Myth: Your compiler is always smarter than you
Truth: No it's not for some "you", and you CAN acquire the knowledge to (frequently) beat it in your domain.
A 2h demonstration by @cmuratori
https://www.youtube.com/watch?v=R5tBY9Zyw6o …
I'd love a commentary by @rygorous on the inidividual findings.
-
-
(Although obviously I support the original intent of the tweet as well, since the only way you know when the optimizer is doing poorly is if you know better than it in the first place :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
To be more specific: there's been a few bits (e.g. Clang generating multiple jumps) that baffled me, and you seemed to also not know why it even would lay out like that and had to intuit it would pessimize. I'd be curious to hear if someone can rationalize it.
-
It is not a front-end issue, apparently, because it does pass a memcpy down to the actual codegen. Someone posted a llvm-dev question here (http://lists.llvm.org/pipermail/llvm-dev/2020-September/145367.html …), but nobody responded.
End of conversation
New conversation -
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.