Return to BSD News archive
Xref: sserve comp.os.386bsd.misc:3925 comp.os.386bsd.questions:14180 comp.os.linux.development:18512 comp.os.linux.misc:28636 sci.electronics:83077 Newsgroups: comp.os.386bsd.misc,comp.os.386bsd.questions,comp.os.linux.development,comp.os.linux.misc,sci.electronics Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!bunyip.cc.uq.oz.au!munnari.oz.au!spool.mu.edu!howland.reston.ans.net!EU.net!sun4nl!rnzll3!sys3.pe1chl!rob From: rob@pe1chl.ampr.org (Rob Janssen) Subject: Re: 16550 detection Reply-To: pe1chl@rabo.nl Organization: PE1CHL Message-ID: <CyntEv.Fy@pe1chl.ampr.org> References: <m0r2XH0-000MhFC@kitana> Date: Wed, 2 Nov 1994 21:37:42 GMT Lines: 18 In <m0r2XH0-000MhFC@kitana> kitana!sysop@caprica.com (JL Gomez) writes: >I've e-mail someone using a BOCA 2016 using 8 ports at 115K what their >load was on the CPU. >He replied 3% under Linux. Says alot about the serial driver. The key question in such cases is "how did he measure it"... when it is only the system load imposed by the applications that actually read the characters, it is not the full story. The actual load caused by the interrupts is harder to measure, and probably above 3%. Rob -- ------------------------------------------------------------------------- | Rob Janssen | AMPRnet: rob@pe1chl.ampr.org | | e-mail: pe1chl@rabo.nl | AX.25 BBS: PE1CHL@PI8UTR.#UTR.NLD.EU | -------------------------------------------------------------------------