From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: [QEMU-traditional] ACPI AML code races with QEMU updating the vCPU count when hotplugging more than ~14 VCPUs. Date: Fri, 10 May 2013 11:01:44 -0400 Message-ID: <20130510150144.GM19520@phenom.dumpdata.com> References: <20130509204723.GA5713@localhost.localdomain> <20130509235047.659a5867@thinkpad> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <20130509235047.659a5867@thinkpad> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Igor Mammedov Cc: xen-devel@lists.xen.org, ross.philipson@citrix.com, stefano.stabellini@eu.citrix.com List-Id: xen-devel@lists.xenproject.org > > Then if the other events are triggered we can just check the > > gpe_state.cpus_state against what XenBus thinks and if they > > are the same just return without doing the qemu_irq_pulse. > > > > Thoughts? > Could you check if switching from level to edge handling in AML helps? > http://git.qemu.org/?p=seabios.git;a=commit;h=9c6635bd48d39a1d17d0a73df6e577ef6bd0037c Sadly did not help. I am thinking just to try out the outline of the code I wrote and see how it works. BTW, this is with the traditional Xen QEMU. I couldn't get the SeaBIOS QEMU to work with this and I think that is b/c the XenStore monitoring of the CPUs availability is just not there. Ah, it might be via the QMP layer thingy. I should look in that. > > > -- > Regards, > Igor > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >