Re: algorithm performance, was Add nested-function support in a language the based on a stack-machine

anton@mips.complang.tuwien.ac.at (Anton Ertl)
Sat, 24 Mar 2018 14:06:31 GMT

          From comp.compilers
Related articles
Re: algorithm performance robin51@dodo.com.au (Robin Vowels) (2018-03-27)
| List of all articles for this month |
From: anton@mips.complang.tuwien.ac.at (Anton Ertl)
Newsgroups: comp.compilers
Date: Sat, 24 Mar 2018 14:06:31 GMT
Organization: Institut fuer Computersprachen, Technische Universitaet Wien
References: <6effed5e-6c90-f5f4-0c80-a03c61fd2127@gkc.org.uk> 18-03-042 18-03-047 18-03-075 18-03-077 18-03-090 18-03-092
Injection-Info: gal.iecc.com; posting-host="news.iecc.com:2001:470:1f07:1126:0:676f:7373:6970"; logging-data="34913"; mail-complaints-to="abuse@iecc.com"
Keywords: performance, comment
Posted-Date: 24 Mar 2018 18:50:24 EDT

Andy Walker <anw@cuboid.co.uk> writes:
> Somewhat more to the point, as this is "comp.compilers",
>it seems unlikely that either the compiler or the source being
>compiled have grown by anything like the growth in RAM, despite the
>unfortunate modern [ie last 40-odd years!] tendency towards bloat.


One of the papers I saw during this recent discussion mentions that
one of Wirth's early compilers (PL/360 IIRC) self-compiled in IIRC 25s
on a machine of the time, i.e., late 1960s (which the author of the
paper found quite impressive). In 1992, I heard a keynote address by
Wirth (at CC'92), where he contrasted Oberon with C++, and used
self-compilation time as one metric: The Oberon compiler compiled
itself in IIRC 6s, which he contrasted with IIRC 20 minutes for the
C++ compiler.


The last few times that I built a recent gcc, it took several hours to
compile itself (three stages, and it included several front-ends, but
still); interestingly, I rebuilt gcc-2.7 in 2015, and that took only a
few minutes.


Software bloats to fill the memory and run-time that is deemed
acceptable. Whether that means more or less sophistication in the
algorithms varies from case to case.


- anton
--
M. Anton Ertl
anton@mips.complang.tuwien.ac.at
http://www.complang.tuwien.ac.at/anton/
[Back in the 1970s, the Dartmouth BASIC compiler was so fast that
nobody bothered to save object code. It rarely took more than a few
clock ticks from source code to starting the executable. -John]

Post a followup to this message

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