https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99529

--- Comment #1 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Regarding calling newunit_alloc, that's due to libgfortran/io/unit.c's:

get_unit (st_parameter_dt *dtp, int do_create)
{
  gfc_unit *unit;

  if ((dtp->common.flags & IOPARM_DT_HAS_INTERNAL_UNIT) != 0)
    {
      int kind;
      if (dtp->common.unit == GFC_INTERNAL_UNIT)
        kind = 1;
      else if (dtp->common.unit == GFC_INTERNAL_UNIT4)
        kind = 4;
      else
        internal_error (&dtp->common, "get_unit(): Bad internal unit KIND");

      dtp->u.p.unit_is_internal = 1;
      dtp->common.unit = newunit_alloc ();
      unit = get_gfc_unit (dtp->common.unit, do_create);
      set_internal_unit (dtp, unit, kind);
      fbuf_init (unit, 128);
      return unit;
    }

 * * *

>From reading the report & looking at the code superficially, it seems as if a
patch is needed like:

diff --git a/libgfortran/io/transfer.c b/libgfortran/io/transfer.c
index 27bee9d4e01..a6f115d5803 100644
--- a/libgfortran/io/transfer.c
+++ b/libgfortran/io/transfer.c
@@ -4360,3 +4360,5 @@ st_read_done_worker (st_parameter_dt *dtp)
            }
+         LOCK (&unit_lock);
          newunit_free (dtp->common.unit);
+         UNLOCK (&unit_lock);
        }
@@ -4448,3 +4450,5 @@ st_write_done_worker (st_parameter_dt *dtp)
            }
+         LOCK (&unit_lock);
          newunit_free (dtp->common.unit);
+         UNLOCK (&unit_lock);
        }

Reply via email to