This patch fixes 9 unexpected failures in the gfortran testsuite on
nvptx-none. The issue is that gfortran's EQUIVALENCE internally uses
symbols such as "equiv.0" even on platforms that define NO_DOT_IN_LABEL.
On nvptx-none, this then results in the following error message(s):
ptxas application ptx input, fatal: Parsing error near '.0': syntax error
ptxas fatal : Ptx assembly aborted due to errors
The fix is to tweak trans-common.cc to respect the target's NO_DOT_IN_LABEL
(and NO_DOLLAR_IN_LABEL) when generating internal equiv.%d symbols.
Only the nvptx, mmix and xtensa backends define NO_DOT_IN_LABEL which
explains why no-one has spotted/fixed this issue since the problematic
code was last changed back in 2005(!).
This patch has been tested on nvptx-none hosted by x86_64-pc-linux-gnu
with make and make -k check with no new failures, and the nine fewer
failures in the gfortran testsuite described above. Ok for mainline?
2022-02-10 Roger Sayle <[email protected]>
gcc/fortran/ChangeLog
* trans-common.cc (GFC_EQUIV_FMT): New macro respecting the
target's NO_DOT_IN_LABEL and NO_DOLLAR_IN_LABEL preferences.
(build_equiv_decl): Use GFC_EQUIV_FMT here.
Thanks in advance,
Roger
--
diff --git a/gcc/fortran/trans-common.cc b/gcc/fortran/trans-common.cc
index 7b4d198..184a976 100644
--- a/gcc/fortran/trans-common.cc
+++ b/gcc/fortran/trans-common.cc
@@ -338,6 +338,13 @@ build_field (segment_info *h, tree union_type,
record_layout_info rli)
h->field = field;
}
+#if !defined (NO_DOT_IN_LABEL)
+#define GFC_EQUIV_FMT "equiv.%d"
+#elif !defined (NO_DOLLAR_IN_LABEL)
+#define GFC_EQUIV_FMT "equiv$%d"
+#else
+#define GFC_EQUIV_FMT "equiv_%d"
+#endif
/* Get storage for local equivalence. */
@@ -356,7 +363,7 @@ build_equiv_decl (tree union_type, bool is_init, bool
is_saved, bool is_auto)
return decl;
}
- snprintf (name, sizeof (name), "equiv.%d", serial++);
+ snprintf (name, sizeof (name), GFC_EQUIV_FMT, serial++);
decl = build_decl (input_location,
VAR_DECL, get_identifier (name), union_type);
DECL_ARTIFICIAL (decl) = 1;