From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58504) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dmM2f-0006s1-U4 for qemu-devel@nongnu.org; Mon, 28 Aug 2017 11:33:34 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dmM2b-0005K4-1t for qemu-devel@nongnu.org; Mon, 28 Aug 2017 11:33:33 -0400 Received: from mx1.redhat.com ([209.132.183.28]:36308) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dmM2a-0005Ju-Rz for qemu-devel@nongnu.org; Mon, 28 Aug 2017 11:33:28 -0400 Date: Mon, 28 Aug 2017 17:33:24 +0200 From: Cornelia Huck Message-ID: <20170828173324.5723a231.cohuck@redhat.com> In-Reply-To: <1503907487-2764-4-git-send-email-zyimin@linux.vnet.ibm.com> References: <1503907487-2764-1-git-send-email-zyimin@linux.vnet.ibm.com> <1503907487-2764-4-git-send-email-zyimin@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 3/4] s390x/pci: fixup ind_offset of msix routing entry List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Yi Min Zhao Cc: qemu-devel@nongnu.org, borntraeger@de.ibm.com, pasic@linux.vnet.ibm.com, pmorel@linux.vnet.ibm.com, agraf@suse.de, richard.henderson@linaro.org On Mon, 28 Aug 2017 10:04:46 +0200 Yi Min Zhao wrote: > The aibvo of zpci device should be constant after issued mpcifc > registering irqs instruction. Each msix vector should offset from the > aibvo. But for flic adapter interrupt, we should use the absolute > offset within the aibv. So let's use the aibvo+vector to fixup msix > routing entry. This makes sense, but I would tweak the description a bit. "The guest uses the mpcifc instruction to register the aibvo of a zpci device, which is the starting offset of indicators in the indicator area and thus remains constant. Each msix vector is an offset from the aibvo. When we map a msix route to an adapter route, we should not modify the starting offset, but instead add the vector to the starting offset to get the absolute offset in the specific route." I'm wondering how this was ever supposed to work? > > Signed-off-by: Yi Min Zhao > --- > target/s390x/kvm.c | 4 +--- > 1 file changed, 1 insertion(+), 3 deletions(-) > > diff --git a/target/s390x/kvm.c b/target/s390x/kvm.c > index e348bfb7cc..c08b7757e7 100644 > --- a/target/s390x/kvm.c > +++ b/target/s390x/kvm.c > @@ -2515,14 +2515,12 @@ int kvm_arch_fixup_msi_route(struct kvm_irq_routing_entry *route, > return -ENODEV; > } > > - pbdev->routes.adapter.ind_offset = vec; > - > route->type = KVM_IRQ_ROUTING_S390_ADAPTER; > route->flags = 0; > route->u.adapter.summary_addr = pbdev->routes.adapter.summary_addr; > route->u.adapter.ind_addr = pbdev->routes.adapter.ind_addr; > route->u.adapter.summary_offset = pbdev->routes.adapter.summary_offset; > - route->u.adapter.ind_offset = pbdev->routes.adapter.ind_offset; > + route->u.adapter.ind_offset = pbdev->routes.adapter.ind_offset + vec; > route->u.adapter.adapter_id = pbdev->routes.adapter.adapter_id; > return 0; > }