From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?windows-1252?Q?Philippe_R=E9tornaz?= Subject: Re: [PATCH 01/44] kernel: Add support for poweroff handler call chain Date: Tue, 07 Oct 2014 09:46:59 +0200 Message-ID: <54339A73.40107@gmail.com> References: <1412659726-29957-1-git-send-email-linux@roeck-us.net> <1412659726-29957-2-git-send-email-linux@roeck-us.net> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1412659726-29957-2-git-send-email-linux@roeck-us.net> Sender: linux-m68k-owner@vger.kernel.org List-Archive: List-Post: To: Guenter Roeck , linux-kernel@vger.kernel.org Cc: linux-m32r-ja@ml.linux-m32r.org, linux-mips@linux-mips.org, linux-efi@vger.kernel.org, linux-ia64@vger.kernel.org, Heiko Stuebner , Len Brown , linux-xtensa@linux-xtensa.org, Pavel Machek , devel@driverdev.osuosl.org, linux-s390@vger.kernel.org, lguest@lists.ozlabs.org, linux-c6x-dev@linux-c6x.org, linux-hexagon@vger.kernel.org, linux-sh@vger.kernel.org, Alexander Graf , linux-acpi@vger.kernel.org, Geert Uytterhoeven , xen-devel@lists.xenproject.org, devicetree@vger.kernel.org, user-mode-linux-devel@lists.sourceforge.net, linux-pm@vger.kernel.org, adi-buildroot-devel@lists.sourceforge.net, linux-m68k@lists.linux-m68k.org, linux-am33-list@redhat.com, linux-tegra@vger.kernel.org, openipmi-developer@lists.sourceforge.net, linux-metag@vger.kernel.org, linux-arm-kernel@lists.infradead List-ID: Hello This seems exactly what I would need on the mc13783 to handle cleanly the poweroff, but after reading this patchset I have the following question: [...] > +/* > + * Notifier list for kernel code which wants to be called > + * to power off the system. > + */ > +static ATOMIC_NOTIFIER_HEAD(poweroff_handler_list); [...] > +void do_kernel_poweroff(void) > +{ > + atomic_notifier_call_chain(&poweroff_handler_list, 0, NULL); > +} > + It seems that the poweroff callback needs to be atomic as per _atomic_notifier_call_chain documentation: "Calls each function in a notifier chain in turn. The functions run in an atomic context" But this is a problem for many MFD (mc13783, twl4030 etc ...) which are accessible on only a blocking bus (SPI, I2C). What I am missing here ? Thanks, Philippe