Re: Wrestling with phase 1 of a C compiler

gah4 <gah4@u.washington.edu>
Wed, 14 Sep 2022 15:40:33 -0700 (PDT)

          From comp.compilers
Related articles
Wrestling with phase 1 of a C compiler luser.droog@gmail.com (luser droog) (2022-09-07)
Re: Wrestling with phase 1 of a C compiler luser.droog@gmail.com (luser droog) (2022-09-09)
Re: Wrestling with phase 1 of a C compiler luser.droog@gmail.com (luser droog) (2022-09-11)
Wrestling with phase 1 of a C compiler christopher.f.clark@compiler-resources.com (Christopher F Clark) (2022-09-12)
Re: Wrestling with phase 1 of a C compiler gah4@u.washington.edu (gah4) (2022-09-12)
Re: Wrestling with phase 1 of a C compiler christopher.f.clark@compiler-resources.com (Christopher F Clark) (2022-09-13)
Re: Wrestling with phase 1 of a C compiler luser.droog@gmail.com (luser droog) (2022-09-14)
Re: Wrestling with phase 1 of a C compiler gah4@u.washington.edu (gah4) (2022-09-14)
Re: Wrestling with phase 1 of a C compiler luser.droog@gmail.com (luser droog) (2022-09-15)
| List of all articles for this month |
From: gah4 <gah4@u.washington.edu>
Newsgroups: comp.compilers
Date: Wed, 14 Sep 2022 15:40:33 -0700 (PDT)
Organization: Compilers Central
References: 22-09-001 22-09-004 22-09-005 22-09-006
Injection-Info: gal.iecc.com; posting-host="news.iecc.com:2001:470:1f07:1126:0:676f:7373:6970"; logging-data="98209"; mail-complaints-to="abuse@iecc.com"
Keywords: performance, history
Posted-Date: 15 Sep 2022 12:18:26 EDT
In-Reply-To: 22-09-006

On Wednesday, September 14, 2022 at 1:25:39 PM UTC-7, christoph...@compiler-resources.com wrote:


(snip)


> On a related note, I have heard stories from C++ compiler
> implementors about the various template libraries that have been created
> which attempt to do "Turing machine" style (NP-complete) computations via
> types and parameters, where the users wonder why the compilation process
> takes much longer than running the resultant program.


Many years ago, there was a story about a Fortran benchmark program,
very complicated with lots of statement functions evaluating many
complicated expressions.


(I probably don't have to mention the problems with
designing good benchmarks here. This was years ago.)


Then it was run though the IBM OS/360 Fortran H compiler.
Among others, it expands statement functions inline, and does
constant expression evaluation. It did the whole thing at
compile time, except printing out a single number.


Fortran now requires many complicated constant expressions
be evaluated at compile time, and some have come up with some,
though not intentionally, that evaluate very slowly. (OK, maybe
some were intentional.)

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.