Write-Only Memory

... but what about “write-only memory?”

mprotect(2) comes to mind though whether and how much this works may vary. Usually this is used to prevent writing and execution (W^X) of particular bits of memory, as an attacker may wish to write something and then execute what they have just written, and other people may wish to prevent such arbitrary code execution. This is complicated by legitimate needs to write and then execute memory, for example when SBCL compiles the LISP to assembly somewhere in memory.

Run a Slab of Code

gemini://gemi.dev/cgi-bin/wp.cgi/view?W%5EX

http://man.openbsd.org/man2/mprotect.2

https://nvd.nist.gov/vuln/detail/CVE-2024-0519

Hopefully you can view the details about that CVE without having to execute something dangerous like JavaScript. No? Well...