Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!newstop!texsun!convex!convex.COM
From: tchrist@convex.COM (Tom Christiansen)
Newsgroups: comp.lang.perl
Subject: Re: fast grep?
Message-ID: <107296@convex.convex.com>
Date: 18 Oct 90 06:01:25 GMT
References: <1990Oct15.211219.8543@wrl.dec.com> <107211@convex.convex.com> <1990Oct17.044839.21709@iwarp.intel.com>
Sender: news@convex.com
Reply-To: tchrist@convex.COM (Tom Christiansen)
Organization: CONVEX Software Development, Richardson, TX
Lines: 15

In article <1990Oct17.044839.21709@iwarp.intel.com> merlyn@iwarp.intel.com 
(Randal Schwartz) writes, replying to me:

|Tom, go back and rerun your test with some slightly-more-optimal (I hope)
|Perl code:

No Randal, it's not so, unless I broke something whne I clipped your code.
It's 10 times slower, and that's on the 100k file.  On the 4Meg one
it gobbled up 30 meg (most resident) and spun for 40 minutes on my 
quarter-gig convex 220 before I killed it.

Larry, how come we only get BM speed on constant strings?  Can't it know?
Having to know to do the eval is too bad.

--tom