linux-kernel - Re: [PATCH] pid: Add a judgment for ns null in pid_nr

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] [day] [month] [year] [list]

Message-ID: <20250802084316.GA31711@redhat.com>
Date: Sat, 2 Aug 2025 10:43:17 +0200
From: Oleg Nesterov <oleg@...hat.com>
To: Al Viro <viro@...iv.linux.org.uk>
Cc: Xiang Gao <gxxa03070307@...il.com>, brauner@...nel.org,
	mjguzik@...il.com, Liam.Howlett@...cle.com,
	joel.granados@...nel.org, lorenzo.stoakes@...cle.com,
	linux-kernel@...r.kernel.org, gaoxiang17 <gaoxiang17@...omi.com>
Subject: Re: [PATCH] pid: Add a judgment for ns null in pid_nr_ns

On 08/02, Al Viro wrote:
>
> In which conditions does that happen?
>
> > 	__task_pid_nr_ns+0x74/0xd0
> > 	...
> > 	__handle_irq_event_percpu+0xd4/0x284
> > 	handle_irq_event+0x48/0xb0
>
> 	Huh?  Just what is it doing inside an IRQ handler?
> Hell, the notion of current process is not usable in those,
> let alone any properties of such...

Well, at least get_curent() should work in this case...

But we can forget about IRQ. I guess the problem is that if the exiting
task has already passed exit_notify() and it was (auto)reaped, then, say,
task_pid_vnr(&init_task) will crash.

Not that I think the exiting task should do anything like this...

Oleg.