Re: GCC gprof statistics

[Andrew Pinski <pinskia at physics dot uc dot edu>]

On Sunday, Jun 29, 2003, at 06:46 US/Eastern, ishikawa@yk.rim.or.jp 
wrote:
> For example, I looked at how for_each_rtx, worst offender in terms of
> time spent in flat profile, is called so many times.  It was not so
> evident how possibly to reduce the calls.

Using Shark from the CHUD tools, sampling for 30 seconds, here are the 
top 3 functions which call expr_equiv_p:
	8.2%	  ldst_entry	cc1	
	3.1%	  expr_equiv_p	cc1		<--- itself (already lowered)
	2.1%	  trim_ld_motion_mems	cc1	


Note this version of gcc which I tested includes a patch which optimize 
the common case of "ee" for the rtl/rtx:
3.6%	1906	  if ( GET_RTX_LENGTH (code) == 2 && fmt[0] == 'e' && fmt[1] 
== 'e')		
	1907	    {		
0.2%	1908	      if (! expr_equiv_p (XEXP (x, 1), XEXP (y, 1)))		
	1909		return 0;	!Invariant	
	1910	      return expr_equiv_p (XEXP (x, 0), XEXP (y, 0));		
	1911	    }		

which causes expr_equiv_p to be able changed into a loop for that case.
This patch has also been applied (for me) to for_each_rtx.

I also have a patch which causes the number of calls to expr_equiv_p to 
be lowered by switching around
(in find_rtx_in_ldst) the left hand with the right in the "and".


Thanks,
Andrew Pinski