Date: Mon, 12 Aug 1996 12:42:02 -0500 (CDT) From: Bob Willcox <bob@luke.pmr.com> To: 100626.3506@CompuServe.COM (Jan Knepper) Cc: khetan@iafrica.com, freebsd-current@FreeBSD.ORG Subject: Re: compile time Message-ID: <199608121742.MAA28495@luke.pmr.com> In-Reply-To: <960806095732_100626.3506_BHL105-1@CompuServe.COM> from Jan Knepper at "Aug 6, 96 05:57:33 am"
next in thread | previous in thread | raw e-mail | index | archive | help
Jan Knepper wrote:
> Kheten Gajjar wrote:
>
> /* > A kernel recompile on 2.1.0 takes for me:
> > text data bss dec hex
> > 954368 61440 74356 1090164 10a274
> > 257.32 real 162.45 user 18.24 sys
>
> Wow! What machine have you got ? */
>
> <ugh> A Pentium Pro 200, 64 MB RAM, NCR SCSI-2 2.1 GB HD, 6x SCSI CD-ROM...
> I am running my own company and have a couple of machine's. Only P's though. The
> PP200 is the latest toy. I think it is in for almost 2 months now, may be longer
> already. I usually develop in C++ for Window NT/95, 3.1x, OS/2 Warp, Novell
> NetWare. Header files and other stuff is getting so *huge* that a faster
> machines really pays off these days...
> I used a P120 before. The compile time turned out to be LESS than half what is
> was before! That is what I call improvement. <grin>
>
> Don't worry, be Kneppie,
> Jan
>
Its a bit of an apple to oranges comparison, but my Pentium Pro
200 with Natom chipset, 96MB RAM, NFS mounted /usr/src over a
10BaseT ethernet link, compiling a customized -current kernel with
the -pipe option gives me:
text data bss dec hex
876544 57344 84876 1018764 f8b8c
207.21s real 150.37s user 14.31s system
Note that these times do _not_ include "make depend". That adds
just under a minute of elapsed time to the results.
--
Bob Willcox politics, n:
bob@luke.pmr.com A strife of interests masquerading as a contest of
Austin, TX principles. The conduct of public affairs for private
advantage. -- Ambrose Bierce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199608121742.MAA28495>