Nat! bio photo


Senior Mull

Twitter Github Twitch

Comparing build speed of gcc vs clang

Just a few datapoints, as this is too large for a tweet.

Build and install llvm, mulle-clang, mulle-lldb (Release)

gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0 and ld

real  11m29,759s
user  309m45,687s
sys   19m24,405s

clang version 6.0.0-1ubuntu2 (tags/RELEASE_600/final) and ld

real  9m59,042s
user  277m56,055s
sys   9m25,066s

Build and install llvm, clang, lldb (Debug)

gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0 and ld

It uses too much memory on my system (32 GB per CPU available) and goes into swapping

clang version 6.0.0-1ubuntu2 (tags/RELEASE_600/final) and lld

real  32m20,432s
user  235m14,486s
sys   14m7,109s

clang version 6.0.0-1ubuntu2 (tags/RELEASE_600/final) and gold

real  32m44,193s
user  231m46,067s
sys   11m51,780s


Installation is a really slow process, I don’t know why. Building everything in DEBUG even with my dual CPU/ 20 core setup seems not feasible for development work.

Post a comment

All comments are held for moderation; basic HTML formatting accepted.

E-mail: (not published)