Thread (148 messages) 148 messages, 20 authors, 2019-03-12

Re: [PATCH 3/3] mm/mincore: provide mapped status when cached status is not allowed

From: Jiri Kosina <jikos@kernel.org>
Date: 2019-02-06 20:14:35
Also in: linux-mm, lkml

On Fri, 1 Feb 2019, Vlastimil Babka wrote:
quoted
Well, but rss update will not tell you that the page has been faulted in
which is the most interesting part.
Sure, but the patch doesn't add back that capability neither. It allows
to recognize page being reclaimed, and I argue you can infer that from
rss change as well. That change is mentioned in the last paragraph in
changelog, and I thought "add a hard to evaluate side channel" in your
reply referred to that. It doesn't add back the "original" side channel
to detect somebody else accessed a page.
On Fri, 1 Feb 2019, Vlastimil Babka wrote:
quoted
Is this really worth it? Do we know about any specific usecase that
would benefit from this change? TBH I would rather wait for the report
than add a hard to evaluate side channel.
Well it's not that complicated IMHO. Linus said it's worth trying, so
let's see how he likes the result. The side channel exists anyway as
long as process can e.g. check if its rss shrinked, and I doubt we are
going to remove that possibility.
Linus, do you have any opinion here?

I have a hunch that mm maintainers are keeping this on a backburner 
because there might still open question(s) in the air.

Thanks,

-- 
Jiri Kosina
SUSE Labs
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help