Date: Mon, 7 Dec 2009 15:05:26 -0600 From: Brooks Davis <brooks@FreeBSD.org> To: Luigi Rizzo <luigi@FreeBSD.org> Cc: svn-src-stable@FreeBSD.org, svn-src-all@FreeBSD.org, src-committers@FreeBSD.org, svn-src-stable-8@FreeBSD.org Subject: Re: svn commit: r200184 - stable/8/sbin/ipfw Message-ID: <20091207210526.GD77256@lor.one-eyed-alien.net> In-Reply-To: <20091207201744.GA77256@lor.one-eyed-alien.net> References: <200912061901.nB6J1X7L028816@svn.freebsd.org> <20091207201744.GA77256@lor.one-eyed-alien.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--sXc4Kmr5FA7axrvy Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Dec 07, 2009 at 02:17:44PM -0600, Brooks Davis wrote: > On Sun, Dec 06, 2009 at 07:01:33PM +0000, Luigi Rizzo wrote: > > Author: luigi > > Date: Sun Dec 6 19:01:33 2009 > > New Revision: 200184 > > URL: http://svn.freebsd.org/changeset/base/200184 > >=20 > > Log: > > mfc r200056 > > use qsort_r instead of heapsort; >=20 > This seems like a move in the wrong direction. The use of a recursive > algorithm in the kernel seems generally ill-advised and I was rather > disturbed when I found we had a quick sort implementation in libkern. >=20 > Unless you know the number of items is small, quick sort isn't safe and > if the number is small, heap sort is plenty efficient. Woops, it was pointed out to me that this was the userspace part where qsort is just fine. :) -- Brooks --sXc4Kmr5FA7axrvy Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (FreeBSD) iD8DBQFLHW4VXY6L6fI4GtQRAvLYAJ9dLBrE1AdVvoR3duv4DWzwUqfexQCbBtYi fZ9IB0YuH41DdOS1D+24u+Q= =ghbb -----END PGP SIGNATURE----- --sXc4Kmr5FA7axrvy--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20091207210526.GD77256>