[3/7] Emit macro expansion related diagnostics

From: Dodji Seketeli <dodji@redhat.com>

I am replying to several emails in one go here, as the discussion was a
bit long on this particular patch.

Jason Merrill <jason@redhat.com> writes:

> It seems that this function steps out until we hit the spelling
> location, and then we have a real source location and stop, which is
> why the loop in maybe_unwind_expanded_macro_loc needs to use
> linemap_macro_map_loc_to_exp_point as well.  Right?  In the example,
> once we hit << in SHIFTL unwinding needs to switch to the expansion
> point.
>
> It seems to me that we should encapsulate all of this in a function
> that expresses this unwinding operation, say
> "linemap_macro_map_loc_unwind_once".  So the loop would look like
>
> +  do
> +    {
> +      loc.where = where;
> +      loc.map = map;
> +
> +      VEC_safe_push (loc_t, heap, loc_vec, &loc);
> +
> +      /* WHERE is the location of a token inside the expansion of a
> +         macro.  MAP is the map holding the locations of that macro
> +         expansion.  Let's get the location of the token in the
> +         context that triggered the expansion of this macro.
> +         This is basically how we go "down" in the trace of macro
> +         expansions that led to WHERE.  */
> +      where = linemap_unwind_once (where, &map);
> +    } while (linemap_macro_expansion_map_p (map));
>

I have done this.  Though I have called the function
linemap_step_out_once to avoid confusion with
linemap_macro_map_loc_unwind_once above.

> I think that linemap_macro_loc_to_def_point when called with false
> returns the unwound spelling location of the token (and thus is used
> for LRK_SPELLING_LOCATION), or when called with true returns the
> (not-unwound) location in the expanded macro (and so I think
> LRK_MACRO_PARM_REPLACEMENT_POINT should be renamed to
> LRK_MACRO_EXPANDED_LOCATION or something similar).

>> Would LRK_MACRO_DEFINITION_LOCATION be better then?  :-)
>
> Yes.  :)

Done.  Note that I had to mechanically update two other patches that
were OKed previously.

>
> I'm having trouble thinking of a less misleading name for
> linemap_macro_map_loc_to_def_point, since the two locations serve
> completely different purposes.  Maybe something vague like
> linemap_macro_map_loc_get_stored_loc.  Or split it into two functions
> like linemap_virtual_loc_unwind_once_toward_spelling and
> linemap_virtual_loc_get_expanded_location or something like that.

>> So would a function named linemap_macro_map_loc_to_def_point that
>> returns the second location (yI) only, and a second function
>> linemap_macro_map_loc_unwind_once be less confusing to you?
>
> Yes, that sounds good.

I have done this.

> It seems to me that either we should split those functions apart in to
> two functions with clearer names, or bundle them and
> linemap_macro_loc_to_exp_point into linemap_resolve_location; if
> linemap_location_in_system_header_p used linemap_resolve_location it
> would be more readable anyway.

I have essentially made this too.  linemap_macro_loc_to_exp_point,
linemap_macro_map_loc_unwind_once and linemap_macro_map_loc_to_def_point
are now static functions that are only subroutines of
linemap_resolve_location.  They are not part of the public line map API
anymore. linemap_location_in_system_header_p uses
linemap_resolve_location now, and I have updated the other parts of the
compiler that were using those to use linemap_resolve_location instead.

> What do you think?

It's should be less confusing now, hopefully.  :)

Now I am re-posting the two additional patches that I had to
mechanically update due to this changes.  I have bootstrapped and tested
each patch one by one, while having its dependent patches installed.
This was done on x86_64-unknown-linux-gnu against trunk for all
languages including Ada.

From: Dodji Seketeli <dodji@redhat.com>
Date: Sat, 4 Dec 2010 16:31:35 +0100
Subject: [PATCH 3/7] Emit macro expansion related diagnostics

In this third instalment the diagnostic machinery -- when faced with
the virtual location of a token resulting from macro expansion -- uses
the new linemap APIs to unwind the stack of macro expansions that led
to that token and emits a [hopefully] more useful message than what we
have today.

diagnostic_report_current_module has been slightly changed to use the
location given by client code instead of the global input_location
variable.  This results in more precise diagnostic locations in
general but then the patch adjusts some C++ tests which output changed
as a result of this.

Three new regression tests have been added.

The mandatory screenshot goes like this:

[dodji@adjoa gcc]$ cat -n test.c
     1    #define OPERATE(OPRD1, OPRT, OPRD2) \
     2      OPRD1 OPRT OPRD2;
     3
     4    #define SHIFTL(A,B) \
     5      OPERATE (A,<<,B)
     6
     7    #define MULT(A) \
     8      SHIFTL (A,1)
     9
    10    void
    11    g ()
    12    {
    13      MULT (1.0);/* 1.0 << 1; <-- so this is an error.  */
    14    }

[dodji@adjoa gcc]$ ./cc1 -quiet -ftrack-macro-expansion test.c
test.c: In function 'g':
test.c:5:14: erreur: invalid operands to binary << (have 'double' and 'int')
test.c:2:9: note: in expansion of macro 'OPERATE'
test.c:5:3: note: expanded from here
test.c:5:14: note: in expansion of macro 'SHIFTL'
test.c:8:3: note: expanded from here
test.c:8:3: note: in expansion of macro 'MULT2'
test.c:13:3: note: expanded from here

The combination of this patch and the previous ones boostrapped with
--enable-languages=all,ada and passed regression tests on
x86_64-unknown-linux-gnu.

gcc/
	* gcc/diagnostic.h (diagnostic_report_current_module): Add a
	location parameter.
	* diagnostic.c (diagnostic_report_current_module): Add a location
	parameter to the function definition.  Use it instead of
	input_location.  Resolve the virtual location rather than just
	looking up its map and risking to touch a resulting macro map.
	(default_diagnostic_starter): Pass the relevant diagnostic
	location to diagnostic_report_current_module.
	* tree-diagnostic.c (maybe_unwind_expanded_macro_loc): New.
	(virt_loc_aware_diagnostic_finalizer): Likewise.
	(diagnostic_report_current_function): Pass the
	relevant location to diagnostic_report_current_module.
	* tree-diagnostic.h (virt_loc_aware_diagnostic_finalizer): Declare
	new function.
	* toplev.c (general_init): By default, use the new
	virt_loc_aware_diagnostic_finalizer as diagnostic finalizer.
	* Makefile.in: Add vec.h dependency to tree-diagnostic.c.

gcc/cp/

	* error.c (cp_diagnostic_starter): Pass the relevant location to
	diagnostic_report_current_module.
	(cp_diagnostic_finalizer): Call virt_loc_aware_diagnostic_finalizer.

gcc/testsuite/

	* lib/prune.exp (prune_gcc_output):  Prune output referring to
	included files.
	* gcc.dg/cpp/macro-exp-tracking-1.c: New test.
	* gcc.dg/cpp/macro-exp-tracking-2.c: Likewise.
	* gcc.dg/cpp/macro-exp-tracking-3.c: Likewise.
	* gcc.dg/cpp/pragma-diagnostic-2.c: Likewise.

Addendum patch 0003

gcc/

	* diagnostic.c (diagnostic_report_current_module): Adjust for the
	renaming of LRK_MACRO_PARM_REPLACEMENT_POINT into
	LRK_MACRO_DEFINITION_LOCATION.
	* tree-diagnostic.c (maybe_unwind_expanded_macro_loc): Use the new
	linemap_step_out_once.  Adjust like above.
---
 gcc/Makefile.in                                 |    3 +-
 gcc/cp/error.c                                  |    5 +-
 gcc/diagnostic.c                                |   13 +-
 gcc/diagnostic.h                                |    2 +-
 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-1.c |   21 +++
 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-2.c |   21 +++
 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-3.c |   14 ++
 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-4.c |   14 ++
 gcc/testsuite/gcc.dg/cpp/pragma-diagnostic-2.c  |   34 +++++
 gcc/testsuite/lib/prune.exp                     |    1 +
 gcc/toplev.c                                    |    3 +
 gcc/tree-diagnostic.c                           |  182 ++++++++++++++++++++++-
 gcc/tree-diagnostic.h                           |    3 +-
 13 files changed, 305 insertions(+), 11 deletions(-)
 create mode 100644 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-1.c
 create mode 100644 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-2.c
 create mode 100644 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-3.c
 create mode 100644 gcc/testsuite/gcc.dg/cpp/macro-exp-tracking-4.c
 create mode 100644 gcc/testsuite/gcc.dg/cpp/pragma-diagnostic-2.c

[3/7] Emit macro expansion related diagnostics

Commit Message

Comments

Patch