[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20220824163100.224449-1-david@redhat.com>
Date: Wed, 24 Aug 2022 18:30:58 +0200
From: David Hildenbrand <david@...hat.com>
To: linux-kernel@...r.kernel.org
Cc: linux-mm@...ck.org, linux-doc@...r.kernel.org,
kexec@...ts.infradead.org, David Hildenbrand <david@...hat.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Ingo Molnar <mingo@...nel.org>,
David Laight <David.Laight@...LAB.COM>,
Jonathan Corbet <corbet@....net>,
Andy Whitcroft <apw@...onical.com>,
Joe Perches <joe@...ches.com>,
Dwaipayan Ray <dwaipayanray1@...il.com>,
Lukas Bulwahn <lukas.bulwahn@...il.com>,
Baoquan He <bhe@...hat.com>, Vivek Goyal <vgoyal@...hat.com>,
Dave Young <dyoung@...hat.com>
Subject: [PATCH RFC 0/2] coding-style.rst: document BUG() and WARN() rules
As it seems to be rather unclear if/when to use BUG(), BUG_ON(),
VM_BUG_ON(), WARN_ON_ONCE(), ... let's try to document the result of a
recent discussion.
Details can be found in patch #1.
--------------------------------------------------------------------------
Here is some braindump after thinking about BUG_ON(), WARN_ON(), ... and
how it interacts with kdump.
I was wondering what the expectation on a system with armed kdump are,
for example, after we removed most BUG_ON() instances and replaced them
by WARN_ON_ONCE(). I would assume that we actually want to panic in some
cases to capture a proper system dump instead of continuing and eventually
ending up with a completely broken system where it's hard to extract any
useful debug information. We'd have to enable panic_on_warn. But we'd only
want to do that in case kdump is actually armed after boot.
So one idea would be to have some kind of "panic_on_warn_with_kdump" mode.
But then, we'd actually crash+kdump even on the most harmless WARN_ON()
conditions, because they all look alike. To compensate, we would need
some kind of "severity" levels of a warning -- at least some kind of
"this is harmless and we can easily recover, but please tell the
developers" vs. "this is real bad and unexpected, capture a dump
immediately instead of trying to recover and eventually failing miserably".
But then, maybe we really want something like BUG_ON() -- let's call it
CBUG_ON() for simplicity -- but be able to make it be usable in
conditionals (to implement recovery code if easily possible) and make the
runtime behavior configurable.
if (CBUG_ON(whatever))
try_to_recover()
Whereby, for example, "panic_on_cbug" and "panic_on_cbug_with_kdump"
could control the runtime behavior.
But this is just a braindump and I assume people reading along have other,
better ideas. Especially, a better name for CBUG.
Cc: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Ingo Molnar <mingo@...nel.org>
Cc: David Laight <David.Laight@...LAB.COM>
Cc: Jonathan Corbet <corbet@....net>
Cc: Andy Whitcroft <apw@...onical.com>
Cc: Joe Perches <joe@...ches.com>
Cc: Dwaipayan Ray <dwaipayanray1@...il.com>
Cc: Lukas Bulwahn <lukas.bulwahn@...il.com>
Cc: Baoquan He <bhe@...hat.com>
Cc: Vivek Goyal <vgoyal@...hat.com>
Cc: Dave Young <dyoung@...hat.com>
David Hildenbrand (2):
coding-style.rst: document BUG() and WARN() rules ("do not crash the
kernel")
checkpatch: warn on usage of VM_BUG_ON() and friends
Documentation/process/coding-style.rst | 27 ++++++++++++++++++++++++++
scripts/checkpatch.pl | 6 +++---
2 files changed, 30 insertions(+), 3 deletions(-)
--
2.37.1
Powered by blists - more mailing lists