linux-kernel - [PATCH v2 0/6] x86/entry,lockdep: Improve IRQ state tracking

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: <20200528140535.206916549@infradead.org>
Date:   Thu, 28 May 2020 16:05:35 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     mingo@...nel.org, will@...nel.org, tglx@...utronix.de
Cc:     x86@...nel.org, linux-kernel@...r.kernel.org,
        a.darwish@...utronix.de, rostedt@...dmis.org,
        bigeasy@...utronix.de, peterz@...radead.org
Subject: [PATCH v2 0/6] x86/entry,lockdep: Improve IRQ state tracking

Ahmed and Sebastian wanted additional lockdep_assert*() macros and ran
into header hell.

Move the IRQ state into per-cpu variables, which removes the dependency on
task_struct, which is what generated the header-hell.

And fix IRQ state tracking to not be affected by lockdep_off() (it really
should not have been anyway) and extends IRQ state tracking across (x86)
NMIs.

They now survive a kbuild+perf-top load.

Changes since v1:

 - this_cpu_inc_return() fix
 - reordered patches
 - fixed up changelogs
 - fixed the actual NMI bits; we can't use IF on exit.