[c++-pthreads] Re: thread-safety definition

[Dave Butenhof]

Alexander Terekhov wrote:

>Fergus Henderson wrote:
>  
>
>>On 13-Jan-2004, Alexander Terekhov <boo at terekhov.de> wrote:
>>    
>>
>>>Ted Baker wrote:
>>>      
>>>
>>>>On Mon, Jan 12, 2004 at 07:19:42PM +0100, Alexander Terekhov wrote:
>>>>        
>>>>
>>>>>Now, in your model with cancellation ALWAYS disabled while running
>>>>>destructors (not only when acting upon a cancel request delivery...
>>>>>thread_exit aside for a moment), I'd have to add enable/disable RAII
>>>>>          
>>>>>
>>>>Please translate "RAII" for me?
>>>>        
>>>>
>>>http://www.terekhov.de/DESIGN-futex-CV.cpp
>>>      
>>>
>>Alexander Terekhov again fails to communicate,
>>and when pressed substitutes a unrelated URL.
>>    
>>
>Related URL.
>  
>
Yes, related; but indirectly, and with no context or explanation of the 
relationship. Only those who already understand the abbreviation could 
have understood the relevance, and they didn't need the example. That's 
why Fergus claims (correctly) that you "failed to communicate". 
Communication is not the trivialized process of throwing "data balls" at 
someone, Alexander. Information content is required, and it must further 
be structured in a way that leads to understanding.

>>Ted, when he says "RAII", he means "Resource Acquision is Initialization".
>>    
>>
>To me, this doesn't say much at all in the context of enabling/disabling 
>and/or disabling/enabling cancellation via "RAII guard objects". The link 
>OTOH does illustrate it (e.g. see "class cancel_off_guard").
>  
>
OK, so let's be more specific, Alexander. You would have been fine if 
you'd explained what the abbreviation stands for, what that phrase 
means, AND offered the link to some code showing how it's used. But 
while your link used the abbreviation "RAII" once or twice, it never 
DEFINED oir EXPLAINED the term and was therefore not particularly useful 
to anyone not already familiar with it.

>>This is the name of a C++ idiom where code which acquires/releases
>>a resource is implemented as the constructor/destructor of a local object.
>>    
>>
>http://www.artima.com/intv/modern3.html
>  
>
Did you notice the interviewer didn't even get the full phrase right in 
the initial question? ;-)

-- 
/--------------------[ David.Butenhof at hp.com ]--------------------\
| Hewlett-Packard Company       Tru64 UNIX & VMS Thread Architect |
|     My book: http://www.awl.com/cseng/titles/0-201-63392-2/     |
\----[ http://homepage.mac.com/dbutenhof/Threads/Threads.html ]---/