strict aliasing is even worse than I thoughthttps://twitter.com/ch3root/status/697939734165901312 …
-
-
Replying to @johnregehr
@johnregehr wait, isn't that plain old undefined behavior? so why would one expect the compiler to do "the right thing"?1 reply 0 retweets 1 like -
Replying to @H2CO3_iOS
@H2CO3_iOS effective type should be determined by the most recent store1 reply 0 retweets 0 likes -
Replying to @johnregehr
@johnregehr so we are allowed to *change* the effective type? I though that already counted as a violation of strict aliasing.1 reply 0 retweets 0 likes -
Replying to @H2CO3_iOS
@H2CO3_iOS yes, storage from malloc can have different effective types at different times2 replies 1 retweet 2 likes -
Replying to @johnregehr
@johnregehr@H2CO3_iOS Where is the language that says effective type can change? Sounds like wishful thinking on your part.3 replies 0 retweets 0 likes -
Replying to @RichFelker
@RichFelker sorry to insist, but are you saying that the arguments in https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69776#c2 … are wrong?2 replies 0 retweets 1 like -
Replying to @volatile_void
@spun_off No, just that I was unaware that it's actually well-defined to repurpose allocated memory after it gets an effective type.
1 reply 0 retweets 0 likes
@spun_off I wrongly assumed it was a sort of "gray area" people just assumed would work without justification.
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.