Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Lawrence D'Oliveiro Newsgroups: comp.arch Subject: Re: Article on new mainframe use Date: Mon, 2 Sep 2024 03:28:52 -0000 (UTC) Organization: A noiseless patient Spider Lines: 11 Message-ID: References: <5GkAO.84916$%Go3.29106@fx12.iad> <20240830183742.000065c5@yahoo.com> <2377cad7ee947ad71c9c3a8afbcdc26e@www.novabbs.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Mon, 02 Sep 2024 05:28:52 +0200 (CEST) Injection-Info: dont-email.me; posting-host="0461de2c4cabc231a24885f70066d9c2"; logging-data="1935273"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+KhvocLz/XONvteUAIFYlm" User-Agent: Pan/0.160 (Toresk; ) Cancel-Lock: sha1:LGV7EQNLsSdUaROcvOee2CRd2SU= Bytes: 1784 On Sun, 1 Sep 2024 00:52:34 +0000, MitchAlsup1 wrote: > Imagine trying to fit LLVM or GCC into a PDP/11 address space. Pretty much from the moment the PDP-11 range was introduced, it was obvious the 16-bit address space was going to be a significant limitation. The main OS I was familiar with on those machines was RSTS/E. In its later versions it introduced some wondrous hacks to try to increase the available address space for programs, making use of various features of the PDP-11 in ways they were never quite meant to be used.