linux-kernel - [PATCH 0/3] various irq handling fixes/docu updates

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-next>] [day] [month] [year] [list]

Message-Id: <20221216150155.200389-1-manfred@colorfullife.com>
Date:   Fri, 16 Dec 2022 16:01:52 +0100
From:   Manfred Spraul <manfred@...orfullife.com>
To:     LKML <linux-kernel@...r.kernel.org>,
        Andrew Morton <akpm@...ux-foundation.org>
Cc:     1vier1@....de, Manfred Spraul <manfred@...orfullife.com>
Subject: [PATCH 0/3] various irq handling fixes/docu updates

Hi,

I found three patches from two topics in my outbox.

* 0001-lib-percpu_counter-percpu_counter_add_batch-overflow.patch
* 0002-include-linux-percpu_counter.h-Race-in-uniprocessor-.patch

  The percpu_counter code does take interrupt into account properly,
  this could result in corrupted counters.

0003-kernel-irq-manage.c-disable_irq-might-sleep.patch

  The disable_irq() documentation does not take threaded interrupt
  handlers into account.

I've not added cc stable, the races are not really likely:

- #002 is probably the most likely case: UP systems that use
  percpu_counters from interrupt should observe corruptions.

- #001 is fairly theoretical

- #003 is a docu update and thus out of scope for stable.

@Andrew: Could you add them to -mm and -next?
Especially #003 should be in -next for a few months, to check what the
might_sleep() encounters.

--
	Manfred