On 1/16/2019 8:52 AM, Josh Snyder wrote:

On Tue, Jan 15, 2019 at 10:34 PM Dominique Martinet [off-list ref]
wrote:

quoted

There is a difference with your previous patch though, that used to list no
page in core when it didn't know; this patch lists pages as in core when it
refuses to tell. I don't think that's very important, though.

I've argued previously that reporting false positives (as your patch does)
should be better, otherwise there might be somebody trying to fault in their
pages in a loop until mincore reports positive, which would become an endless
loop. So agreed with your change.

Or maybe we could resort to the 5.0-rc1 page table check (that is now being
reverted) but only in cases when we are not allowed the page cache residency
check? Or would that be needlessly complicated? And it would be able to leak if
a page was evicted from the page cache...

Is there a reason not to return -EPERM in this case?

That would definitely break somebody.

quoted

If anything, the 0400 user-owner file might be a problem in some edge
case (e.g. if you're preloading git directories, many objects are 0444);
should we *also* check ownership?...

Yes, this seems valuable. Some databases with immutable files (e.g. git, as
you've mentioned) conceivably operate this way.

Josh