Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Paul Rubin <no.email@nospam.invalid>
Newsgroups: comp.lang.forth
Subject: Re: polyForth multitasker question
Date: Tue, 11 Apr 2017 18:58:42 -0700
Organization: A noiseless patient Spider
Lines: 16
Message-ID: <877f2ql73h.fsf@nightsong.com>
References: <ocafkt$1929$1@gioia.aioe.org> <koGdneROYv-Yg3TFnZ2dnUU7-KvNnZ2d@supernews.com> <ocehe0$19ks$1@gioia.aioe.org> <CM2dnXuhx9lfTnfFnZ2dnUU7-TfNnZ2d@supernews.com> <ocfehb$467$1@dont-email.me> <bIydnby3AqzK03bFnZ2dnUU7-VfNnZ2d@supernews.com> <ocfkhb$8ig$1@dont-email.me> <ochdnu$1r70$1@gioia.aioe.org> <ocile6$7ro$1@dont-email.me> <e11a25dd-5c4c-404e-bbfa-3ec67c7b1d08@googlegroups.com> <2ef60cba-5577-4813-ae93-3946cd4b415b@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain
Injection-Info: mx02.eternal-september.org; posting-host="0de1f87b5972fac17e91256555e4d05f"; logging-data="22157"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18o16/5WPJkAqSTYpKeZZnn"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux)
Cancel-Lock: sha1:6gqYQoZwlD1iJOnLxTcbsuJ3npQ= sha1:4qhU++/gL6iNZo1alU5zkrPXnCY=
Xref: csiph.com comp.lang.forth:57085

foxaudioresearch@gmail.com writes:
> I have built small databases in applications without formal record
> locking because I could control the record update routine to not yield
> until complete.  It becomes so simple.

Do you mean your db would lock up the system til disk i/o had completed?
That doesn't sound so great.

> Seems to have fallen out of favour.

I don't know about falling out of favour (i.e. over time) but in more
complicated systems, programmers seem to prefer pre-emptive
multitasking, maybe because it's harder to hang the system through a
single task locking up, things like your database don't stop the
scheduler, etc.  I've never used a cooperative tasker so don't know when
the issues start appearing.