elliot@alfred.UUCP (Elliot Dierksen) (01/12/90)
I was looking through the documentation today (Amazing what's in there :-)) and I stumbled on the service & parts ordering information. In there was listed: 105160253 Math Acc. Unit MC 68881 Comments anybody?? Also, in the reference manual there are 2 tape boards listed: Interface board for tape backup (floppy tape) and Interface board for tape backup (QIC-02) Can anyone shed some light on this subject??? -- Elliot Dierksen "I don't care if my lettuce has DDT on it, as long as it's crisp!!" -- Jorma Kaukonen Work) {att,codas}!candi!fang!ebd (407) 660-3377 Home) {peora,uunet,ucf-cs}!tarpit!alfred!elliot (407) 290-9744
thad@cup.portal.com (Thad P Floryan) (01/14/90)
elliot@alfred.UUCP (Elliot Dierksen) in <607@alfred.UUCP> writes: I was looking through the documentation today (Amazing what's in there :-)) and I stumbled on the service & parts ordering information. In there was listed: 105160253 Math Acc. Unit MC 68881 Comments anybody?? A repost from the past: >From: jbm@uncle.UUCP (John B. Milton) >Newsgroups: unix-pc.general,comp.sys.att >Subject: Could a 68881 Math co-processor go fast?; HwNote07 >Message-ID: <372@uncle.UUCP> >Date: 23 Oct 88 19:12:19 GMT >Date-Received: 25 Oct 88 06:18:41 GMT > >Parts of this article are copied from mail to David Wexelblat >(dwex@mtgzz.att.com), who suggested a 68881 when I asked for hw ideas. The >rest of you just don't seem to have any ideas :-! > >I was thinking about that one a while ago. AT&T actually made a 68881 board, >they even have a part number, 105160253. I talked to some guy at AT&T, and >he says the board was never release due to poor performance. Since the 68010 >does not have a coprocessor interface, the 68881 would have to operate as an >I/O device, so it would work fine as an expansion card. David tells me that >the software interface was implemented using an F-line exception handler. Whil e >this is clean from the compiler point of view (you just generate 68881 >instructions), it is expensive: >1. Execute instruction >2. Get F-line exception >3. Stack everything away >4. Start running the common kernel exception code >5. Pull all the parameters from the user PC and talk to the 68881 through I/O >6. Defer to another process while the instruction runs on the 68881 >7. Get an interrupt, get the status >8. Call the kernel code to return from exception. > >Steps 6&7 could be replaced with >6. Loop polling the 68881 for completion >7. Get the status > >I don't know what they did about having an expanded process context for procs >using the 68881. ...other material deleted... Thad Floryan [ thad@cup.portal.com (OR) ..!sun!portal!cup.portal.com!thad ]