* [patch 3/4] decimal floating point support for expressions
[not found] <20070323030737.475073862@br.ibm.com>
@ 2007-03-27 16:30 ` Thiago Jung Bauermann
2007-03-27 16:30 ` [patch 1/4] libdecnumber support Thiago Jung Bauermann
2007-03-27 16:30 ` [patch 4/4] decimal floating point testcases Thiago Jung Bauermann
2 siblings, 0 replies; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-03-27 16:30 UTC (permalink / raw)
To: gdb-patches
plain text document attachment (expressions.diff)
2007-03-23 Thiago Jung Bauermann <bauerman@br.ibm.com>
* c-exp.y (YYSTYPE): Add typed_val_decfloat for decimal
floating point in YYSTYPE union.
Add token DECFLOAT for typed_val_decfloat.
Add expression element handling code for DECFLOAT.
(parse_number): Parse DFP constants, which end with suffix 'df',
'dd' or 'dl'. Return DECFLOAT.
* eval.c (evaluate_subexp_standard): Call value_from_decfloat to
handle OP_DECFLOAT.
* expression.h (enum exp_opcode): Add an opcode (OP_DECFLOAT)
for DFP constants.
(union exp_element): Add decfloatconst to represent DFP
elements, which is 16 bytes by default.
* parse.c (write_exp_elt_decfloatcst): New function to write a
decimal float const into the expression.
(operator_length_standard): Set operator length for OP_DECFLOAT
to 4.
* parser-defs.h (write_exp_elt_decfloatcst): Prototype.
* valarith.c (value_neg): Add some code to handle the negation
operation of DFP values.
* value.c (value_from_decfloat): New function to get the value
from a decimal floating point.
* value.h (value_from_decfloat): Prototype.
Index: gdb/c-exp.y
===================================================================
--- gdb/c-exp.y.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/c-exp.y 2007-03-20 14:07:37.000000000 -0300
@@ -52,6 +52,7 @@ Boston, MA 02110-1301, USA. */
#include "charset.h"
#include "block.h"
#include "cp-support.h"
+#include "dfp.h"
/* Remap normal yacc parser interface names (yyparse, yylex, yyerror, etc),
as well as gratuitiously global symbol names, so we can have multiple
@@ -130,6 +131,10 @@ void yyerror (char *);
DOUBLEST dval;
struct type *type;
} typed_val_float;
+ struct {
+ gdb_byte val[16];
+ struct type *type;
+ } typed_val_decfloat;
struct symbol *sym;
struct type *tval;
struct stoken sval;
@@ -162,6 +167,7 @@ static int parse_number (char *, int, in
%token <typed_val_int> INT
%token <typed_val_float> FLOAT
+%token <typed_val_decfloat> DECFLOAT
/* Both NAME and TYPENAME tokens represent symbols in the input,
and both convey their data as strings.
@@ -496,6 +502,13 @@ exp : FLOAT
write_exp_elt_opcode (OP_DOUBLE); }
;
+exp : DECFLOAT
+ { write_exp_elt_opcode (OP_DECFLOAT);
+ write_exp_elt_type ($1.type);
+ write_exp_elt_decfloatcst ($1.val);
+ write_exp_elt_opcode (OP_DECFLOAT); }
+ ;
+
exp : variable
;
@@ -1077,6 +1090,40 @@ parse_number (p, len, parsed_float, puti
char saved_char = p[len];
p[len] = 0; /* null-terminate the token */
+
+ /* If it ends at "df", "dd" or "dl", take it as type of decimal floating
+ point. Return DECFLOAT. */
+
+ if (p[len - 2] == 'd' && p[len - 1] == 'f')
+ {
+ p[len - 2] = '\0';
+ putithere->typed_val_decfloat.type =
+ builtin_type (current_gdbarch)->builtin_decfloat;
+ decimal_from_string (putithere->typed_val_decfloat.val, 4, p);
+ p[len] = saved_char;
+ return (DECFLOAT);
+ }
+
+ if (p[len - 2] == 'd' && p[len - 1] == 'd')
+ {
+ p[len - 2] = '\0';
+ putithere->typed_val_decfloat.type =
+ builtin_type (current_gdbarch)->builtin_decdouble;
+ decimal_from_string (putithere->typed_val_decfloat.val, 8, p);
+ p[len] = saved_char;
+ return (DECFLOAT);
+ }
+
+ if (p[len - 2] == 'd' && p[len - 1] == 'l')
+ {
+ p[len - 2] = '\0';
+ putithere->typed_val_decfloat.type =
+ builtin_type (current_gdbarch)->builtin_declong;
+ decimal_from_string (putithere->typed_val_decfloat.val, 16, p);
+ p[len] = saved_char;
+ return (DECFLOAT);
+ }
+
num = sscanf (p, DOUBLEST_SCAN_FORMAT "%s",
&putithere->typed_val_float.dval, s);
p[len] = saved_char; /* restore the input stream */
Index: gdb/eval.c
===================================================================
--- gdb/eval.c.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/eval.c 2007-03-20 14:07:37.000000000 -0300
@@ -458,6 +458,11 @@ evaluate_subexp_standard (struct type *e
return value_from_double (exp->elts[pc + 1].type,
exp->elts[pc + 2].doubleconst);
+ case OP_DECFLOAT:
+ (*pos) += 3;
+ return value_from_decfloat (expect_type, exp->elts[pc + 1].type,
+ exp->elts[pc + 2].decfloatconst);
+
case OP_VAR_VALUE:
(*pos) += 3;
if (noside == EVAL_SKIP)
Index: gdb/expression.h
===================================================================
--- gdb/expression.h.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/expression.h 2007-03-20 14:07:37.000000000 -0300
@@ -334,6 +334,11 @@ enum exp_opcode
/* A F90 array range operator (for "exp:exp", "exp:", ":exp" and ":"). */
OP_F90_RANGE,
+ /* OP_DECFLOAT is followed by a type pointer in the next exp_element
+ and a dec long constant value in the following exp_element.
+ Then comes another OP_DECFLOAT. */
+ OP_DECFLOAT,
+
/* First extension operator. Individual language modules define
extra operators they need as constants with values
OP_LANGUAGE_SPECIFIC0 + k, for k >= 0, using a separate
@@ -361,6 +366,7 @@ union exp_element
struct symbol *symbol;
LONGEST longconst;
DOUBLEST doubleconst;
+ gdb_byte decfloatconst[16];
/* Really sizeof (union exp_element) characters (or less for the last
element of a string). */
char string;
Index: gdb/parse.c
===================================================================
--- gdb/parse.c.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/parse.c 2007-03-20 14:07:38.000000000 -0300
@@ -251,6 +251,18 @@ write_exp_elt_dblcst (DOUBLEST expelt)
}
void
+write_exp_elt_decfloatcst (gdb_byte expelt[16])
+{
+ union exp_element tmp;
+ int index;
+
+ for (index = 0; index < 16; index++)
+ tmp.decfloatconst[index] = expelt[index];
+
+ write_exp_elt (tmp);
+}
+
+void
write_exp_elt_type (struct type *expelt)
{
union exp_element tmp;
@@ -892,6 +904,7 @@ operator_length_standard (struct express
case OP_LONG:
case OP_DOUBLE:
+ case OP_DECFLOAT:
case OP_VAR_VALUE:
oplen = 4;
break;
Index: gdb/parser-defs.h
===================================================================
--- gdb/parser-defs.h.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/parser-defs.h 2007-03-20 14:07:38.000000000 -0300
@@ -121,6 +121,8 @@ extern void write_exp_elt_longcst (LONGE
extern void write_exp_elt_dblcst (DOUBLEST);
+extern void write_exp_elt_decfloatcst (gdb_byte *);
+
extern void write_exp_elt_type (struct type *);
extern void write_exp_elt_intern (struct internalvar *);
Index: gdb/valarith.c
===================================================================
--- gdb/valarith.c.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/valarith.c 2007-03-20 14:13:55.000000000 -0300
@@ -1378,6 +1378,20 @@ value_neg (struct value *arg1)
type = check_typedef (value_type (arg1));
+ if (TYPE_CODE (type) == TYPE_CODE_DECFLOAT)
+ {
+ struct value *val = allocate_value (result_type);
+ int len = TYPE_LENGTH (type);
+ gdb_byte *decbytes = (gdb_byte *) value_contents (arg1);
+
+ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_LITTLE)
+ decbytes[len-1] = decbytes[len - 1] | 0x80;
+ else
+ decbytes[0] = decbytes[0] | 0x80;
+ memcpy (value_contents_raw (val), decbytes, 16);
+ return val;
+ }
+
if (TYPE_CODE (type) == TYPE_CODE_FLT)
return value_from_double (result_type, -value_as_double (arg1));
else if (is_integral_type (type))
Index: gdb/value.c
===================================================================
--- gdb/value.c.orig 2007-03-13 14:56:27.000000000 -0300
+++ gdb/value.c 2007-03-20 14:07:38.000000000 -0300
@@ -37,6 +37,7 @@
#include "gdb_assert.h"
#include "regcache.h"
#include "block.h"
+#include "dfp.h"
/* Prototypes for exported functions. */
@@ -1607,6 +1608,27 @@ value_from_double (struct type *type, DO
}
struct value *
+value_from_decfloat (struct type *expect_type, struct type *type,
+ gdb_byte decbytes[16])
+{
+ struct value *val = allocate_value (type);
+ int len = TYPE_LENGTH (type);
+
+ if (expect_type)
+ {
+ int expect_len = TYPE_LENGTH (expect_type);
+ char decstr[128];
+ int real_len;
+
+ decimal_to_string (decbytes, len, decstr);
+ decimal_from_string (decbytes, expect_len, decstr);
+ }
+
+ memcpy (value_contents_raw (val), decbytes, len);
+ return val;
+}
+
+struct value *
coerce_ref (struct value *arg)
{
struct type *value_type_arg_tmp = check_typedef (value_type (arg));
Index: gdb/value.h
===================================================================
--- gdb/value.h.orig 2007-03-20 14:07:37.000000000 -0300
+++ gdb/value.h 2007-03-20 14:07:38.000000000 -0300
@@ -274,6 +274,9 @@ extern LONGEST unpack_field_as_long (str
extern struct value *value_from_longest (struct type *type, LONGEST num);
extern struct value *value_from_pointer (struct type *type, CORE_ADDR addr);
extern struct value *value_from_double (struct type *type, DOUBLEST num);
+extern struct value *value_from_decfloat (struct type *expect_type,
+ struct type *type,
+ gdb_byte decbytes[16]);
extern struct value *value_from_string (char *string);
extern struct value *value_at (struct type *type, CORE_ADDR addr);
--
^ permalink raw reply [flat|nested] 10+ messages in thread
* [patch 1/4] libdecnumber support
[not found] <20070323030737.475073862@br.ibm.com>
2007-03-27 16:30 ` [patch 3/4] decimal floating point support for expressions Thiago Jung Bauermann
@ 2007-03-27 16:30 ` Thiago Jung Bauermann
2007-03-27 16:30 ` [patch 4/4] decimal floating point testcases Thiago Jung Bauermann
2 siblings, 0 replies; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-03-27 16:30 UTC (permalink / raw)
To: gdb-patches
plain text document attachment (libdecnumber-support.diff)
This patch provides functions to interact with libdecnumber and changes
the Makefile to compile it.
2007-03-23 Thiago Jung Bauermann <bauerman@br.ibm.com>
* Makefile.in (LIBDECNUMBER_DIR, LIBDECNUMBER, LIBDECNUMBER_SRC
LIBDECNUMBER_CFLAGS): New macros for libdecnumber.
(INTERNAL_CFLAGS_BASE): Add LIBDECNUMBER_CFLAGS in.
(INSTALLED_LIBS): Add -ldecnumber in.
(CLIBS): Add LIBDECNUMBER in.
(decimal128_h, decimal64_h, decimal32_h): New macros for decimal
headers.
(dfp_h): New macros for decimal floating point.
(dfp.o): New target.
(COMMON_OBS): Add dfp.o in.
(valprint.o): Add dfp_h as a new dependence.
* dfp.h: New header file for decimal floating point support in
GDB.
* dfp.c: New source file for decimal floating point support in
GDB. Implement decimal_from_string and decimal_to_string based
on libdecnumber API.
Index: gdb/Makefile.in
===================================================================
--- gdb/Makefile.in.orig 2007-03-13 14:56:28.000000000 -0300
+++ gdb/Makefile.in 2007-03-13 14:56:33.000000000 -0300
@@ -122,6 +122,12 @@ BFD = $(BFD_DIR)/libbfd.a
BFD_SRC = $(srcdir)/$(BFD_DIR)
BFD_CFLAGS = -I$(BFD_DIR) -I$(BFD_SRC)
+# Where is the decnumber library? Typically in ../libdecnumber.
+LIBDECNUMBER_DIR = ../libdecnumber
+LIBDECNUMBER = $(LIBDECNUMBER_DIR)/libdecnumber.a
+LIBDECNUMBER_SRC = $(srcdir)/$(LIBDECNUMBER_DIR)
+LIBDECNUMBER_CFLAGS = -I$(LIBDECNUMBER_DIR) -I$(LIBDECNUMBER_SRC)
+
# Where is the READLINE library? Typically in ../readline.
READLINE_DIR = ../readline
READLINE = $(READLINE_DIR)/libreadline.a
@@ -354,7 +360,7 @@ CXXFLAGS = -g -O
INTERNAL_CFLAGS_BASE = \
$(CFLAGS) $(GLOBAL_CFLAGS) $(PROFILE_CFLAGS) \
$(GDB_CFLAGS) $(OPCODES_CFLAGS) $(READLINE_CFLAGS) \
- $(BFD_CFLAGS) $(INCLUDE_CFLAGS) \
+ $(BFD_CFLAGS) $(INCLUDE_CFLAGS) $(LIBDECNUMBER_CFLAGS) \
$(INTL_CFLAGS) $(ENABLE_CFLAGS) $(INTERNAL_CPPFLAGS)
INTERNAL_WARN_CFLAGS = $(INTERNAL_CFLAGS_BASE) $(GDB_WARN_CFLAGS)
INTERNAL_CFLAGS = $(INTERNAL_WARN_CFLAGS) $(GDB_WERROR_CFLAGS)
@@ -377,10 +383,10 @@ INTERNAL_LDFLAGS = $(CFLAGS) $(GLOBAL_CF
# LIBIBERTY appears twice on purpose.
# If you have the Cygnus libraries installed,
# you can use 'CLIBS=$(INSTALLED_LIBS)' 'CDEPS='
-INSTALLED_LIBS=-lbfd -lreadline -lopcodes -liberty \
+INSTALLED_LIBS=-lbfd -lreadline -lopcodes -liberty -ldecnumber \
$(XM_CLIBS) $(TM_CLIBS) $(NAT_CLIBS) $(GDBTKLIBS) @LIBS@ \
-lintl -liberty
-CLIBS = $(SIM) $(READLINE) $(OPCODES) $(BFD) $(INTL) $(LIBIBERTY) \
+CLIBS = $(SIM) $(READLINE) $(OPCODES) $(BFD) $(INTL) $(LIBIBERTY) $(LIBDECNUMBER) \
$(XM_CLIBS) $(TM_CLIBS) $(NAT_CLIBS) $(GDBTKLIBS) @LIBS@ \
$(LIBICONV) $(LIBEXPAT) \
$(LIBIBERTY) $(WIN32LIBS)
@@ -620,6 +626,10 @@ safe_ctype_h = $(INCLUDE_DIR)/safe-ctyp
hashtab_h = $(INCLUDE_DIR)/hashtab.h
filenames_h = $(INCLUDE_DIR)/filenames.h
+decimal128_h = $(LIBDECNUMBER_DIR)/decimal128.h
+decimal64_h = $(LIBDECNUMBER_DIR)/decimal64.h
+decimal32_h = $(LIBDECNUMBER_DIR)/decimal32.h
+
#
# $BUILD/ headers
#
@@ -687,6 +697,7 @@ dictionary_h = dictionary.h
disasm_h = disasm.h
doublest_h = doublest.h $(floatformat_h)
dummy_frame_h = dummy-frame.h
+dfp_h = dfp.h
dwarf2expr_h = dwarf2expr.h
dwarf2_frame_h = dwarf2-frame.h
dwarf2loc_h = dwarf2loc.h
@@ -937,7 +948,7 @@ COMMON_OBS = $(DEPFILES) $(CONFIG_OBS) $
auxv.o \
bfd-target.o \
blockframe.o breakpoint.o findvar.o regcache.o \
- charset.o disasm.o dummy-frame.o \
+ charset.o disasm.o dummy-frame.o dfp.o \
source.o value.o eval.o valops.o valarith.o valprint.o printcmd.o \
block.o symtab.o symfile.o symmisc.o linespec.o dictionary.o \
infcall.o \
@@ -1966,6 +1977,7 @@ dsrec.o: dsrec.c $(defs_h) $(serial_h) $
dummy-frame.o: dummy-frame.c $(defs_h) $(dummy_frame_h) $(regcache_h) \
$(frame_h) $(inferior_h) $(gdb_assert_h) $(frame_unwind_h) \
$(command_h) $(gdbcmd_h) $(gdb_string_h)
+dfp.o: dfp.c $(defs_h) $(dfp_h) $(decimal128_h) $(decimal64_h) $(decimal32_h)
dwarf2expr.o: dwarf2expr.c $(defs_h) $(symtab_h) $(gdbtypes_h) $(value_h) \
$(gdbcore_h) $(elf_dwarf2_h) $(dwarf2expr_h)
dwarf2-frame.o: dwarf2-frame.c $(defs_h) $(dwarf2expr_h) $(elf_dwarf2_h) \
@@ -2867,7 +2879,7 @@ valops.o: valops.c $(defs_h) $(symtab_h)
valprint.o: valprint.c $(defs_h) $(gdb_string_h) $(symtab_h) $(gdbtypes_h) \
$(value_h) $(gdbcore_h) $(gdbcmd_h) $(target_h) $(language_h) \
$(annotate_h) $(valprint_h) $(floatformat_h) $(doublest_h) \
- $(exceptions_h)
+ $(exceptions_h) $(dfp_h)
value.o: value.c $(defs_h) $(gdb_string_h) $(symtab_h) $(gdbtypes_h) \
$(value_h) $(gdbcore_h) $(command_h) $(gdbcmd_h) $(target_h) \
$(language_h) $(scm_lang_h) $(demangle_h) $(doublest_h) \
Index: gdb/dfp.c
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/dfp.c 2007-03-13 15:19:43.000000000 -0300
@@ -0,0 +1,122 @@
+/* Decimal floating point support for GDB.
+
+ Copyright 2006 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 "dfp.h"
+#include <ctype.h>
+
+/* The order of the following headers is important for making sure
+ decNumber structure is large enough to hold decimal128 digits. */
+
+#include "decimal128.h"
+#include "decimal64.h"
+#include "decimal32.h"
+#include "decNumber.h"
+
+/* In gdb, we are using an array of gdb_byte to represent decimal values. The
+ byte order of our representation might be different than that of underlying
+ architecture. This routine does the conversion if it is necessary. */
+static void
+exchange_dfp (const gdb_byte *valaddr, int len, gdb_byte *dec_val)
+{
+ int index;
+
+ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_LITTLE)
+ for (index = 0; index < len; index++)
+ dec_val[index] = valaddr[len - index - 1];
+ else
+ for (index = 0; index < len; index++)
+ dec_val[index] = valaddr[index];
+
+ return;
+}
+
+/* Convert decimal type to its string representation. LEN is the length
+ of the decimal type, 4 bytes for decimal32, 8 bytes for decimal64 and
+ 16 bytes for decimal128. */
+void
+decimal_to_string (const uint8_t *decbytes, int len, char *s)
+{
+ uint8_t *dec = (uint8_t *)malloc (len);
+
+ exchange_dfp (decbytes, len, dec);
+ switch (len)
+ {
+ case 4:
+ decimal32ToString ((decimal32 *) dec, s);
+ return;
+ case 8:
+ decimal64ToString ((decimal64 *) dec, s);
+ return;
+ case 16:
+ decimal128ToString ((decimal128 *) dec, s);
+ return;
+ default:
+ return;
+ }
+
+ free (dec);
+}
+
+/* Convert the string form of a decimal value to its decimal representation.
+ LEN is the length of the decimal type, 4 bytes for decimal32, 8 bytes for
+ decimal64 and 16 bytes for decimal128. */
+int
+decimal_from_string (uint8_t *decbytes, int len, char *string)
+{
+ decNumber dn;
+ decContext set;
+ int index;
+ uint8_t *dec = (uint8_t *)malloc (len);
+
+ switch (len)
+ {
+ case 4:
+ decContextDefault (&set, DEC_INIT_DECIMAL32);
+ break;
+ case 8:
+ decContextDefault (&set, DEC_INIT_DECIMAL64);
+ break;
+ case 16:
+ decContextDefault (&set, DEC_INIT_DECIMAL128);
+ break;
+ }
+
+ set.traps = 0;
+
+ decNumberFromString (&dn, string, &set);
+ switch (len)
+ {
+ case 4:
+ decimal32FromNumber ((decimal32 *) dec, &dn, &set);
+ break;
+ case 8:
+ decimal64FromNumber ((decimal64 *) dec, &dn, &set);
+ break;
+ case 16:
+ decimal128FromNumber ((decimal128 *) dec, &dn, &set);
+ break;
+ }
+
+ exchange_dfp (dec, len, decbytes);
+ free (dec);
+ return 1;
+}
Index: gdb/dfp.h
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/dfp.h 2007-03-13 14:56:33.000000000 -0300
@@ -0,0 +1,40 @@
+/* Decimal floating point support for GDB.
+
+ Copyright 2006 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. */
+
+/* Decimal floating point is one of the extension to IEEE 754, which is
+ described in http://grouper.ieee.org/groups/754/revision.html and
+ http://www2.hursley.ibm.com/decimal/. It completes binary floating
+ point by representing floating point more exactly. */
+
+/* There is a project intended to add DFP support into GCC, described in
+ http://gcc.gnu.org/wiki/Decimal%20Floating-Point. This file is intended
+ to add DFP support into GDB. */
+
+#ifndef DFP_H
+#define DFP_H
+#include <string.h>
+#include <stdio.h>
+#include <stdint.h>
+
+extern void decimal_to_string (const uint8_t *, int, char *);
+extern int decimal_from_string (uint8_t *, int, char *);
+
+#endif
--
^ permalink raw reply [flat|nested] 10+ messages in thread
* [patch 4/4] decimal floating point testcases
[not found] <20070323030737.475073862@br.ibm.com>
2007-03-27 16:30 ` [patch 3/4] decimal floating point support for expressions Thiago Jung Bauermann
2007-03-27 16:30 ` [patch 1/4] libdecnumber support Thiago Jung Bauermann
@ 2007-03-27 16:30 ` Thiago Jung Bauermann
2 siblings, 0 replies; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-03-27 16:30 UTC (permalink / raw)
To: gdb-patches
plain text document attachment (testcases.diff)
2007-03-23 Thiago Jung Bauermann <bauerman@br.ibm.com>
* gdb.base/dfp-exprs.exp: new file adding tests for Decimal
Floating Point expressions.
* gdb.base/dfp-test.exp: new file adding tests for Decimal
Floating Point variables.
* gdb.base/dfp-test.c: new file containing program with Decimal
Floating variables, used by gdb.base/dfp-test.exp.
Index: gdb/testsuite/gdb.base/dfp-exprs.exp
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/testsuite/gdb.base/dfp-exprs.exp 2007-03-13 14:47:56.000000000 -0300
@@ -0,0 +1,95 @@
+# Copyright (C) 2006 Free Software Foundation, Inc.
+
+# 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.
+
+# This file was written by Wu Zhou. (woodzltc@cn.ibm.com)
+
+# This file is part of the gdb testsuite. It contains test for evaluating
+# simple decimal floating point (DFP) expression.
+
+if $tracelevel then {
+ strace $tracelevel
+}
+
+proc test_dfp_literals_accepted {} {
+
+ # Test various dfp values, covering 32-bit, 64-bit and 128-bit ones
+
+ # _Decimal32 constants, which can support up to 7 digits
+ gdb_test "p 1.2df" " = 1.2"
+ gdb_test "p -1.2df" " = -1.2"
+ gdb_test "p 1.234567df" " = 1.234567"
+ gdb_test "p -1.234567df" " = -1.234567"
+ gdb_test "p 1234567.df" " = 1234567"
+ gdb_test "p -1234567.df" " = -1234567"
+
+ gdb_test "p 1.2E1df" " = 12"
+ gdb_test "p 1.2E10df" " = 1.2E\\+10"
+ gdb_test "p 1.2E-10df" " = 1.2E-10"
+
+ # The largest exponent for 32-bit dfp value is 96.
+ gdb_test "p 1.2E96df" " = 1.200000E\\+96"
+
+ # _Decimal64 constants, which can support up to 16 digits
+ gdb_test "p 1.2dd" " = 1.2"
+ gdb_test "p -1.2dd" " = -1.2"
+ gdb_test "p 1.234567890123456dd" " = 1.234567890123456"
+ gdb_test "p -1.234567890123456dd" " = -1.234567890123456"
+ gdb_test "p 1234567890123456.dd" " = 1234567890123456"
+ gdb_test "p -1234567890123456.dd" " = -1234567890123456"
+
+ gdb_test "p 1.2E1dd" " = 12"
+ gdb_test "p 1.2E10dd" " = 1.2E\\+10"
+ gdb_test "p 1.2E-10dd" " = 1.2E-10"
+
+ # The largest exponent for 64-bit dfp value is 384.
+ gdb_test "p 1.2E384dd" " = 1.200000000000000E\\+384"
+
+ # _Decimal128 constants, which can support up to 34 digits
+ gdb_test "p 1.2dl" " = 1.2"
+ gdb_test "p -1.2dl" " = -1.2"
+ gdb_test "p 1.234567890123456789012345678901234dl" " = 1.234567890123456789012345678901234"
+ gdb_test "p -1.234567890123456789012345678901234dl" " = -1.234567890123456789012345678901234"
+ gdb_test "p 1234567890123456789012345678901234.dl" " = 1234567890123456789012345678901234"
+ gdb_test "p -1234567890123456789012345678901234.dl" " = -1234567890123456789012345678901234"
+
+ gdb_test "p 1.2E1dl" " = 12"
+ gdb_test "p 1.2E10dl" " = 1.2E\\+10"
+ gdb_test "p 1.2E-10dl" " = 1.2E-10"
+
+ # The largest exponent for 128-bit dfp value is 6144.
+ gdb_test "p 1.2E6144dl" " = 1.200000000000000000000000000000000E\\+6144"
+}
+
+proc test_dfp_arithmetic_expressions {} {
+
+# Arithmetic operations for DFP types are not yet implemented in GDB.
+# These tests are to verify that they will generate expected error messages.
+
+ gdb_test "p 1.4df + 1.2df" "Argument to arithmetic operation not a number or boolean.*"
+ gdb_test "p 1.4df - 1.2df" ".*Argument to arithmetic operation not a number or boolean.*"
+ gdb_test "p 1.4df * 1.2df" "Argument to arithmetic operation not a number or boolean.*"
+ gdb_test "p 1.4df / 1.2df" "Argument to arithmetic operation not a number or boolean.*"
+
+}
+
+# Start with a fresh gdb.
+
+gdb_exit
+gdb_start
+gdb_reinitialize_dir $srcdir/$subdir
+
+test_dfp_literals_accepted
+test_dfp_arithmetic_expressions
Index: gdb/testsuite/gdb.base/dfp-test.c
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/testsuite/gdb.base/dfp-test.c 2007-03-13 14:47:56.000000000 -0300
@@ -0,0 +1,95 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+ Copyright 2006 Free Software Foundation, Inc.
+
+ 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 <stdio.h>
+#include <stdlib.h>
+
+volatile _Decimal32 d32;
+volatile _Decimal64 d64;
+volatile _Decimal128 d128;
+
+struct decstruct
+{
+ int int4;
+ long long8;
+ _Decimal32 dec32;
+ _Decimal64 dec64;
+ _Decimal128 dec128;
+} ds;
+
+static _Decimal32
+arg0_32 (_Decimal32 arg0, _Decimal32 arg1, _Decimal32 arg2,
+ _Decimal32 arg3, _Decimal32 arg4, _Decimal32 arg5)
+{
+ return arg0;
+}
+
+static _Decimal64
+arg0_64 (_Decimal64 arg0, _Decimal64 arg1, _Decimal64 arg2,
+ _Decimal64 arg3, _Decimal64 arg4, _Decimal64 arg5)
+{
+ return arg0;
+}
+
+static _Decimal128
+arg0_128 (_Decimal128 arg0, _Decimal128 arg1, _Decimal128 arg2,
+ _Decimal128 arg3, _Decimal128 arg4, _Decimal128 arg5)
+{
+ return arg0;
+}
+
+int main()
+{
+ /* An finite 32-bits decimal floating point. */
+ d32 = 1.2345df; /* Initialize d32. */
+
+ /* Non-finite 32-bits decimal floating point: infinity and NaN. */
+ d32 = __builtin_infd32(); /* Positive infd32. */
+ d32 = -__builtin_infd32(); /* Negative infd32. */
+ d32 = __builtin_nand32("");
+
+ /* An finite 64-bits decimal floating point. */
+ d64 = 1.2345dd; /* Initialize d64. */
+
+ /* Non-finite 64-bits decimal floating point: infinity and NaN. */
+ d64 = __builtin_infd64(); /* Positive infd64. */
+ d64 = -__builtin_infd64(); /* Negative infd64. */
+ d64 = __builtin_nand64("");
+
+ /* An finite 128-bits decimal floating point. */
+ d128 = 1.2345dl; /* Initialize d128. */
+
+ /* Non-finite 128-bits decimal floating point: infinity and NaN. */
+ d128 = __builtin_infd128(); /* Positive infd128. */
+ d128 = -__builtin_infd128(); /* Negative infd128. */
+ d128 = __builtin_nand128("");
+
+ /* Functions with decimal floating point as parameter and return value. */
+ d32 = arg0_32 (0.1df, 1.0df, 2.0df, 3.0df, 4.0df, 5.0df);
+ d64 = arg0_64 (0.1dd, 1.0dd, 2.0dd, 3.0dd, 4.0dd, 5.0dd);
+ d128 = arg0_128 (0.1dl, 1.0dl, 2.0dl, 3.0dl, 4.0dl, 5.0dl);
+
+ ds.int4 = 1;
+ ds.long8 = 2;
+ ds.dec32 = 1.2345df;
+ ds.dec64 = 1.2345dd;
+ ds.dec128 = 1.2345dl;
+
+ return 0; /* Exit point. */
+}
Index: gdb/testsuite/gdb.base/dfp-test.exp
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/testsuite/gdb.base/dfp-test.exp 2007-03-13 14:47:56.000000000 -0300
@@ -0,0 +1,247 @@
+# Copyright 2006 Free Software Foundation, Inc.
+
+# 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.
+
+# This file was written by Wu Zhou. (woodzltc@cn.ibm.com)
+
+# This file is part of the gdb testsuite. It is intended to test that
+# gdb could correctly hane decimal floating point introduced in IEEE 754R.
+
+proc d32_set_tests {} {
+
+ gdb_test "p d32=123.45df" " = 123.45"
+ gdb_test "p d32=12345.df" " = 12345"
+ gdb_test "p d32=12345.67df" " = 12345.67"
+ gdb_test "p d32=1234567.df" " = 1234567"
+
+ gdb_test "p d32=1.234567E0df" " = 1.234567"
+ gdb_test "p d32=1.234567E10df" " = 1.234567E\\+10"
+ gdb_test "p d32=1.234567E+96df" " = 1.234567E\\+96"
+
+ # Test that gdb could handle the max, normalized min and subnormalized min.
+ gdb_test "p d32=9.999999E96df" " = 9.999999E\\+96"
+ gdb_test "p d32=1.0E-95df" " = 1.0E\\-95"
+ gdb_test "p d32=1.E-101df" " = 1E\\-101"
+ gdb_test "p d32=0.000001E-95df" " = 1E\\-101"
+
+ # Test that gdb could detect coefficient/exponent out of range.
+ # The coefficient out of range will be rounded to its nearest value.
+ # And the exponent out of range will be handled as infinity.
+ gdb_test "p d32=1.2345678df" " = 1.234568" "1.2345678 is rounded to 1.234568"
+ gdb_test "p d32=1.0E-101df" " = 1E-101" "1.0E-101 is rounded to 1E-101"
+ gdb_test "p d32=1.234567E+97df" " = Infinity" "1.234567E+97 is Infinity"
+
+ # Test that gdb could detect the errors in the string representation of _Decimal32
+ gdb_test "p d32=12345.df" " = 12345" "12345. is an valid number"
+ gdb_test "p d32=12345df" ".*Invalid number.*" "12345 is an invalid number"
+ gdb_test "p d32=1.23Edf" " = NaN" "1.23E is NaN (not a number)"
+ gdb_test "p d32=1.23E45Adf" " = NaN" "1.23E45A is NaN (not a number)"
+}
+
+proc d64_set_tests {} {
+
+ gdb_test "p d64=123.45dd" " = 123.45"
+ gdb_test "p d64=12345.dd" " = 12345"
+ gdb_test "p d64=12345.67dd" " = 12345.67"
+ gdb_test "p d64=1.234567890123456dd" " = 1.234567890123456"
+
+ gdb_test "p d64=1.234567890123456E10dd" " = 12345678901.23456"
+ gdb_test "p d64=1.234567890123456E100dd" " = 1.234567890123456E\\+100"
+ gdb_test "p d64=1.234567890123456E384dd" " = 1.234567890123456E\\+384"
+
+ # Test that gdb could handle the max, normalized min and subnormalized min.
+ gdb_test "p d64=9.999999999999999E384dd" " = 9.999999999999999E\\+384"
+ gdb_test "p d64=1.E-383dd" " = 1E\\-383"
+ gdb_test "p d64=1.E-398dd" " = 1E\\-398"
+ gdb_test "p d64=0.000000000000001E-383dd" " = 1E\\-398"
+
+ # Test that gdb could detect coefficient/exponent out of range.
+ # The coefficient out of range will be rounded to its nearest value.
+ # And the exponent out of range will be handled as infinity.
+ gdb_test "p d64=1.2345678901234567dd" " = 1.234567890123457" "1.2345678901234567 is rounded to 1.234567890123457"
+ gdb_test "p d64=9.9999999999999999E384dd" " = Infinity" "d64=9.9999999999999999E384 is Infinity"
+ gdb_test "p d64=1.234567890123456E385dd" " = Infinity" "d64=1.234567890123456E385 is Infinity"
+
+ # Test that gdb could detect the errors in the string representation of _Decimal64
+ gdb_test "p d64=12345dd" ".*Invalid number.*" "12345dd is an invalid number"
+ gdb_test "p d64=1.23Edd" " = NaN" "1.23E is NaN (not a number)"
+ gdb_test "p d64=1.23E45Add" "= NaN" "1.23E45A is NaN (not a number)"
+}
+
+proc d128_set_tests {} {
+
+ gdb_test "p d128=123.45dl" " = 123.45"
+ gdb_test "p d128=12345.dl" " = 12345"
+ gdb_test "p d128=12345.67dl" " = 12345.67"
+ gdb_test "p d128=1.234567890123456789012345678901234dl" " = 1.234567890123456789012345678901234"
+
+ gdb_test "p d128=1.234567890123456E10dl" " = 12345678901.23456"
+ gdb_test "p d128=1.234567890123456E100dl" " = 1.234567890123456E\\+100"
+ gdb_test "p d128=1.234567890123456E1000dl" " = 1.234567890123456E\\+1000"
+
+ # Test that gdb could handle the max, normalized min and subnormalized min.
+ gdb_test "p d128=9.999999999999999999999999999999999E6144dl" " = 9.999999999999999999999999999999999E\\+6144"
+ gdb_test "p d128=1.E-6143dl" " = 1E\\-6143"
+ gdb_test "p d128=1.E-6176dl" " = 1E\\-6176"
+ gdb_test "p d128=0.000000000000000000000000000000001E-6143dl" " = 1E\\-6176"
+
+ # Test that gdb could detect coefficient/exponent out of range.
+ # The coefficient out of range will be rounded to its nearest value.
+ # And the exponent out of range will be handled as infinity.
+ gdb_test "p d128=1.2345678901234567890123456789012345dl" "1.234567890123456789012345678901234" "1.2345678901234567890123456789012345 is rounded to 1.234567890123456789012345678901234"
+ gdb_test "p d128=1.234567890123456E6145dl" "Infinity" "d128=1.234567890123456E6145 is Infinity"
+
+ # Test that gdb could detect the errors in the string representation of _Decimal128
+ gdb_test "p d128=12345dl" ".*Invalid number.*" "12345dl is an invalid number"
+ gdb_test "p d128=1.23Edl" " = NaN" "1.23E is NaN (not a number)"
+ gdb_test "p d128=1.23E45Adl" "= NaN" "1.23E45A is NaN (not a number)"
+}
+
+
+if $tracelevel {
+ strace $tracelevel
+}
+
+set testfile "dfp-test"
+set srcfile ${testfile}.c
+set binfile ${objdir}/${subdir}/${testfile}
+if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable "compiler=/opt/biarch/gcc411-P6DFP4/bin/gcc debug"] != "" } {
+ untested "Couldn't compile ${srcfile}"
+ return -1
+}
+
+gdb_exit
+gdb_start
+gdb_reinitialize_dir $srcdir/$subdir
+gdb_load ${binfile}
+
+if ![runto_main] then {
+ perror "couldn't run to breakpoint"
+ continue
+}
+
+# Different tests on 32-bits decimal floating point, including the printing
+# of finite numbers, infinite and NaN, and also the setting of different
+# decimal value.
+
+if [gdb_test "next" \
+ ".*Positive infd32.*" \
+ "next after initializing d32"] then { gdb_suppress_tests }
+gdb_test "print d32" "1.2345" "d32 is initialized to 1.2345"
+
+if [gdb_test "next" \
+ ".*Negative infd32.*" \
+ "next after assigning builtin infinity to d32"] then { gdb_suppress_tests }
+gdb_test "print d32" "Infinity" "d32 is positive Infinity"
+
+if [gdb_test "next" \
+ ".*__builtin_nand32.*" \
+ "next after assigning negative builtin infinity to d32"] then { gdb_suppress_tests }
+gdb_test "print d32" "-Infinity" "d32 is negative Infinity"
+
+if [gdb_test "next" \
+ ".*d64 = 1.2345.*" \
+ "next after assigning builtin NaN to d32"] then { gdb_suppress_tests }
+gdb_test "print d32" "NaN" "d32 is NaN"
+
+d32_set_tests
+
+
+# Different tests on 64-bits decimal floating point, including the display
+# of finite number, infinite and NaN, and also the setting of different
+# decimal value.
+
+if [gdb_test "next" \
+ ".*Positive infd64.*" \
+ "next after initializing d64"] then { gdb_suppress_tests }
+gdb_test "print d64" "1.2345" "d64 is initialized to 1.2345"
+
+if [gdb_test "next" \
+ ".*Negative infd64.*" \
+ "next after assigning builtin infinity to d64"] then { gdb_suppress_tests }
+gdb_test "print d64" "Infinity" "d64 is positive Infinity"
+
+if [gdb_test "next" \
+ ".*__builtin_nand64.*" \
+ "next after assigning negative builtin infinity to d64"] then { gdb_suppress_tests }
+gdb_test "print d64" "-Infinity" "d64 is negative Infinity"
+
+if [gdb_test "next" \
+ ".*d128 = 1.2345.*" \
+ "next after assigning builtin NaN to d64"] then { gdb_suppress_tests }
+gdb_test "print d64" "NaN" "d64 is NaN"
+
+d64_set_tests
+
+
+# Different tests on 128-bits decimal floating point, including the display
+# of finite number, infinite and NaN, and also the setting of different
+# decimal value.
+
+if [gdb_test "next" \
+ ".*Positive infd128.*" \
+ "next after initializing d128"] then { gdb_suppress_tests }
+gdb_test "print d128" "1.2345" "d128 is initialized to 1.2345"
+
+d128_set_tests
+
+if [gdb_test "next" \
+ ".*Negative infd128.*" \
+ "next after assigning builtin infinity to d128"] then { gdb_suppress_tests }
+gdb_test "print d128" "Infinity" "d128 is positive Infinity"
+
+if [gdb_test "next" \
+ ".*__builtin_nand128.*" \
+ "next after assigning negative builtin infinity to d128"] then { gdb_suppress_tests }
+gdb_test "print d128" "-Infinity" "d128 is negative Infinity"
+
+if [gdb_test "next" \
+ ".*arg0_32.*" \
+ "next after assigning builtin NaN to d128"] then { gdb_suppress_tests }
+gdb_test "print d128" "NaN" "d128 is NaN"
+
+# The following tests are intended to verify that gdb can correctly handle
+# DFP types in function arguments.
+
+gdb_breakpoint arg0_32
+gdb_continue_to_breakpoint "entry to arg0_32"
+gdb_test "backtrace" ".*arg0_32 \\(arg0=0.1, arg1=1.0, arg2=2.0, arg3=3.0, arg4=4.0, arg5=5.0\\).*" "backtrace at arg0_32"
+
+gdb_breakpoint arg0_64
+gdb_continue_to_breakpoint "entry to arg0_64"
+gdb_test "backtrace" ".*arg0_64 \\(arg0=0.1, arg1=1.0, arg2=2.0, arg3=3.0, arg4=4.0, arg5=5.0\\).*" "backtrace at arg0_64"
+
+gdb_breakpoint arg0_128
+gdb_continue_to_breakpoint "entry to arg0_128"
+gdb_test "backtrace" ".*arg0_128 \\(arg0=0.1, arg1=1.0, arg2=2.0, arg3=3.0, arg4=4.0, arg5=5.0\\).*" "backtrace at arg0_128"
+
+# The following tests are intended to verify that gdb can handle DFP types
+# correctly in struct.
+
+gdb_breakpoint [gdb_get_line_number "Exit point"]
+gdb_continue_to_breakpoint "Setting a decimal struct"
+gdb_test "print ds.dec32" " = 1.2345"
+gdb_test "print ds.dec64" " = 1.2345"
+gdb_test "print ds.dec128" " = 1.2345"
+
+# The following tests are intended to verify that gdb can handle "d1=d2"
+# and "d1=-d2" correctly.
+
+gdb_test "print ds.dec32=d32" " = 0.1"
+gdb_test "print ds.dec64=d64" " = 0.1"
+gdb_test "print ds.dec128=d128" " = 0.1"
+gdb_test "print ds.dec32 = -d32" " = -0.1"
+gdb_test "print ds.dec64 = -d64" " = -0.1"
+gdb_test "print ds.dec128 = -d128" " = -0.1"
--
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [patch 1/4] libdecnumber support
2007-10-15 18:38 ` Daniel Jacobowitz
@ 2007-10-25 18:23 ` Thiago Jung Bauermann
0 siblings, 0 replies; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-10-25 18:23 UTC (permalink / raw)
To: Daniel Jacobowitz; +Cc: gdb-patches
On Mon, 2007-10-15 at 14:31 -0400, Daniel Jacobowitz wrote:
> Our policy is to always use leading tabs; so whenever you have eight
> spaces at the line it becomes a tab. This comes, of course, from
> emacs :-)
>
> Aside from leading spaces, and using autoconf 2.59, this version is
> OK. I added libdecnumber last week, so all should be ready now.
Converted to leading tabs, used autoconf 2.59 and commited. Thanks!
--
[]'s
Thiago Jung Bauermann
Software Engineer
IBM Linux Technology Center
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [patch 1/4] libdecnumber support
2007-10-15 18:11 ` Thiago Jung Bauermann
@ 2007-10-15 18:38 ` Daniel Jacobowitz
2007-10-25 18:23 ` Thiago Jung Bauermann
0 siblings, 1 reply; 10+ messages in thread
From: Daniel Jacobowitz @ 2007-10-15 18:38 UTC (permalink / raw)
To: Thiago Jung Bauermann; +Cc: gdb-patches
On Mon, Oct 15, 2007 at 04:10:57PM -0200, Thiago Jung Bauermann wrote:
> Ok, I'm using it now. When I run autoreconf inside the src/gdb
> directory, I get the following:
>
> $ autoreconf
> autoreconf: cannot create `cd $srcdir;pwd`/../../../..: No such file or
> directory
> autoreconf: cannot create `cd $srcdir;pwd`/../../..: No such file or
> directory
> autoreconf: cannot create `cd $srcdir;pwd`/..: No such file or directory
> $ echo $?
> 0
>
> Is this a problem? Things still seem to work fine. I used autoconf
> version 2.61.
I have not tried autoreconf. It's helpful to use the same version
that was used last time, when not deliberately upgrading autoconf;
in this case that's 2.59. I keep a separate installation of it around
for this purpose, and run autoconf + autoheader by hand.
> > > + break;
> > > + default:
> > > + free (dec);
> >
> > Spaces and tabs mixed up? xfree, like xmalloc.
>
> Fixed. No '\t' in the file now. By the way, what's the policy in GDB
> w.r.t. tabs and spaces? Never use tabs?
Our policy is to always use leading tabs; so whenever you have eight
spaces at the line it becomes a tab. This comes, of course, from
emacs :-)
Aside from leading spaces, and using autoconf 2.59, this version is
OK. I added libdecnumber last week, so all should be ready now.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [patch 1/4] libdecnumber support
2007-10-11 16:07 ` Daniel Jacobowitz
@ 2007-10-15 18:32 ` Thiago Jung Bauermann
0 siblings, 0 replies; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-10-15 18:32 UTC (permalink / raw)
To: Daniel Jacobowitz; +Cc: gdb-patches
On Thu, 2007-10-11 at 12:00 -0400, Daniel Jacobowitz wrote:
> Oh, I also noticed some GPLv2 copyright headers; please update to
> GPLv3.
Yeah, forgot to update headers. This is done in the new patches I'm
posting now, for the new files created by this patch series.
--
[]'s
Thiago Jung Bauermann
Software Engineer
IBM Linux Technology Center
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [patch 1/4] libdecnumber support
2007-10-11 15:57 ` Daniel Jacobowitz
2007-10-11 16:07 ` Daniel Jacobowitz
@ 2007-10-15 18:11 ` Thiago Jung Bauermann
2007-10-15 18:38 ` Daniel Jacobowitz
1 sibling, 1 reply; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-10-15 18:11 UTC (permalink / raw)
To: Daniel Jacobowitz; +Cc: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 2968 bytes --]
On Thu, 2007-10-11 at 11:54 -0400, Daniel Jacobowitz wrote:
> I'll review all these now.
Thank you very much.
> Before you check them in, though, give
> me a chance to get libdecnumber merged into the src repository. It
> shouldn't take long.
Thanks for taking care of this.
> On Thu, Sep 20, 2007 at 06:54:11PM -0300, Thiago Jung Bauermann wrote:
> > +#include <ctype.h>
> > +#include <endian.h>
> > +#include "defs.h"
> > +#include "dfp.h"
>
> What do you need ctype.h for? That will tell me whether you really
> want ctype.h or libiberty's locale-independent safe-ctype.h.
>
> defs.h should always be first.
It turned out that ctype.h was not needed at all. I just removed it and
it made no difference. Probably a left-over from an older version of the
code.
> endian.h is not portable. You should probably use the autoconf macro
> AC_C_BIGENDIAN.
Ok, I'm using it now. When I run autoreconf inside the src/gdb
directory, I get the following:
$ autoreconf
autoreconf: cannot create `cd $srcdir;pwd`/../../../..: No such file or
directory
autoreconf: cannot create `cd $srcdir;pwd`/../../..: No such file or
directory
autoreconf: cannot create `cd $srcdir;pwd`/..: No such file or directory
$ echo $?
0
Is this a problem? Things still seem to work fine. I used autoconf
version 2.61.
> > +/* Convert decimal type to its string representation. LEN is the length
> > + of the decimal type, 4 bytes for decimal32, 8 bytes for decimal64 and
> > + 16 bytes for decimal128. */
> > +void
> > +decimal_to_string (const uint8_t *decbytes, int len, char *s)
> > +{
> > + uint8_t *dec = (uint8_t *)malloc (len);
>
> You didn't include anything that would give you uint8_t; use
> gdb_byte.
I included dfp.h, which included stdint.h. I see now that GDB doesn't
use C99 yet, so I changed to gdb_byte.
> Also we never use malloc, only xmalloc. Or you could avoid
> the allocation by using gdb_byte dec[16].
Good idea. Just changed to use temporary array as suggested.
> > + break;
> > + default:
> > + free (dec);
>
> Spaces and tabs mixed up? xfree, like xmalloc.
Fixed. No '\t' in the file now. By the way, what's the policy in GDB
w.r.t. tabs and spaces? Never use tabs?
> > + error(_("Unknown decimal floating point type.\n"));
>
> Space before parentheses please.
Fixed.
> > +/* There is a project intended to add DFP support into GCC, described in
> > + http://gcc.gnu.org/wiki/Decimal%20Floating-Point. This file is intended
> > + to add DFP support into GDB. */
>
> This comment doesn't belong here. The file is part of GDB once the
> patch is applied.
Ok, removed comment.
> > +#ifndef DFP_H
> > +#define DFP_H
> > +#include <string.h>
> > +#include <stdio.h>
> > +#include <stdint.h>
>
> Don't include standard headers here; stick to what defs.h provides.
> Just as well, since neither string.h nor stdint.h is portable.
Ok, removed.
--
[]'s
Thiago Jung Bauermann
Software Engineer
IBM Linux Technology Center
[-- Attachment #2: libdecnumber-support.diff --]
[-- Type: text/x-patch, Size: 11282 bytes --]
Subject: libdecnumber support
This patch provides functions to interact with libdecnumber and changes
the Makefile to compile it.
2007-10-15 Wu Zhou <woodzltc@cn.ibm.com>
Thiago Jung Bauermann <bauerman@br.ibm.com>
* Makefile.in (LIBDECNUMBER_DIR, LIBDECNUMBER, LIBDECNUMBER_SRC
LIBDECNUMBER_CFLAGS): New macros for libdecnumber.
(INTERNAL_CFLAGS_BASE): Add LIBDECNUMBER_CFLAGS in.
(INSTALLED_LIBS): Add -ldecnumber in.
(CLIBS): Add LIBDECNUMBER in.
(decimal128_h, decimal64_h, decimal32_h): New macros for decimal
headers.
(dfp_h): New macros for decimal floating point.
(dfp.o): New target.
(COMMON_OBS): Add dfp.o in.
(c-exp.o): Add dfp_h as dependency.
(valprint.o): Add dfp_h as dependency.
(value.o): Add dfp_h as dependency.
* dfp.h: New header file for decimal floating point support in
GDB.
* dfp.c: New source file for decimal floating point support in
GDB. Implement decimal_from_string and decimal_to_string based
on libdecnumber API.
* configure.ac: Add AC_C_BIGENDIAN test.
Index: gdb/Makefile.in
===================================================================
--- gdb/Makefile.in.orig 2007-10-15 14:25:14.000000000 -0200
+++ gdb/Makefile.in 2007-10-15 14:35:06.000000000 -0200
@@ -125,6 +125,12 @@ BFD = $(BFD_DIR)/libbfd.a
BFD_SRC = $(srcdir)/$(BFD_DIR)
BFD_CFLAGS = -I$(BFD_DIR) -I$(BFD_SRC)
+# Where is the decnumber library? Typically in ../libdecnumber.
+LIBDECNUMBER_DIR = ../libdecnumber
+LIBDECNUMBER = $(LIBDECNUMBER_DIR)/libdecnumber.a
+LIBDECNUMBER_SRC = $(srcdir)/$(LIBDECNUMBER_DIR)
+LIBDECNUMBER_CFLAGS = -I$(LIBDECNUMBER_DIR) -I$(LIBDECNUMBER_SRC)
+
# Where is the READLINE library? Typically in ../readline.
READLINE_DIR = ../readline
READLINE_SRC = $(srcdir)/$(READLINE_DIR)
@@ -358,7 +364,7 @@ CXXFLAGS = -g -O
INTERNAL_CFLAGS_BASE = \
$(CFLAGS) $(GLOBAL_CFLAGS) $(PROFILE_CFLAGS) \
$(GDB_CFLAGS) $(OPCODES_CFLAGS) $(READLINE_CFLAGS) \
- $(BFD_CFLAGS) $(INCLUDE_CFLAGS) \
+ $(BFD_CFLAGS) $(INCLUDE_CFLAGS) $(LIBDECNUMBER_CFLAGS) \
$(INTL_CFLAGS) $(ENABLE_CFLAGS) $(INTERNAL_CPPFLAGS)
INTERNAL_WARN_CFLAGS = $(INTERNAL_CFLAGS_BASE) $(GDB_WARN_CFLAGS)
INTERNAL_CFLAGS = $(INTERNAL_WARN_CFLAGS) $(GDB_WERROR_CFLAGS)
@@ -381,10 +387,10 @@ INTERNAL_LDFLAGS = $(CFLAGS) $(GLOBAL_CF
# LIBIBERTY appears twice on purpose.
# If you have the Cygnus libraries installed,
# you can use 'CLIBS=$(INSTALLED_LIBS)' 'CDEPS='
-INSTALLED_LIBS=-lbfd -lreadline -lopcodes -liberty \
+INSTALLED_LIBS=-lbfd -lreadline -lopcodes -liberty -ldecnumber \
$(XM_CLIBS) $(TM_CLIBS) $(NAT_CLIBS) $(GDBTKLIBS) @LIBS@ \
-lintl -liberty
-CLIBS = $(SIM) $(READLINE) $(OPCODES) $(BFD) $(INTL) $(LIBIBERTY) \
+CLIBS = $(SIM) $(READLINE) $(OPCODES) $(BFD) $(INTL) $(LIBIBERTY) $(LIBDECNUMBER) \
$(XM_CLIBS) $(TM_CLIBS) $(NAT_CLIBS) $(GDBTKLIBS) @LIBS@ \
$(LIBICONV) $(LIBEXPAT) \
$(LIBIBERTY) $(WIN32LIBS)
@@ -624,6 +630,10 @@ safe_ctype_h = $(INCLUDE_DIR)/safe-ctyp
hashtab_h = $(INCLUDE_DIR)/hashtab.h
filenames_h = $(INCLUDE_DIR)/filenames.h
+decimal128_h = $(LIBDECNUMBER_DIR)/dpd/decimal128.h
+decimal64_h = $(LIBDECNUMBER_DIR)/dpd/decimal64.h
+decimal32_h = $(LIBDECNUMBER_DIR)/dpd/decimal32.h
+
#
# $BUILD/ headers
#
@@ -690,6 +700,7 @@ dictionary_h = dictionary.h
disasm_h = disasm.h
doublest_h = doublest.h $(floatformat_h)
dummy_frame_h = dummy-frame.h
+dfp_h = dfp.h
dwarf2expr_h = dwarf2expr.h
dwarf2_frame_h = dwarf2-frame.h
dwarf2loc_h = dwarf2loc.h
@@ -938,7 +949,7 @@ COMMON_OBS = $(DEPFILES) $(CONFIG_OBS) $
auxv.o \
bfd-target.o \
blockframe.o breakpoint.o findvar.o regcache.o \
- charset.o disasm.o dummy-frame.o \
+ charset.o disasm.o dummy-frame.o dfp.o \
source.o value.o eval.o valops.o valarith.o valprint.o printcmd.o \
block.o symtab.o symfile.o symmisc.o linespec.o dictionary.o \
infcall.o \
@@ -1871,7 +1882,7 @@ buildsym.o: buildsym.c $(defs_h) $(bfd_h
$(cp_support_h) $(dictionary_h) $(buildsym_h) $(stabsread_h)
c-exp.o: c-exp.c $(defs_h) $(gdb_string_h) $(expression_h) $(value_h) \
$(parser_defs_h) $(language_h) $(c_lang_h) $(bfd_h) $(symfile_h) \
- $(objfiles_h) $(charset_h) $(block_h) $(cp_support_h)
+ $(objfiles_h) $(charset_h) $(block_h) $(cp_support_h) $(dfp_h)
charset.o: charset.c $(defs_h) $(charset_h) $(gdbcmd_h) $(gdb_assert_h) \
$(gdb_string_h)
c-lang.o: c-lang.c $(defs_h) $(symtab_h) $(gdbtypes_h) $(expression_h) \
@@ -1954,6 +1965,7 @@ dsrec.o: dsrec.c $(defs_h) $(serial_h) $
dummy-frame.o: dummy-frame.c $(defs_h) $(dummy_frame_h) $(regcache_h) \
$(frame_h) $(inferior_h) $(gdb_assert_h) $(frame_unwind_h) \
$(command_h) $(gdbcmd_h) $(gdb_string_h)
+dfp.o: dfp.c $(defs_h) $(dfp_h) $(decimal128_h) $(decimal64_h) $(decimal32_h)
dwarf2expr.o: dwarf2expr.c $(defs_h) $(symtab_h) $(gdbtypes_h) $(value_h) \
$(gdbcore_h) $(elf_dwarf2_h) $(dwarf2expr_h)
dwarf2-frame.o: dwarf2-frame.c $(defs_h) $(dwarf2expr_h) $(elf_dwarf2_h) \
@@ -2815,11 +2827,11 @@ valops.o: valops.c $(defs_h) $(symtab_h)
valprint.o: valprint.c $(defs_h) $(gdb_string_h) $(symtab_h) $(gdbtypes_h) \
$(value_h) $(gdbcore_h) $(gdbcmd_h) $(target_h) $(language_h) \
$(annotate_h) $(valprint_h) $(floatformat_h) $(doublest_h) \
- $(exceptions_h)
+ $(exceptions_h) $(dfp_h)
value.o: value.c $(defs_h) $(gdb_string_h) $(symtab_h) $(gdbtypes_h) \
$(value_h) $(gdbcore_h) $(command_h) $(gdbcmd_h) $(target_h) \
$(language_h) $(demangle_h) $(doublest_h) \
- $(gdb_assert_h) $(regcache_h) $(block_h)
+ $(gdb_assert_h) $(regcache_h) $(block_h) $(dfp_h)
varobj.o: varobj.c $(defs_h) $(exceptions_h) $(value_h) $(expression_h) \
$(frame_h) $(language_h) $(wrapper_h) $(gdbcmd_h) $(block_h) \
$(gdb_assert_h) $(gdb_string_h) $(varobj_h) $(vec_h)
Index: gdb/dfp.c
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/dfp.c 2007-10-15 14:35:06.000000000 -0200
@@ -0,0 +1,113 @@
+/* Decimal floating point support for GDB.
+
+ Copyright 2007 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 3 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, see <http://www.gnu.org/licenses/>. */
+
+#include "defs.h"
+
+/* The order of the following headers is important for making sure
+ decNumber structure is large enough to hold decimal128 digits. */
+
+#include "dpd/decimal128.h"
+#include "dpd/decimal64.h"
+#include "dpd/decimal32.h"
+
+/* In GDB, we are using an array of gdb_byte to represent decimal values.
+ They are stored in host byte order. This routine does the conversion if
+ the target byte order is different. */
+static void
+match_endianness (const gdb_byte *from, int len, gdb_byte *to)
+{
+ int i;
+
+#if WORDS_BIGENDIAN
+#define OPPOSITE_BYTE_ORDER BFD_ENDIAN_LITTLE
+#else
+#define OPPOSITE_BYTE_ORDER BFD_ENDIAN_BIG
+#endif
+
+ if (gdbarch_byte_order (current_gdbarch) == OPPOSITE_BYTE_ORDER)
+ for (i = 0; i < len; i++)
+ to[i] = from[len - i - 1];
+ else
+ for (i = 0; i < len; i++)
+ to[i] = from[i];
+
+ return;
+}
+
+/* Convert decimal type to its string representation. LEN is the length
+ of the decimal type, 4 bytes for decimal32, 8 bytes for decimal64 and
+ 16 bytes for decimal128. */
+void
+decimal_to_string (const gdb_byte *decbytes, int len, char *s)
+{
+ gdb_byte dec[16];
+
+ match_endianness (decbytes, len, dec);
+ switch (len)
+ {
+ case 4:
+ decimal32ToString ((decimal32 *) dec, s);
+ break;
+ case 8:
+ decimal64ToString ((decimal64 *) dec, s);
+ break;
+ case 16:
+ decimal128ToString ((decimal128 *) dec, s);
+ break;
+ default:
+ error (_("Unknown decimal floating point type.\n"));
+ break;
+ }
+}
+
+/* Convert the string form of a decimal value to its decimal representation.
+ LEN is the length of the decimal type, 4 bytes for decimal32, 8 bytes for
+ decimal64 and 16 bytes for decimal128. */
+int
+decimal_from_string (gdb_byte *decbytes, int len, const char *string)
+{
+ decContext set;
+ gdb_byte dec[16];
+
+ switch (len)
+ {
+ case 4:
+ decContextDefault (&set, DEC_INIT_DECIMAL32);
+ set.traps = 0;
+ decimal32FromString ((decimal32 *) dec, string, &set);
+ break;
+ case 8:
+ decContextDefault (&set, DEC_INIT_DECIMAL64);
+ set.traps = 0;
+ decimal64FromString ((decimal64 *) dec, string, &set);
+ break;
+ case 16:
+ decContextDefault (&set, DEC_INIT_DECIMAL128);
+ set.traps = 0;
+ decimal128FromString ((decimal128 *) dec, string, &set);
+ break;
+ default:
+ error (_("Unknown decimal floating point type.\n"));
+ break;
+ }
+
+ match_endianness (dec, len, decbytes);
+
+ return 1;
+}
Index: gdb/dfp.h
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/dfp.h 2007-10-15 14:35:06.000000000 -0200
@@ -0,0 +1,35 @@
+/* Decimal floating point support for GDB.
+
+ Copyright 2007 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 3 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, see <http://www.gnu.org/licenses/>. */
+
+/* Decimal floating point is one of the extension to IEEE 754, which is
+ described in http://grouper.ieee.org/groups/754/revision.html and
+ http://www2.hursley.ibm.com/decimal/. It completes binary floating
+ point by representing floating point more exactly. */
+
+#ifndef DFP_H
+#define DFP_H
+
+/* When using decimal128, this is the maximum string length + 1
+ * (value comes from libdecnumber's DECIMAL128_String constant). */
+#define MAX_DECIMAL_STRING 43
+
+extern void decimal_to_string (const gdb_byte *, int, char *);
+extern int decimal_from_string (gdb_byte *, int, const char *);
+
+#endif
Index: gdb/configure.ac
===================================================================
--- gdb/configure.ac.orig 2007-10-01 14:20:02.000000000 -0300
+++ gdb/configure.ac 2007-10-15 14:35:06.000000000 -0200
@@ -496,6 +496,7 @@ AC_CHECK_TYPES(uintptr_t, [], [], [#incl
AC_C_CONST
AC_C_INLINE
+AC_C_BIGENDIAN
# ------------------------------ #
# Checks for library functions. #
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [patch 1/4] libdecnumber support
2007-10-11 15:57 ` Daniel Jacobowitz
@ 2007-10-11 16:07 ` Daniel Jacobowitz
2007-10-15 18:32 ` Thiago Jung Bauermann
2007-10-15 18:11 ` Thiago Jung Bauermann
1 sibling, 1 reply; 10+ messages in thread
From: Daniel Jacobowitz @ 2007-10-11 16:07 UTC (permalink / raw)
To: Thiago Jung Bauermann, gdb-patches
Oh, I also noticed some GPLv2 copyright headers; please update to
GPLv3.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [patch 1/4] libdecnumber support
2007-09-20 21:57 ` [patch 1/4] libdecnumber support Thiago Jung Bauermann
@ 2007-10-11 15:57 ` Daniel Jacobowitz
2007-10-11 16:07 ` Daniel Jacobowitz
2007-10-15 18:11 ` Thiago Jung Bauermann
0 siblings, 2 replies; 10+ messages in thread
From: Daniel Jacobowitz @ 2007-10-11 15:57 UTC (permalink / raw)
To: Thiago Jung Bauermann; +Cc: gdb-patches
I'll review all these now. Before you check them in, though, give
me a chance to get libdecnumber merged into the src repository. It
shouldn't take long.
On Thu, Sep 20, 2007 at 06:54:11PM -0300, Thiago Jung Bauermann wrote:
> +#include <ctype.h>
> +#include <endian.h>
> +#include "defs.h"
> +#include "dfp.h"
What do you need ctype.h for? That will tell me whether you really
want ctype.h or libiberty's locale-independent safe-ctype.h.
defs.h should always be first.
endian.h is not portable. You should probably use the autoconf macro
AC_C_BIGENDIAN.
> +/* Convert decimal type to its string representation. LEN is the length
> + of the decimal type, 4 bytes for decimal32, 8 bytes for decimal64 and
> + 16 bytes for decimal128. */
> +void
> +decimal_to_string (const uint8_t *decbytes, int len, char *s)
> +{
> + uint8_t *dec = (uint8_t *)malloc (len);
You didn't include anything that would give you uint8_t; use
gdb_byte. Also we never use malloc, only xmalloc. Or you could avoid
the allocation by using gdb_byte dec[16].
> + break;
> + default:
> + free (dec);
Spaces and tabs mixed up? xfree, like xmalloc.
> + error(_("Unknown decimal floating point type.\n"));
Space before parentheses please.
> +/* There is a project intended to add DFP support into GCC, described in
> + http://gcc.gnu.org/wiki/Decimal%20Floating-Point. This file is intended
> + to add DFP support into GDB. */
This comment doesn't belong here. The file is part of GDB once the
patch is applied.
> +#ifndef DFP_H
> +#define DFP_H
> +#include <string.h>
> +#include <stdio.h>
> +#include <stdint.h>
Don't include standard headers here; stick to what defs.h provides.
Just as well, since neither string.h nor stdint.h is portable.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 10+ messages in thread
* [patch 1/4] libdecnumber support
2007-09-20 21:57 [patch 0/4] decimal floating point support Thiago Jung Bauermann
@ 2007-09-20 21:57 ` Thiago Jung Bauermann
2007-10-11 15:57 ` Daniel Jacobowitz
0 siblings, 1 reply; 10+ messages in thread
From: Thiago Jung Bauermann @ 2007-09-20 21:57 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: libdecnumber-support.diff --]
[-- Type: text/plain, Size: 10523 bytes --]
This patch provides functions to interact with libdecnumber and changes
the Makefile to compile it.
2007-09-20 Wu Zhou <woodzltc@cn.ibm.com>
Thiago Jung Bauermann <bauerman@br.ibm.com>
* Makefile.in (LIBDECNUMBER_DIR, LIBDECNUMBER, LIBDECNUMBER_SRC
LIBDECNUMBER_CFLAGS): New macros for libdecnumber.
(INTERNAL_CFLAGS_BASE): Add LIBDECNUMBER_CFLAGS in.
(INSTALLED_LIBS): Add -ldecnumber in.
(CLIBS): Add LIBDECNUMBER in.
(decimal128_h, decimal64_h, decimal32_h): New macros for decimal
headers.
(dfp_h): New macros for decimal floating point.
(dfp.o): New target.
(COMMON_OBS): Add dfp.o in.
(valprint.o): Add dfp_h as dependency.
(value.o): Add dfp_h as dependency.
* dfp.h: New header file for decimal floating point support in
GDB.
* dfp.c: New source file for decimal floating point support in
GDB. Implement decimal_from_string and decimal_to_string based
on libdecnumber API.
Index: gdb/Makefile.in
===================================================================
--- gdb/Makefile.in.orig 2007-09-20 14:21:39.000000000 -0300
+++ gdb/Makefile.in 2007-09-20 14:37:29.000000000 -0300
@@ -125,6 +125,12 @@ BFD = $(BFD_DIR)/libbfd.a
BFD_SRC = $(srcdir)/$(BFD_DIR)
BFD_CFLAGS = -I$(BFD_DIR) -I$(BFD_SRC)
+# Where is the decnumber library? Typically in ../libdecnumber.
+LIBDECNUMBER_DIR = ../libdecnumber
+LIBDECNUMBER = $(LIBDECNUMBER_DIR)/libdecnumber.a
+LIBDECNUMBER_SRC = $(srcdir)/$(LIBDECNUMBER_DIR)
+LIBDECNUMBER_CFLAGS = -I$(LIBDECNUMBER_DIR) -I$(LIBDECNUMBER_SRC)
+
# Where is the READLINE library? Typically in ../readline.
READLINE_DIR = ../readline
READLINE_SRC = $(srcdir)/$(READLINE_DIR)
@@ -358,7 +364,7 @@ CXXFLAGS = -g -O
INTERNAL_CFLAGS_BASE = \
$(CFLAGS) $(GLOBAL_CFLAGS) $(PROFILE_CFLAGS) \
$(GDB_CFLAGS) $(OPCODES_CFLAGS) $(READLINE_CFLAGS) \
- $(BFD_CFLAGS) $(INCLUDE_CFLAGS) \
+ $(BFD_CFLAGS) $(INCLUDE_CFLAGS) $(LIBDECNUMBER_CFLAGS) \
$(INTL_CFLAGS) $(ENABLE_CFLAGS) $(INTERNAL_CPPFLAGS)
INTERNAL_WARN_CFLAGS = $(INTERNAL_CFLAGS_BASE) $(GDB_WARN_CFLAGS)
INTERNAL_CFLAGS = $(INTERNAL_WARN_CFLAGS) $(GDB_WERROR_CFLAGS)
@@ -381,10 +387,10 @@ INTERNAL_LDFLAGS = $(CFLAGS) $(GLOBAL_CF
# LIBIBERTY appears twice on purpose.
# If you have the Cygnus libraries installed,
# you can use 'CLIBS=$(INSTALLED_LIBS)' 'CDEPS='
-INSTALLED_LIBS=-lbfd -lreadline -lopcodes -liberty \
+INSTALLED_LIBS=-lbfd -lreadline -lopcodes -liberty -ldecnumber \
$(XM_CLIBS) $(TM_CLIBS) $(NAT_CLIBS) $(GDBTKLIBS) @LIBS@ \
-lintl -liberty
-CLIBS = $(SIM) $(READLINE) $(OPCODES) $(BFD) $(INTL) $(LIBIBERTY) \
+CLIBS = $(SIM) $(READLINE) $(OPCODES) $(BFD) $(INTL) $(LIBIBERTY) $(LIBDECNUMBER) \
$(XM_CLIBS) $(TM_CLIBS) $(NAT_CLIBS) $(GDBTKLIBS) @LIBS@ \
$(LIBICONV) $(LIBEXPAT) \
$(LIBIBERTY) $(WIN32LIBS)
@@ -624,6 +630,10 @@ safe_ctype_h = $(INCLUDE_DIR)/safe-ctyp
hashtab_h = $(INCLUDE_DIR)/hashtab.h
filenames_h = $(INCLUDE_DIR)/filenames.h
+decimal128_h = $(LIBDECNUMBER_DIR)/dpd/decimal128.h
+decimal64_h = $(LIBDECNUMBER_DIR)/dpd/decimal64.h
+decimal32_h = $(LIBDECNUMBER_DIR)/dpd/decimal32.h
+
#
# $BUILD/ headers
#
@@ -690,6 +700,7 @@ dictionary_h = dictionary.h
disasm_h = disasm.h
doublest_h = doublest.h $(floatformat_h)
dummy_frame_h = dummy-frame.h
+dfp_h = dfp.h
dwarf2expr_h = dwarf2expr.h
dwarf2_frame_h = dwarf2-frame.h
dwarf2loc_h = dwarf2loc.h
@@ -938,7 +949,7 @@ COMMON_OBS = $(DEPFILES) $(CONFIG_OBS) $
auxv.o \
bfd-target.o \
blockframe.o breakpoint.o findvar.o regcache.o \
- charset.o disasm.o dummy-frame.o \
+ charset.o disasm.o dummy-frame.o dfp.o \
source.o value.o eval.o valops.o valarith.o valprint.o printcmd.o \
block.o symtab.o symfile.o symmisc.o linespec.o dictionary.o \
infcall.o \
@@ -1953,6 +1964,7 @@ dsrec.o: dsrec.c $(defs_h) $(serial_h) $
dummy-frame.o: dummy-frame.c $(defs_h) $(dummy_frame_h) $(regcache_h) \
$(frame_h) $(inferior_h) $(gdb_assert_h) $(frame_unwind_h) \
$(command_h) $(gdbcmd_h) $(gdb_string_h)
+dfp.o: dfp.c $(defs_h) $(dfp_h) $(decimal128_h) $(decimal64_h) $(decimal32_h)
dwarf2expr.o: dwarf2expr.c $(defs_h) $(symtab_h) $(gdbtypes_h) $(value_h) \
$(gdbcore_h) $(elf_dwarf2_h) $(dwarf2expr_h)
dwarf2-frame.o: dwarf2-frame.c $(defs_h) $(dwarf2expr_h) $(elf_dwarf2_h) \
@@ -2814,11 +2826,11 @@ valops.o: valops.c $(defs_h) $(symtab_h)
valprint.o: valprint.c $(defs_h) $(gdb_string_h) $(symtab_h) $(gdbtypes_h) \
$(value_h) $(gdbcore_h) $(gdbcmd_h) $(target_h) $(language_h) \
$(annotate_h) $(valprint_h) $(floatformat_h) $(doublest_h) \
- $(exceptions_h)
+ $(exceptions_h) $(dfp_h)
value.o: value.c $(defs_h) $(gdb_string_h) $(symtab_h) $(gdbtypes_h) \
$(value_h) $(gdbcore_h) $(command_h) $(gdbcmd_h) $(target_h) \
$(language_h) $(demangle_h) $(doublest_h) \
- $(gdb_assert_h) $(regcache_h) $(block_h)
+ $(gdb_assert_h) $(regcache_h) $(block_h) $(dfp_h)
varobj.o: varobj.c $(defs_h) $(exceptions_h) $(value_h) $(expression_h) \
$(frame_h) $(language_h) $(wrapper_h) $(gdbcmd_h) $(block_h) \
$(gdb_assert_h) $(gdb_string_h) $(varobj_h) $(vec_h)
Index: gdb/dfp.c
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/dfp.c 2007-09-20 14:21:45.000000000 -0300
@@ -0,0 +1,123 @@
+/* Decimal floating point support for GDB.
+
+ Copyright 2007 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 <ctype.h>
+#include <endian.h>
+#include "defs.h"
+#include "dfp.h"
+
+/* The order of the following headers is important for making sure
+ decNumber structure is large enough to hold decimal128 digits. */
+
+#include "dpd/decimal128.h"
+#include "dpd/decimal64.h"
+#include "dpd/decimal32.h"
+
+/* In GDB, we are using an array of gdb_byte to represent decimal values.
+ They are stored in host byte order. This routine does the conversion if
+ the target byte order is different. */
+static void
+match_endianness (const gdb_byte *from, int len, gdb_byte *to)
+{
+ int i;
+
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+#define OPPOSITE_BYTE_ORDER BFD_ENDIAN_BIG
+#else
+#define OPPOSITE_BYTE_ORDER BFD_ENDIAN_LITTLE
+#endif
+
+ if (gdbarch_byte_order (current_gdbarch) == OPPOSITE_BYTE_ORDER)
+ for (i = 0; i < len; i++)
+ to[i] = from[len - i - 1];
+ else
+ for (i = 0; i < len; i++)
+ to[i] = from[i];
+
+ return;
+}
+
+/* Convert decimal type to its string representation. LEN is the length
+ of the decimal type, 4 bytes for decimal32, 8 bytes for decimal64 and
+ 16 bytes for decimal128. */
+void
+decimal_to_string (const uint8_t *decbytes, int len, char *s)
+{
+ uint8_t *dec = (uint8_t *)malloc (len);
+
+ match_endianness (decbytes, len, dec);
+ switch (len)
+ {
+ case 4:
+ decimal32ToString ((decimal32 *) dec, s);
+ break;
+ case 8:
+ decimal64ToString ((decimal64 *) dec, s);
+ break;
+ case 16:
+ decimal128ToString ((decimal128 *) dec, s);
+ break;
+ default:
+ free (dec);
+ error(_("Unknown decimal floating point type.\n"));
+ break;
+ }
+
+ free (dec);
+}
+
+/* Convert the string form of a decimal value to its decimal representation.
+ LEN is the length of the decimal type, 4 bytes for decimal32, 8 bytes for
+ decimal64 and 16 bytes for decimal128. */
+int
+decimal_from_string (uint8_t *decbytes, int len, const char *string)
+{
+ decContext set;
+ uint8_t *dec = (uint8_t *)malloc (len);
+
+ switch (len)
+ {
+ case 4:
+ decContextDefault (&set, DEC_INIT_DECIMAL32);
+ set.traps = 0;
+ decimal32FromString ((decimal32 *) dec, string, &set);
+ break;
+ case 8:
+ decContextDefault (&set, DEC_INIT_DECIMAL64);
+ set.traps = 0;
+ decimal64FromString ((decimal64 *) dec, string, &set);
+ break;
+ case 16:
+ decContextDefault (&set, DEC_INIT_DECIMAL128);
+ set.traps = 0;
+ decimal128FromString ((decimal128 *) dec, string, &set);
+ break;
+ default:
+ free (dec);
+ error(_("Unknown decimal floating point type.\n"));
+ break;
+ }
+
+ match_endianness (dec, len, decbytes);
+ free (dec);
+
+ return 1;
+}
Index: gdb/dfp.h
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ gdb/dfp.h 2007-09-20 14:21:45.000000000 -0300
@@ -0,0 +1,40 @@
+/* Decimal floating point support for GDB.
+
+ Copyright 2007 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. */
+
+/* Decimal floating point is one of the extension to IEEE 754, which is
+ described in http://grouper.ieee.org/groups/754/revision.html and
+ http://www2.hursley.ibm.com/decimal/. It completes binary floating
+ point by representing floating point more exactly. */
+
+/* There is a project intended to add DFP support into GCC, described in
+ http://gcc.gnu.org/wiki/Decimal%20Floating-Point. This file is intended
+ to add DFP support into GDB. */
+
+#ifndef DFP_H
+#define DFP_H
+#include <string.h>
+#include <stdio.h>
+#include <stdint.h>
+
+extern void decimal_to_string (const uint8_t *, int, char *);
+extern int decimal_from_string (uint8_t *, int, const char *);
+
+#endif
--
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2007-10-25 18:22 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <20070323030737.475073862@br.ibm.com>
2007-03-27 16:30 ` [patch 3/4] decimal floating point support for expressions Thiago Jung Bauermann
2007-03-27 16:30 ` [patch 1/4] libdecnumber support Thiago Jung Bauermann
2007-03-27 16:30 ` [patch 4/4] decimal floating point testcases Thiago Jung Bauermann
2007-09-20 21:57 [patch 0/4] decimal floating point support Thiago Jung Bauermann
2007-09-20 21:57 ` [patch 1/4] libdecnumber support Thiago Jung Bauermann
2007-10-11 15:57 ` Daniel Jacobowitz
2007-10-11 16:07 ` Daniel Jacobowitz
2007-10-15 18:32 ` Thiago Jung Bauermann
2007-10-15 18:11 ` Thiago Jung Bauermann
2007-10-15 18:38 ` Daniel Jacobowitz
2007-10-25 18:23 ` Thiago Jung Bauermann
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox