This is the mail archive of the gcc-patches@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: cpplib: Separate out translation phases 1 and 2


Neil Booth wrote:-

> Neil Booth wrote:-
> 
> > Actually, WTF, I think I'll do this, though like I said it's not really
> > a fair comparison.  I'll report the results tomorrow to "make bootstrap"
> > GCC before and after the patch with --enable-languages=c and
> > --disable-checking; the latter setting will also magnify any slowdown.

For the curious, I've done some more timings, including 3 in the
original state, and 3 with the current CVS.

Orig:
real	30m55.332s
user	28m18.960s
sys	1m48.990s

real	30m51.564s
user	28m13.780s
sys	1m48.220s

real	31m25.388s
user	28m10.560s
sys	1m48.640s

After Initial Patch:
real	31m32.023s
user	28m36.130s
sys	1m52.130s

After latest lexer improvements:
real	31m27.710s
user	28m22.360s
sys	1m51.660s

real	30m42.745s
user	28m23.700s
sys	1m50.300s

real	30m53.925s
user	28m28.000s
sys	1m52.270s

These last 3 elide the new file cppcharset.c as it distorts the picture.
My conclulsion is that "real" time is not a useful indicator, and that
we've already clawed back half of the loss of the initial patch.

Neil.


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