"BOLT" post-link optimizer gives 15% speed boost to Clang
Johan Engelen
j at j.nl
Tue Oct 23 22:07:21 UTC 2018
Hi all,
"Post-link optimization", what? Yes, indeed, optimization
_after_ the _linker_ has generated a _binary_.
Read about this interesting work and the discussion here:
https://lists.llvm.org/pipermail/llvm-dev/2018-October/126985.html
When applied to clang, the performance gain is 15% faster
execution (note, that's the result of applying BOLT on a clang
binary that was already built with PGO and LTO !)
Cheers,
Johan
More information about the Digitalmars-d
mailing list