From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoffer Dall Subject: Re: [PATCH] KVM: arm64: ITS: move ITS registration into first VCPU run Date: Wed, 3 Aug 2016 19:13:52 +0200 Message-ID: <20160803171352.GZ32244@cbox> References: <20160803145745.7556-1-andre.przywara@arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org, Marc Zyngier , Paolo Bonzini , kvmarm@lists.cs.columbia.edu, linux-arm-kernel@lists.infradead.org To: Andre Przywara Return-path: Content-Disposition: inline In-Reply-To: <20160803145745.7556-1-andre.przywara@arm.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: kvmarm-bounces@lists.cs.columbia.edu Sender: kvmarm-bounces@lists.cs.columbia.edu List-Id: kvm.vger.kernel.org On Wed, Aug 03, 2016 at 03:57:45PM +0100, Andre Przywara wrote: > Currently we register ITS devices upon userland issuing the CTRL_INIT > ioctl to mark initialization of the ITS as done. > This deviates from the initialization sequence of the existing GIC > devices and does not play well with the way QEMU handles things. > To be more in line with what we are used to, register the ITS(es) just > before the first VCPU is about to run, so in the map_resources() call. > This involves iterating through the list of KVM devices and handle each > ITS that we find. > > Signed-off-by: Andre Przywara > --- > Hi, > > this is based upon next-20160728 plus Christoffer's kvm_device locking Since my fix was horribly broken, I suggest you just base this on next and let it be similarly broken to the vfio-device that we already have in the kernel. That way, we get the userspace ABI right and we fix KVM in general for -rc2. Paolo/Radim, are you ok with this? Thanks, -Christoffer