Sunday, November 17, 2019

Digest for comp.lang.c++@googlegroups.com - 14 updates in 5 topics

Sam <sam@email-scan.com>: Nov 17 07:59AM -0500

Bonita Montero writes:
 
 
>> Right. I clearly remember how everyone in the schoolyard always pretended to
>> be experts on heap allocation strategies.
 
> I was talking about you, and you know that.
 
Highly-trained professionals have a word for this. It's called "projection".
Bonita Montero <Bonita.Montero@gmail.com>: Nov 17 04:06PM +0100


>> I was talking about you, and you know that.
 
> Highly-trained professionals have a word for this. It's called
> "projection".
 
You're not so professional to determine when someone projects
something. You're simply an idiot.
Sam <sam@email-scan.com>: Nov 17 04:27PM -0500

Bonita Montero writes:
 
 
>> Highly-trained professionals have a word for this. It's called "projection".
 
> You're not so professional to determine when someone projects
> something. You're simply an idiot.
 
And what exactly would be your professional qualifications for diagnosing
someone to be an idiot? If you're going to stake a position that
professional qualifications are required for a psychological evaluation –
such as whether someone's projecting, or not – shirley you then have to
prove your professional qualifications for diagnosing a case of idiocy.
 
I'm very convinced that if you demand others to uphold to the highest
standards of professional qualifications, in order to accuse someone of
projection, you most certainly will meet your own standards and be able to
demonstrate your own professional qualifications for determining if
someone's an idiot.
 
So, let's see them (asking for a friend).
Manfred <noname@invalid.add>: Nov 17 04:46PM +0100

On 11/17/19 12:51 AM, Alf P. Steinbach wrote:
> {
>     foo( "Baluba!" );
> }
 
Bansky coding?
"Alf P. Steinbach" <alf.p.steinbach+usenet@gmail.com>: Nov 17 12:28PM +0100

On 17.11.2019 02:54, Mr Flibble wrote:
> Hi!
> [snip] if Brazil is a good film or not?
 
It's a classic, but to enjoy it one must have at least some residual
freedom urge and residual ability to think about things.
 
One person in my family, with an amazing almost photographic memory, but
old now and throughout her life a strict flock follower, i.e. she has a
very strong herd instinct, disliked it intensely. I think her instincts
made her blind to many aspects of the story in the movie, so much that
it became meaningless to her. That phenomenon has also happened in this
group, then about other things: herd instinct makes blind, so that the
responses are not even wrong, just technically completely baffling.
 
Btw. this is not the elderly lady that Francis Glassborow used to
test-read his introductory C++ text book. I've never met Francis. That's
a different family (if his reader was family, which I suspect).
 
 
- Alf
Mr Flibble <flibbleREMOVETHISBIT@i42.co.uk>: Nov 17 03:21PM

On 17/11/2019 11:28, Alf P. Steinbach wrote:
 
> Btw. this is not the elderly lady that Francis Glassborow used to
> test-read his introductory C++ text book. I've never met Francis. That's a
> different family (if his reader was family, which I suspect).
 
Oh I love the film Brazil which is why it doesn't require much thought on
my part on deciding if it is a good film or not. :D
 
/Flibble
 
--
"Snakes didn't evolve, instead talking snakes with legs changed into
snakes." - Rick C. Hodgin
 
"You won't burn in hell. But be nice anyway." – Ricky Gervais
 
"I see Atheists are fighting and killing each other again, over who
doesn't believe in any God the most. Oh, no..wait.. that never happens." –
Ricky Gervais
 
"Suppose it's all true, and you walk up to the pearly gates, and are
confronted by God," Byrne asked on his show The Meaning of Life. "What
will Stephen Fry say to him, her, or it?"
"I'd say, bone cancer in children? What's that about?" Fry replied.
"How dare you? How dare you create a world to which there is such misery
that is not our fault. It's not right, it's utterly, utterly evil."
"Why should I respect a capricious, mean-minded, stupid God who creates a
world that is so full of injustice and pain. That's what I would say."
"Öö Tiib" <ootiib@hot.ee>: Nov 17 01:52AM -0800

On Sunday, 17 November 2019 10:17:23 UTC+2, Bonita Montero wrote:
> > itself to decide. ...
 
> We're not talking about certain implementations but the language.
> And unaligned accesses simply aren't portable.
 
I wasn't tslking about concrete compiler. Standard is specification
of C++ compilers and I was writing what standard allows and requires
conforming compilers to do. The compilers that did not do what
standard requires have gradually lost their market share.
 
> 2. std::atomic<T> thereby has also a natural alignment.
> 3. is_lock_free couldn't be dynamic because of runtime alignment
> properties because C++ doesn't support unaligned acceses anyway.
 
I understand very well what you write. It is just wrong.
That "natural alignment" does not exist in C++. Quote from where you
manufactured that misconception? Bald, groundless and wrong assertion
in 1. is making also conclusions from it in 2. and 3. wrong and whole
chain of your logic nonsense.
Bonita Montero <Bonita.Montero@gmail.com>: Nov 17 11:23AM +0100

>> properties because C++ doesn't support unaligned acceses anyway.
 
> I understand very well what you write. It is just wrong.
> That "natural alignment" does not exist in C++.
 
The implementation-defined natural alignment can be determined with
alignof(). If you write your code so that it will satisfy this alignment
on one platform and not another it is not portable. So each platform
knows its own alignment-requirements and these are also maintained for
the atomic-types. As long as you use proper alignment for your atomics
there coudln't be any alignement-issues. So there's no need for the
standad to mandate support for unaligned atomics through is_lock_free.
"Öö Tiib" <ootiib@hot.ee>: Nov 17 03:22AM -0800

On Sunday, 17 November 2019 12:23:58 UTC+2, Bonita Montero wrote:
> > That "natural alignment" does not exist in C++.
 
> The implementation-defined natural alignment can be determined with
> alignof().
 
Then read up on alignof in [basic.align]. It is returning alignment
requirement for complete object (and not for subobject).
 
> If you write your code so that it will satisfy this alignment
> on one platform and not another it is not portable.
 
Sure, I (as programmer) may not start life-time of whole object
of type T in buffer that is more weakly aligned than alignof(T)
requires.
 
> So each platform
> knows its own alignment-requirements and these are also maintained for
> the atomic-types.
 
You are mixing up. The alignof(T) is telling what it is requiring
from programmers. It is not promising that subobjects of type T of
composite objects are aligned by alignof(T).
 
> As long as you use proper alignment for your atomics
> there coudln't be any alignement-issues. So there's no need for the
> standad to mandate support for unaligned atomics through is_lock_free.
 
And here it is totally unclear who you address as "you". Are you
addressing programmers, compilers, standards or whole reality that
does not behave by your silly misconceptions?
Bonita Montero <Bonita.Montero@gmail.com>: Nov 17 02:20PM +0100

>> alignof().
 
> Then read up on alignof in [basic.align]. It is returning alignment
> requirement for complete object (and not for subobject).
 
I didn't talk about subobjects.
 
 
> You are mixing up. The alignof(T) is telling what it is requiring
> from programmers. It is not promising that subobjects of type T of
> composite objects are aligned by alignof(T).
 
Of course the compiler meets the requirements it also imposes on the
programmer.
"Öö Tiib" <ootiib@hot.ee>: Nov 17 06:39AM -0800

On Sunday, 17 November 2019 15:20:21 UTC+2, Bonita Montero wrote:
 
> > Then read up on alignof in [basic.align]. It is returning alignment
> > requirement for complete object (and not for subobject).
 
> I didn't talk about subobjects.
 
Indeed why you wiggle away from those for whole week?
Where it is written that atomic<T> is unsuitable to be type of subobject
of other object?
 
> > composite objects are aligned by alignof(T).
 
> Of course the compiler meets the requirements it also imposes on the
> programmer.
 
Standard explicitly tells that the subobjects might be aligned
differently than whole objects and now you say that these "of course"
are required to be aligned by alignof and may not be aligned
differently? :D
 
I think you are just pretending to be so utterly bone-headed so I give
up attempts of teaching you.
Bonita Montero <Bonita.Montero@gmail.com>: Nov 17 03:49PM +0100

>> I didn't talk about subobjects.
 
> Indeed why you wiggle away from those for whole week?
 
That's what I wrote and where you responoded with subobjects-whatever:
 
The implementation-defined natural alignment can be determined with
alignof(). If you write your code so that it will satisfy this alignment
on one platform and not another it is not portable. So each platform
knows its own alignment-requirements and these are also maintained for
the atomic-types. As long as you use proper alignment for your atomics
there coudln't be any alignement-issues. So there's no need for the
standad to mandate support for unaligned atomics through is_lock_free.
 
Where did I talk about subobjects there?
 
> Where it is written that atomic<T> is unsuitable to be type of subobject
> of other object?
 
Quote where Is said this.
 
> differently than whole objects and now you say that these "of course"
> are required to be aligned by alignof and may not be aligned
> differently? :D
 
The standard says that all objects of complete types have a poper
alignment.
Bonita Montero <Bonita.Montero@gmail.com>: Nov 17 03:58PM +0100

> Nonsense. What is aligned or misaligned is up to compiler
> itself to decide.
 
Wrong. If you pack your members tighter than alignof(membertype) allows
this is not conforming.
David Brown <david.brown@hesbynett.no>: Nov 17 11:12AM +0100

On 17/11/2019 09:12, Bonita Montero wrote:
 
>> No, you didn't.  There are many ways to describe you based on your
>> postings - "stupid" is not one of them.
 
> You can't read my mind.
 
Right... so you are stupid, not nasty? Okay, I'll take your word for it.
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)