[PATCH][3/3] ColdFire V2 pipeline description and scheduler hooks

Andreas Schwab schwab@suse.de
Mon Sep 10 12:53:00 GMT 2007


Maxim Kuvyrkov <maxim@codesourcery.com> writes:

> +/* An empty state that is used in m68k_sched_adjust_cost ().  */

Please don't write the parens when referring to functions in a comment.

> +/* Return statistical type of INSN regarding splits.  */
> +static sched_dump_class_def
> +sched_dump_split_class (rtx insn)
> +{
> +  gcc_assert (recog_memoized (insn) >= 0);

Can this have a side effect?  The argument of gcc_assert must not have
side effects since it can be optimized out with --disable-checking.

> +/* Return statistical type of INSN regarding DFA reservations.  */
> +static sched_dump_class_def
> +sched_dump_dfa_class (rtx insn)
> +{
> +  gcc_assert (recog_memoized (insn) >= 0
> +	      && insn_has_dfa_reservation_p (insn));

Likewise.

> +  if (reload_completed)
> +    /* Dump statistics.  */
> +    {
> +      if (sched_verbose >= 6 || msched_split_verbose_flag)
> +	{
> +	  const char *fn;
> +
> +	  if (msched_split_verbose_flag)
> +	    fn = "sched-split.log";
> +	  else
> +	    fn = NULL;
> +
> +	  m68k_sched_dump (sched_dump_split_class, fn);
> +	}

I don't see the point of having a separate dump file, moreover with a
fixed name.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."



More information about the Gcc-patches mailing list