[SystemSafety] Candidates for the firing squad

Andy Ashworth andy at the-ashworths.org
Thu Sep 20 16:21:40 CEST 2018


	
		
		
	
		
		If the claim of a lint-type tool being static analysis is a “WTF Moment” then I dread to think what reaction you will have to this... 
A few years ago, I was reviewing claims of SIL4 compliance (against EN50128) made for some software for a train control system. When challenged as to which static analysis tool had been used, the answer was “peer review”... it wasn’t even a structured Fagan type review, just another engineer reading the code and signing off as acceptable! This is what can happen when a standard is retrospectively applied by non-experts to an existing mature development process. 
Andy


		

		Get Outlook for iOS
	From: 30422465340n behalf of 
Sent: Thursday, September 20, 2018 10:10
To: Olwen Morgan
Cc: systemsafety at lists.techfak.uni-bielefeld.de
Subject: Re: [SystemSafety] Candidates for the firing squad On 2018-09-20 13:55:06 +0100, Olwen Morgan wrote:
>    ... WTF? ....

Several (many) years ago, a static analyser was a program that could
analyse your code, produce call trees, detect recursion and even
perform basic analysis (range checks, paths etc.) of what would happen
during execution (although doing this properly essentially means
solving the halting problem). These days there are even better
versions of these tools - still with limitations, but computers
are much faster and have more memory the standard can be higher.

Simple source code checkers (lint etc.) were called just that.
MISRA-C checkers would have fallen into this category, had they
existed.

These days, the name "static analysis" seems to have been misappropriated
to cover the whole gamut of source code verification tools.

The problem with that is, when the standard says:
   "Perform static analysis - highly recommended"
the under-skilled people who are charged with verifying that the
development process follows the standard put this in their
Excel sheet:
   "MISRA-C checker  [tick]"

Another WTF moment :-(

>    Now that my lithium is kicking in (seriously, I do have to take it),
>    the manic postings will subside.

Oh, I hope not ...

:-)

Dave


-- 
David Haworth   OS Kernel Developer              david.haworth at elektrobit.com
Elektrobit Automotive GmbH                            Tel: +49 9131 7701-6154
Am Wolfsmantel 46, 91058 Erlangen, Germany            Fax: +49 9131 7701-6333
Geschäftsführer: Alexander Kocher, Gregor Zink     Amtsgericht Fürth HRB 4886
_______________________________________________
The System Safety Mailing List
systemsafety at TechFak.Uni-Bielefeld.DE


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20180920/f6ffec73/attachment.html>


More information about the systemsafety mailing list