Date: Thu, 24 Apr 2008 21:58:22 +0200 From: David Naylor <naylor.b.david@gmail.com> To: freebsd-questions@freebsd.org Subject: Performance research Message-ID: <200804242158.27132.naylor.b.david@gmail.com>
next in thread | raw e-mail | index | archive | help
--nextPart4248026.QdsbbB2Shf Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hi, (If you want to skip to the questions (since this is a questions mailing=20 list :-) please see the 2nd last paragraph) I am interested in the performance difference of using either -Os or -O2. = =20 =46rom my preliminary research I have found that in most cases -Os=20 outperforms -O2 (and produces smalled binaries). =20 However this is only in a very isolated situation and does not make any cas= e=20 for a system wide change therefore I would like to run some experiments to= =20 measure the difference between the two flags. =20 I plan to use a live CD (with a system using 3GB) and having both the sourc= e=20 and object tree in memory (so as to avoid the harddrive influencing the=20 times). Since the CD is very small I am expecting caching to eliminate any= =20 influence of the slow CD read times of any of the files required to build=20 world. =20 The tests will be automated by scripts with two CD's, the first with -Os an= d=20 the second with -O2 binaries for the kernel and world (both amd64). I am=20 running an Intel Core 2 Duo 2.2 and 3GB of ram (as mentioned before). I am= =20 hoping that only the execution speed will be measured without any other=20 factors influencing (such a lack of memory [and thus -Os having the=20 advantage]) the results. =20 Onto the questions: 1) Is there any other (better) tool I should use to measure the performance= of=20 building world (other than time)? 2) Since (I think) all of FreeBSD sources of aliassing safe (I have been=20 running a system without -fno-strict-aliassing and optimised for a while=20 without a problem) should I leave it out of the build flags (i.e.=20 CFLAGS=3D-Os -pipe) 3) Any other suggestions on refining this experiment? 4) Should this experiment produce potential, or conclusive, results any oth= er=20 suggestions of experiments to run (perhaps a database or other flags?)=20 I will be posting my results (possible wiki page, or a particular mailing=20 list) after this weekend. =20 Thank you for your time. =20 David --nextPart4248026.QdsbbB2Shf Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQBIEOZjUaaFgP9pFrIRAvX7AJ4++t7wp6tNmnIwv2rgxJvYL5iWIQCaA62d m8PikFqzg7WPXRUIiqJs/m4= =ftWu -----END PGP SIGNATURE----- --nextPart4248026.QdsbbB2Shf--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200804242158.27132.naylor.b.david>