Return to BSD News archive
Newsgroups: comp.os.386bsd.development Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!haven.umd.edu!news.umbc.edu!eff!news.kei.com!ub!acsu.buffalo.edu!jones From: jones@acsu.buffalo.edu (T. Jones) Subject: Re: [FreeBSD 1.0R] DMA Problems? Message-ID: <CJKBzG.4sL@acsu.buffalo.edu> Sender: nntp@acsu.buffalo.edu Nntp-Posting-Host: hyades.cs.buffalo.edu Organization: CEDAR Research Group. References: <JTW.94Jan3234318@pmws.lcs.mit.edu> <jmonroyCJ3u56.1us@netcom.com> <2gd27p$dc6@u.cc.utah.edu> Date: Thu, 13 Jan 1994 10:04:28 GMT Lines: 35 In article <2gd27p$dc6@u.cc.utah.edu> of comp.os.386bsd.development terry@cs.weber.edu (A Wizard of Earth C) writes: >Yes, you should take it very seriously. Here, John is demonstrating a >better knowledge of solid state physics than most coders who dabble in >hardware drivers. > >Cosmic Rays are high energy particles from outer space -- you may have >heard of "the solar wind", and if you have ever worked on satellite or >other aerospace systems, you will probably be aware of a process called >"hardening", whose intent it is to make the harware more tolerant of >radiation events. A more proper term for an earthy engineer might have >been "background radiation", which includes both radiated particles as >decay products of matter and what cosmic radiation hits us through the >magnetosphere. I agree although the terminology may have been somewhat leading. With that it looks as though Mr. Monroy took the bait and came back asking for seconds. Alpha particle and nuetron radiation sources are known to cause transient ionizing effects in semiconductor devices, like DRAMs. They can also impart permanent effects if the event is significant enough, such as in the case of EMP emitted from a nuclear detonation. I hope that we can disregard that case for this discussion. Alpha sources are present in the device packaging itself as well as ocurring "naturally" in the form of background radiation. Bulk memories tend to be especially prone to errors from these events due in part to the high gate count per unit die area, as well as the reliance on stored charge. NMOS devices tend to be the least reliable, with GaAs being about the most reliable in terms of tolerating radiation effects. Of course you could shield your system in a cube of lead, but the alpha sources present in the device packaging would still get you. For me I think I would like to leave my DRAM refresh alone, as long as it conforms to the device manufacturer's specifications. Terry Jones jones@cedar.buffalo.edu