From mboxrd@z Thu Jan 1 00:00:00 1970 From: catalin.marinas@arm.com (Catalin Marinas) Date: Mon, 7 Jan 2013 10:56:18 +0000 Subject: [PATCH v4 10/19] irqchip: Move ARM gic.h to include/linux/irqchip/arm-gic.h In-Reply-To: <50EA35AF.1000308@gmail.com> References: <1357235668-9450-1-git-send-email-robherring2@gmail.com> <1357235668-9450-11-git-send-email-robherring2@gmail.com> <201301032144.52079.arnd@arndb.de> <50E601A7.8010801@gmail.com> <50EA35AF.1000308@gmail.com> Message-ID: <20130107105618.GB29030@arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Mon, Jan 07, 2013 at 02:40:47AM +0000, Rob Herring wrote: > On 01/06/2013 01:07 PM, Catalin Marinas wrote: > > On 3 January 2013 22:09, Rob Herring wrote: > >> On 01/03/2013 03:44 PM, Arnd Bergmann wrote: > >>> On Thursday 03 January 2013, Rob Herring wrote: > >>>> From: Rob Herring > >>>> > >>>> Now that we have GIC moved to drivers/irqchip and all GIC DT init for > >>>> platforms using irqchip_init, move gic.h and update the remaining > >>>> includes. > >>> > >>> Can you elaborate on why this is done? Does any other code rely on the > >>> header file being globally visible, or is there a more general reason > >>> to move header files out of asm/ ? > >> > >> KVM needs the register definitions for GIC distributor emulation. > >> Platform SMP code still needs gic_secondary_init. > > > > While I'm not arguing against this patch, for gic_secondary_init() we > > should use a CPU notifier. All platforms calling this function only > > pass 0 as argument (and it makes sense, since the GIC CPU interface is > > part of primary GIC). > > I'm waiting for you to add that support for arm64. :) Then we can > convert arm platforms to use it. I'll do it, no problem ;) -- Catalin