Intel Fully Embracing LLVM For Their C/C++ Compilers

[grigi]

No, there is not some big grand conspiracy by corporations to not contribute to open source. Virtually all major corporations are contributing to open source, even more so now. These days you see nothing but swaths of projects being published under MIT or BSD or similar licenses by every software and service company out there. The reason nobody wants to contribute to GCC is because it's GPL which is, from a legal standpoint, the most annoying and obnoxious license ever conceived and appreciated by nobody except literal software cultists who think software deserves more rights than human beings.
People contribute to projects published under licenses that respect their rights and freedom because they want the freedom to do what they want, not whatever the original publisher wants. GPL is not open source, it's explicitly, vocally not open source, it's practically an educational license or EULA which is why you see Windows installers for GPL'd software put the GPL as a "EULA section" in the installer. When you realize this, you'll see people actively fighting for open source software by trying to make new tools free from the cancer of GPL.

I didn't say anything about a conspiracy. There is no conspiracy. Just human nature to keep control.

[discordian]

Good point, particularly taking into account the main point of ICC is it used various IS/Microcode hacks/tweaks to squeeze out that extra performance for Intel CPU's where as LLVM is specifically designed to be ISA agnostic.

Guess the most interesting part is that you can just plug your stuff in immediately without spending time getting upstream approval. Given all the various places LLVM/Clang is used, hunting down regressions must be a nightmare.

[uid313]

Now that Intel is burying ICC in the graveyard I would like to see they open source it.
Maybe someone can study it and learn something from it.

[Rallos Zek]

No, there is not some big grand conspiracy by corporations to not contribute to open source. Virtually all major corporations are contributing to open source, even more so now. These days you see nothing but swaths of projects being published under MIT or BSD or similar licenses by every software and service company out there. The reason nobody wants to contribute to GCC is because it's GPL which is, from a legal standpoint, the most annoying and obnoxious license ever conceived and appreciated by nobody except literal software cultists who think software deserves more rights than human beings.
People contribute to projects published under licenses that respect their rights and freedom because they want the freedom to do what they want, not whatever the original publisher wants. GPL is not open source, it's explicitly, vocally not open source, it's practically an educational license or EULA which is why you see Windows installers for GPL'd software put the GPL as a "EULA section" in the installer. When you realize this, you'll see people actively fighting for open source software by trying to make new tools free from the cancer of GPL.

MIT and BSD licenses are for corporate bootlickers and people who do not care for user freedom. MIT and BSD license promoters are the slavery apologists of the 21st century.

[Gonk]

<snip>
When you realize this, you'll see people actively fighting for open source software by trying to make new tools free from the cancer of GPL.

I'd characterize it as more like an STD.

[thebear]

So I tried this on a project swapping CXX to icpx. I note that C++ concepts not yet supported, but that's OK. I simply added a bunch of conditional macros based on __cpp_concepts >= 201907L.

What was more surprising though was that __FAST_MATH__ macro was defined even for -O0 (I have some code which is invalidated if compiled with imprecise floating point arithmetics). So I had to explicitly pass -fp-model precise for the macro not to be defined. It would be a shame if those reported floating point performance gains over gcc/clang just boils down to more aggressive default optimization levels...

[icaci]

Here's a strange detail. In the last q/a of the techdecoded webinar, "Introducing the Next Gen of Intel® Parallel Studio: Transitioning to the Latest HPC Software Development Suite

", they state that icx and ifx will still need to be paired with gcc headers and glibc library.

Why is that strange? GLIBC is the standard system library on Linux and GCC is the standard system compiler there. Intel doesn't need to constantly follow up and replicate what is already provided by the GCC/GLIBC combo. If Intel were to make their own system library that is completely hosted in their own tools, you won't need GCC or GLIBC headers.

[coats]

...The reason nobody wants to contribute to GCC is because it's GPL which is, from a legal standpoint, the most annoying and obnoxious license ever conceived and appreciated by nobody except literal software cultists who think software deserves more rights than human beings...

No, the reason I use GPL is that I don't want 419701 lines of my (environmental modeling) code stolen, with my attribution removed.

[carewolf]

So I tried this on a project swapping CXX to icpx. I note that C++ concepts not yet supported, but that's OK. I simply added a bunch of conditional macros based on __cpp_concepts >= 201907L.

What was more surprising though was that __FAST_MATH__ macro was defined even for -O0 (I have some code which is invalidated if compiled with imprecise floating point arithmetics). So I had to explicitly pass -fp-model precise for the macro not to be defined. It would be a shame if those reported floating point performance gains over gcc/clang just boils down to more aggressive default optimization levels...

They are.

The other advantages icc used to have, like autovectorization are largely caught up to. (except good vectorization of FP commands require fast-math)

[aspen]

...The reason nobody wants to contribute to GCC is because it's GPL which is, from a legal standpoint, the most annoying and obnoxious license ever conceived and appreciated by nobody except literal software cultists who think software deserves more rights than human beings...

No, the reason I use GPL is that I don't want 419701 lines of my (environmental modeling) code stolen, with my attribution removed.

I mean... GPL isn't the only license which forbids that? I personally prefer MPLv2, which is similar to LGPL, but not viral to the entire project - if you change my MPL'd code, you have to open source your changes to it, but idc about your code, go do whatever with it.

There's the Zlib license, which pretty much forbids people from pretending that they made something that they didn't. Pretty sure most BSD licenses require attribution of some sort, too.