linux-kernel - Re: userfaultfd: two-step UFFDIO

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <4689f014-2885-42b9-91a4-ff8b8133599f@yandex.ru>
Date: Mon, 25 Nov 2024 21:01:47 +0300
From: stsp <stsp2@...dex.ru>
To: Peter Xu <peterx@...hat.com>
Cc: Linux kernel <linux-kernel@...r.kernel.org>,
 Muhammad Usama Anjum <usama.anjum@...labora.com>
Subject: Re: userfaultfd: two-step UFFDIO_API always gives -EINVAL

25.11.2024 20:44, Peter Xu пишет:
> Apps who tracks snapshots needs the unmodified pages before being written.
> Those cannot rely on kernel resolution because it needs more than "if the
> page is written" - it also needs the page data before being written.
Say I am writing a frame grabber
(not exactly, but very close to).
I monitor the video buffer of another
process, and "snapshot" it with some
frequency. I only need to know what
pages were modified, to reduce the
bandwidth to an absolute minimum,
and if the process is not playing - then
to not grab anything until it resumes.
UFFD_FEATURE_PAGEFAULT_FLAG_WP
works quite well for me already, but
I envision a huge boost with
UFFD_FEATURE_WP_ASYNC.
What would you suggest for that usage
scenario?