On today's @handmade_hero, we saw the fallacy of the inline "zero cost abstraction" play out dramatically. 50c on a ~170c routine saved!
-
-
Replying to @cmuratori
@handmade_hero Literally all we did was copy two 1-line inline functions into the body of a routine that called them.4 replies 2 retweets 6 likes -
Replying to @cmuratori
@cmuratori@handmade_hero Weird. Is it inlining it? Maybe use /Ox (full optimization) and see what happens?1 reply 0 retweets 0 likes -
Replying to @ssylvan
@ssylvan@cmuratori Yes, it inlined, but after inlining, it apparently didn't see the more global optimizations it could have done.2 replies 0 retweets 0 likes -
Replying to @cmuratori
@cmuratori@ssylvan Using better compiler helps. This is all that clang produces for old code (with function calls)pic.twitter.com/Xvu78qrmf1
1 reply 0 retweets 0 likes -
Replying to @mmozeiko
@cmuratori@ssylvan Basically old code with clang is as fast as MSVC with todays manually inlined code.1 reply 0 retweets 0 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.