[Fortran-Experiments]: patch

[Christopher D. Rickett]

hi all,

i've attached another patch for the fortran-experiments branch.  it
contains quite a bit of whitespace fixes and new test cases.  it also has
improved error messages for a few things and new error checking on binding
labels.  it also fixes some minor bugs, such as c_size_t and the required
parens for subroutines marked with bind(c).  bootstrapped and regtested on
x86 with no new failures.

ChangeLog entry:

2007-02-24  Christopher D. Rickett  <crickett@lanl.gov>
	* gcc/testsuite/gfortran.dg/test_bind_c_parens.f03: New test case
	to make sure gfortran requires the parens for a subroutine that is
	marked BIND(C).
	* gcc/testsuite/gfortran.dg/bind_c_usage_5.f03: New test case to
	verify that gfortran catches duplicate BIND(C) attributes.
	* gcc/testsuite/gfortran.dg/bind_c_usage_6.f03: New test case for
	BIND(C) attribute only being applied to variables or common blocks.
	* gcc/testsuite/gfortran.dg/c_size_t_driver.c: Main program for
	c_size_t_test.
	* gcc/testsuite/gfortran.dg/binding_label_tests_2.f03: Updated
	expected error messages.
	* gcc/testsuite/gfortran.dg/c_size_t_test.f03: New test case to
	verify that c_size_t equals sizeof(size_t).
	* gcc/testsuite/gfortran.dg/binding_label_tests_3.f03: New test
	case for verifying that binding labels don't collide with other
	binding labels or global symbol names.
	* gcc/testsuite/gfortran.dg/binding_label_tests_4.f03: Ditto.
	* gcc/testsuite/gfortran.dg/binding_label_tests_5.f03: Ditto.
	* gcc/testsuite/gfortran.dg/binding_label_tests_6.f03: Ditto.
	* gcc/testsuite/gfortran.dg/binding_label_tests_7.f03: Ditto.
	* gcc/testsuite/gfortran.dg/binding_label_tests_8.f03: Ditto.
	* gcc/testsuite/gfortran.dg/bind_c_usage_3.f03: Whitespace.
	* gcc/fortran/symbol.c: Added check to gfc_add_is_bind_c to make
	sure the symbol isn't already marked as BIND(C).  Added flag to
	gfc_add_is_bind_c to say if the BIND(C) is coming from a procedure
	declaration.  Fixed whitespace.
	* gcc/fortran/decl.c: Removed call to gfc_add_is_bind_c from
	gfc_match_bind_c; the call is now in the routines that call
	gfc_match_bind_c. Added another check for preventing multiple
	identifiers for a BIND(C) that has a NAME= specifier (there was
	one case that wasn't being checked).  Split
	gfc_match_attr_spec_stmt into gfc_match_bind_c_stmt and
	gfc_match_proc_decl_stmt.  The two new functions are now called
	from parse.c.  Added error checking for BIND(C).  Improved error
	messages for gfc_match_bind_c.  Fixed whitespace.
	* gcc/fortran/gfortran.h: Updated prototype for
	gfc_add_is_bind_c.  Added a sym_name field to gfc_gsymbol for use
	with global symbols created for binding labels.
	* gcc/fortran/iso-c-binding.def: Set c_size_t equal to
	gfc_index_integer_kind because size_type_node is not set in time
	for gfc_init_kinds.
	* gcc/fortran/resolve.c: Added gfc_verify_binding_labels to
	report errors if binding labels collide with another binding label
	or global symbol name.
	* gcc/fortran/match.c: Improved gfc_match_name_C for better error
	reporting.
	* gcc/fortran/match.h: Removed gfc_match_attr_spec_stmt.  Added
	gfc_match_bind_c_stmt and gfc_match_proc_decl_stmt.
	* gcc/fortran/parse.c: Added calls to gfc_match_bind_c_stmt and
	gfc_match_proc_decl_stmt in decode_statement.

Chris
-------------- next part --------------
A non-text attachment was scrubbed...
Name: svn_diff_ws.txt.gz
Type: application/x-gzip
Size: 10591 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20070224/4c6adb4f/attachment.bin>