[PATCH v2] [PR debug/67192] Fix C loops' back-jump location

Andreas Arnez arnez@linux.vnet.ibm.com
Thu Oct 29 13:16:00 GMT 2015


Ping:

  https://gcc.gnu.org/ml/gcc-patches/2015-10/msg02393.html

On Fri, Oct 23 2015, Andreas Arnez wrote:

> After parsing an unconditional "while"- or "for"-loop, the C front-end
> generates a backward-goto statement and implicitly sets its location to
> the current input_location.  But in some cases the parser peeks ahead
> first, such that input_location already points to the line after the
> loop and the generated backward-goto gets the wrong line number.
>
> One way this can occur is with a loop body consisting of an "if"
> statement, because then the parser peeks for an optional "else" before
> finishing the loop.
>
> Another way occurs since r223098 ("Implement -Wmisleading-indentation"),
> even with a loop body enclosed in braces.  This is because the check for
> misleading indentation always peeks ahead one token as well.
>
> This patch adds a new parameter to c_finish_loop that expclitly
> specifies the location to be used for the loop iteration.  All calls to
> c_finish_loop are adjusted accordingly.
>
> gcc/c/ChangeLog:
>
> 	PR debug/67192
> 	* c-typeck.c (c_finish_loop): Replace implicit use of
> 	input_location by new parameter iter_locus.
> 	* c-tree.h (c_finish_loop): Adjust prototype.
> 	* c-array-notation.c (fix_builtin_array_notation_fn): Explicitly
> 	pass input_location as the new parameter to c_finish_loop.
> 	(build_array_notation_expr): Likewise.
> 	(fix_conditional_array_notations_1): Likewise.
> 	(fix_array_notation_expr): Likewise.
> 	(fix_array_notation_call_expr): Likewise.
> 	* c-parser.c (c_parser_while_statement): Choose iter_locus
> 	depending on whether the loop body is enclosed in braces, and pass
> 	it to c_finish_loop.
> 	(c_parser_for_statement): Likewise.
> 	(c_parser_do_statement): Use the final semicolon's location for
> 	iter_locus and pass it to c_finish_loop.
>
> gcc/testsuite/ChangeLog:
>
> 	PR debug/67192
> 	* gcc.dg/guality/pr67192.c: New test.



More information about the Gcc-patches mailing list