Date: Fri, 19 Dec 1997 18:21:18 +0900 (JST) From: Michael Hancock <michaelh@cet.co.jp> To: John-Mark Gurney <gurney_j@resnet.uoregon.edu> Cc: A Joseph Koshy <koshy@india.hp.com>, freebsd-hackers@FreeBSD.ORG Subject: Re: converting drivers to dynamic memory Message-ID: <Pine.SV4.3.95.971219181416.6307A-100000@parkplace.cet.co.jp> In-Reply-To: <19971218204334.41263@hydrogen.nike.efn.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 18 Dec 1997, John-Mark Gurney wrote: > A Joseph Koshy scribbled this message on Dec 19: > > >>>>>> John-Mark Gurney <gurney_j@efn.org> said: > > If you are proposing new in-kernel structures, it may be convenient to > > choose those that can be later be easily parallelized. This can then avoid > > unnecessary serialization on the data structure if/when the degree of > > processor parallelism supported by FreeBSD increases in the future. > > hmm... good point... I'll have to talk with a professor of mine that > knows data structures a bit better than I for suggestions... this isn't > something that they normally cover in a 300 or 400 level classes... :( > Here's an interesting link which covers why some data structures are better than others for use in an OS kernel, http://www-dsg.stanford.edu/michaelg/. Algorithms in C by Robert Sedgewick, a former student of Donald Knuth, is also pretty good. The book is in the bibliography of the BSD book by McKusick, et al. The book covers radix tries which are in the networking code. Regards, Mike Hancock
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.SV4.3.95.971219181416.6307A-100000>