http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53437
Richard Guenther <rguenth at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |ASSIGNED Last reconfirmed| |2012-05-21 Component|regression |middle-end AssignedTo|unassigned at gcc dot |rguenth at gcc dot gnu.org |gnu.org | Target Milestone|--- |4.8.0 Summary|FAIL: |[4.8 Regression] FAIL: |gcc.dg/guality/inline-param |gcc.dg/guality/inline-param |s.c -O0 |s.c -O0 Ever Confirmed|0 |1 --- Comment #1 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-05-21 13:50:51 UTC --- Mine. Index: gcc/tree-inline.c =================================================================== --- gcc/tree-inline.c (revision 187709) +++ gcc/tree-inline.c (working copy) @@ -2701,7 +2701,8 @@ setup_one_parameter (copy_body_data *id, STRIP_USELESS_TYPE_CONVERSION (rhs); /* If we are in SSA form properly remap the default definition - or omit the initialization if the parameter is unused. */ + or assign to a dummy SSA name if the parameter is unused and + we are not optimizing. */ if (gimple_in_ssa_p (cfun) && is_gimple_reg (p)) { if (def) @@ -2711,6 +2712,11 @@ setup_one_parameter (copy_body_data *id, SSA_NAME_IS_DEFAULT_DEF (def) = 0; set_default_def (var, NULL); } + else if (!optimize) + { + def = make_ssa_name (var, NULL); + init_stmt = gimple_build_assign (def, rhs); + } } else init_stmt = gimple_build_assign (var, rhs);