> would moving to another libc that supports the necessary C11 extensions
be a possibility?

Dont think so. glibc has such a monopoly on unix systems. And musl only
supports Linux as it seems.

The core problem I see is that threads and even atomic are an optional C11
feature. While I could live with the fact that we switch to C11 requiring a
toolchain that supports atomics, also requiring threads would vastly
decrease the amount of available toolchains (cross platform) as it seems
(at least nowadays).

So I really think that going with C++11 threads is the better option (would
require installation of MS Visual C++ library framework stuff, wouldnt
it?). For everything else we could use C11 (which would make building with
VisualStudio completely impossible??). I'm also lacking cross platform
experience with C11 and C++11.


Tom


2017-10-10 20:01 GMT+02:00 Marcus Weseloh <mweselo...@gmail.com>:

> Sorry, sent off too early. Let me try again...
>
> Hi all,
>
> Just a quick thought: would moving to another libc that supports the
> necessary C11 extensions be a possibility? I've read that musl has C11
> support, including the threading extensions. [1]
>
> Not sure how big the repercussions would be for the Linux and Mac OS
> ports, and if older targets like OS/2 would still be a possibility, though.
> Does anybody have experience with musl and can say anything about the glibc
> compatibility? musl itself has some documentation on that, which doesn't
> look too scary. [2]
>
> [1] http://www.etalabs.net/compare_libcs.html
> [2] http://wiki.musl-libc.org/wiki/Functional_differences_from_glibc
>
> Cheers,
>
>    Marcus
>
> Hi all,
>>
>> Just a quick thought: would moving to another libc that supports the
>> necessary C11 extensions be a possibility? I've read that musl has C11
>> support, including the threading extensions.
>>
>>
>> Not sure how big the repercussions would be for the Linux and Mac OS
>> ports,
>>
>> Am 10.10.2017 19:32 schrieb "Tom M." <tom.m...@googlemail.com>:
>>
>>> Without having seen your changes, you have probably done something like
>>> what existed before glib:
>>> https://github.com/FluidSynth/fluidsynth/blob/b49458e817ed09
>>> aedd948a6ea1831965373adc80/fluidsynth/src/fluid_sys.h#L142
>>>
>>> This is not an option because it's unmaintainable. See the related
>>> discussion here for possible solutions:
>>> http://lists.nongnu.org/archive/html/fluid-dev/2017-09/msg00031.html
>>>
>>>
>>> Tom
>>>
>>>
>>> 2017-10-10 18:57 GMT+02:00 Carlo Bramini <carlo.bra...@libero.it>:
>>>
>>>> Hello,
>>>> as I wrote in the subject, I worked a bit on Fluidsynth and I made a
>>>> version that works fine without external dependencies. I compiled it on the
>>>> ancient VisualStudio 6.0 and my version currentl runs on Windows.
>>>> However, it should not be difficult to include a new driver for adding
>>>> again support to GLIB at compile time or to add support for other
>>>> platforms, especially direct pthreads support.
>>>> I'm wondering if you would be interested to evaluate these changes and
>>>> how to do it if it could be possible to add them to the original source
>>>> code.
>>>> I have still some work to do on it, because some functions are still
>>>> not implemented and my coding started almost from version 1.1.7 (I cloned
>>>> the repository a bit before the release of the latest stable) but in the
>>>> meanwhile new patches have been submitted, so I should also align my code
>>>> at latest revision.
>>>>
>>>> Sincerely.
>>>>
>>>> _______________________________________________
>>>> fluid-dev mailing list
>>>> fluid-dev@nongnu.org
>>>> https://lists.nongnu.org/mailman/listinfo/fluid-dev
>>>>
>>>
>>>
>>> _______________________________________________
>>> fluid-dev mailing list
>>> fluid-dev@nongnu.org
>>> https://lists.nongnu.org/mailman/listinfo/fluid-dev
>>>
>>>
>
>
> --
> Marcus Weseloh
> Elektronische Musikinstrumente
> Fuhlendorfweg 27a
> 22589 Hamburg
> USt-ID: DE295961146
>
> mweselo...@gmail.com
> +49 (0)176 48816340 <+49%20176%2048816340>
> +49 (0)40 67301933 <+49%2040%2067301933> (nur Abends / only in the
> evening)
>
> _______________________________________________
> fluid-dev mailing list
> fluid-dev@nongnu.org
> https://lists.nongnu.org/mailman/listinfo/fluid-dev
>
>
_______________________________________________
fluid-dev mailing list
fluid-dev@nongnu.org
https://lists.nongnu.org/mailman/listinfo/fluid-dev

Reply via email to