From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754837AbbFRJ6O (ORCPT ); Thu, 18 Jun 2015 05:58:14 -0400 Received: from cantor2.suse.de ([195.135.220.15]:55974 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751001AbbFRJ5g (ORCPT ); Thu, 18 Jun 2015 05:57:36 -0400 From: Michal Hocko To: Andrew Morton Cc: David Rientjes , , LKML Subject: [PATCH 0/2] oom: sysrq+f shouldn't not panic the system + cleanup Date: Thu, 18 Jun 2015 11:57:25 +0200 Message-Id: <1434621447-21175-1-git-send-email-mhocko@suse.cz> X-Mailer: git-send-email 2.1.4 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, I have split the patch sent previously http://marc.info/?l=linux-mm&m=143323521519798&w=2 into two parts. The first patch prevents from the panic when OOM killer is sysrq triggered. This is an obvious bug fix and hopefuly not controversial. I still believe that combining the regular and the sysrq triggered OOM paths is ugly, error prone and it deserves a split up which is done in the second patch. There are no functional changes introduced there. I have dropped __oom_kill_process part because this one turned out to be harmless for for the sysrq+f path - I couldn't have found any interruptible sleep after exit_signals. I find the resulting code easier to follow (35 (+), 22 (-) sounds like a reasonable code overhead for that purpose).