From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34317) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZRxZQ-0002ZF-Od for qemu-devel@nongnu.org; Wed, 19 Aug 2015 03:14:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZRxZN-00021y-Df for qemu-devel@nongnu.org; Wed, 19 Aug 2015 03:14:00 -0400 Received: from mail-wi0-f180.google.com ([209.85.212.180]:35063) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZRxZN-00021E-7e for qemu-devel@nongnu.org; Wed, 19 Aug 2015 03:13:57 -0400 Received: by wicne3 with SMTP id ne3so112442415wic.0 for ; Wed, 19 Aug 2015 00:13:56 -0700 (PDT) Message-ID: <55D42C84.5070408@linaro.org> Date: Wed, 19 Aug 2015 09:13:08 +0200 From: Eric Auger MIME-Version: 1.0 References: <76a96dc06cc55504cbe2f46165b2b171712f78e9.1439904588.git.p.fedin@samsung.com> <55D35512.1000903@linaro.org> <00d201d0da49$5bb60e30$13222a90$@samsung.com> In-Reply-To: <00d201d0da49$5bb60e30$13222a90$@samsung.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v10 2/5] intc/gic: Extract some reusable vGIC code List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Pavel Fedin , qemu-devel@nongnu.org Cc: 'Peter Maydell' , 'Shlomo Pongratz' , 'Christoffer Dall' , 'Shlomo Pongratz' Hi Pavel, On 08/19/2015 08:36 AM, Pavel Fedin wrote: > Hello! > >> I think it would be worth justifying the changes in signature: >> removal of GICState* due to the introduction of GICV3State and also >> justify replacement of uint32_t *val into void*. > > I described it in the cover letter. OK thanks for the explanations. I think they are useful to understand the content of the patch and I would advise you to put them in the commit message since the cover letter then is discarded when committing the series. Right now you don't see usage for it, but here i started > prototyping live migration, and in some cases 'val' is going to be uint64_t. This is because > GICD_IROUTER registers are 64 bits wide. kvm_gic_access() by itself does not dereference the > pointer, just passes it to the kernel, therefore i decided to make it type-agnostic. > >> what is the point of moving kvm_gicd_access and kvm_gicc_access here? If >> I am not mistaken, they only are used in arm_gic_kvm.c? I think they can >> stay static in arm_gic_kvm.c? > > They will be used in future for live migration, at least kvm_gicd_access. kvm_gicc_access just > accompanies it to keep a single style for both. OK. Then either explain it in the commit msg or move them later when live migration gets implemented. Best Regards Eric > > Kind regards, > Pavel Fedin > Expert Engineer > Samsung Electronics Research center Russia > >