Re: [PATCH v3 2/5] mm/mseal: update madvise() logic

[David Hildenbrand]

On 25.07.25 00:47, David Hildenbrand wrote:
> On 25.07.25 00:29, Kees Cook wrote:
> > On Thu, Jul 24, 2025 at 11:41:04PM +0200, David Hildenbrand wrote:
> > > On 24.07.25 23:32, David Hildenbrand wrote:
> > > > > As an aside, why should discard work in this case even without step 4?
> > > > > Wouldn't setting "read-only" imply you don't want the memory to change
> > > > > out from under you? I guess I'm not clear on the semantics: how do memory
> > > > > protection bits map to madvise actions like this?
> > > >
> > > > They generally don't affect MADV_DONTNEED behavior. The only documented
> > > > (man page) reason for EPERM in the man page is related to MADV_HWPOISON.
> > >
> > > (Exception: MADV_POPULATE_READ/MADV_POPULATE_WRITE requires corresponding
> > > permissions)
> >
> > Shouldn't an MADV action that changes memory contents require the W bit
> > though?

Pondering about this some more, at least MADV_DONTNEED is mostly a 
cheaper way of doing mmap(MAP_FIXED): in other word, zap everything but 
leave the original mapping unchanged.

So if you allow for mmap(MAP_FIXED) -- ignore any permissions bits, of 
course -- nothing really wrong about allowing MADV_DONTNEED.

With mseal(), it got all weird I am afraid, because we have this 
exception list, and apparently, it has holes. :(

--
Cheers,

David / dhildenb