>(((compilers)))

>std::vector buffer (64);
>buffer.clear();
How will C-tards cope?

>Imagine still coding in C++

Nothing wrong with C or C++ as long as you adhere to procedural style and not the OOP meme.

>How do we fix this?
Don't use memset - make your own function to do what you want it to do... or choose a different compiler.

>How do we fix this?
Use Rust

there's probably some compiler directive you can use to prevent something being optimised out. I dont know, I dont write code I have to worry about in this manner

explicit_bzero, or use libsodium for other handy functions
there is nothing wrong with OP's pic. if you work with sensitive data you should approach it differently even regarding allocation. malloc data or stack can leak do coredumps or swap on disk or get allocated without cleaning to another process.

>retarded niggers still zero out the memory when their operating system will always zero out the memory before handing it off to any other unprivileged process
first of all, if you're "defensively" zeroing out memory in your own process to "mitigate" bugs like heartbleed, you have no right or any authority telling me what a secure system looks like, dumb nigger monkey, kill yourself

Attached: 1639952119895.jpg (958x1080, 115.18K)

that only sets size to 0 retard, it doesn't zero out the memory nor deallocates it
>and that's a good thing

>How do we fix this?
So... suppose you make a volatile pointer to the buffer, and then use that with memset. Since it's volatile, the compiler has to treat every write as deliberate, no?

now suppose instead of doing that, you stopped zeroing out the memory pointlessly and instead fixed the bugs that cause your memory to be leaked to 3rd party programs and maybe even, entities on the other end of the wire

>wasting cpu cycles on pointless security theatre
based compiler btfos fag

And what if I don't know there's a bug? What if there's not a bug, but some other program uses rowhammer to force the data out of memory with no software being exploited to get it? Honestly, a little bit of redundant security never hurt anyone.

git.musl-libc.org/cgit/musl/tree/src/string/explicit_bzero.c
common technique is to just fake read with inline assembly
this is like the worst topic to discuss online. every SO and other forums are just bloated with uneducated wrong answers

whoah something about zeroing memory really triggers you

then you're a dumb nigger monkey and zeroing out memory won't help
something about dumb niggers wasting time on bandaiding the real problem instead of fixing it really does trigger me, I just can't figure out what exactly