-
-
Replying to @FioraAeterna
Who got the idiotic idea that static is either necessary or sufficient (it's neither) to force the order?!
1 reply 0 retweets 0 likes -
-
Replying to @FioraAeterna
The static should just be optimized out since its address doesn't leak and the value is never used before storing to it.
1 reply 0 retweets 0 likes -
Replying to @RichFelker @FioraAeterna
I assume it happens to work in practice, but if so it's a bug (shortcoming) in the optimizer, not a real constraint.
1 reply 0 retweets 0 likes -
-
Replying to @FioraAeterna @RichFelker
the actual implementation doesn't do it that way, just the comment
1 reply 0 retweets 0 likes -
Replying to @FioraAeterna @RichFelker
I think olde x87 GCC would care about that named var though. AKA when you know your semantics jumped the shark.
1 reply 0 retweets 1 like -
Replying to @pkhuong @FioraAeterna
Modern gcc in standards-conforming mode (as opposed to -ffast-math or something) will always do it in the right order
1 reply 0 retweets 0 likes
But excess-precision issues and double rounding due to them still could mess you up.
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.
