Malloc doesn't guarantee zero filling memory, but ObjC and Swift currently need to zero memory immediately after malloc returns it, and they both share C's malloc pool.

This makes a lot more performance sense for those languages.

One advantage is that when you allocate the memory back out (either by malloc, or calloc which requires zeroing), you don't have to do anything.

It is mentioned that if you write to freed memory, that could lead to later callocs not returning zeroed memory, which shows they are relying on this.

The advantage of 0xDEADBEEF would be that it reduces the chance of software starting to rely on that value, compared to zero, isn't it?

Better for memory compression, at least.

It's faster: https://community.arm.com/support-forums/f/architectures-and...

Less likely to cause crashes, probably.