From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43138) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dpA74-0003uh-Ku for qemu-devel@nongnu.org; Tue, 05 Sep 2017 05:25:48 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dpA6z-0004nK-Fu for qemu-devel@nongnu.org; Tue, 05 Sep 2017 05:25:42 -0400 Received: from mx1.redhat.com ([209.132.183.28]:56024) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dpA6z-0004mT-77 for qemu-devel@nongnu.org; Tue, 05 Sep 2017 05:25:37 -0400 Date: Tue, 5 Sep 2017 11:25:33 +0200 From: Cornelia Huck Message-ID: <20170905112533.31e576b5.cohuck@redhat.com> In-Reply-To: <65b4c93c-fc2e-7ba3-b6e2-aa6c2312df3b@linux.vnet.ibm.com> References: <1504239778-29893-1-git-send-email-zyimin@linux.vnet.ibm.com> <1504239778-29893-2-git-send-email-zyimin@linux.vnet.ibm.com> <20170905102928.6b23a28a.cohuck@redhat.com> <0923f97e-4f69-0bf3-b9ab-8e5f69a56bbb@linux.vnet.ibm.com> <20170905105041.3409c8cd.cohuck@redhat.com> <07ef5826-6b09-5ac8-7d51-d00db7170437@linux.vnet.ibm.com> <20170905111540.027058d1.cohuck@redhat.com> <65b4c93c-fc2e-7ba3-b6e2-aa6c2312df3b@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v2 1/3] s390x/pci: remove idx from msix msg data 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 Tue, 5 Sep 2017 17:21:52 +0800 Yi Min Zhao wrote: > =E5=9C=A8 2017/9/5 =E4=B8=8B=E5=8D=885:15, Cornelia Huck =E5=86=99=E9=81= =93: > > On Tue, 5 Sep 2017 17:08:14 +0800 > > Yi Min Zhao wrote: > > =20 > >> =E5=9C=A8 2017/9/5 =E4=B8=8B=E5=8D=884:50, Cornelia Huck =E5=86=99=E9= =81=93: =20 > >>> On Tue, 5 Sep 2017 16:44:37 +0800 > >>> Yi Min Zhao wrote: > >>> =20 > >>>> =E5=9C=A8 2017/9/5 =E4=B8=8B=E5=8D=884:29, Cornelia Huck =E5=86=99= =E9=81=93: =20 > >>>>> On Fri, 1 Sep 2017 06:22:56 +0200 > >>>>> Yi Min Zhao wrote: > >>>>> =20 > >>>>>> PCIDevice pointer has been a parameter of kvm_arch_fixup_msi_route= (). > >>>>>> So we don't need to store zpci idx in msix message data to find ou= t the > >>>>>> specific zpci device. Instead, we could use pci device id to find = its > >>>>>> corresponding zpci device. > >>>>>> > >>>>>> Signed-off-by: Yi Min Zhao > >>>>>> --- > >>>>>> hw/s390x/s390-pci-bus.c | 16 +++++----------- > >>>>>> hw/s390x/s390-pci-bus.h | 2 ++ > >>>>>> hw/s390x/s390-pci-inst.c | 24 ------------------------ > >>>>>> hw/s390x/s390-pci-stub.c | 6 ++++++ > >>>>>> target/s390x/kvm.c | 7 +++++-- > >>>>>> 5 files changed, 18 insertions(+), 37 deletions(-) > >>>>>> > >>>>>> diff --git a/target/s390x/kvm.c b/target/s390x/kvm.c > >>>>>> index 1338c29528..3d490c5e4b 100644 > >>>>>> --- a/target/s390x/kvm.c > >>>>>> +++ b/target/s390x/kvm.c > >>>>>> @@ -2533,10 +2533,13 @@ int kvm_arch_fixup_msi_route(struct kvm_ir= q_routing_entry *route, > >>>>>> uint64_t address, uint32_t data, = PCIDevice *dev) > >>>>>> { > >>>>>> S390PCIBusDevice *pbdev; > >>>>>> - uint32_t idx =3D data >> ZPCI_MSI_VEC_BITS; > >>>>>> uint32_t vec =3D data & ZPCI_MSI_VEC_MASK; > >>>>>> =20 > >>>>>> - pbdev =3D s390_pci_find_dev_by_idx(s390_get_phb(), idx); > >>>>>> + if (!dev) { > >>>>>> + return -ENODEV; =20 > >>>>> Can this actually happen? =20 > >>>> I think this cannot happen. But I'm afraid that I miss something. > >>>> So I added this to avoid NULL pointer. But from the code and > >>>> my test, there has not been NULL pointer happened. =20 > >>> I'm wondering if that is in the same category as the instance I > >>> commented on above. Do you want to log something? > >>> =20 > >> For the case above, I ensure that zpci device must exist. But here, I'm > >> not sure. > >> Because it's called from outside. I'm not sure if the caller might call > >> kvm_irqchip_add_msi_route() with NULL as pci device argument. > >> > >> Although msix ctrl mr is accessed from outside. But its initialization > >> is controled by our code and the pointer to zpci device is saved as > >> mr's opaque. =20 > > OK. Maybe add a DPRINTF as for the condition below? =20 > OK. How about DPRINTF("add_msi_route no pci device\n")? > And change the DPRINTF for the below condition to > DPRINTF("add_msi_route no zpci device\n"). works for me > > =20 > >>>>> =20 > >>>>>> + } > >>>>>> + > >>>>>> + pbdev =3D s390_pci_find_dev_by_target(s390_get_phb(), DEVICE(= dev)->id); > >>>>>> if (!pbdev) { > >>>>>> DPRINTF("add_msi_route no dev\n"); > >>>>>> return -ENODEV; =20 > >>>>> =20 > >>> =20 > > =20 >=20