[Oberon] Numerical CASE Statements in Project Oberon

[John R. Strohm]

Most “modern” compilers generate truly terrible code if optimization is turned off, and rely on the optimizers to clean it up.

I can’t remember now who wrote it, but Wirth’s compilers demonstrated pretty solidly that it was a lot easier, all around, to generate good code in the first place.  This reduced the ability of the optimizer to “improve” it, but the final code was frequently better than what was produced by the “modern” “production quality” compilers.  (The piece mentioned that only IBM’s RS-6000 compiler did better than Wirth’s Oberon compilers.)

The ARM instruction set is somewhat of a mess.  (It isn’t nearly as bad as the x86, but still...)

--John R. Strohm

From: Jan de Kruyf 
Sent: Wednesday, November 18, 2015 4:29 AM
To: ETH Oberon and related systems 
Subject: Re: [Oberon] Numerical CASE Statements in Project Oberon

I just picked up something else in the ADA best practice: 

"Use a sequence of assignments for an aggregation when measured performance indicates."


And I remember now looking at ARM machine code produced by the Ada compiler: The mess produced was quite drastic.

j.


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.inf.ethz.ch/pipermail/oberon/attachments/20151118/828a0dcf/attachment.html>