Anyone have insight into the origin of this text? It's bothered me for years because I'm pretty much 100% sure it's wrong, but keep wondering if there's some subtlety I missed each time I run across it again.
Conversation
Replying to
Tried to figure that out from Git but it's not clear. From newest to oldest changes:
* github.com/mkerrisk/man-p
* github.com/mkerrisk/man-p
* github.com/mkerrisk/man-p
It's problematic that official documentation is scarce. Linux man-pages is an interpretation of that scarce info.
1
The man-pages project is often out-of-sync with upstream code and the upstream stances on various things. The random(4) and random(7) pages have been major examples. As a current example, the blocking pool was removed and /dev/random blocks until init at boot like getrandom.
1
The GRND_RANDOM parameter for getrandom is simply a no-op. I don't think there's upstream documentation on the random interfaces aside from the code. For futex, there's actually some upstream documentation. Linux should really have official system call man pages, etc...
Generally how things go with the Linux kernel. Semi-official downstream projects have to take care of most of the testing and documentation. It's up to those projects to figure out how things work despite lack of information from upstream. It's frustrating...