On Tue, Aug 26, 2025 at 09:37:23AM +0100, Lorenzo Stoakes wrote: > Harry - one brief very nitty note - could you do a cover letter even for 2 patch > series? > > This is a subjective thing and literally just my taste but I prefer it :P > obviously this is optional as a result, but I feel it's neater. > No problem! will do cover letter from next time. > On Tue, Aug 26, 2025 at 01:22:03AM -0600, Jonathan Corbet wrote: > > Harry Yoo <harry.yoo@xxxxxxxxxx> writes: > > > > > While move_ptes() has a comment explaining why rmap locks are needed, > > > Documentation/mm/process_addrs.rst does not. Without being aware of that > > > comment, I spent hours figuring out how things could go wrong and why, > > > in some cases, rmap locks can be safely skipped. > > > > > > Add a more comprehensive explanation to the documentation to save time > > > for others. > > > > > > Signed-off-by: Harry Yoo <harry.yoo@xxxxxxxxxx> > > > --- > > > Documentation/mm/process_addrs.rst | 32 ++++++++++++++++++++++++++++++ > > > 1 file changed, 32 insertions(+) > > > > > > diff --git a/Documentation/mm/process_addrs.rst b/Documentation/mm/process_addrs.rst > > > index be49e2a269e4..ee7c0dba339e 100644 > > > --- a/Documentation/mm/process_addrs.rst > > > +++ b/Documentation/mm/process_addrs.rst > > > @@ -744,6 +744,38 @@ You can observe this in the :c:func:`!mremap` implementation in the functions > > > :c:func:`!take_rmap_locks` and :c:func:`!drop_rmap_locks` which perform the rmap > > > side of lock acquisition, invoked ultimately by :c:func:`!move_page_tables`. > > > > > > +.. note:: If :c:func:`!mremap()` -> :c:func:`!move_ptes()` does not take rmap > > > + locks, :c:func:`!rmap_walk()` may miss a pte for the folio. > > > + > > > + The problematic sequence is as follows: > > > > Please don't use :c:func: - just write function() and all the right > > things will happen. (For extra credit, fix the existing usages :) Hi Jonathan and Lorenzo, **blaming myself for thinking** "Hmmm it's already there, it should be fine to use it..." > Yeah sorry Jon on latter bit, I did mean to get to that but workload > been... well you can see on lore :P > > I have a real backlog even more than usual right now too due to daring to take a > day off on a national holiday here in the UK :)) > > Harry - more than happy for you to do the above as part of this series or > separately, will sling you some tags accordingly. Okay, I'll do as a part of the series (process_addrs.rst and memory-model.rst). > If you're not already doing it (expect you are) you can generate docs via: > > make SPHINXDIRS=mm htmldocs > > Then get access to generated HTML in a browser locally in Documentation/output/ Thanks and yeah I'm doing it! :) > > > > Thanks, > > > > jon > > Cheers, Lorenzo -- Cheers, Harry / Hyeonggon
