Re: [PATCH v2 2/7] Dwarf: Fortran, support DW_TAG_entry_point.

[Andrew Burgess <andrew.burgess@embecosm.com>]

* Pawel Wodkowski <pwodkowski@pl.sii.eu> [2018-11-19 22:38:45 +0100]:

> From: Bernhard Heckel <bernhard.heckel@intel.com>
> 
> Fortran provides additional entry-points to an subprogram.
> Those entry-points may have only a subset of parameters
> of the original subprogram as well.
> Add support for parsing DW_TAG_entry_point's for Fortran.

As with patch #1 it's probably worth documenting which compilers you
see DW_TAG_entry_point in, along with an example of what the generated
DWARF looks like.

This means that in the future if/when others start to use this DWARF
feature we can figure out which other compilers we need to test.

> 
> 2016-06-01  Bernhard Heckel  <bernhard.heckel@intel.com>
> 
> gdb/Changelog:
> 	* gdb/dwarf2read.c (add_partial_symbol): Handle DW_TAG_entry_point.
> 	(add_partial_entry_point): New.
> 	(add_partial_subprogram): Search for entry_points.
> 	(process_die): Handle DW_TAG_entry_point.
> 	(dwarf2_get_pc_bounds): Update low pc from DWARF.
> 	(load_partial_dies): Save DW_TAG_entry_point's.
> 	(load_partial_dies): Save DW_TAG_entry_point to hash table.
> 	(load_partial_dies): Look into child's of DW_TAG_sub_program
> 	for fortran.
> 	(new_symbol_full): Process DW_TAG_entry_point.
> 	(read_type_die_1): Handle DW_TAG_entry_point.
> 
> gdb/Testsuite/Changelog:
> 	* gdb.fortran/entry_point.f90: New.
> 	* gdb.fortran/entry_point.exp: New.
> ---
>  gdb/dwarf2read.c                          | 98 ++++++++++++++++++++++++++++++-
>  gdb/testsuite/gdb.fortran/entry-point.exp | 70 ++++++++++++++++++++++
>  gdb/testsuite/gdb.fortran/entry-point.f90 | 48 +++++++++++++++
>  3 files changed, 215 insertions(+), 1 deletion(-)
>  create mode 100644 gdb/testsuite/gdb.fortran/entry-point.exp
>  create mode 100644 gdb/testsuite/gdb.fortran/entry-point.f90
> 
> diff --git a/gdb/dwarf2read.c b/gdb/dwarf2read.c
> index 89fd4ae15e80..88e57d7ab68e 100644
> --- a/gdb/dwarf2read.c
> +++ b/gdb/dwarf2read.c
> @@ -1473,6 +1473,10 @@ static void add_partial_module (struct partial_die_info *pdi, CORE_ADDR *lowpc,
>  static void add_partial_enumeration (struct partial_die_info *enum_pdi,
>  				     struct dwarf2_cu *cu);
>  
> +static void add_partial_entry_point (struct partial_die_info *pdi,
> +				     CORE_ADDR *lowpc, CORE_ADDR *highpc,
> +				     int need_pc, struct dwarf2_cu *cu);
> +
>  static void add_partial_subprogram (struct partial_die_info *pdi,
>  				    CORE_ADDR *lowpc, CORE_ADDR *highpc,
>  				    int need_pc, struct dwarf2_cu *cu);
> @@ -8876,6 +8880,32 @@ add_partial_symbol (struct partial_die_info *pdi, struct dwarf2_cu *cu)
>  
>    switch (pdi->tag)
>      {
> +    case DW_TAG_entry_point:
> +      /* Don't know any other language than fortran which is
> +	 using DW_TAG_entry_point.  */
> +      if (cu->language == language_fortran)
> +	{

I'm not sure the language check is needed here.  The description for
DW_TAG_entry_point in the DWARF spec seems pretty generic.  I don't
see how a different language would expect significantly different
results, and so, I would suggest we should just add this as general
purpose code.

> +	  addr = gdbarch_adjust_dwarf2_addr (gdbarch, pdi->lowpc + baseaddr);

When compared to DW_TAG_subprogram, this line doesn't include a '-
baseaddr'.  I think that deserves a comment explaining why.

> +	  /* DW_TAG_entry_point provides an additional entry_point to an
> +	     existing sub_program. Therefore, we inherit the "external"
> +	     attribute from the sub_program to which the entry_point
> +	     belongs to.  */
> +	  if (pdi->die_parent->is_external)
> +	    add_psymbol_to_list (actual_name, strlen (actual_name),
> +				 built_actual_name != nullptr,
> +				 VAR_DOMAIN, LOC_BLOCK,
> +				 SECT_OFF_TEXT (objfile),
> +				 &objfile->global_psymbols,
> +				 addr, cu->language, objfile);
> +	  else
> +	    add_psymbol_to_list (actual_name, strlen (actual_name),
> +				 built_actual_name != nullptr,
> +				 VAR_DOMAIN, LOC_BLOCK,
> +				 SECT_OFF_TEXT (objfile),
> +				 &objfile->static_psymbols,
> +				 addr, cu->language, objfile);

I think these two add_psymbol_to_list calls, and the two for
DW_TAG_subprogram should be factored out.  Just my opinion though...

> +	}
> +      break;
>      case DW_TAG_inlined_subroutine:
>      case DW_TAG_subprogram:
>        addr = (gdbarch_adjust_dwarf2_addr (gdbarch, pdi->lowpc + baseaddr)
> @@ -9082,6 +9112,17 @@ add_partial_module (struct partial_die_info *pdi, CORE_ADDR *lowpc,
>      scan_partial_symbols (pdi->die_child, lowpc, highpc, set_addrmap, cu);
>  }
>  
> +static void
> +add_partial_entry_point (struct partial_die_info *pdi,
> +			 CORE_ADDR *p_lowpc, CORE_ADDR *p_highpc,
> +			 int set_addrmap, struct dwarf2_cu *cu)
> +{
> +  if (pdi->name == nullptr)
> +    complaint (_("DW_TAG_entry_point must have a name"));
> +  else
> +    add_partial_symbol (pdi, cu);
> +}

This function should have a header comment, however, a quick look at
add_partial_subprogram seems to indicate that at this point GDB is not
complaining, but just ignoring weird looking DWARF.  For example this
code in add_partial_subprogram:

     /* Ignore subprogram DIEs that do not have a name, they are
        illegal.  Do not emit a complaint at this point, we will
        do so when we convert this psymtab into a symtab.  */
     if (pdi->name)
       add_partial_symbol (pdi, cu);

I think the same logic should apply for DW_TAG_entry_point maybe? So,
add_partial_entry_point is possibly redundant.

> +
>  /* Read a partial die corresponding to a subprogram or an inlined
>     subprogram and create a partial symbol for that subprogram.
>     When the CU language allows it, this routine also defines a partial
> @@ -9158,6 +9199,16 @@ add_partial_subprogram (struct partial_die_info *pdi,
>  	  pdi = pdi->die_sibling;
>  	}
>      }
> +  else if (cu->language == language_fortran)
> +    {

Like before, I'm not convinced we should tie this to Fortran...

> +      pdi = pdi->die_child;
> +      while (pdi != nullptr)
> +	{
> +	  if (pdi->tag == DW_TAG_entry_point)
> +	    add_partial_entry_point (pdi, lowpc, highpc, set_addrmap, cu);

You can probably just make the add_partial_symbol call directly here
(after a check that pdi->name is not null.

> +	  pdi = pdi->die_sibling;
> +	}
> +    }
>  }
>  
>  /* Read a partial die corresponding to an enumeration type.  */
> @@ -10596,6 +10647,7 @@ process_die (struct die_info *die, struct dwarf2_cu *cu)
>      case DW_TAG_type_unit:
>        read_type_unit_scope (die, cu);
>        break;
> +    case DW_TAG_entry_point:
>      case DW_TAG_subprogram:
>      case DW_TAG_inlined_subroutine:
>        read_func_scope (die, cu);
> @@ -14669,6 +14721,26 @@ dwarf2_get_pc_bounds (struct die_info *die, CORE_ADDR *lowpc,
>    CORE_ADDR high = 0;
>    enum pc_bounds_kind ret;
>  
> +  if (die->tag == DW_TAG_entry_point)
> +    {
> +      /* Entry_point is embedded in an subprogram. Therefore, we can use
> +	 the highpc from it's enveloping subprogram and get the
> +	 lowpc from DWARF.  */
> +      if (0 == dwarf2_get_pc_bounds (die->parent, lowpc, highpc, cu, pst))

Compare to 'PC_BOUNDS_NOT_PRESENT' not 0.

> +	return PC_BOUNDS_NOT_PRESENT;
> +
> +      attr = dwarf2_attr (die, DW_AT_low_pc, cu);
> +      if (!attr)
> +	{
> +	  complaint (_("DW_TAG_entry_point is missing DW_AT_low_pc"));
> +	  return PC_BOUNDS_NOT_PRESENT;
> +	}
> +      low = attr_value_as_address (attr);
> +      *lowpc = low;
> +
> +      return PC_BOUNDS_HIGH_LOW;
> +    }
> +
>    attr_high = dwarf2_attr (die, DW_AT_high_pc, cu);
>    if (attr_high)
>      {
> @@ -18399,6 +18471,7 @@ load_partial_dies (const struct die_reader_specs *reader,
>  	  && !is_type_tag_for_partial (abbrev->tag)
>  	  && abbrev->tag != DW_TAG_constant
>  	  && abbrev->tag != DW_TAG_enumerator
> +	  && abbrev->tag != DW_TAG_entry_point
>  	  && abbrev->tag != DW_TAG_subprogram
>  	  && abbrev->tag != DW_TAG_inlined_subroutine
>  	  && abbrev->tag != DW_TAG_lexical_block
> @@ -18529,6 +18602,7 @@ load_partial_dies (const struct die_reader_specs *reader,
>  
>        if (load_all
>  	  || abbrev->tag == DW_TAG_constant
> +	  || abbrev->tag == DW_TAG_entry_point
>  	  || abbrev->tag == DW_TAG_subprogram
>  	  || abbrev->tag == DW_TAG_variable
>  	  || abbrev->tag == DW_TAG_namespace
> @@ -18570,7 +18644,9 @@ load_partial_dies (const struct die_reader_specs *reader,
>  		      || last_die->tag == DW_TAG_union_type))
>  	      || (cu->language == language_ada
>  		  && (last_die->tag == DW_TAG_subprogram
> -		      || last_die->tag == DW_TAG_lexical_block))))
> +		      || last_die->tag == DW_TAG_lexical_block))
> +	      || (cu->language == language_fortran
> +		  && last_die->tag == DW_TAG_subprogram)))
>  	{
>  	  nesting_level++;
>  	  parent_die = last_die;
> @@ -21442,6 +21518,25 @@ new_symbol (struct die_info *die, struct type *type, struct dwarf2_cu *cu,
>  	  SYMBOL_ACLASS_INDEX (sym) = LOC_LABEL;
>  	  dw2_add_symbol_to_list (sym, cu->list_in_scope);
>  	  break;
> +	case DW_TAG_entry_point:
> +	  /* Don't know any other language than fortran which is
> +	     using DW_TAG_entry_point.  */
> +	  if (cu->language == language_fortran)
> +	    {
> +	      /* SYMBOL_BLOCK_VALUE (sym) will be filled in later by
> +		 finish_block.  */
> +	      SYMBOL_ACLASS_INDEX (sym) = LOC_BLOCK;
> +	      /* DW_TAG_entry_point provides an additional entry_point to an
> +		 existing sub_program. Therefore, we inherit the "external"
> +		 attribute from the sub_program to which the entry_point
> +		 belongs to.  */
> +	      attr2 = dwarf2_attr (die->parent, DW_AT_external, cu);
> +	      if (attr2 && (DW_UNSND (attr2) != 0))
> +		list_to_add = cu->builder->get_global_symbols ();
> +	      else
> +		list_to_add = cu->list_in_scope;
> +	    }
> +	  break;
>  	case DW_TAG_subprogram:
>  	  /* SYMBOL_BLOCK_VALUE (sym) will be filled in later by
>  	     finish_block.  */
> @@ -22129,6 +22224,7 @@ read_type_die_1 (struct die_info *die, struct dwarf2_cu *cu)
>      case DW_TAG_enumeration_type:
>        this_type = read_enumeration_type (die, cu);
>        break;
> +    case DW_TAG_entry_point:
>      case DW_TAG_subprogram:
>      case DW_TAG_subroutine_type:
>      case DW_TAG_inlined_subroutine:
> diff --git a/gdb/testsuite/gdb.fortran/entry-point.exp b/gdb/testsuite/gdb.fortran/entry-point.exp
> new file mode 100644
> index 000000000000..a1f3f2bebdb9
> --- /dev/null
> +++ b/gdb/testsuite/gdb.fortran/entry-point.exp
> @@ -0,0 +1,70 @@
> +# 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 <http://www.gnu.org/licenses/>.
> +
> +if { [skip_fortran_tests] } { return -1 }
> +
> +standard_testfile .f90
> +load_lib "fortran.exp"
> +
> +if {[prepare_for_testing $testfile.exp $testfile $srcfile {debug f90}]} {
> +    return -1
> +}
> +
> +if ![runto MAIN__] then {
> +    perror "couldn't run to breakpoint MAIN__"
> +    continue
> +}
> +
> +# Test if we can set a breakpoint via entry-point name
> +set ept_name "foo"
> +gdb_breakpoint $ept_name
> +gdb_test "continue" \
> +    [multi_line "Breakpoint $decimal, $ept_name \\(j=1, k=2, l=3, i1=4\\) at .*" \
> +                ".*"] \
> +    "continue to breakpoint: $ept_name"
> +
> +gdb_test "print j" "= 1" "print j, entered via $ept_name"
> +gdb_test "print k" "= 2" "print k, entered via $ept_name"
> +gdb_test "print l" "= 3" "print l, entered via $ept_name"
> +gdb_test "print i1" "= 4" "print i1, entered via $ept_name"
> +gdb_test "info args" \
> +  [multi_line "j = 1" \
> +              "k = 2" \
> +              "l = 3" \
> +              "i1 = 4"] \
> +   "info args, entered via $ept_name"
> +
> +# Test if we can set a breakpoint via function name
> +set ept_name "bar"
> +gdb_breakpoint $ept_name
> +gdb_test "continue" \
> +    [multi_line "Breakpoint $decimal, $ept_name \\(i=4, j=5, k=6, i1=7\\) at .*" \
> +                ".*"] \
> +    "continue to breakpoint: $ept_name"
> +
> +gdb_test "print i" "= 4" "print i, entered via $ept_name"
> +gdb_test "print j" "= 5" "print j, entered via $ept_name"
> +gdb_test "print k" "= 6" "print k, entered via $ept_name"
> +gdb_test "print i1" "= 7" "print i1, entered via $ept_name"
> +
> +set ept_name "tim"
> +gdb_breakpoint $ept_name
> +gdb_test "continue" \
> +    [multi_line "Breakpoint $decimal, $ept_name \\(j=1\\) at .*" \
> +                ".*"] \
> +    "continue to breakpoint: $ept_name"
> +
> +gdb_test "print j" "= 1" "print j, entered via $ept_name"
> +gdb_test "info args" "j = 1" "info args, entered via $ept_name"
> diff --git a/gdb/testsuite/gdb.fortran/entry-point.f90 b/gdb/testsuite/gdb.fortran/entry-point.f90
> new file mode 100644
> index 000000000000..cb663b956982
> --- /dev/null
> +++ b/gdb/testsuite/gdb.fortran/entry-point.f90
> @@ -0,0 +1,48 @@
> +! 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 <http://www.gnu.org/licenses/>.
> +
> +program TestEntryPoint
> +
> +  call foo(1,2,3,4)
> +  call bar(4,5,6,7)
> +  call tim(1)
> +
> +end program TestEntryPoint
> +
> +  subroutine bar(I,J,K,I1)
> +    INTEGER I,J,K,L,I1
> +    INTEGER A
> +    REAL    C
> +
> +    A = 0
> +    C = 0.0
> +
> +    A = I + K + I1
> +    goto 1000
> +
> +    entry foo(J,K,L,I1)
> +    A = J + K + L + I1
> +
> +200 C = J
> +    goto 1000
> +
> +    entry tim(J)
> +    goto 200
> +
> +1000 A = C + 1
> +     C = J * 1.5
> +
> +    return
> +  end subroutine
> -- 
> 2.7.4
> 


Thanks,
Andrew