"C Is Not a Low-level Language"

[Lynn McGuire]

"C Is Not a Low-level Language"
https://queue.acm.org/detail.cfm?id=3212479

"Your computer is not a fast PDP-11."

Sigh, another proponent of "C sucks".

Lynn

[Öö Tiib]

Thanks, I had strong feeling of becoming stupider as result of reading
that strange rant. Wasn't it irrelevant if violator and victim processes
of Meltdown/Spectre family of vulnerabilities were written in JavaScript
or in C? The author seemed to imply that it all was somehow fault of C. :/

My understanding so far was that the manufacturers of processors have
done their branch predictions, speculative executions and out-of-order
executions in dirty manner that leave side effects. If to carefully
dig that dirt then it is possible to get data of other processes or
kernel out of it. Attempts to compensate such vulnerabilities on software
level have been controversial.
"As it is, the patches are COMPLETE AND UTTER GARBAGE."
Linus Torvalds https://lkml.org/lkml/2018/1/21/192

[Alf P. Steinbach]

On 05.05.2018 11:37, Öö Tiib wrote:
> On Thursday, 3 May 2018 00:21:56 UTC+3, Lynn McGuire wrote:
>> "C Is Not a Low-level Language"
>> https://queue.acm.org/detail.cfm?id=3212479
>>
>> "Your computer is not a fast PDP-11."
>>
>> Sigh, another proponent of "C sucks".
>
> Thanks, I had strong feeling of becoming stupider as result of reading
> that strange rant. Wasn't it irrelevant if violator and victim processes
> of Meltdown/Spectre family of vulnerabilities were written in JavaScript
> or in C? The author seemed to imply that it all was somehow fault of C. :/

The author's main point seems to be that there is an architectural
monopoly situation for main processing, similar to (my analogy) DOS that
was horrible but dominated because for most users it was good enough and
there was a supporting infrastructure of knowledge and tools and
compatibility with others. He just notes (rightly) that C is adapted to
that architecture, and (incorrectly) that current programming practices
are adapted to C. He never mentions "von Neumann architecture", which is
what it is: the blame for the idea of total centralization of
processing, a conceptual single thread, should be placed on the Moore
school and Johann von Neumann, roughly 1946. Well, if we don't go
further back to Euclid, some 300 BC or thereabouts.

> My understanding so far was that the manufacturers of processors have
> done their branch predictions, speculative executions and out-of-order
> executions in dirty manner that leave side effects. If to carefully
> dig that dirt then it is possible to get data of other processes or
> kernel out of it. Attempts to compensate such vulnerabilities on software
> level have been controversial.
> "As it is, the patches are COMPLETE AND UTTER GARBAGE."
> Linus Torvalds https://lkml.org/lkml/2018/1/21/192

I wonder what became of all the wonderful parallel processing ideas of
the 1980's, like Linda and Lucid. I liked those. :)

Cheers!,

- Alf

[woodb...@gmail.com]

On Saturday, May 5, 2018 at 6:05:18 AM UTC-5, Alf P. Steinbach wrote:
> On 05.05.2018 11:37, Öö Tiib wrote:
> > On Thursday, 3 May 2018 00:21:56 UTC+3, Lynn McGuire wrote:
> >> "C Is Not a Low-level Language"
> >> https://queue.acm.org/detail.cfm?id=3212479
> >>
> >> "Your computer is not a fast PDP-11."
> >>
> >> Sigh, another proponent of "C sucks".
> >
> > Thanks, I had strong feeling of becoming stupider as result of reading
> > that strange rant. Wasn't it irrelevant if violator and victim processes
> > of Meltdown/Spectre family of vulnerabilities were written in JavaScript
> > or in C? The author seemed to imply that it all was somehow fault of C. :/
>
> The author's main point seems to be that there is an architectural
> monopoly situation for main processing, similar to (my analogy) DOS that
> was horrible but dominated because for most users it was good enough and
> there was a supporting infrastructure of knowledge and tools and
> compatibility with others. He just notes (rightly) that C is adapted to
> that architecture, and (incorrectly) that current programming practices
> are adapted to C. He never mentions "von Neumann architecture", which is
> what it is: the blame for the idea of total centralization of
> processing, a conceptual single thread, should be placed on the Moore
> school and Johann von Neumann, roughly 1946. Well, if we don't go
> further back to Euclid, some 300 BC or thereabouts.

Call me old school, but single threaded applications are still
cool in 2018.


Brian
Ebenezer Enterprises
http://webEbenezer.net

[wyn...@gmail.com]

Lynn McGuire於 2018年5月3日星期四 UTC+8上午5時21分56秒寫道：

I used to think C as a high-level assembly language,
not bothered with what level it is attributed, as long as
it "close to the metal" and general enough, and some other
language feature (RAII) is desired, though.