Re: [PATCH] Add newline when checking path profitability.

[Richard Biener <richard.guenther@gmail.com>]

On Tue, Aug 23, 2022 at 12:12 PM Aldy Hernandez <aldyh@redhat.com> wrote:
>
> It looks like we're missing a newline for cases where we don't print
> anything.
>
> OK?

OK

> gcc/ChangeLog:
>
>         * tree-ssa-threadbackward.cc (possibly_profitable_path_p): Always
>         add newline.
>         (profitable_path_p): Same.
> ---
>  gcc/tree-ssa-threadbackward.cc | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/gcc/tree-ssa-threadbackward.cc b/gcc/tree-ssa-threadbackward.cc
> index 3218ad931ef..9725f50e639 100644
> --- a/gcc/tree-ssa-threadbackward.cc
> +++ b/gcc/tree-ssa-threadbackward.cc
> @@ -719,7 +719,11 @@ back_threader_profitability::possibly_profitable_path_p
>               gimple *stmt = gsi_stmt (gsi);
>               if (gimple_call_internal_p (stmt, IFN_UNIQUE)
>                   || gimple_call_builtin_p (stmt, BUILT_IN_CONSTANT_P))
> -               return false;
> +               {
> +                 if (dump_file && (dump_flags & TDF_DETAILS))
> +                   fputc ('\n', dump_file);
> +                 return false;
> +               }
>               /* Do not count empty statements and labels.  */
>               if (gimple_code (stmt) != GIMPLE_NOP
>                   && !is_gimple_debug (stmt))
> @@ -821,6 +825,8 @@ back_threader_profitability::possibly_profitable_path_p
>                     && (m_n_insns * param_fsm_scale_path_stmts
>                         >= param_max_jump_thread_duplication_stmts));
>
> +  if (dump_file && (dump_flags & TDF_DETAILS))
> +    fputc ('\n', dump_file);
>    return true;
>  }
>
> @@ -947,6 +953,8 @@ back_threader_profitability::profitable_path_p (const vec<basic_block> &m_path,
>                  "non-empty latch\n");
>        return false;
>      }
> +  if (dump_file && (dump_flags & TDF_DETAILS))
> +    fputc ('\n', dump_file);
>    return true;
>  }
>
> --
> 2.37.1
>