https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44491
--- Comment #7 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
(In reply to Dominique d'Humieres from comment #6)
> Related to pr69544. Note that compiling the test with 5.3.0 or trunk (6.0)
> gives now an ICE, r218570 gives the error and r218658 gives the ICE.
Hi, if you want to fix the ICEs for release builds even if you don't get a good
location, then do:
--- error.c (revision 232580)
+++ error.c (working copy)
@@ -934,7 +934,15 @@
loc = &gfc_current_locus;
else
loc = va_arg (*text->args_ptr, locus *);
- gcc_assert (loc->nextc - loc->lb->line >= 0);
+ gcc_assert_checking (loc->nextc - loc->lb->line >= 0);
+ /* FIXME: This should not happen, but it does: See PR44491 */
+ if (loc == NULL || loc->lb == NULL
+ || loc->nextc - loc->lb->line < 0)
+ {
+ text->set_location (0, UNKNOWN_LOCATION, true);
+ pp_string (pp, result[0]);
+ return true;
+ }
unsigned int offset = loc->nextc - loc->lb->line;
/* If location[0] != UNKNOWN_LOCATION means that we already
processed one of %C/%L. */
