From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [PATCH for 4.5] xen/arm: Initialize the domain vgic lock Date: Thu, 18 Dec 2014 12:26:02 +0000 Message-ID: <5492C7DA.8050301@linaro.org> References: <1418830815-4720-1-git-send-email-julien.grall@linaro.org> <1418896057.11882.0.camel@citrix.com> <5492C310.2050903@linaro.org> <1418904755.11882.43.camel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: Received: from mail6.bemta14.messagelabs.com ([193.109.254.103]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1Y1a9e-00032j-Np for xen-devel@lists.xenproject.org; Thu, 18 Dec 2014 12:26:07 +0000 Received: by mail-wi0-f177.google.com with SMTP id l15so1606549wiw.10 for ; Thu, 18 Dec 2014 04:26:04 -0800 (PST) In-Reply-To: <1418904755.11882.43.camel@citrix.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Campbell Cc: xen-devel@lists.xenproject.org, tim@xen.org, stefano.stabellini@citrix.com List-Id: xen-devel@lists.xenproject.org On 18/12/2014 12:12, Ian Campbell wrote: > On Thu, 2014-12-18 at 12:05 +0000, Julien Grall wrote: >> Hi Ian, >> >> On 18/12/2014 09:47, Ian Campbell wrote: >>> On Wed, 2014-12-17 at 15:40 +0000, Julien Grall wrote: >>>> The domain vgic lock is used uninitialized. >>>> >>>> Signed-off-by: Julien Grall >>> >>> Acked-by: Ian Campbell >>> >>>> --- >>>> This is a bug fix for Xen 4.5 and Xen 4.4. The vgic lock is used >>>> unitialized. Luckily we only use the field "raw" which is reset to 0 >>>> during the domain allocation. >>>> >>>> There is no harm to apply for Xen 4.5 because it will correctly set >>>> the spin_lock structure for a later usage. >>> >>> By your above reasoning there is also no point, is there? That said, I >>> think we should take this since as you say it is harmless and good >>> practice to initialise spinlocks even if not strictly necessary. >> >> It's necessary to initialize spinlocks, not all the field of the >> spinlock is using the value 0 at initialization time. > > You said "...we only use the field "raw" which is reset to 0...". Was > that statement inaccurate? It's accurate in the current. I should have give more details earlier, sorry. Regards, -- Julien Grall