From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7B629C10F14 for ; Fri, 12 Apr 2019 07:09:19 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 4252020850 for ; Fri, 12 Apr 2019 07:09:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727067AbfDLHJO (ORCPT ); Fri, 12 Apr 2019 03:09:14 -0400 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:44710 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726024AbfDLHJO (ORCPT ); Fri, 12 Apr 2019 03:09:14 -0400 Received: from pps.filterd (m0098409.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.27/8.16.0.27) with SMTP id x3C753Pn002175 for ; Fri, 12 Apr 2019 03:09:13 -0400 Received: from e06smtp07.uk.ibm.com (e06smtp07.uk.ibm.com [195.75.94.103]) by mx0a-001b2d01.pphosted.com with ESMTP id 2rtkund5wf-1 (version=TLSv1.2 cipher=AES256-GCM-SHA384 bits=256 verify=NOT) for ; Fri, 12 Apr 2019 03:09:12 -0400 Received: from localhost by e06smtp07.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Fri, 12 Apr 2019 08:09:10 +0100 Received: from b06cxnps3075.portsmouth.uk.ibm.com (9.149.109.195) by e06smtp07.uk.ibm.com (192.168.101.137) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; (version=TLSv1/SSLv3 cipher=AES256-GCM-SHA384 bits=256/256) Fri, 12 Apr 2019 08:09:07 +0100 Received: from d06av24.portsmouth.uk.ibm.com (d06av24.portsmouth.uk.ibm.com [9.149.105.60]) by b06cxnps3075.portsmouth.uk.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id x3C795Yp27394154 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Fri, 12 Apr 2019 07:09:05 GMT Received: from d06av24.portsmouth.uk.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 913E04204B; Fri, 12 Apr 2019 07:09:05 +0000 (GMT) Received: from d06av24.portsmouth.uk.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 68C4542042; Fri, 12 Apr 2019 07:09:04 +0000 (GMT) Received: from [10.0.2.15] (unknown [9.145.19.216]) by d06av24.portsmouth.uk.ibm.com (Postfix) with ESMTP; Fri, 12 Apr 2019 07:09:04 +0000 (GMT) Subject: Re: [PATCH 6/7] s390: vfio-ap: handle dynamic config/deconfig of AP adapter To: Tony Krowiak , linux-s390@vger.kernel.org, linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: borntraeger@de.ibm.com, cohuck@redhat.com, frankja@linux.ibm.com, david@redhat.com, schwidefsky@de.ibm.com, heiko.carstens@de.ibm.com, pmorel@linux.ibm.com, pasic@linux.ibm.com, alex.williamson@redhat.com, kwankhede@nvidia.com References: <1555016604-2008-1-git-send-email-akrowiak@linux.ibm.com> <1555016604-2008-7-git-send-email-akrowiak@linux.ibm.com> From: Harald Freudenberger Date: Fri, 12 Apr 2019 09:09:04 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.6.1 MIME-Version: 1.0 In-Reply-To: <1555016604-2008-7-git-send-email-akrowiak@linux.ibm.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Content-Language: en-US X-TM-AS-GCONF: 00 x-cbid: 19041207-0028-0000-0000-000003600452 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 19041207-0029-0000-0000-0000241F2E39 Message-Id: <735b94b2-9c03-2757-a2d7-043d46958d70@linux.ibm.com> X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2019-04-12_05:,, signatures=0 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 malwarescore=0 suspectscore=0 phishscore=0 bulkscore=0 spamscore=0 clxscore=1015 lowpriorityscore=0 mlxscore=0 impostorscore=0 mlxlogscore=999 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1810050000 definitions=main-1904120048 Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org On 11.04.19 23:03, Tony Krowiak wrote: > Once an APQN is assigned to an mdev device it will remained assigned until > it is explicitly unassigned from the mdev device. The associated AP queue > devices, however, can come and go due to failures or deliberate actions by > a sysadmin. For example, a sysadmin can dynamically remove an AP adapter > card using the SE or by executing an SCLP command. This patch refactors > the probe and remove callbacks of the vfio_ap driver to handle dynamic > changes as follows: > > * Probe callback changes: > > If the APQN of the queue being probed is assigned to an mdev device, the > mdev device is in use by a guest, and the APQN is not set in the guest's > CRYCB, the CRYCB will be dynamically updated to give the guest access to > the queue. > > * Remove callback changes: > > If the APQN of the queue being removed is assigned to an mdev device, > the mdev > device is in use by a guest, and the APQN is set in the guest's CRYCB, > the CRYCB will be dynamically updated to remove the guest's access to > the adapter card associated with the queue. Keep in mind, the > architecture does not provide a way to remove access to a single queue > unless only one queue is in the guest's configuration, so it was decided > that it makes more sense to unplug the adapter from the guest. > > The APQN of the queue being removed will remain assigned to the mdev > device should the queue be dynamically returned to the configuration. > The queue will also be reset prior to returning control to the caller > (a.k.a., the AP bus). Didn't we agree that the reset is done by the AP bus anyway. This was why I did the rework around the remove callback in the AP bus code. > Signed-off-by: Tony Krowiak > --- > arch/s390/include/asm/kvm_host.h | 2 ++ > arch/s390/kvm/kvm-s390.c | 37 +++++++++++++++++++ > drivers/s390/crypto/vfio_ap_drv.c | 16 +++++++-- > drivers/s390/crypto/vfio_ap_ops.c | 67 +++++++++++++++++++++++++++++++++-- > drivers/s390/crypto/vfio_ap_private.h | 2 ++ > 5 files changed, 120 insertions(+), 4 deletions(-) > > diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h > index c47e22bba87f..0ce5d9b0df59 100644 > --- a/arch/s390/include/asm/kvm_host.h > +++ b/arch/s390/include/asm/kvm_host.h > @@ -895,6 +895,8 @@ void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, > void kvm_arch_crypto_clear_masks(struct kvm *kvm); > void kvm_arch_crypto_set_masks(struct kvm *kvm, unsigned long *apm, > unsigned long *aqm, unsigned long *adm); > +int kvm_arch_crypto_test_masks(struct kvm *kvm, unsigned long *apm, > + unsigned long *aqm, unsigned long *adm); > > extern int sie64a(struct kvm_s390_sie_block *, u64 *); > extern char sie_exit; > diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c > index 4638303ba6a8..5f423cdd29ba 100644 > --- a/arch/s390/kvm/kvm-s390.c > +++ b/arch/s390/kvm/kvm-s390.c > @@ -2217,6 +2217,43 @@ static void kvm_s390_set_crycb_format(struct kvm *kvm) > kvm->arch.crypto.crycbd |= CRYCB_FORMAT1; > } > > +int kvm_arch_crypto_test_masks(struct kvm *kvm, unsigned long *apm, > + unsigned long *aqm, unsigned long *adm) > +{ > + int ret; > + struct kvm_s390_crypto_cb *crycb = kvm->arch.crypto.crycb; > + > + switch (kvm->arch.crypto.crycbd & CRYCB_FORMAT_MASK) { > + case CRYCB_FORMAT2: /* APCB1 use 256 bits */ > + ret = bitmap_equal(apm, (unsigned long *)crycb->apcb1.apm, 256); > + VM_EVENT(kvm, 3, "TEST CRYCB: apm %016lx %016lx %016lx %016lx", > + apm[0], apm[1], apm[2], apm[3]); > + ret &= bitmap_equal(aqm, > + (unsigned long *)crycb->apcb1.aqm, 256); > + VM_EVENT(kvm, 3, "TEST CRYCB: aqm %016lx %016lx %016lx %016lx", > + aqm[0], aqm[1], aqm[2], aqm[3]); > + ret &= bitmap_equal(adm, > + (unsigned long *)crycb->apcb1.adm, 256); > + VM_EVENT(kvm, 3, "TEST CRYCB: adm %016lx %016lx %016lx %016lx", > + adm[0], adm[1], adm[2], adm[3]); > + break; > + case CRYCB_FORMAT1: > + case CRYCB_FORMAT0: /* Fall through both use APCB0 */ > + ret = bitmap_equal(apm, (unsigned long *)crycb->apcb1.apm, 64); > + ret &= bitmap_equal(aqm, (unsigned long *)crycb->apcb1.aqm, 16); > + ret &= bitmap_equal(adm, (unsigned long *)crycb->apcb1.adm, 16); > + VM_EVENT(kvm, 3, "TEST CRYCB: apm %016lx aqm %04x adm %04x", > + apm[0], *((unsigned short *)aqm), > + *((unsigned short *)adm)); > + break; > + default: /* Can not happen */ > + ret = 0; > + break; > + } > + return ret; > +} > +EXPORT_SYMBOL_GPL(kvm_arch_crypto_test_masks); > + > void kvm_arch_crypto_set_masks(struct kvm *kvm, unsigned long *apm, > unsigned long *aqm, unsigned long *adm) > { > diff --git a/drivers/s390/crypto/vfio_ap_drv.c b/drivers/s390/crypto/vfio_ap_drv.c > index f340a28c1d65..2a79d27d9730 100644 > --- a/drivers/s390/crypto/vfio_ap_drv.c > +++ b/drivers/s390/crypto/vfio_ap_drv.c > @@ -42,12 +42,24 @@ MODULE_DEVICE_TABLE(vfio_ap, ap_queue_ids); > > static int vfio_ap_queue_dev_probe(struct ap_device *apdev) > { > - return 0; > + struct ap_queue *apq = to_ap_queue(&apdev->device); > + unsigned long apid = AP_QID_CARD(apq->qid); > + unsigned long apqi = AP_QID_QUEUE(apq->qid); > + > + mutex_lock(&matrix_dev->lock); > + vfio_ap_mdev_probe_queue(apid, apqi); > + mutex_unlock(&matrix_dev->lock); > } > > static void vfio_ap_queue_dev_remove(struct ap_device *apdev) > { > - /* Nothing to do yet */ > + struct ap_queue *apq = to_ap_queue(&apdev->device); > + unsigned long apid = AP_QID_CARD(apq->qid); > + unsigned long apqi = AP_QID_QUEUE(apq->qid); > + > + mutex_lock(&matrix_dev->lock); > + vfio_ap_mdev_remove_queue(apid, apqi); > + mutex_unlock(&matrix_dev->lock); > } > > static void vfio_ap_matrix_dev_release(struct device *dev) > diff --git a/drivers/s390/crypto/vfio_ap_ops.c b/drivers/s390/crypto/vfio_ap_ops.c > index ade2c150fe6b..8a70707bf870 100644 > --- a/drivers/s390/crypto/vfio_ap_ops.c > +++ b/drivers/s390/crypto/vfio_ap_ops.c > @@ -683,8 +683,8 @@ static void vfio_ap_mdev_wait_for_qempty(unsigned long apid, unsigned long apqi) > msleep(20); > break; > default: > - pr_warn("%s: tapq err %02x: 0x%04x may not be empty\n", > - __func__, status.response_code, q->apqn); > + pr_warn("%s: tapq err %02x: %02lx%04lx may not be empty\n", > + __func__, status.response_code, apid, apqi); > return; > } > } while (--retry); > @@ -840,3 +840,66 @@ void vfio_ap_mdev_unregister(void) > { > mdev_unregister_device(&matrix_dev->device); > } > + > +static struct ap_matrix_mdev *vfio_ap_mdev_find_matrix_mdev(unsigned long apid, > + unsigned long apqi) > +{ > + struct ap_matrix_mdev *matrix_mdev; > + > + list_for_each_entry(matrix_mdev, &matrix_dev->mdev_list, node) { > + if (test_bit_inv(apid, matrix_mdev->matrix.apm) && > + test_bit_inv(apqi, matrix_mdev->matrix.aqm)) > + return matrix_mdev; > + } > + > + return NULL; > +} > + > +void vfio_ap_mdev_remove_queue(unsigned long apid, unsigned long apqi) > +{ > + struct ap_matrix_mdev *matrix_mdev; > + > + matrix_mdev = vfio_ap_mdev_find_matrix_mdev(apid, apqi); > + > + /* > + * If the queue is assigned to the mdev device and the mdev device > + * is in use by a guest > + */ > + if (matrix_mdev && matrix_mdev->kvm) { > + /* > + * If the queue is plugged into the guest, unplug the adapter > + * but keep the adapter assigned to the mdev device > + */ > + if (!kvm_arch_crypto_test_masks(matrix_mdev->kvm, > + matrix_mdev->matrix.apm, > + matrix_mdev->matrix.aqm, > + matrix_mdev->matrix.adm)) { > + clear_bit_inv(apid, matrix_mdev->matrix.apm); > + vfio_ap_mdev_update_crycb(matrix_mdev); > + set_bit_inv(apid, matrix_mdev->matrix.apm); > + } > + } > + > + vfio_ap_mdev_reset_queue(apid, apqi); > +} > + > +void vfio_ap_mdev_probe_queue(unsigned long apid, unsigned long apqi) > +{ > + struct ap_matrix_mdev *matrix_mdev; > + > + matrix_mdev = vfio_ap_mdev_find_matrix_mdev(apid, apqi); > + > + /* > + * If the queue is assigned to the mdev device and the mdev device > + * is in use by a guest > + */ > + if (matrix_mdev && matrix_mdev->kvm) { > + /* If the queue is not plugged into the guest, plug it in */ > + if (!kvm_arch_crypto_test_masks(matrix_mdev->kvm, > + matrix_mdev->matrix.apm, > + matrix_mdev->matrix.aqm, > + matrix_mdev->matrix.adm)) { > + vfio_ap_mdev_update_crycb(matrix_mdev); > + } > + } > +} > diff --git a/drivers/s390/crypto/vfio_ap_private.h b/drivers/s390/crypto/vfio_ap_private.h > index 76b7f98e47e9..acdd5bfabaaf 100644 > --- a/drivers/s390/crypto/vfio_ap_private.h > +++ b/drivers/s390/crypto/vfio_ap_private.h > @@ -85,5 +85,7 @@ struct ap_matrix_mdev { > > extern int vfio_ap_mdev_register(void); > extern void vfio_ap_mdev_unregister(void); > +void vfio_ap_mdev_remove_queue(unsigned long apid, unsigned long apqi); > +void vfio_ap_mdev_probe_queue(unsigned long apid, unsigned long apqi); > > #endif /* _VFIO_AP_PRIVATE_H_ */