3 ms·
strcpy is optimized, memccpy not. First use -march=native on a new clang (>5), and see if clang can optimize memccpy by itself. gcc probably not. And BTW, gcc
by rurban 7y ago
strcpy is optimized, memccpy not.
First use -march=native on a new clang (>5), and see if clang can optimize memccpy by itself. gcc probably not.
And BTW, gcc-9 is still broken and should be blacklisted everywhere.
- johnisgood 7y agoThanks for the tip. I made some changes to the source code so the one above is a bit outdated. I should probably try the two different implementations found on the website, too. > gcc-9 is still broken and should be blacklisted everywhere. Yes, it seems to be the case. I ran into a few peculiarities, to say the least.