From 4daa5054c599c8aec70f1264a01341a15921c145 Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Wed, 30 Nov 2022 11:24:36 +0000 Subject: vhost: enable vrings in vhost_dev_start() for vhost-user devices Commit 02b61f38d3 ("hw/virtio: incorporate backend features in features") properly negotiates VHOST_USER_F_PROTOCOL_FEATURES with the vhost-user backend, but we forgot to enable vrings as specified in docs/interop/vhost-user.rst: If ``VHOST_USER_F_PROTOCOL_FEATURES`` has not been negotiated, the ring starts directly in the enabled state. If ``VHOST_USER_F_PROTOCOL_FEATURES`` has been negotiated, the ring is initialized in a disabled state and is enabled by ``VHOST_USER_SET_VRING_ENABLE`` with parameter 1. Some vhost-user front-ends already did this by calling vhost_ops.vhost_set_vring_enable() directly: - backends/cryptodev-vhost.c - hw/net/virtio-net.c - hw/virtio/vhost-user-gpio.c But most didn't do that, so we would leave the vrings disabled and some backends would not work. We observed this issue with the rust version of virtiofsd [1], which uses the event loop [2] provided by the vhost-user-backend crate where requests are not processed if vring is not enabled. Let's fix this issue by enabling the vrings in vhost_dev_start() for vhost-user front-ends that don't already do this directly. Same thing also in vhost_dev_stop() where we disable vrings. [1] https://gitlab.com/virtio-fs/virtiofsd [2] https://github.com/rust-vmm/vhost/blob/240fc2966/crates/vhost-user-backend/src/event_loop.rs#L217 Fixes: 02b61f38d3 ("hw/virtio: incorporate backend features in features") Reported-by: German Maglione Tested-by: German Maglione Signed-off-by: Stefano Garzarella Acked-by: Raphael Norwitz Message-Id: <20221123131630.52020-1-sgarzare@redhat.com> Signed-off-by: Alex Bennée Reviewed-by: Michael S. Tsirkin Message-Id: <20221130112439.2527228-3-alex.bennee@linaro.org> Signed-off-by: Michael S. Tsirkin --- hw/virtio/vhost-user-gpio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'hw/virtio/vhost-user-gpio.c') diff --git a/hw/virtio/vhost-user-gpio.c b/hw/virtio/vhost-user-gpio.c index 5851cb3bc9..0b40ebd15a 100644 --- a/hw/virtio/vhost-user-gpio.c +++ b/hw/virtio/vhost-user-gpio.c @@ -81,7 +81,7 @@ static int vu_gpio_start(VirtIODevice *vdev) */ vhost_ack_features(&gpio->vhost_dev, feature_bits, vdev->guest_features); - ret = vhost_dev_start(&gpio->vhost_dev, vdev); + ret = vhost_dev_start(&gpio->vhost_dev, vdev, false); if (ret < 0) { error_report("Error starting vhost-user-gpio: %d", ret); goto err_guest_notifiers; @@ -139,7 +139,7 @@ static void vu_gpio_stop(VirtIODevice *vdev) return; } - vhost_dev_stop(vhost_dev, vdev); + vhost_dev_stop(vhost_dev, vdev, false); ret = k->set_guest_notifiers(qbus->parent, vhost_dev->nvqs, false); if (ret < 0) { -- cgit v1.2.3-55-g7522 From 060f4a944072ecf37cece0f16a0609babfb679b8 Mon Sep 17 00:00:00 2001 From: Alex Bennée Date: Wed, 30 Nov 2022 11:24:37 +0000 Subject: hw/virtio: add started_vu status field to vhost-user-gpio As per the fix to vhost-user-blk in f5b22d06fb (vhost: recheck dev state in the vhost_migration_log routine) we really should track the connection and starting separately. Signed-off-by: Alex Bennée Reviewed-by: Michael S. Tsirkin Message-Id: <20221130112439.2527228-4-alex.bennee@linaro.org> Signed-off-by: Michael S. Tsirkin --- hw/virtio/vhost-user-gpio.c | 11 ++++------- include/hw/virtio/vhost-user-gpio.h | 10 ++++++++++ 2 files changed, 14 insertions(+), 7 deletions(-) (limited to 'hw/virtio/vhost-user-gpio.c') diff --git a/hw/virtio/vhost-user-gpio.c b/hw/virtio/vhost-user-gpio.c index 0b40ebd15a..be9be08b4c 100644 --- a/hw/virtio/vhost-user-gpio.c +++ b/hw/virtio/vhost-user-gpio.c @@ -86,6 +86,7 @@ static int vu_gpio_start(VirtIODevice *vdev) error_report("Error starting vhost-user-gpio: %d", ret); goto err_guest_notifiers; } + gpio->started_vu = true; /* * guest_notifier_mask/pending not used yet, so just unmask @@ -126,16 +127,12 @@ static void vu_gpio_stop(VirtIODevice *vdev) struct vhost_dev *vhost_dev = &gpio->vhost_dev; int ret; - if (!k->set_guest_notifiers) { + if (!gpio->started_vu) { return; } + gpio->started_vu = false; - /* - * We can call vu_gpio_stop multiple times, for example from - * vm_state_notify and the final object finalisation. Check we - * aren't already stopped before doing so. - */ - if (!vhost_dev_is_started(vhost_dev)) { + if (!k->set_guest_notifiers) { return; } diff --git a/include/hw/virtio/vhost-user-gpio.h b/include/hw/virtio/vhost-user-gpio.h index 4fe9aeecc0..a9305c5e6c 100644 --- a/include/hw/virtio/vhost-user-gpio.h +++ b/include/hw/virtio/vhost-user-gpio.h @@ -28,7 +28,17 @@ struct VHostUserGPIO { VhostUserState vhost_user; VirtQueue *command_vq; VirtQueue *interrupt_vq; + /** + * There are at least two steps of initialization of the + * vhost-user device. The first is a "connect" step and + * second is a "start" step. Make a separation between + * those initialization phases by using two fields. + * + * @connected: see vu_gpio_connect()/vu_gpio_disconnect() + * @started_vu: see vu_gpio_start()/vu_gpio_stop() + */ bool connected; + bool started_vu; /*< public >*/ }; -- cgit v1.2.3-55-g7522 From 71e076a07dc195129fe25d90d4b276be3b2f12d8 Mon Sep 17 00:00:00 2001 From: Alex Bennée Date: Wed, 30 Nov 2022 11:24:38 +0000 Subject: hw/virtio: generalise CHR_EVENT_CLOSED handling ..and use for both virtio-user-blk and virtio-user-gpio. This avoids the circular close by deferring shutdown due to disconnection until a later point. virtio-user-blk already had this mechanism in place so generalise it as a vhost-user helper function and use for both blk and gpio devices. While we are at it we also fix up vhost-user-gpio to re-establish the event handler after close down so we can reconnect later. Signed-off-by: Alex Bennée Reviewed-by: Michael S. Tsirkin Reviewed-by: Raphael Norwitz Message-Id: <20221130112439.2527228-5-alex.bennee@linaro.org> Signed-off-by: Michael S. Tsirkin --- hw/block/vhost-user-blk.c | 41 +++--------------------- hw/virtio/vhost-user-gpio.c | 11 ++++++- hw/virtio/vhost-user.c | 71 ++++++++++++++++++++++++++++++++++++++++++ include/hw/virtio/vhost-user.h | 18 +++++++++++ 4 files changed, 104 insertions(+), 37 deletions(-) (limited to 'hw/virtio/vhost-user-gpio.c') diff --git a/hw/block/vhost-user-blk.c b/hw/block/vhost-user-blk.c index 1177064631..aff4d2b8cb 100644 --- a/hw/block/vhost-user-blk.c +++ b/hw/block/vhost-user-blk.c @@ -369,17 +369,10 @@ static void vhost_user_blk_disconnect(DeviceState *dev) vhost_user_blk_stop(vdev); vhost_dev_cleanup(&s->dev); -} -static void vhost_user_blk_chr_closed_bh(void *opaque) -{ - DeviceState *dev = opaque; - VirtIODevice *vdev = VIRTIO_DEVICE(dev); - VHostUserBlk *s = VHOST_USER_BLK(vdev); - - vhost_user_blk_disconnect(dev); + /* Re-instate the event handler for new connections */ qemu_chr_fe_set_handlers(&s->chardev, NULL, NULL, vhost_user_blk_event, - NULL, opaque, NULL, true); + NULL, dev, NULL, true); } static void vhost_user_blk_event(void *opaque, QEMUChrEvent event) @@ -398,33 +391,9 @@ static void vhost_user_blk_event(void *opaque, QEMUChrEvent event) } break; case CHR_EVENT_CLOSED: - if (!runstate_check(RUN_STATE_SHUTDOWN)) { - /* - * A close event may happen during a read/write, but vhost - * code assumes the vhost_dev remains setup, so delay the - * stop & clear. - */ - AioContext *ctx = qemu_get_current_aio_context(); - - qemu_chr_fe_set_handlers(&s->chardev, NULL, NULL, NULL, NULL, - NULL, NULL, false); - aio_bh_schedule_oneshot(ctx, vhost_user_blk_chr_closed_bh, opaque); - - /* - * Move vhost device to the stopped state. The vhost-user device - * will be clean up and disconnected in BH. This can be useful in - * the vhost migration code. If disconnect was caught there is an - * option for the general vhost code to get the dev state without - * knowing its type (in this case vhost-user). - * - * FIXME: this is sketchy to be reaching into vhost_dev - * now because we are forcing something that implies we - * have executed vhost_dev_stop() but that won't happen - * until vhost_user_blk_stop() gets called from the bh. - * Really this state check should be tracked locally. - */ - s->dev.started = false; - } + /* defer close until later to avoid circular close */ + vhost_user_async_close(dev, &s->chardev, &s->dev, + vhost_user_blk_disconnect); break; case CHR_EVENT_BREAK: case CHR_EVENT_MUX_IN: diff --git a/hw/virtio/vhost-user-gpio.c b/hw/virtio/vhost-user-gpio.c index be9be08b4c..b7b82a1099 100644 --- a/hw/virtio/vhost-user-gpio.c +++ b/hw/virtio/vhost-user-gpio.c @@ -233,6 +233,8 @@ static int vu_gpio_connect(DeviceState *dev, Error **errp) return 0; } +static void vu_gpio_event(void *opaque, QEMUChrEvent event); + static void vu_gpio_disconnect(DeviceState *dev) { VirtIODevice *vdev = VIRTIO_DEVICE(dev); @@ -245,6 +247,11 @@ static void vu_gpio_disconnect(DeviceState *dev) vu_gpio_stop(vdev); vhost_dev_cleanup(&gpio->vhost_dev); + + /* Re-instate the event handler for new connections */ + qemu_chr_fe_set_handlers(&gpio->chardev, + NULL, NULL, vu_gpio_event, + NULL, dev, NULL, true); } static void vu_gpio_event(void *opaque, QEMUChrEvent event) @@ -262,7 +269,9 @@ static void vu_gpio_event(void *opaque, QEMUChrEvent event) } break; case CHR_EVENT_CLOSED: - vu_gpio_disconnect(dev); + /* defer close until later to avoid circular close */ + vhost_user_async_close(dev, &gpio->chardev, &gpio->vhost_dev, + vu_gpio_disconnect); break; case CHR_EVENT_BREAK: case CHR_EVENT_MUX_IN: diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c index abe23d4ebe..8f635844af 100644 --- a/hw/virtio/vhost-user.c +++ b/hw/virtio/vhost-user.c @@ -21,6 +21,7 @@ #include "qemu/error-report.h" #include "qemu/main-loop.h" #include "qemu/sockets.h" +#include "sysemu/runstate.h" #include "sysemu/cryptodev.h" #include "migration/migration.h" #include "migration/postcopy-ram.h" @@ -2670,6 +2671,76 @@ void vhost_user_cleanup(VhostUserState *user) user->chr = NULL; } + +typedef struct { + vu_async_close_fn cb; + DeviceState *dev; + CharBackend *cd; + struct vhost_dev *vhost; +} VhostAsyncCallback; + +static void vhost_user_async_close_bh(void *opaque) +{ + VhostAsyncCallback *data = opaque; + struct vhost_dev *vhost = data->vhost; + + /* + * If the vhost_dev has been cleared in the meantime there is + * nothing left to do as some other path has completed the + * cleanup. + */ + if (vhost->vdev) { + data->cb(data->dev); + } + + g_free(data); +} + +/* + * We only schedule the work if the machine is running. If suspended + * we want to keep all the in-flight data as is for migration + * purposes. + */ +void vhost_user_async_close(DeviceState *d, + CharBackend *chardev, struct vhost_dev *vhost, + vu_async_close_fn cb) +{ + if (!runstate_check(RUN_STATE_SHUTDOWN)) { + /* + * A close event may happen during a read/write, but vhost + * code assumes the vhost_dev remains setup, so delay the + * stop & clear. + */ + AioContext *ctx = qemu_get_current_aio_context(); + VhostAsyncCallback *data = g_new0(VhostAsyncCallback, 1); + + /* Save data for the callback */ + data->cb = cb; + data->dev = d; + data->cd = chardev; + data->vhost = vhost; + + /* Disable any further notifications on the chardev */ + qemu_chr_fe_set_handlers(chardev, + NULL, NULL, NULL, NULL, NULL, NULL, + false); + + aio_bh_schedule_oneshot(ctx, vhost_user_async_close_bh, data); + + /* + * Move vhost device to the stopped state. The vhost-user device + * will be clean up and disconnected in BH. This can be useful in + * the vhost migration code. If disconnect was caught there is an + * option for the general vhost code to get the dev state without + * knowing its type (in this case vhost-user). + * + * Note if the vhost device is fully cleared by the time we + * execute the bottom half we won't continue with the cleanup. + */ + vhost->started = false; + } +} + static int vhost_user_dev_start(struct vhost_dev *dev, bool started) { if (!virtio_has_feature(dev->protocol_features, diff --git a/include/hw/virtio/vhost-user.h b/include/hw/virtio/vhost-user.h index c6e693cd3f..191216a74f 100644 --- a/include/hw/virtio/vhost-user.h +++ b/include/hw/virtio/vhost-user.h @@ -68,4 +68,22 @@ bool vhost_user_init(VhostUserState *user, CharBackend *chr, Error **errp); */ void vhost_user_cleanup(VhostUserState *user); +/** + * vhost_user_async_close() - cleanup vhost-user post connection drop + * @d: DeviceState for the associated device (passed to callback) + * @chardev: the CharBackend associated with the connection + * @vhost: the common vhost device + * @cb: the user callback function to complete the clean-up + * + * This function is used to handle the shutdown of a vhost-user + * connection to a backend. We handle this centrally to make sure we + * do all the steps and handle potential races due to VM shutdowns. + * Once the connection is disabled we call a backhalf to ensure + */ +typedef void (*vu_async_close_fn)(DeviceState *cb); + +void vhost_user_async_close(DeviceState *d, + CharBackend *chardev, struct vhost_dev *vhost, + vu_async_close_fn cb); + #endif -- cgit v1.2.3-55-g7522