LLVM Clang 16 vs. GCC 13 Compiler Performance On Intel Raptor Lake

[akostadinov]

Wanted to see what compiles faster, ont the resulting binaries...

[DanglingPointer]

When are we getting this... ???
11.4 vs 12.2 vs 12.3 vs 13.1 vs llvm-15 vs llvm-16
with:

• O2
• O3
• O2 -flto
• O3 -flto
• O3 -flto=thin (llvm)
• march=native O2
• march=native O3
• march=native O2 -flto
• march=native O3 -flto
• march=native O3 -flto=thin (llvm)

with AMD

[brad0]

Looks like OpenMandriva's idea of switching the entire OS to building with clang will finally pay off in terms of performance.

It already has for some time.

[berolinux]

Looks like OpenMandriva's idea of switching the entire OS to building with clang will finally pay off in terms of performance.

[Weasel]

TBH this whole ELF interposition should've never been the default but an explicit opt-in. Its not used anywhere outside of hooking stuff into libc

It's due to ELF's garbage global symbols. PE in contrast can't even store them. You can still hook there, it just requires more effort, and who cares? Most people who hook are experienced enough for it to not matter, unlike 99.999% of users who will suffer because of ELF's stupid design decisions.

[brad0]

Does clang compile kernel yet? If not it can lick GCC boots.

It does and with LTO. GCC is the one licking Clang's boots.

[Volta]

marios, discordian

It would be interesting to see Linux GCC vs Linux Clang then.

[discordian]

Both GCC and Clang require -fvisibility=hidden to not export the symbol. Making -fvisibility=hidden a default would break a lot of existing packages that does not mark exported symbols.

Even more packages wont care, and the remaining deserve to be fixed (the C/C++ standard knows nothing about public/hidden symbols, and depending on compiler historic compiler behavior aint a good idea). TBH, I thought the visibility default were a configuration option when compiling gcc, seems I remembered this wrong - using mostly toolchains with wrappers and buildsystems where -fvisibility=hidden is used transparently.

If you compile the testcase with -fvisibility=hidden, GCC will optimize return to 1. GCC has quite careful logic about what can be interposed:

1. Interposition is possible only with -fpic or weak symbols
   

1. And that comes with a cost: inefficent code and relocations.
   

2. Inline functions or comdats can not be be semantically interposed (but can be interposed to different, equivalent definition possibly optimized with different optimization flags or compiled by other compiler)
3. With -flto, one can use resolution file to hide symbols at link-time (as you suggest) and GCC will be informed about this (it will see that the symbol is not LDPR_PREVAILING_DEF_IRONLY and optimize accordingly)