This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
RE: [PATCH 4/6] Dwarf: Fortran, support DW_TAG_entry_point.
- From: "Wiederhake, Tim" <tim dot wiederhake at intel dot com>
- To: Yao Qi <qiyaoltc at gmail dot com>
- Cc: "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>
- Date: Mon, 7 Aug 2017 08:46:04 +0000
- Subject: RE: [PATCH 4/6] Dwarf: Fortran, support DW_TAG_entry_point.
- Authentication-results: sourceware.org; auth=none
- Dlp-product: dlpe-windows
- Dlp-reaction: no-action
- Dlp-version: 10.0.102.7
- References: <1500629040-12972-1-git-send-email-tim.wiederhake@intel.com> <1500629040-12972-5-git-send-email-tim.wiederhake@intel.com> <86tw1q2jt3.fsf@gmail.com> <9676A094AF46E14E8265E7A3F4CCE9AF5AC20F28@irsmsx105.ger.corp.intel.com>
See below.
> -----Original Message-----
> From: gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] On Behalf Of Wiederhake, Tim
> Sent: Friday, August 4, 2017 1:04 PM
> To: Yao Qi <qiyaoltc@gmail.com>
> Cc: gdb-patches@sourceware.org
> Subject: RE: [PATCH 4/6] Dwarf: Fortran, support DW_TAG_entry_point.
>
> Hi Yao!
>
> > -----Original Message-----
> > From: gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> > owner@sourceware.org] On Behalf Of Yao Qi
> > Sent: Wednesday, August 2, 2017 3:15 PM
> > To: Wiederhake, Tim <tim.wiederhake@intel.com>
> > Cc: gdb-patches@sourceware.org; Bernhard Heckel
> > <bernhard.heckel@intel.com>
> > Subject: Re: [PATCH 4/6] Dwarf: Fortran, support DW_TAG_entry_point.
> >
> > Tim Wiederhake <tim.wiederhake@intel.com> writes:
> >
> > > 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.
> >
> > Looks gfortran doesn't generate DW_TAG_entry_point now. Could you
> > confirm that this patch fix PR 8043 and PR 9279?
>
> Tested and confirmed.
>
> > >
> > > xxxx-yy-zz Bernhard Heckel <bernhard.heckel@intel.com>
> > > Tim Wiederhake <tim.wiederhake@intel.com>
> > >
> > > gdb/ChangeLog:
> >
> > PR fortran/8043
> > PR fortran/9279
> > > * gdb/dwarf2read.c (add_partial_symbol): Handle DW_TAG_entry_point.
> > > (add_partial_entry_point): New function.
> > > (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 file.
> > > * gdb.fortran/entry_point.exp: New file.
> > >
> > >
> > > ---
> > > gdb/dwarf2read.c | 100
> > +++++++++++++++++++++++++++++-
> > > gdb/testsuite/gdb.fortran/entry_point.exp | 70 +++++++++++++++++++++
> > > gdb/testsuite/gdb.fortran/entry_point.f90 | 48 ++++++++++++++
> > > 3 files changed, 217 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 6d38d70..c3c01a7 100644
> > > --- a/gdb/dwarf2read.c
> > > +++ b/gdb/dwarf2read.c
> > > @@ -1556,6 +1556,10 @@ 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);
> > > @@ -7132,6 +7136,32 @@ 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)
> > > + {
> > > + addr = gdbarch_adjust_dwarf2_addr (gdbarch, pdi->lowpc +
> > baseaddr);
> > > + /* 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 != NULL,
> > > + VAR_DOMAIN, LOC_BLOCK,
> > > + &objfile->global_psymbols,
> > > + addr, cu->language, objfile);
> > > + else
> > > + add_psymbol_to_list (actual_name, strlen (actual_name),
> > > + built_actual_name != NULL,
> > > + VAR_DOMAIN, LOC_BLOCK,
> > > + &objfile->static_psymbols,
> > > + addr, cu->language, objfile);
> >
> > Nit, can you do
> >
> > add_psymbol_to_list (actual_name, strlen (actual_name),
> > built_actual_name != NULL,
> > VAR_DOMAIN, LOC_BLOCK,
> > pdi->die_parent->is_external ?
> > &objfile->global_psymbols : &objfile-
> > >static_psymbols,
> > addr, cu->language, objfile);
> >
> > or
> >
> > auto psymbols = pdi->die_parent->is_external ?
> > &objfile->global_psymbols : &objfile-
> > >static_psymbols;
> >
> > add_psymbol_to_list (actual_name, strlen (actual_name),
> > built_actual_name != NULL,
> > VAR_DOMAIN, LOC_BLOCK,
> > psymbols,
> > addr, cu->language, objfile);
>
> Changed locally.
>
> > > + }
> > > + break;
> > > + }
> > > case DW_TAG_subprogram:
> > > addr = gdbarch_adjust_dwarf2_addr (gdbarch, pdi->lowpc +
> > baseaddr);
> > > if (pdi->is_external || cu->language == language_ada)
> > > @@ -7332,6 +7362,18 @@ 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 == NULL)
> > > + complaint (&symfile_complaints,
> > > + _("DW_TAG_entry_point have to have a name"));
> > > + else
> > > + add_partial_symbol (pdi, cu);
> > > +}
> > > +
> > > /* Read a partial die corresponding to a subprogram and create a
> > partial
> > > symbol for that subprogram. When the CU language allows it, this
> > > routine also defines a partial symbol for each nested subprogram
> > > @@ -7402,6 +7444,16 @@ add_partial_subprogram (struct partial_die_info
> > *pdi,
> > > pdi = pdi->die_sibling;
> > > }
> > > }
> > > + else if (cu->language == language_fortran)
> > > + {
> > > + pdi = pdi->die_child;
> > > + while (pdi != NULL)
> > > + {
> > > + if (pdi->tag == DW_TAG_entry_point)
> > > + add_partial_entry_point (pdi, lowpc, highpc, set_addrmap, cu);
> >
> > In Fortran, is it possible to define "nested functions" or "contained
> > routines" within an entry point of a subprogram? like,
> >
> > subroutine bar
> >
> > entry foo
> >
> > contains
> > subroutine foo_sub
> > entry foo_sub_entry
> > end subroutine foo_sub
> > entry bar
> >
> > end subroutine
>
> Will test and report.
$ gfortran -g -o test test.f90
test.f90:5:15:
entry foo_sub_entry
1
Error: ENTRY statement at (1) cannot appear in a contained procedure
Regards,
Tim
> > can gdb correctly handle these nested functions and nested entry points?
> > My point is that it is better that add_partial_subprogram handles both
> > DW_TAG_subprogram and DW_TAG_entry_point first, and then, recursively
> > call itself to handle DIE children.
> >
> > add_partial_subprogram ()
> > {
> > if (pdi->tag == DW_TAG_subprogram)
> > {}
> > else if (pdi->tag == DW_TAG_entry_point)
> > {}
> >
> > if (! pdi->has_children)
> > return;
> >
> > if (cu->language == language_ada || cu->language == language_fortran)
> > {
> > pdi = pdi->die_child;
> > while (pdi != NULL)
> > {
> > fixup_partial_die (pdi, cu);
> > if (pdi->tag == DW_TAG_subprogram
> > || pdi->tag == DW_TAG_lexical_block
> > || pdi->tag == DW_TAG_entry_point)
> > add_partial_subprogram (pdi, lowpc, highpc, set_addrmap, cu);
> > pdi = pdi->die_sibling;
> > }
> > }
> > }
> >
> > > + pdi = pdi->die_sibling;
> > > + }
> > > + }
> > > }
> > >
> > > /* Read a partial die corresponding to an enumeration type. */
> > > @@ -8508,6 +8560,7 @@ 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);
> > > @@ -12499,6 +12552,27 @@ 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 its enveloping subprogram and get the lowpc from DWARF.
> > */
> > > + ret = dwarf2_get_pc_bounds (die->parent, lowpc, highpc, cu,
> pst);
> > > + if (ret == PC_BOUNDS_NOT_PRESENT || ret == PC_BOUNDS_INVALID)
> > > + return ret;
> > > +
> > > + attr = dwarf2_attr (die, DW_AT_low_pc, cu);
> > > + if (!attr)
> > > + {
> > > + complaint (&symfile_complaints,
> > > + _("DW_TAG_entry_point is missing DW_AT_low_pc"));
> > > + return PC_BOUNDS_INVALID;
> > > + }
> > > + low = attr_value_as_address (attr);
> > > + *lowpc = low;
> > > +
> > > + return PC_BOUNDS_HIGH_LOW;
> >
> > Why return PC_BOUNDS_HIGH_LOW, which means both DW_AT_low_pc and
> > DW_AT_high_pc are found. However, DW_TAG_entry_point doesn't have
> > DW_AT_high_pc. The question is why do we call dwarf2_get_pc_bounds for
> > DW_TAG_entry_point. Is it because we call read_func_scope for
> > DW_TAG_entry_point?
>
> I may be misunderstanding you here. Yes, DW_TAG_entry_point doesn't
> have DW_AT_high_pc but we know that value implicitly from the surrounding
> subprogram, as explained in the comment above.
>
> > > + }
> > > +
> > > attr_high = dwarf2_attr (die, DW_AT_high_pc, cu);
> > > if (attr_high)
> > > {
> > > @@ -16029,6 +16103,7 @@ load_partial_dies (const struct
> die_reader_specs
> > *reader,
> > > && abbrev->tag != DW_TAG_constant
> > > && abbrev->tag != DW_TAG_enumerator
> > > && abbrev->tag != DW_TAG_subprogram
> > > + && abbrev->tag != DW_TAG_entry_point
> > > && abbrev->tag != DW_TAG_lexical_block
> > > && abbrev->tag != DW_TAG_variable
> > > && abbrev->tag != DW_TAG_namespace
> > > @@ -16155,6 +16230,7 @@ load_partial_dies (const struct
> die_reader_specs
> > *reader,
> > > if (load_all
> > > || abbrev->tag == DW_TAG_constant
> > > || abbrev->tag == DW_TAG_subprogram
> > > + || abbrev->tag == DW_TAG_entry_point
> >
> > Could you update the comments above this block?
>
>
> Sorry, which comments specifically?
> The comment directly above the last block states:
> DW_AT_abstract_origin refers to functions (and many things under the
> function DIE [...])"
>
> > > || abbrev->tag == DW_TAG_variable
> > > || abbrev->tag == DW_TAG_namespace
> > > || part_die->is_declaration)
> > > @@ -16197,7 +16273,9 @@ 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)))
> >
> > Likewise, update comments above. Do we need to check both
> > DW_TAG_subprogram and DW_TAG_entry_point?
>
> Changed locally.
>
> > > {
> > > nesting_level++;
> > > parent_die = last_die;
> > > @@ -19058,6 +19136,25 @@ new_symbol_full (struct die_info *die, struct
> > type *type, struct dwarf2_cu *cu,
> > > SYMBOL_ACLASS_INDEX (sym) = LOC_LABEL;
> > > 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))
> >
> > if (attr2 != NULL && DW_UNSND (attr2) != 0)
>
> Changed locally.
>
> > > + list_to_add = &global_symbols;
> > > + else
> > > + list_to_add = cu->list_in_scope;
> > > + }
> > > + break;
> > > case DW_TAG_subprogram:
> >
> >
> > Can we merge to case block for DW_TAG_subprogram and DW_TAG_subprogram?
> > They are quite similar.
>
> Merged locally.
> >
> > --
> > Yao (齐尧)
>
> Regards,
> Tim
> Intel Deutschland GmbH
> Registered Address: Am Campeon 10-12, 85579 Neubiberg, Germany
> Tel: +49 89 99 8853-0, www.intel.de
> Managing Directors: Christin Eisenschmid, Christian Lamprechter
> Chairperson of the Supervisory Board: Nicole Lau
> Registered Office: Munich
> Commercial Register: Amtsgericht Muenchen HRB 186928
Intel Deutschland GmbH
Registered Address: Am Campeon 10-12, 85579 Neubiberg, Germany
Tel: +49 89 99 8853-0, www.intel.de
Managing Directors: Christin Eisenschmid, Christian Lamprechter
Chairperson of the Supervisory Board: Nicole Lau
Registered Office: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928