If you read the standards (even old versions) & rationale it's clear this was just due to bad impls.
-
-
I was referring to UB in general.
-
Well to make a C impl where all UB has defined behavior you need to incur huge runtime costs, equiv to ASan+UBSan+much more.
-
Um... no. Sanitizers are useful, but so is just having behavior defined in a way appropriate to platform.
-
You missed the point. Defining most UB requires determining when it happens, which requires very expensive tracking.
-
Or the code is written in such a way that it can absorb whatever naturally happens on the platform (without a check).
-
"Whatever naturally happens" cannot be defined consistently when you have out-of-bounds pointer arithmetic, etc.
-
One usually does not need to worry about supporting an infinite number of unknown platforms.
-
Like, some of us just do x86_64 posix & windows, and it's fine.
- 5 more replies
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.