This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: mainline "exploding"


> Here is some data for testcase from PR8361:
> 
> 	  	-O0 	-O1	-O2	-O3
> ----------------------------------------------
> GCC 3.2.3 	28.98	54.27	70.90	73.9
> GCC 3.3.2 	30.13	54.59	76.39	77.89
> mainline  	31.91	113.22	105.29	106.30
>  w/o checking   26.37   90.78   87.88   89.16
> tree-ssa        32.60   71.49   93.78   99.14
> 
> Mainline (w/o checking) is nearly twice as slow as 3.3.2 for -O1.

Hi, here are timmings of all sollutions I've proposed.

1 Current mainline                               0m57s
2 unit-at-a-time by default                      0m32s
3 Using different limits with old heuristics     0m27s
4 unit-at-a-time by default + small limits at O1 0m24s
5 New non-unit-at-a-time inlining heruistics     0m25s

So all of them cuts the time to roughtly half of it.  My preference is
either 5 or 2, but I would be happy with any of it (3 is IMO worst one).

Honza


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]