kern/64573: mmap with PROT_NONE, but still could be read

[Gregory Bond]

> The same bug happens mmap()-ing a file with PROT_WRITE, you still get
> read access.  In this case, PROT_NONE was just the simplest way to
> demonstrate the bug.

IIRC, in the original mmap man page lo these many years ago, there was a
proviso about "hardware willing" (with the "PROT_WRITE implies PROT_READ" 
case explicitly mentioned).  

I wonder if the implementation still favours efficient implementation over
semantic correctness, and if this proviso might be noted in the man page?
Is it even possible to implement PROT_WRITE&~PROT_READ or PROT_NONE on i386?