linux-kernel - Re: [PATCH v2 27/28] binder: reverse locking order in shrinker callback

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 4 Dec 2023 14:45:42 +0000
From:   Carlos Llamas <cmllamas@...gle.com>
To:     Alice Ryhl <aliceryhl@...gle.com>
Cc:     arve@...roid.com, brauner@...nel.org, gregkh@...uxfoundation.org,
        joel@...lfernandes.org, kernel-team@...roid.com,
        linux-kernel@...r.kernel.org, maco@...roid.com, surenb@...gle.com,
        tkjos@...roid.com
Subject: Re: [PATCH v2 27/28] binder: reverse locking order in shrinker
 callback

On Mon, Dec 04, 2023 at 11:57:42AM +0000, 'Alice Ryhl' via kernel-team wrote:
> > +	trace_binder_unmap_kernel_start(alloc, index);
> > +
> > +	page_to_free = page->page_ptr;
> > +	page->page_ptr = NULL;
> > +
> > +	trace_binder_unmap_kernel_end(alloc, index);
> 
> What are these trace calls used for? Previously they wrapped the call to
> __free_page, and happened after the `trace_binder_unmap_user_*` calls.
> 

It also used to wrap an unmap_kernel_range() call, which explains the
naming of the tracepoint I suppose. It looked like this:

	trace_binder_unmap_kernel_start(alloc, index);

	unmap_kernel_range(page_addr, PAGE_SIZE);
	__free_page(page->page_ptr);
	page->page_ptr = NULL;

	trace_binder_unmap_kernel_end(alloc, index);

My thinking was that we care more about the page->page_ptr clearing but
it sounds like we could just drop the tracepoint, we no longer have a
kernel mapped area.

--
Carlos Llamas