Date: Mon, 24 Jul 1995 16:27:44 -0700 From: asami@cs.berkeley.edu (Satoshi Asami) To: knarf@nasim.cube.net Cc: freebsd-ports@freebsd.org Subject: Re: diffs and checksums Message-ID: <199507242327.QAA00806@forgery.CS.Berkeley.EDU> In-Reply-To: <3v11tt$ik@nasim.nasim.cube.net> (knarf@nasim.cube.net)
next in thread | previous in thread | raw e-mail | index | archive | help
* > I mean we put tarfiles that have been znew --best'd on wcarchive, and
* > every mirror site will pick them up. So the CPU is used only on one
* > site (probably thud), and all the mirror sites can save space. How's
* > that now? :)
*
* Yes, but every machine who wants to build the port will need more CPU
* time also.
Um, why is that? If anything, gzip --best'ed files uncompress
slightly *faster* than their gzip --fast'ed files. For instance:
>> time gzip --fast XMAIL ; lg XMAIL.gz; time gunzip XMAIL
6.30 real 4.48 user 0.42 sys
928 -rw------- 1 asami grad 940979 Jul 24 14:00 XMAIL.gz
2.92 real 1.54 user 0.52 sys
>> time gzip --best XMAIL ; lg XMAIL.gz; time gunzip XMAIL
15.66 real 11.63 user 0.31 sys
768 -rw------- 1 asami grad 778181 Jul 24 14:00 XMAIL.gz
2.60 real 1.44 user 0.43 sys
Maybe it's due to the smaller file size.
For people who want to do znew --best by themselves to save disk
space...I'll add a variable to bsd.port.mk that will enable znew
--best when set as part of the fetch target, with a logic like:
fetch file
if ZNEW_BEST is set
test the fetched file against znew --best version of md5
if not equal
do znew --best
fi
fi
Comments?
* > The only reasonable solution to this I can think of is to include a
* > md5 for the original (compressed) tarball and the znew --best version,
* > and let bsd.port.mk test both of them before complaining. This should
* > salvage mirror sites that do a recompression (assuming they are doing
* > znew --best...if they aren't, to hell to them! ;).
*
* Will you implement it then?
Yes, unless someone has a better idea. You'll have to wait a few days
though, bsd.port.mk has a huge pending update (re: the mail I sent out
last night), and I don't want to mess with it now.
Satoshi
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199507242327.QAA00806>