linux-kernel - Re: [GIT PULL] KVM changes for Linux 6.14

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: <CAHk-=wiOSyfW3sgccrfVtanZGUSnjFidSbaP3tg9wapydb-u6g@mail.gmail.com>
Date: Sun, 26 Jan 2025 10:34:16 -0800
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Oleg Nesterov <oleg@...hat.com>
Cc: Paolo Bonzini <pbonzini@...hat.com>, "Michael S. Tsirkin" <mst@...hat.com>, 
	Christian Brauner <brauner@...nel.org>, "Eric W. Biederman" <ebiederm@...ssion.com>, linux-kernel@...r.kernel.org, 
	kvm@...r.kernel.org
Subject: Re: [GIT PULL] KVM changes for Linux 6.14

On Sun, 26 Jan 2025 at 06:21, Oleg Nesterov <oleg@...hat.com> wrote:
>
> > Should we just add some flag to say "don't show this thread in this
> > context"?
>
> Not sure I understand... Looking at is_single_threaded() above I guess
> something like below should work (incomplete, in particular we need to
> chang first_tid() as well).

So yes, I was thinking something similar, but:

> But a PF_HIDDEN sub-thread will still be visible via /proc/$pid_of_PF_HIDDEN
>
> > We obviously still want to see it for management purposes,
> > so it's not like the thing should be entirely invisible,
>
> Can you explain?

I was literally thinking that instead of a "hidden" flag, it would be
a "self-hidden" flag.

So if somebody _else_ (notably the sysadmin) does "ps" they see the
kernel thread as a subthread.

But when you look at your own /proc/self/task/ listing, you only see
your own explicit threads. So that "is_singlethreaded()" logic works.

Maybe that's just too ugly for words, and the kvm workaround is better.

            Linus