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,URIBL_BLOCKED autolearn=ham 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 26DC2C0044C for ; Wed, 7 Nov 2018 22:32:11 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id CA3572086C for ; Wed, 7 Nov 2018 22:32:10 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org CA3572086C Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.ibm.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728381AbeKHIEg (ORCPT ); Thu, 8 Nov 2018 03:04:36 -0500 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:58660 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728361AbeKHIEf (ORCPT ); Thu, 8 Nov 2018 03:04:35 -0500 Received: from pps.filterd (m0098394.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.22/8.16.0.22) with SMTP id wA7MSTOJ033976 for ; Wed, 7 Nov 2018 17:32:08 -0500 Received: from e06smtp03.uk.ibm.com (e06smtp03.uk.ibm.com [195.75.94.99]) by mx0a-001b2d01.pphosted.com with ESMTP id 2nm4myaggy-1 (version=TLSv1.2 cipher=AES256-GCM-SHA384 bits=256 verify=NOT) for ; Wed, 07 Nov 2018 17:32:07 -0500 Received: from localhost by e06smtp03.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 7 Nov 2018 22:32:05 -0000 Received: from b06cxnps3074.portsmouth.uk.ibm.com (9.149.109.194) by e06smtp03.uk.ibm.com (192.168.101.133) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; (version=TLSv1/SSLv3 cipher=AES256-GCM-SHA384 bits=256/256) Wed, 7 Nov 2018 22:32:01 -0000 Received: from d06av24.portsmouth.uk.ibm.com (d06av24.portsmouth.uk.ibm.com [9.149.105.60]) by b06cxnps3074.portsmouth.uk.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id wA7MW01W61603946 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL); Wed, 7 Nov 2018 22:32:00 GMT Received: from d06av24.portsmouth.uk.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 029C142042; Wed, 7 Nov 2018 22:32:00 +0000 (GMT) Received: from d06av24.portsmouth.uk.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 5FE274204F; Wed, 7 Nov 2018 22:31:59 +0000 (GMT) Received: from [9.145.168.249] (unknown [9.145.168.249]) by d06av24.portsmouth.uk.ibm.com (Postfix) with ESMTP; Wed, 7 Nov 2018 22:31:59 +0000 (GMT) Reply-To: pmorel@linux.ibm.com Subject: Re: [PATCH v1 4/7] vfio: ap: AP Queue Interrupt Control VFIO ioctl calls To: Tony Krowiak , borntraeger@de.ibm.com Cc: alex.williamson@redhat.com, cohuck@redhat.com, linux-kernel@vger.kernel.org, linux-s390@vger.kernel.org, kvm@vger.kernel.org, frankja@linux.ibm.com, pasic@linux.ibm.com, david@redhat.com, schwidefsky@de.ibm.com, heiko.carstens@de.ibm.com, freude@linux.ibm.com, mimu@linux.ibm.com References: <1541009577-29656-1-git-send-email-pmorel@linux.ibm.com> <1541009577-29656-5-git-send-email-pmorel@linux.ibm.com> <7e4cbc97-8c77-b393-efdd-6fd8550c15f1@linux.ibm.com> From: Pierre Morel Date: Wed, 7 Nov 2018 23:31:59 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.2.1 MIME-Version: 1.0 In-Reply-To: <7e4cbc97-8c77-b393-efdd-6fd8550c15f1@linux.ibm.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 8bit X-TM-AS-GCONF: 00 x-cbid: 18110722-0012-0000-0000-000002C6D834 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 18110722-0013-0000-0000-000020FB24A3 Message-Id: X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2018-11-07_18:,, 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=8.0.1-1807170000 definitions=main-1811070199 Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 06/11/2018 21:21, Tony Krowiak wrote: > On 10/31/18 2:12 PM, Pierre Morel wrote: >> This is the implementation of the VFIO ioctl calls to handle >> the AQIC interception and use GISA to handle interrupts. >> >> Signed-off-by: Pierre Morel >> --- >>   drivers/s390/crypto/vfio_ap_ops.c | 95 +++++++++++++++++++++++++++++++ >>   1 file changed, 95 insertions(+) >> >> diff --git a/drivers/s390/crypto/vfio_ap_ops.c >> b/drivers/s390/crypto/vfio_ap_ops.c >> index 272ef427dcc0..f68102163bf4 100644 >> --- a/drivers/s390/crypto/vfio_ap_ops.c >> +++ b/drivers/s390/crypto/vfio_ap_ops.c >> @@ -895,12 +895,107 @@ static int >> vfio_ap_mdev_get_device_info(unsigned long arg) >>       return copy_to_user((void __user *)arg, &info, minsz); >>   } >> +static int ap_ioctl_setirq(struct ap_matrix_mdev *matrix_mdev, > > In keeping with the other function names in this file, how about > naming this function vfio_ap_mdev_setirq??? OK, agreed. > >> +               struct vfio_ap_aqic *parm) >> +{ >> +    struct aqic_gisa aqic_gisa = reg2aqic(0); >> +    struct kvm_s390_gisa *gisa = matrix_mdev->kvm->arch.gisa; >> +    struct ap_status ap_status = reg2status(0); >> +    unsigned long p; >> +    int ret = -1; >> +    int apqn; >> +    uint32_t gd; >> + >> +    apqn = (int)(parm->cmd & 0xffff); >> + >> +    gd = matrix_mdev->kvm->vcpus[0]->arch.sie_block->gd; >> +    if (gd & 0x01) >> +        aqic_gisa.f = 1; >> +    aqic_gisa.gisc = matrix_mdev->gisc; > > Can't you get this value from parm? I don't see any relationship > between the mdev device and gisc, why store it there? The idea is that we may want to report this value to the admin or as debug information, so I wanted to keep track of it. > >> +    aqic_gisa.isc = GAL_ISC; >> +    aqic_gisa.ir = 1; >> +    aqic_gisa.gisao = gisa->next_alert >> 4; >> + >> +    p = (unsigned long) page_address(matrix_mdev->map->page); >> +    p += (matrix_mdev->map->guest_addr & 0x0fff); >> + >> +    ret = ap_host_aqic((uint64_t)apqn, aqic2reg(aqic_gisa), p); >> +    parm->status = ret; >> + >> +    ap_status = reg2status(ret); >> +    return (ap_status.rc) ? -EIO : 0; >> +} >> + >> +static int ap_ioctl_clrirq(struct ap_matrix_mdev *matrix_mdev, >> +               struct vfio_ap_aqic *parm) > > In keeping with the other function names in this file, how about > naming this function vfio_ap_mdev_clrirq, or better yet, > vfio_ap_mdev_clear_irq??? agreed too. > >> +{ >> +    struct aqic_gisa aqic_gisa = reg2aqic(0); >> +    struct ap_status ap_status = reg2status(0) > +    int apqn; >> +    int retval; >> +    uint32_t gd; >> + >> +    apqn = (int)(parm->cmd & 0xffff); >> + >> +    gd = matrix_mdev->kvm->vcpus[0]->arch.sie_block->gd; >> +    if (gd & 0x01) >> +        aqic_gisa.f = 1; >> +    aqic_gisa.ir = 0; >> + >> +    retval = ap_host_aqic((uint64_t)apqn, aqic2reg(aqic_gisa), 0); >> +    parm->status = retval; >> + >> +    ap_status = reg2status(retval); >> +    return (ap_status.rc) ? -EIO : 0; >> +} >> + >>   static ssize_t vfio_ap_mdev_ioctl(struct mdev_device *mdev, >>                       unsigned int cmd, unsigned long arg) >>   { >>       int ret; >> +    struct ap_matrix_mdev *matrix_mdev = mdev_get_drvdata(mdev); >> +    struct s390_io_adapter *adapter; >> +    struct vfio_ap_aqic parm; >> +    struct s390_map_info *map; >> +    int apqn, found = 0; >>       switch (cmd) { >> +    case VFIO_AP_SET_IRQ: >> +        if (copy_from_user(&parm, (void __user *)arg, sizeof(parm))) >> +            return -EFAULT; >> +        apqn = (int)(parm.cmd & 0xffff); >> +        parm.status &= 0x00000000ffffffffUL; >> +        matrix_mdev->gisc = parm.status & 0x07; > > It seems that the only reason for the 'gisc' field in matrix_mdev > is to pass the value to the ap_ioctl_setirq() function. Since the > gisc has nothing to do with the mdev device and the 'parm' is being > passed to ap_ioctl_setirq(), why not just eliminate the > matrix_mdev->gisc field and get it from the 'parm' parameter in > ap_ioctl_setirq()? OK, seems better. > >> +        /* find the adapter */ap_ioctl_setirq() >> +        adapter = matrix_mdev->kvm->arch.adapters[parm.adapter_id]; >> +        if (!adapter) >> +            return -ENOENT; >> +        down_write(&adapter->maps_lock); >> +        list_for_each_entry(map, &adapter->maps, list) { >> +            if (map->guest_addr == parm.nib) { >> +                found = 1; >> +                break; >> +            } >> +        } >> +        up_write(&adapter->maps_lock); >> + >> +        if (!found) >> +            return -EINVAL; >> + >> +        matrix_mdev->map = map; > > See my comment above about matrix_mdev->gisc. Can't we just get rid > of the matrix_mdev->map field and pass the map into the > ap_ioctl_setirq() function? or calculate it from parm... as we give parm as argument to this function > >> +        ret = ap_ioctl_setirq(matrix_mdev, &parm); >> +        parm.status &= 0x00000000ffffffffUL; >> +        if (copy_to_user((void __user *)arg, &parm, sizeof(parm))) >> +            return -EFAULT; >> + >> +        break; > > IMHO, the case statements should only determine which ioctl is being > invoked and call the appropriate function to handle it. All of the above > code could be in an intermediate function called from this case > statement, thus reducing the case to calling the intermediate function. OK, I can do so, however I would like to let the __user access here. > >> +    case VFIO_AP_CLEAR_IRQ: >> +        if (copy_from_user(&parm, (void __user *)arg, sizeof(parm))) >> +            return -EFAULT; >> +        ret = ap_ioctl_clrirq(matrix_mdev, &parm); >> +        if (copy_to_user((void __user *)arg, &parm, sizeof(parm))) >> +            return -EFAULT; >> +        break; >>       case VFIO_DEVICE_GET_INFO: >>           ret = vfio_ap_mdev_get_device_info(arg); >>           break; >> > Thanks Pierre -- Pierre Morel Linux/KVM/QEMU in Böblingen - Germany