Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!feeder.erje.net!1.eu.feeder.erje.net!newsfeed.fsmpi.rwth-aachen.de!newsfeed.straub-nv.de!eternal-september.org!feeder.eternal-september.org!mx02.eternal-september.org!.POSTED!not-for-mail From: D Finnigan Newsgroups: comp.sys.apple2.programmer Subject: Re: Marina, now with Berkeley sockets Date: Wed, 17 Jun 2015 21:28:21 +0000 (UTC) Organization: Mac GUI Lines: 23 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Wed, 17 Jun 2015 21:28:21 +0000 (UTC) Injection-Info: mx02.eternal-september.org; posting-host="ac9e0e4d7dd3957fb97803ee6801bab8"; logging-data="18684"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX193WxmYGj5bIy5jLqlxx3MQ" User-Agent: Mac GUI Usenet In-Reply-To: Cancel-Lock: sha1:AxnOQ9l3IdfHkQQWHw6c3KT2ac4= Xref: csiph.com comp.sys.apple2.programmer:1701 David Schmidt wrote: > On 6/16/2015 4:54 PM, D Finnigan wrote: >> The June 2015 edition of Marina is here, featuring a Berkeley sockets >> API. > > Glad to see the source code. In its current form, how much memory > footprint is required (understanding the send buffer is variable - so > let's just say 1492 for now)? > Code is a little over 9,000 bytes, but significant space is taken by the DHCP, DNS, and ARP subsystems that all use dedicated buffers for send operations. Incoming buffer should be 1514 bytes, and outgoing buffer is 1500 because the protocol headers are stored separate from the payload (see ETH.S on disk 2). In future, I will probably have to introduce support for multiple receive buffers in order to support simultaneous TCP and UDP sockets. -- ]DF$ The Marina IP stack for Apple II-- http://marina.a2hq.com/