C11 Compiler Test Suite

[James Kuyper]

....
The point is not whether you can perform some task with previous C
standards or not. It is whether you can perform those tasks in an easier
manner.

"Allows performing tasks to be performed more easily" != "essential".
There's only a few specialized tasks that can be performed more easily
in C2011 than in C99. It's not even as big of a change as C99 was, and
that was a fairly minor update.

For instance threading in C11 promises to be much easier as you
will have a threading library that is cross platform compatible built
into the language itself rather than having to rely on third party
implementations. Writing code that works on Windows and UNIX will be
much easier.

Will it? I don't write multi-threaded code at all, so I have no idea.
But I know that <threads.h> only implements a sub-set of the features of
either POSIX pthreads, or the Windows equivalent (I've no idea what it's
name is). What fraction of current multi-threaded programs use only
those features of POSIX (or Windows) for which there's a <threads.h>
equivalent?

....

I would be highly surprised if an upgrade to C11 in a compiler would
break existing code especially as most compilers (all that I know of)
allow you to specify the standard you wish to compile your code against
(with the -std=cxx command line switch).

You haven't actually upgraded to C2011, even if your compiler supports
it, if you need to use -std=c99 to make your code compile.

If it was shown that a particular compiler broke existing code due to an
upgrade to C11 I would imagine that the compiler vendor would treat that
problem with the up-most priority.

Many of the new features of C99 that were mandatory are now optional in
C2011. If a vendor took advantage of that fact to cease supporting one
of those features, I doubt that they would consider the lack of support
to be a problem of any kind, much less a high priority one. That
wouldn't make code that used those features any less broken.

....

Let us not forget that C compiler vendors are used to implementing new C
standards. ...

Not really - for most of them, the last time they fully implemented a
new C standard was 1995 or earlier. Even the more up-to-date compilers
haven't had to implement a new C standard since 2000, which is a fairly
long time in the IT world. Many of the people who did the design work in
2000 may have either retired, been laid off, or been promoted into
management by now.

 

[88888 Dihedral]

(e-mail address removed)æ–¼ 2012å¹´10月6日星期六UTC+8上åˆ6時15分35秒寫é“：

I suppose it depends on how you count, but the only mandatory C99

features that are optional in C11 are complex support and VLAs.



I must admit I don't get the opposition to complex support (and that

was even optional for freestanding C99 implementations). The compiler

support needed is so minimal. For VLAs I can at least imagine

platforms where support would be excessively difficult.

Well if there are no new cpus with some new instruction sets in a few years,
the GCC 90 and 99 standards are good enough for those old instruction sets..

 

[Chicken McNuggets]

Truth be told, pthreads already provided a way to write multi-threaded code
in an OS-independent fashion for nearly a decade, now. The only thing that
blocked pthreads' adoption was Microsoft's unwillingness to support it.
Now, looking at Microsoft's notorious unwillingness to comply with the C
standard, I believe that C11 won't change that as well.

Unfortunately Windows is a required platform for a lot of developers and
the only way to get a decent C compiler on Windows is to use MinGW or
the Intel C and C++ compiler suite (I have heard that Clang is making
progress on the Windows front as well so that may well be an option in
the not too distant future as well).

All of the above options should support C11 given enough time. The Intel
compiler in particular was very good at C99 support and was one of the
first (if I remember correctly) to support the vast majority of the
standard. So fingers crossed the same will be true for C11.

 

[Robert Miles]

On 10/05/2012 02:33 PM, Chicken McNuggets wrote: [snip]

(Apologies if the quoting above appears broken. Thunderbird isn't the
best NNTP client in the world)

The quoting does not appear to be broken. Have you seen Thunderbird
generate quotation errors? If, of what kind?

I've seen Thunderbird give two kinds of errors or peculiarities:

1. It does not follow the Google Groups method of marking ends of
lines well.

2. It does not follow the quoting by indentation method often used in
this newsgroup, but rarely if ever in the other newsgroups I read.

It works better in other newsgroups, when replying to threads that
contain no Google Groups posts.

 

[James Kuyper]

On 10/05/2012 02:33 PM, Chicken McNuggets wrote: [snip]

(Apologies if the quoting above appears broken. Thunderbird isn't the
best NNTP client in the world)

The quoting does not appear to be broken. Have you seen Thunderbird
generate quotation errors? If, of what kind?

I've seen Thunderbird give two kinds of errors or peculiarities:

1. It does not follow the Google Groups method of marking ends of
lines well.

That's arguably a problem with Google Groups, rather than Thunderbird.
What are those two methods?

2. It does not follow the quoting by indentation method often used in
this newsgroup, but rarely if ever in the other newsgroups I read.

Could you demonstrate the two different quotation methods you're
referring to? I'm using Thunderbird to create my messages, and my
quotation of your message, your quotation of my message, and my earlier
quotation of Chicken McNuggets' message all appear, in Thunderbird, to
use the same consistent quotation method: in the display window,
vertical lines of multiple colors run along the left edge of the quoted
material, the number of lines indicating the depth of quotation. In the
message editing window, those lines are replaced by the same number of
'>' characters. That's the same method I see in when using Google Groups
to view your message. I believe I once read that there was an ISO
standard specifying that method of quotation, though I'm not sure how to
confirm that.
Reviewing the messages in this thread, I do find one other quotation
method, one that puts spaces between the '>' characters. It appears in
one of my own messages, but I've no idea why.

 

[Keith Thompson]

On 10/05/2012 02:33 PM, Chicken McNuggets wrote: [snip]

(Apologies if the quoting above appears broken. Thunderbird isn't the
best NNTP client in the world)

The quoting does not appear to be broken. Have you seen Thunderbird
generate quotation errors? If, of what kind?

I've seen Thunderbird give two kinds of errors or peculiarities:

1. It does not follow the Google Groups method of marking ends of
lines well.

What method is that? Many posts I've seen from Google Groups *don't*
mark ends of lines.

2. It does not follow the quoting by indentation method often used in
this newsgroup, but rarely if ever in the other newsgroups I read.

What "quoting by indentation" method? Quotes from previous articles are
marked with ">", "> >", etc., here and in other newsgroups.

It works better in other newsgroups, when replying to threads that
contain no Google Groups posts.

Google Groups is badly broken; I don't think Thunderbird can be blamed.

 

[Stephen Sprunk]

On 10/05/2012 02:33 PM, Chicken McNuggets wrote: [snip]

(Apologies if the quoting above appears broken. Thunderbird isn't the
best NNTP client in the world)

The quoting does not appear to be broken. Have you seen Thunderbird
generate quotation errors? If, of what kind?

I've seen Thunderbird give two kinds of errors or peculiarities:

1. It does not follow the Google Groups method of marking ends of
lines well.

There is some disagreement in the community about flowed format, which I
suspect is what you're talking about. That's not limited to GG, and
T-bird's behavior is configurable.

2. It does not follow the quoting by indentation method often used in
this newsgroup, but rarely if ever in the other newsgroups I read.

It doesn't? My T-bird does that just fine, though I often have to make
use of Edit->Rewrap to make it look pretty.

It works better in other newsgroups, when replying to threads that
contain no Google Groups posts.

I've never noticed any problems specific to GG posts.

S