From: Andrew Burgess Date: Thu, 13 Dec 2018 18:25:25 +0000 (+0000) Subject: gdb: Allow struct fields named double X-Git-Url: http://drtracing.org/?a=commitdiff_plain;h=6f0ffe50c8f395ccb45ba9e801906b49d18c1af9;p=deliverable%2Fbinutils-gdb.git gdb: Allow struct fields named double The 64-bit RISC-V target currently models the floating point registers as having the following type: union riscv_double { builtin_type_ieee_single float; builtin_type_ieee_double double; } Notice the choice of names for the fields of this struct, possibly not ideal choices, as these are not valid field names in C. However, this type is only ever defined within GDB (or in the target description), and no restriction seems to exist on the field names in that case. The problem though is that currently: (gdb) info registers $ft0 ft0 {float = 0, double = 0} (raw 0x0000000000000000) (gdb) p $ft0.float $1 = 0 (gdb) p $ft0.double A syntax error in expression, near `double'. We can access the 'float' field, but not the 'double' field. This is because the string 'double' is handled differently to the string 'float' in c-exp.y. In both cases the string '$ft0' is parsed as a VARIABLE expression. In the 'float' case, the string 'float' becomes a generic NAME token in 'lex_one_token', which then allows the rule "exp '.' name" to match and the field name lookup to occur. The 'double' case is different. In order to allow parsing of the type string 'long double', the 'double' string becomes the token DOUBLE_KEYWORD. At this point there's no rule to match "exp '.' DOUBLE_KEYWORD", so we can never lookup the field named 'double'. We could rename the fields for RISC-V, and maybe that would be the best solution. However, its not hard to allow for fields named 'double', which is what this patch does. A new case is added to the 'field_name' rule to match the DOUBLE_KEYWORD, and create a suitable 'struct stoken'. With this done the "exp '.' field_name" pattern can now match, and we can lookup the double field. With this patch in place I now see this behaviour: (gdb) info registers $ft0 ft0 {float = 0, double = 0} (raw 0x0000000000000000) (gdb) p $ft0.float $1 = 0 (gdb) p $ft0.double $2 = 0 I've gone ahead and handled INT_KEYWORD, LONG, SHORT, SIGNED_KEYWORD, and UNSIGNED as well within field_name. I've added a new test for this functionality. This change was tested on x86-64 GNU/Linux with no regressions. gdb/ChangeLog: * c-exp.y (field_name): Allow DOUBLE_KEYWORD, INT_KEYWORD, LONG, SHORT, SIGNED_KEYWORD, and UNSIGNED tokens to act as a field names. (typename_stoken): New function. gdb/testsuite/ChangeLog: * gdb.dwarf2/dw2-unusual-field-names.c: New file. * gdb.dwarf2/dw2-unusual-field-names.exp: New file. --- diff --git a/gdb/ChangeLog b/gdb/ChangeLog index 5c5699e757..c37e281740 100644 --- a/gdb/ChangeLog +++ b/gdb/ChangeLog @@ -1,3 +1,10 @@ +2018-12-24 Andrew Burgess + + * c-exp.y (field_name): Allow DOUBLE_KEYWORD, INT_KEYWORD, LONG, + SHORT, SIGNED_KEYWORD, and UNSIGNED tokens to act as a field + names. + (typename_stoken): New function. + 2018-12-24 Andrew Burgess * c-exp.y (field_name): New %token, and new rule. diff --git a/gdb/c-exp.y b/gdb/c-exp.y index 98a0df3e19..996152d21f 100644 --- a/gdb/c-exp.y +++ b/gdb/c-exp.y @@ -113,6 +113,7 @@ static int type_aggregate_p (struct type *); static int parse_number (struct parser_state *par_state, const char *, int, int, YYSTYPE *); static struct stoken operator_stoken (const char *); +static struct stoken typename_stoken (const char *); static void check_parameter_typelist (VEC (type_ptr) *); static void write_destructor_name (struct parser_state *par_state, struct stoken); @@ -1645,9 +1646,21 @@ oper: OPERATOR NEW } ; - +/* This rule exists in order to allow some tokens that would not normally + match the 'name' rule to appear as fields within a struct. The example + that initially motivated this was the RISC-V target which models the + floating point registers as a union with fields called 'float' and + 'double'. The 'float' string becomes a TYPENAME token and can appear + anywhere a 'name' can, however 'double' is its own token, + DOUBLE_KEYWORD, and doesn't match the 'name' rule.*/ field_name : name + | DOUBLE_KEYWORD { $$ = typename_stoken ("double"); } + | INT_KEYWORD { $$ = typename_stoken ("int"); } + | LONG { $$ = typename_stoken ("long"); } + | SHORT { $$ = typename_stoken ("short"); } + | SIGNED_KEYWORD { $$ = typename_stoken ("signed"); } + | UNSIGNED { $$ = typename_stoken ("unsigned"); } ; name : NAME { $$ = $1.stoken; } @@ -1720,6 +1733,16 @@ operator_stoken (const char *op) return st; }; +/* Returns a stoken of the type named TYPE. */ + +static struct stoken +typename_stoken (const char *type) +{ + struct stoken st = { type, 0 }; + st.length = strlen (type); + return st; +}; + /* Return true if the type is aggregate-like. */ static int diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog index fda5318990..47572ba14a 100644 --- a/gdb/testsuite/ChangeLog +++ b/gdb/testsuite/ChangeLog @@ -1,3 +1,8 @@ +2018-12-24 Andrew Burgess + + * gdb.dwarf2/dw2-unusual-field-names.c: New file. + * gdb.dwarf2/dw2-unusual-field-names.exp: New file. + 2018-12-24 Philippe Waroquiers * gdb.ada/bp_fun_addr/a.adb (a): Rename to bp_fun_addr. diff --git a/gdb/testsuite/gdb.dwarf2/dw2-unusual-field-names.c b/gdb/testsuite/gdb.dwarf2/dw2-unusual-field-names.c new file mode 100644 index 0000000000..091e475fb3 --- /dev/null +++ b/gdb/testsuite/gdb.dwarf2/dw2-unusual-field-names.c @@ -0,0 +1,32 @@ +/* This testcase is part of GDB, the GNU debugger. + + Copyright 2018 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 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 . */ + +struct foo +{ + int field; +}; + +struct foo obj = { 0 }; + +struct foo *ptr = &obj; + +int +main (void) +{ + return obj.field; +} + diff --git a/gdb/testsuite/gdb.dwarf2/dw2-unusual-field-names.exp b/gdb/testsuite/gdb.dwarf2/dw2-unusual-field-names.exp new file mode 100644 index 0000000000..6443acee24 --- /dev/null +++ b/gdb/testsuite/gdb.dwarf2/dw2-unusual-field-names.exp @@ -0,0 +1,132 @@ +# Copyright 2018 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 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 . + +# Test that GDB can support accesing fields of a structure if the +# fields have non-standard names. Specifically, if the names are +# reserved C type names like 'double', 'float', 'int', etc. +# +# We don't expect to that such structures should be seen in real C +# code, but in some cases GDB will generate artificial structures, and +# in some cases, these type names are the obvious choice for field +# names. +# +# One specific example is RISC-V, the 64-bit floating point registers +# are represented as a structure with the field names 'float' and +# 'double'. + +load_lib dwarf.exp + +# This test can only be run on targets which support DWARF-2 and use +# gas. +if {![dwarf2_support]} { + return 0 +} + +standard_testfile dw2-unusual-field-names.c dw2-unusual-field-names.S +set asm_file [standard_output_file $srcfile2] + +# We need to know the size of integer and address types in order to +# write some of the debugging info we'd like to generate. +# +# For that, we ask GDB by debugging our test program. Any program +# would do, but since we already have one specifically for this +# testcase, might as well use that. +if [prepare_for_testing "failed to prepare" ${testfile} ${srcfile}] { + return -1 +} +set int_size [get_sizeof "int" -1] + +# Rebuild the test binary with the single field within the structure +# renamed to FIELD_NAME, then test that we can access the field +# through '.' and through '->'. +proc run_test { field_name } { + global asm_file testfile srcfile subdir srcdir + global int_size + + Dwarf::assemble $asm_file { + global srcdir subdir srcfile + global field_name int_size + + cu {} { + DW_TAG_compile_unit { + {DW_AT_language @DW_LANG_C} + {DW_AT_name dw2-unusual-field-names.c} + {DW_AT_comp_dir /tmp} + } { + declare_labels itype ptype stype + + itype: DW_TAG_base_type { + {DW_AT_byte_size $int_size DW_FORM_sdata} + {DW_AT_encoding @DW_ATE_signed} + {DW_AT_name int} + } + + stype: DW_TAG_structure_type { + {DW_AT_name "foo"} + {DW_AT_byte_size $int_size DW_FORM_sdata} + } { + member { + {name $field_name} + {type :$itype} + {data_member_location 0 data1} + } + } + + ptype: DW_TAG_pointer_type { + {DW_AT_type :$stype} + } + + DW_TAG_variable { + {DW_AT_name obj} + {DW_AT_type :$stype} + {DW_AT_location { + DW_OP_addr [gdb_target_symbol obj] + } SPECIAL_expr} + {external 1 flag} + } + + DW_TAG_variable { + {DW_AT_name ptr} + {DW_AT_type :$ptype} + {DW_AT_location { + DW_OP_addr [gdb_target_symbol ptr] + } SPECIAL_expr} + {external 1 flag} + } + } + } + } + + if { [prepare_for_testing "failed to prepare" ${testfile} \ + [list $srcfile $asm_file] {nodebug}] } { + return -1 + } + + if ![runto_main] { + return -1 + } + + gdb_test "p obj.$field_name" " = 0" \ + "access a field named '$field_name' directly" + + gdb_test "p ptr->$field_name" " = 0" \ + "access a field named '$field_name' through a pointer" + + gdb_exit +} + +foreach field_name { double float char byte long int short unsigned signed } { + run_test $field_name +}