \
(n) + 4)
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) ((write_symbols == DWARF_DEBUG) \ #define DBX_REGISTER_NUMBER(n) ((write_symbols == DWARF_DEBUG) \
? DWARF_DBX_REGISTER_NUMBER(n) \ ? svr4_dbx_register_map[n] \
: STABS_DBX_REGISTER_NUMBER(n)) : dbx_register_map[n])
gcc/config/i386/ptx4-i.h
...@@ -107,82 +107,8 @@ do { long value[3]; \ ...@@ -107,82 +107,8 @@ do { long value[3]; \
fprintf (FILE, "\t.version\t\"01.01\"\n"); \ fprintf (FILE, "\t.version\t\"01.01\"\n"); \
} while (0) } while (0)
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* The routine used to output sequences of byte values. We use a special /* The routine used to output sequences of byte values. We use a special
version of this for most svr4 targets because doing so makes the version of this for most svr4 targets because doing so makes the
......
gcc/config/i386/rtemself.h
...@@ -48,73 +48,8 @@ Boston, MA 02111-1307, USA. */ ...@@ -48,73 +48,8 @@ Boston, MA 02111-1307, USA. */
necessary when compiling PIC code. */ necessary when compiling PIC code. */
#define JUMP_TABLES_IN_TEXT_SECTION (flag_pic) #define JUMP_TABLES_IN_TEXT_SECTION (flag_pic)
/* Copy this from the svr4 specifications... */
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* Output assembler code to FILE to increment profiler label # LABELNO /* Output assembler code to FILE to increment profiler label # LABELNO
for profiling a function entry. */ for profiling a function entry. */
......
gcc/config/i386/sco5.h
...@@ -541,28 +541,8 @@ do { \ ...@@ -541,28 +541,8 @@ do { \
#define DBX_FUNCTION_FIRST 1 #define DBX_FUNCTION_FIRST 1
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) \
((TARGET_ELF) ? \ ((TARGET_ELF) ? svr4_dbx_register_map[n] : dbx_register_map[n])
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1)) \
: \
((n) == 0 ? 0 : \
(n) == 1 ? 2 : \
(n) == 2 ? 1 : \
(n) == 3 ? 3 : \
(n) == 4 ? 6 : \
(n) == 5 ? 7 : \
(n) == 6 ? 4 : \
(n) == 7 ? 5 : \
(n) + 4))
#undef DWARF_DEBUGGING_INFO #undef DWARF_DEBUGGING_INFO
#undef SDB_DEBUGGING_INFO #undef SDB_DEBUGGING_INFO
......
gcc/config/i386/sequent.h
...@@ -64,10 +64,6 @@ Boston, MA 02111-1307, USA. */ ...@@ -64,10 +64,6 @@ Boston, MA 02111-1307, USA. */
* dbx order is ax, dx, cx, st(0), st(1), bx, si, di, st(2), st(3), * dbx order is ax, dx, cx, st(0), st(1), bx, si, di, st(2), st(3),
* st(4), st(5), st(6), st(7), sp, bp */ * st(4), st(5), st(6), st(7), sp, bp */
/* ??? The right thing would be to change the ordering of the
registers to correspond to the conventions of this system,
and get rid of DBX_REGISTER_NUMBER. */
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) \
((n) < 3 ? (n) : (n) < 6 ? (n) + 2 \ ((n) < 3 ? (n) : (n) < 6 ? (n) + 2 \
......
gcc/config/i386/sysv4.h
...@@ -105,82 +105,8 @@ do { long value[3]; \ ...@@ -105,82 +105,8 @@ do { long value[3]; \
fprintf (FILE, "\t.version\t\"01.01\"\n"); \ fprintf (FILE, "\t.version\t\"01.01\"\n"); \
} while (0) } while (0)
/* Define the register numbers to be used in Dwarf debugging information.
The SVR4 reference port C compiler uses the following register numbers
in its Dwarf output code:
0 for %eax (gnu regno = 0)
1 for %ecx (gnu regno = 2)
2 for %edx (gnu regno = 1)
3 for %ebx (gnu regno = 3)
4 for %esp (gnu regno = 7)
5 for %ebp (gnu regno = 6)
6 for %esi (gnu regno = 4)
7 for %edi (gnu regno = 5)
The following three DWARF register numbers are never generated by
the SVR4 C compiler or by the GNU compilers, but SDB on x86/svr4
believes these numbers have these meanings.
8 for %eip (no gnu equivalent)
9 for %eflags (no gnu equivalent)
10 for %trapno (no gnu equivalent)
It is not at all clear how we should number the FP stack registers
for the x86 architecture. If the version of SDB on x86/svr4 were
a bit less brain dead with respect to floating-point then we would
have a precedent to follow with respect to DWARF register numbers
for x86 FP registers, but the SDB on x86/svr4 is so completely
broken with respect to FP registers that it is hardly worth thinking
of it as something to strive for compatibility with.
The version of x86/svr4 SDB I have at the moment does (partially)
seem to believe that DWARF register number 11 is associated with
the x86 register %st(0), but that's about all. Higher DWARF
register numbers don't seem to be associated with anything in
particular, and even for DWARF regno 11, SDB only seems to under-
stand that it should say that a variable lives in %st(0) (when
asked via an `=' command) if we said it was in DWARF regno 11,
but SDB still prints garbage when asked for the value of the
variable in question (via a `/' command).
(Also note that the labels SDB prints for various FP stack regs
when doing an `x' command are all wrong.)
Note that these problems generally don't affect the native SVR4
C compiler because it doesn't allow the use of -O with -g and
because when it is *not* optimizing, it allocates a memory
location for each floating-point variable, and the memory
location is what gets described in the DWARF AT_location
attribute for the variable in question.
Regardless of the severe mental illness of the x86/svr4 SDB, we
do something sensible here and we use the following DWARF
register numbers. Note that these are all stack-top-relative
numbers.
11 for %st(0) (gnu regno = 8)
12 for %st(1) (gnu regno = 9)
13 for %st(2) (gnu regno = 10)
14 for %st(3) (gnu regno = 11)
15 for %st(4) (gnu regno = 12)
16 for %st(5) (gnu regno = 13)
17 for %st(6) (gnu regno = 14)
18 for %st(7) (gnu regno = 15)
*/
#undef DBX_REGISTER_NUMBER #undef DBX_REGISTER_NUMBER
#define DBX_REGISTER_NUMBER(n) \ #define DBX_REGISTER_NUMBER(n) svr4_dbx_register_map[n]
((n) == 0 ? 0 \
: (n) == 1 ? 2 \
: (n) == 2 ? 1 \
: (n) == 3 ? 3 \
: (n) == 4 ? 6 \
: (n) == 5 ? 7 \
: (n) == 6 ? 5 \
: (n) == 7 ? 4 \
: ((n) >= FIRST_STACK_REG && (n) <= LAST_STACK_REG) ? (n)+3 \
: (-1))
/* The routine used to output sequences of byte values. We use a special /* The routine used to output sequences of byte values. We use a special
version of this for most svr4 targets because doing so makes the version of this for most svr4 targets because doing so makes the
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment