Date: Mon, 8 May 2000 14:11:29 -0700 From: Alfred Perlstein <bright@wintelcom.net> To: Ben Smithurst <ben@scientia.demon.co.uk> Cc: Scott <gsi22419@gsaix2.cc.GaSoU.EDU>, Gustavo Vieira Goncalves Coelho Rios <kernel@tdnet.com.br>, shannon stees <shadyshay@hotmail.com>, questions@FreeBSD.ORG Subject: Re: researching FreeBSD Message-ID: <20000508141129.B17425@fw.wintelcom.net> In-Reply-To: <20000508193701.A55455@strontium.scientia.demon.co.uk>; from ben@scientia.demon.co.uk on Mon, May 08, 2000 at 07:37:01PM %2B0100 References: <3916CF65.8BC7835A@tdnet.com.br> <Pine.A41.3.96.1000508140546.27716A-100000@gsaix2.cc.GaSoU.EDU> <20000508193701.A55455@strontium.scientia.demon.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
* Ben Smithurst <ben@scientia.demon.co.uk> [000508 14:06] wrote: > Scott wrote: > > > This question and your answer makes me wonder. Are there no ASM > > optimizations that have been made? > > Yes, but the vast majority of it is C. > > ben@magnesium:~$ locate "/usr/src/*.[Ss]" | wc -l > 243 > ben@magnesium:~$ locate "/usr/src/*.c" | wc -l > 7127 > > Don't take this as meaning the ratio of C to assembler is 30:1, since > the .c files may well be much bigger, the .s files are usually just > one routine which has been optimized (e.g. some libc functions in > /usr/src/lib/libc/i386). There is almost certainly more C and less > assembler than 30:1. It's also important to note that although there is quite a bit of asm (relatively speaking) a bunch of it can be attributed to hand optimized routines for which C equivelants exist for archs that don't want/have-time to hand optimize. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000508141129.B17425>