Re: [PATCH 3/3] mlock: avoid dirtying pages and triggering writeback

[Michel Lespinasse <walken@google.com>]

On Wed, Nov 17, 2010 at 3:11 PM, Dave Chinner <david@fromorbit.com> wrote:
>> Really, my understanding is that not pre-allocating filesystem blocks
>> is just fine. This is, after all, what happens with ext3 and it's
>> never been reported as a bug (that I know of).
>
> It's not ext3 you have to worry about - it's the filesystems that
> need special state set up on their pages/buffers for ->writepage to
> work correctly that are the problem. You need to call
> ->write_begin/->write_end to get the state set up properly.
>
> If this state is not set up properly, silent data loss will occur
> during mmap writes either by ENOSPC or failing to set up writes into
> unwritten extents correctly (i.e. we'll be back to where we were in
> 2.6.15).
>
> I don't think ->page_mkwrite can be worked around - we need that to
> be called on the first write fault of any mmap()d page to ensure it
> is set up correctly for writeback.  If we don't get write faults
> after the page is mlock()d, then we need the ->page_mkwrite() call
> during the mlock() call.

Just to be clear - I'm proposing to skip the entire do_wp_page() call
by doing a read fault rather than a write fault. If the page wasn't
dirty already, it will stay clean and with a non-writable PTE until it
gets actually written to, at which point we'll get a write fault and
do_wp_page will be invoked as usual.

I am not proposing to skip the page_mkwrite() while upgrading the PTE
permissions, which I think is what you were arguing against ?

-- 
Michel "Walken" Lespinasse
A program is never fully debugged until the last user dies.