Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!cs.utexas.edu!uunet!mcsun!sunic!chalmers!cs.chalmers.se!augustss From: augustss@cs.chalmers.se (Lennart Augustsson) Newsgroups: comp.sys.sequent Subject: Re: Sequent Machines, what are their strengths & weaknesses Message-ID: <3235@fnatte.cs.chalmers.se> Date: 9 Sep 89 02:59:29 GMT Organization: Dept. of CS, Chalmers, Sweden Lines: 53 In article <17580@bellcore.bellcore.com> johno@dduck.UUCP (John OBrien) writes: >> >> 5) What are your experiences with bug-fixes and customer-support? > > Very responsive! Bug fixes have been shipped promptly. CUstomer > support is great (although, we had some trouble with the time > difference between Oregon and the east coast (7 AM in NJ is 4 AM > in Oregon and no one is awake.) On this point I have to disagree. I don't know if being far away from Oregon has something to do with it (the Sequent office that is nearest to Sweden is in Amsterdam, Netherlands). I have sent around 10 mailbugs to Sequent, ranging from very serious (at least for me) to just annoying. Out of this I have just gotton a response for one of them (I was quite serious bug for me, so I really begged them for a response). The response was that they admitted that what I complained about really was a bug! I have seen no bug fixes (come to think of it, one of the bugs reported about 1 1/2 years ago has been fixed now) to my problems, it has now been about more than a year since I reported them. If you want to know the problems you can read more below. >> >> 6) Given a free choice, would you buy another one... > > Absolutely!!!!!!!! I agree!!! Despite some problems it is one of the nicest machines I have ever used. -- Lennart Augustsson Description of two serious problems (Sequent, are you listening?): 1. The sigstack system call is supposed to tell the system that a different stack is supposed to be used whenever a signal occurs. Despite this a signal still thrashes two locations on the ordinary stack. This can be very dangerous if you are using the stack pointer in a non-standard way (which I am in a language implementation). The problem is present already in the original BSD 4.2 code, but has been fixed in 4.3. I have now fixed this bug myself (since Sequent seems to be incapable if doing it), by discompiling the sigstack system call and merging in the BSD 4.3 fixes and installing this as a new system call. Anyone who needs this fix badly, just drop me a mail and I'll send you the code (about 50 lines of C). 2. Under some very peculiar circumstances involving instructions with the LOCK prefix crossing page boundaries and causing page faults the program will get the Illegal instruction signal. There are work arounds to this, but they are messy. -- Lennart Augustsson Email: augustss@cs.chalmers.se or augustss@chalmers.csnet