home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!decwrl!esl!cary@esl.com
- From: cary@esl.com (Cary Jamison)
- Newsgroups: comp.sys.hp
- Subject: Update on 380 problem
- Message-ID: <1724@esl.ESL.COM>
- Date: 19 Nov 92 23:21:57 GMT
- Sender: news@esl.ESL.COM
- Organization: ESL, A Subsidiary of TRW Inc
- Lines: 19
- Nntp-Posting-Host: 129.193.147.76
-
- I posted a message a couple weeks ago about a problem I had encountered with
- a set of bad 68040s in our HP 9000/380s. I have just received some more
- information on the problem and thought I would pass it on as a followup,
- in case any one is interested, though I think there are only 7xx users in
- this group :-)
-
- The bug in the 68040 shows up when an interrupt occurs after a particular
- floating point instruction, I believe a set condition codes instruction, if
- that means anything to you '040 assembly hackers out there. If there's
- more than one such instruction, I couldn't tell you which is the bad one.
-
- HP claims this fp instruction is only generated by their Ada compiler (we
- are using HP Ada 35t). So, unless you have an older 380 and use HP Ada,
- you probably don't need to worry about it.
-
- ------------
- Cary Jamison
- cary@esl.com
-
-