Date: Sun, 16 Mar 1997 09:44:42 +0100 (MET) From: Peter Mutsaers <plm@xs4all.nl> To: Wolfgang Helbig <helbig@MX.BA-Stuttgart.De> Cc: current@freebsd.org Subject: Re: why /usr/libexec/cc* static? Message-ID: <199703160844.JAA00453@plm.xs4all.nl> In-Reply-To: <199703151029.LAA01247@helbig.informatik.ba-stuttgart.de> References: <87n2s5msv9.fsf@plm.xs4all.nl> <199703151029.LAA01247@helbig.informatik.ba-stuttgart.de>
next in thread | previous in thread | raw e-mail | index | archive | help
>> On Sat, 15 Mar 1997 11:29:43 +0100 (MET), Wolfgang Helbig
>> <helbig@MX.BA-Stuttgart.De> said:
>> I noted that /usr/libexec/cc* (the C compiler passes) are linked
>> statically, and wonder why? cpp and cc itself are linked dynamically,
>> so it's not as if one could recompile libc in case something would be
>> wrong with libc.so etc.
>>
WH> Here is the answer:
WH> (cvs log Makefile ...)
WH> RCS file: /usr/cvsroot/src/gnu/usr.bin/cc/cc1/Makefile,v
WH> Working file: Makefile
WH> [...]
WH> head: 1.11
WH> [...]
WH> ----------------------------
WH> revision 1.7
WH> date: 1995/10/01 20:16:27; author: davidg; state: Exp; lines: +2 -1
WH> Build cc1 nonshared. This actually results in it consuming 40K *less* disk
WH> space and improves compile times by a few percent.
WH> ----------------------------
Thanks for pointing this out.
Strange. How is it possible that a nonshared cc1 is smaller?
I guess the faster compiles are only true when enough memory is
available. If paging is going on, than the cc1 executable is not fully
cached and using shared code will help (?)
--
/\_/\
( o.o ) Peter Mutsaers | Abcoude (Utrecht), | Trust is a good quality
) ^ ( plm@xs4all.nl | the Netherlands | for other people to have
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199703160844.JAA00453>