home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!usc!elroy.jpl.nasa.gov!ames!sun-barr!news2me.ebay.sun.com!exodus.Eng.Sun.COM!appserv.Eng.Sun.COM!ouroborous.Eng.Sun.COM!limes
- From: limes@ouroborous.Eng.Sun.COM (Greg Limes)
- Newsgroups: comp.arch
- Subject: Re: Sun 600MP Benchmark Anomaly
- Date: 25 Jul 1992 00:51:21 GMT
- Organization: Sun Microsystems, Inc.
- Lines: 19
- Message-ID: <l719c9INNhm8@appserv.Eng.Sun.COM>
- References: <1992Jul23.145502.20975@ncrcae.ColumbiaSC.NCR.COM> <l6tucvINNbqt@appserv.Eng.Sun.COM> <JIM.92Jul24155432@hunter.cs.strath.ac.uk>
- NNTP-Posting-Host: ouroborous
-
- I wrote:
- | Second answer: Only one CPU at a time can receive interrupt
- | requests from devices in the system.
-
- In article <JIM.92Jul24155432@hunter.cs.strath.ac.uk> jim@cs.strath.ac.uk (Jim Reid) writes:
- | It shouldn't be that difficult to arrange for multiple processors to
- | execute multiple interrupt service routines simultaneously.
-
- You get that when you get multithreading, ie. Solaris 2, the version
- of the system that supports the SuperSPARC in MP mode.
-
- In SunOS 4.1.2mp, we opted for maximum compatibility, which includes
- the interactions between the two halves of the device drivers, and to
- as large an extent as possible, not requiring source changes to
- interrupt service routines.
-
- Feel free to let me know if more detail is needed on this; I'm cutting
- it short because (a) it seems obvious, and (b) it's 6pm friday evening
- and I hear a beer calling.
-