From nobody Mon Feb 9 21:12:23 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1681991054; cv=none; d=zohomail.com; s=zohoarc; b=EmTVwAUFEboRpLve8NpbMkx0/DPHNtjfICuGoNdF5uP1MnnDniwE072U4a7OPBh/G8fQIrEOOBzZMx7FeFWwpXaxQf19fDv1MRk1MupAVd0RtwFH9HTnIFm3LeMzNOniQIMkOehnRVkScQCMr2bgSylMCS+2qYxh+amnFyBwHL4= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1681991054; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=V8/JM/4FpUbs1MFSPNPGCdntPIYaUvTzToJc4P5DEws=; b=HeBKxKF8bVRSuXYwc/qE+IZpGYoS0/Np4RDKH68zc68qs1wFjLThH0E7M/Uk91lJtK2XZRGLchwi6dxNxJRM7pRGoYunk3A3+1OGrWyhEz+Cckd0lD01yGxp53S3WNqdqWX+ZX0mp/kOt08B5hgdi16tUKQSiG97dUWmUxMu0F4= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1681991054354795.5815734821466; Thu, 20 Apr 2023 04:44:14 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ppSfo-0002lr-VF; Thu, 20 Apr 2023 07:42:01 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ppSdV-0008AI-FU for qemu-devel@nongnu.org; Thu, 20 Apr 2023 07:39:42 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ppSdN-0001j5-Ev for qemu-devel@nongnu.org; Thu, 20 Apr 2023 07:39:37 -0400 Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-549-iXCyyEoXOhWgcUvuTttC3g-1; Thu, 20 Apr 2023 07:39:02 -0400 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.rdu2.redhat.com [10.11.54.5]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id B2226280BF65; Thu, 20 Apr 2023 11:39:01 +0000 (UTC) Received: from localhost (unknown [10.39.193.254]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9A1EC63A5E; Thu, 20 Apr 2023 11:39:00 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1681990745; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=V8/JM/4FpUbs1MFSPNPGCdntPIYaUvTzToJc4P5DEws=; b=LWD1bXZyTARn5T/RL8QolxwrGjHwamQ+ZCxGi0m6N92MqGDdtLf0nYDRSF4TUtWMijv8Wt R7Gf97xs231fQQ+EOg4JdLTAWv2RREF0EO+9hLvGNFi8icufxExpS/29N9hyeL/pl1EZaq 9AFRCdCRZWOlCZ6+905nUoP+5BPK9cM= X-MC-Unique: iXCyyEoXOhWgcUvuTttC3g-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Peter Lieven , Stefan Hajnoczi , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Paolo Bonzini , =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= , Xie Yongji , Juan Quintela , qemu-block@nongnu.org, Eduardo Habkost , Richard Henderson , David Woodhouse , Stefan Weil , Fam Zheng , Julia Suvorova , Ronnie Sahlberg , xen-devel@lists.xenproject.org, Hanna Reitz , "Dr. David Alan Gilbert" , eesposit@redhat.com, Kevin Wolf , Marcel Apfelbaum , Stefano Stabellini , Paul Durrant , Aarushi Mehta , "Michael S. Tsirkin" , Anthony Perard , "Richard W.M. Jones" , Coiby Xu , Stefano Garzarella Subject: [PATCH v3 18/20] virtio-scsi: implement BlockDevOps->drained_begin() Date: Thu, 20 Apr 2023 07:37:30 -0400 Message-Id: <20230420113732.336620-19-stefanha@redhat.com> In-Reply-To: <20230420113732.336620-1-stefanha@redhat.com> References: <20230420113732.336620-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.5 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, T_SCC_BODY_TEXT_LINE=-0.01, T_SPF_TEMPERROR=0.01 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1681991056413100003 Content-Type: text/plain; charset="utf-8" The virtio-scsi Host Bus Adapter provides access to devices on a SCSI bus. Those SCSI devices typically have a BlockBackend. When the BlockBackend enters a drained section, the SCSI device must temporarily stop submitting new I/O requests. Implement this behavior by temporarily stopping virtio-scsi virtqueue processing when one of the SCSI devices enters a drained section. The new scsi_device_drained_begin() API allows scsi-disk to message the virtio-scsi HBA. scsi_device_drained_begin() uses a drain counter so that multiple SCSI devices can have overlapping drained sections. The HBA only sees one pair of .drained_begin/end() calls. After this commit, virtio-scsi no longer depends on hw/virtio's ioeventfd aio_set_event_notifier(is_external=3Dtrue). This commit is a step towards removing the aio_disable_external() API. Signed-off-by: Stefan Hajnoczi --- include/hw/scsi/scsi.h | 14 ++++++++++++ hw/scsi/scsi-bus.c | 40 +++++++++++++++++++++++++++++++++ hw/scsi/scsi-disk.c | 27 +++++++++++++++++----- hw/scsi/virtio-scsi-dataplane.c | 22 ++++++++++-------- hw/scsi/virtio-scsi.c | 38 +++++++++++++++++++++++++++++++ hw/scsi/trace-events | 2 ++ 6 files changed, 129 insertions(+), 14 deletions(-) diff --git a/include/hw/scsi/scsi.h b/include/hw/scsi/scsi.h index 6f23a7a73e..e2bb1a2fbf 100644 --- a/include/hw/scsi/scsi.h +++ b/include/hw/scsi/scsi.h @@ -133,6 +133,16 @@ struct SCSIBusInfo { void (*save_request)(QEMUFile *f, SCSIRequest *req); void *(*load_request)(QEMUFile *f, SCSIRequest *req); void (*free_request)(SCSIBus *bus, void *priv); + + /* + * Temporarily stop submitting new requests between drained_begin() and + * drained_end(). Called from the main loop thread with the BQL held. + * + * Implement these callbacks if request processing is triggered by a f= ile + * descriptor like an EventNotifier. Otherwise set them to NULL. + */ + void (*drained_begin)(SCSIBus *bus); + void (*drained_end)(SCSIBus *bus); }; =20 #define TYPE_SCSI_BUS "SCSI" @@ -144,6 +154,8 @@ struct SCSIBus { =20 SCSISense unit_attention; const SCSIBusInfo *info; + + int drain_count; /* protected by BQL */ }; =20 /** @@ -213,6 +225,8 @@ void scsi_req_cancel_complete(SCSIRequest *req); void scsi_req_cancel(SCSIRequest *req); void scsi_req_cancel_async(SCSIRequest *req, Notifier *notifier); void scsi_req_retry(SCSIRequest *req); +void scsi_device_drained_begin(SCSIDevice *sdev); +void scsi_device_drained_end(SCSIDevice *sdev); void scsi_device_purge_requests(SCSIDevice *sdev, SCSISense sense); void scsi_device_set_ua(SCSIDevice *sdev, SCSISense sense); void scsi_device_report_change(SCSIDevice *dev, SCSISense sense); diff --git a/hw/scsi/scsi-bus.c b/hw/scsi/scsi-bus.c index 64d7311757..b571fdf895 100644 --- a/hw/scsi/scsi-bus.c +++ b/hw/scsi/scsi-bus.c @@ -1668,6 +1668,46 @@ void scsi_device_purge_requests(SCSIDevice *sdev, SC= SISense sense) scsi_device_set_ua(sdev, sense); } =20 +void scsi_device_drained_begin(SCSIDevice *sdev) +{ + SCSIBus *bus =3D DO_UPCAST(SCSIBus, qbus, sdev->qdev.parent_bus); + if (!bus) { + return; + } + + assert(qemu_get_current_aio_context() =3D=3D qemu_get_aio_context()); + assert(bus->drain_count < INT_MAX); + + /* + * Multiple BlockBackends can be on a SCSIBus and each may begin/end + * draining at any time. Keep a counter so HBAs only see begin/end onc= e. + */ + if (bus->drain_count++ =3D=3D 0) { + trace_scsi_bus_drained_begin(bus, sdev); + if (bus->info->drained_begin) { + bus->info->drained_begin(bus); + } + } +} + +void scsi_device_drained_end(SCSIDevice *sdev) +{ + SCSIBus *bus =3D DO_UPCAST(SCSIBus, qbus, sdev->qdev.parent_bus); + if (!bus) { + return; + } + + assert(qemu_get_current_aio_context() =3D=3D qemu_get_aio_context()); + assert(bus->drain_count > 0); + + if (bus->drain_count-- =3D=3D 1) { + trace_scsi_bus_drained_end(bus, sdev); + if (bus->info->drained_end) { + bus->info->drained_end(bus); + } + } +} + static char *scsibus_get_dev_path(DeviceState *dev) { SCSIDevice *d =3D SCSI_DEVICE(dev); diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c index e01bd84541..2249087d6a 100644 --- a/hw/scsi/scsi-disk.c +++ b/hw/scsi/scsi-disk.c @@ -2360,6 +2360,20 @@ static void scsi_disk_reset(DeviceState *dev) s->qdev.scsi_version =3D s->qdev.default_scsi_version; } =20 +static void scsi_disk_drained_begin(void *opaque) +{ + SCSIDiskState *s =3D opaque; + + scsi_device_drained_begin(&s->qdev); +} + +static void scsi_disk_drained_end(void *opaque) +{ + SCSIDiskState *s =3D opaque; + + scsi_device_drained_end(&s->qdev); +} + static void scsi_disk_resize_cb(void *opaque) { SCSIDiskState *s =3D opaque; @@ -2414,16 +2428,19 @@ static bool scsi_cd_is_medium_locked(void *opaque) } =20 static const BlockDevOps scsi_disk_removable_block_ops =3D { - .change_media_cb =3D scsi_cd_change_media_cb, + .change_media_cb =3D scsi_cd_change_media_cb, + .drained_begin =3D scsi_disk_drained_begin, + .drained_end =3D scsi_disk_drained_end, .eject_request_cb =3D scsi_cd_eject_request_cb, - .is_tray_open =3D scsi_cd_is_tray_open, .is_medium_locked =3D scsi_cd_is_medium_locked, - - .resize_cb =3D scsi_disk_resize_cb, + .is_tray_open =3D scsi_cd_is_tray_open, + .resize_cb =3D scsi_disk_resize_cb, }; =20 static const BlockDevOps scsi_disk_block_ops =3D { - .resize_cb =3D scsi_disk_resize_cb, + .drained_begin =3D scsi_disk_drained_begin, + .drained_end =3D scsi_disk_drained_end, + .resize_cb =3D scsi_disk_resize_cb, }; =20 static void scsi_disk_unit_attention_reported(SCSIDevice *dev) diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi-dataplan= e.c index 81643445ed..1060038e13 100644 --- a/hw/scsi/virtio-scsi-dataplane.c +++ b/hw/scsi/virtio-scsi-dataplane.c @@ -153,14 +153,16 @@ int virtio_scsi_dataplane_start(VirtIODevice *vdev) s->dataplane_starting =3D false; s->dataplane_started =3D true; =20 - aio_context_acquire(s->ctx); - virtio_queue_aio_attach_host_notifier(vs->ctrl_vq, s->ctx); - virtio_queue_aio_attach_host_notifier_no_poll(vs->event_vq, s->ctx); + if (s->bus.drain_count =3D=3D 0) { + aio_context_acquire(s->ctx); + virtio_queue_aio_attach_host_notifier(vs->ctrl_vq, s->ctx); + virtio_queue_aio_attach_host_notifier_no_poll(vs->event_vq, s->ctx= ); =20 - for (i =3D 0; i < vs->conf.num_queues; i++) { - virtio_queue_aio_attach_host_notifier(vs->cmd_vqs[i], s->ctx); + for (i =3D 0; i < vs->conf.num_queues; i++) { + virtio_queue_aio_attach_host_notifier(vs->cmd_vqs[i], s->ctx); + } + aio_context_release(s->ctx); } - aio_context_release(s->ctx); return 0; =20 fail_host_notifiers: @@ -206,9 +208,11 @@ void virtio_scsi_dataplane_stop(VirtIODevice *vdev) } s->dataplane_stopping =3D true; =20 - aio_context_acquire(s->ctx); - aio_wait_bh_oneshot(s->ctx, virtio_scsi_dataplane_stop_bh, s); - aio_context_release(s->ctx); + if (s->bus.drain_count =3D=3D 0) { + aio_context_acquire(s->ctx); + aio_wait_bh_oneshot(s->ctx, virtio_scsi_dataplane_stop_bh, s); + aio_context_release(s->ctx); + } =20 blk_drain_all(); /* ensure there are no in-flight requests */ =20 diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c index a02f9233ec..eba1e84dac 100644 --- a/hw/scsi/virtio-scsi.c +++ b/hw/scsi/virtio-scsi.c @@ -1081,6 +1081,42 @@ static void virtio_scsi_hotunplug(HotplugHandler *ho= tplug_dev, DeviceState *dev, } } =20 +/* Suspend virtqueue ioeventfd processing during drain */ +static void virtio_scsi_drained_begin(SCSIBus *bus) +{ + VirtIOSCSI *s =3D container_of(bus, VirtIOSCSI, bus); + VirtIODevice *vdev =3D VIRTIO_DEVICE(s); + uint32_t total_queues =3D VIRTIO_SCSI_VQ_NUM_FIXED + + s->parent_obj.conf.num_queues; + + if (!s->dataplane_started) { + return; + } + + for (uint32_t i =3D 0; i < total_queues; i++) { + VirtQueue *vq =3D virtio_get_queue(vdev, i); + virtio_queue_aio_detach_host_notifier(vq, s->ctx); + } +} + +/* Resume virtqueue ioeventfd processing after drain */ +static void virtio_scsi_drained_end(SCSIBus *bus) +{ + VirtIOSCSI *s =3D container_of(bus, VirtIOSCSI, bus); + VirtIODevice *vdev =3D VIRTIO_DEVICE(s); + uint32_t total_queues =3D VIRTIO_SCSI_VQ_NUM_FIXED + + s->parent_obj.conf.num_queues; + + if (!s->dataplane_started) { + return; + } + + for (uint32_t i =3D 0; i < total_queues; i++) { + VirtQueue *vq =3D virtio_get_queue(vdev, i); + virtio_queue_aio_attach_host_notifier(vq, s->ctx); + } +} + static struct SCSIBusInfo virtio_scsi_scsi_info =3D { .tcq =3D true, .max_channel =3D VIRTIO_SCSI_MAX_CHANNEL, @@ -1095,6 +1131,8 @@ static struct SCSIBusInfo virtio_scsi_scsi_info =3D { .get_sg_list =3D virtio_scsi_get_sg_list, .save_request =3D virtio_scsi_save_request, .load_request =3D virtio_scsi_load_request, + .drained_begin =3D virtio_scsi_drained_begin, + .drained_end =3D virtio_scsi_drained_end, }; =20 void virtio_scsi_common_realize(DeviceState *dev, diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events index ab238293f0..bdd4e2c7c7 100644 --- a/hw/scsi/trace-events +++ b/hw/scsi/trace-events @@ -6,6 +6,8 @@ scsi_req_cancel(int target, int lun, int tag) "target %d lu= n %d tag %d" scsi_req_data(int target, int lun, int tag, int len) "target %d lun %d tag= %d len %d" scsi_req_data_canceled(int target, int lun, int tag, int len) "target %d l= un %d tag %d len %d" scsi_req_dequeue(int target, int lun, int tag) "target %d lun %d tag %d" +scsi_bus_drained_begin(void *bus, void *sdev) "bus %p sdev %p" +scsi_bus_drained_end(void *bus, void *sdev) "bus %p sdev %p" scsi_req_continue(int target, int lun, int tag) "target %d lun %d tag %d" scsi_req_continue_canceled(int target, int lun, int tag) "target %d lun %d= tag %d" scsi_req_parsed(int target, int lun, int tag, int cmd, int mode, int xfer)= "target %d lun %d tag %d command %d dir %d length %d" --=20 2.39.2