Google Developers Are Looking At Creating A New libc For LLVM

[phoronix]

Phoronix: Google Developers Are Looking At Creating A New libc For LLVM

As part of Google's consolidating their different toolchains around LLVM, they are exploring the possibility of writing a new C library "libc" implementation...

http://www.phoronix.com/scan.php?page=news_item&px=Google-New-LLVM-Libc-Thoughts

[rene]

yay, re-inventing the wheel once again, one has to wonder what is wrong with musl libc though? not good enough, how? https://www.youtube.com/watch?v=NMQrnVDf-rY

[milkylainen]

Umm. Surely you create a libc for an OS with a preferably standardized interface?
I don't get why LLVM has anything to do with it? Why wouldn't one be able to use the new "libc" with GCC for example?

[bug77]

I hope they write it in Rust, not in Go

[Venemo]

I thought google already has their own libc, called “bionic” - isn't that the case?

[crystall]

I thought google already has their own libc, called “bionic” - isn't that the case?

That's what they use on Android. The only design goal was to avoid a GPL-licensed libc. The resulting library is mostly garbage.

[anarsoul]

Don't be evil, heh?

[Filiprino]

I will always side with the GNU.

OTOH this paragraph looks awful:

We intend to build the new libc in a gradual manner. To begin with, the new libc will be a layer sitting between the application and the system libc. Eventually, when the implementation is sufficiently complete, it will be able to replace the system libc at least for some use cases and contexts.

First as an interposer and later a replacement for some use cases. To me it seems like a library for emulation/testing. Except if we take that when it is beyond "sufficently complete", it will be able to replace the whole system libc.

Anyways, Google is trying hard to skip all GPL code as soon as possible.

[MartinN]

I hope they write it in Rust, not in Go

Rust would be the logical choice.

[OpenProgger]

yay, re-inventing the wheel once again, one has to wonder what is wrong with musl libc though? not good enough, how? https://www.youtube.com/watch?v=NMQrnVDf-rY

Altough musl is slower than glibc in some cases, a whole desktop system compiled with musl have nearly 50% less RAM usage. I don't get it why google dislikes dynamic loading/linking but this would be a reason why it wouldn't be an alternative to any current libc libraries. Even musl supports it, though its focused on static linkage (and do it well).