Article View: comp.sys.sun
Article #9894Disabling BRK signal
From: Brent Welch
Date: Sat, 24 Dec 1988 21:31
Date: Sat, 24 Dec 1988 21:31
16 lines
1433 bytes
1433 bytes
My two-cents worth on the BREAK signal that aborts a Sun is that this is a software feature, not an intrinsic hardware feature. The serial line driver in SunOS sees the BREAK and call the PROM abort routine. I haven't actually looked at the SunOS code that does this. However, our Sprite OS that runs on Suns does this same sort of thing, and it is done in software. Hmmm, reexamining a Sun-Spots message imlies that the BREAK signal seems to cause a Watchdog Reset, not a mere abort-to-monitor. That's a little puzzling, but I still maintain that it is software because I've yanked the console line on our Sprite machines with impunity. (The Sprite kernel interprets a BREAK-followed-by-'a' as ABORT, BREAK-followed-by'd' as enter the debugger, etc. etc. We map a bunch of special keystrokes like this so we can poke at lifeless machines.) Brent Welch University of California, Berkeley brent%sprite@gingner.Berkeley.EDU
Technical Details
Message-ID:
<8812161613.AA205100@sprite.Berkeley.EDU>Path:
arc1-ca.pugleaf.net!utzoo!attcan!uunet!husc6!mailrus!cornell!uw-beaver!rice!sun-spots-requestShow Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!attcan!uunet!husc6!mailrus!cornell!uw-beaver!rice!sun-spots-request From: brent%sprite.Berkeley.EDU@ginger.berkeley.edu (Brent Welch) Newsgroups: comp.sys.sun Subject: Disabling BRK signal Message-ID: <8812161613.AA205100@sprite.Berkeley.EDU> Date: 24 Dec 88 21:31:18 GMT Sender: usenet@rice.edu Organization: Sun-Spots Lines: 16 Approved: Sun-Spots@rice.edu Original-Date: Fri, 16 Dec 88 08:13:37 PST X-Sun-Spots-Digest: Volume 7, Issue 75, message 6 of 11