The following patches were built over linux-next which contains various vhost patches in mst's tree and the vhost_task patchset in Christian Brauner's tree: git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git kernel.user_worker branch: https://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git/log/?h=kernel.user_worker The latter patchset handles the review comment for the patches in thread to make sure that worker threads we create are accounted for in the parent process's NPROC limit. The patches are scheduled to be sent to Linus for 6.4. The patches in this patchset allow us to support multiple vhost workers per device. The design is a modified version of Stefan's original idea where userspace has the kernel create a worker and we pass back the pid. In this version instead of passing the pid between user/kernel space we use a worker_id which is just an integer managed by the vhost driver and we allow userspace to create and free workers and then attach them to virtqueues at setup time. All review comments from the past reviews should be handled. If I didn't reply to a review comment, I agreed with the comment and should have handled it in this posting. Let me know if I missed one. Results: -------- fio jobs 1 2 4 8 12 16 ---------------------------------------------------------- 1 worker 160k 488k - - - - worker per vq 160k 310k 620k 1300k 1836k 2326k Notes: 0. This used a simple fio command: fio --filename=/dev/sdb --direct=1 --rw=randrw --bs=4k \ --ioengine=libaio --iodepth=128 --numjobs=$JOBS_ABOVE and I used a VM with 16 vCPUs and 16 virtqueues. 1. The patches were tested with LIO's emulate_pr=0 which drops the LIO PR lock use. This was a bottleneck at around 12 vqs/jobs. 2. Because we have a hard limit of 1024 cmds, if the num jobs * iodepth was greater than 1024, I would decrease iodepth. So 12 jobs used 85 cmds, and 16 used 64. 3. The perf issue above at 2 jobs is because when we only have 1 worker we execute more cmds per vhost_work due to all vqs funneling to one worker. This results in less context switches and better batching without having to tweak any settings. I'm working on patches to add back batching during lio completion and do polling on the submission side. We will still want the threading patches, because if we batch at the fio level plus use the vhost theading patches, we can see a big boost like below. So hopefully doing it at the kernel will allow apps to just work without having to be smart like fio. fio using io_uring and batching with the iodepth_batch* settings: fio jobs 1 2 4 8 12 16 ------------------------------------------------------------- 1 worker 494k 520k - - - - worker per vq 496k 878k 1542k 2436k 2304k 2590k V6: - Rebase against vhost_task patchset. - Used xa instead of idr. V5: - Rebase against user_worker patchset. - Rebase against flush patchset. - Redo vhost-scsi tmf flush handling so it doesn't access vq->worker. V4: - fix vhost-sock VSOCK_VQ_RX use. - name functions called directly by ioctl cmd's to match the ioctl cmd. - break up VHOST_SET_VRING_WORKER into a new, free and attach cmd. - document worker lifetime, and cgroup, namespace, mm, rlimit inheritance, make it clear we currently only support sharing within the device. - add support to attach workers while IO is running. - instead of passing a pid_t of the kernel thread, pass a int allocated by the vhost layer with an idr. V3: - fully convert vhost code to use vq based APIs instead of leaving it half per dev and half per vq. - rebase against kernel worker API. - Drop delayed worker creation. We always create the default worker at VHOST_SET_OWNER time. Userspace can create and bind workers after that. V2: - change loop that we take a refcount to the worker in - replaced pid == -1 with define. - fixed tabbing/spacing coding style issue - use hash instead of list to lookup workers. - I dropped the patch that added an ioctl cmd to get a vq's worker's pid. I saw we might do a generic netlink interface instead.
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 01/11] vhost: add vhost_worker pointer to vhost_virtqueue
This patchset allows userspace to map vqs to different workers. This
patch adds a worker pointer to the vq so we can store that info.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/vhost.c | 24 +++++++++++++-----------
drivers/vhost/vhost.h | 1 +
2 files changed, 14 insertions(+), 11 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 4368ee9b999c..e041e116afee 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -486,6 +486,7 @@ void vhost_dev_init(struct vhost_dev *dev,
vq->log = NULL;
vq->indirect = NULL;
vq->heads = NULL;
+ vq->worker = NULL;
vq->dev = dev;
mutex_init(&vq->mutex);
vhost_vq_reset(dev, vq);
@@ -554,16 +555,15 @@ static void vhost_worker_free(struct vhost_dev *dev)
kfree(worker);
}
-static int vhost_worker_create(struct vhost_dev *dev)
+static struct vhost_worker *vhost_worker_create(struct vhost_dev *dev)
{
struct vhost_worker *worker;
struct vhost_task *vtsk;
char name[TASK_COMM_LEN];
- int ret;
worker = kzalloc(sizeof(*worker), GFP_KERNEL_ACCOUNT);
if (!worker)
- return -ENOMEM;
+ return NULL;
dev->worker = worker;
worker->kcov_handle = kcov_common_handle();
@@ -571,25 +571,24 @@ static int vhost_worker_create(struct vhost_dev *dev)
snprintf(name, sizeof(name), "vhost-%d", current->pid);
vtsk = vhost_task_create(vhost_worker, worker, name);
- if (!vtsk) {
- ret = -ENOMEM;
+ if (!vtsk)
goto free_worker;
- }
worker->vtsk = vtsk;
vhost_task_start(vtsk);
- return 0;
+ return worker;
free_worker:
kfree(worker);
dev->worker = NULL;
- return ret;
+ return NULL;
}
/* Caller should have device mutex */
long vhost_dev_set_owner(struct vhost_dev *dev)
{
- int err;
+ struct vhost_worker *worker;
+ int err, i;
/* Is there an owner already? */
if (vhost_dev_has_owner(dev)) {
@@ -600,9 +599,12 @@ long vhost_dev_set_owner(struct vhost_dev *dev)
vhost_attach_mm(dev);
if (dev->use_worker) {
- err = vhost_worker_create(dev);
- if (err)
+ worker = vhost_worker_create(dev);
+ if (!worker)
goto err_worker;
+
+ for (i = 0; i < dev->nvqs; i++)
+ dev->vqs[i]->worker = worker;
}
err = vhost_dev_alloc_iovecs(dev);
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 0308638cdeee..e72b665ba3a5 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -74,6 +74,7 @@ struct vhost_vring_call {
/* The virtqueue structure describes a queue attached to a device. */
struct vhost_virtqueue {
struct vhost_dev *dev;
+ struct vhost_worker *worker;
/* The actual ring of buffers. */
struct mutex mutex;
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 02/11] vhost, vhost-net: add helper to check if vq has work
In the next patches each vq might have different workers so one could
have work but others do not. For net, we only want to check specific vqs,
so this adds a helper to check if a vq has work pending and converts
vhost-net to use it.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/net.c | 2 +-
drivers/vhost/vhost.c | 6 +++---
drivers/vhost/vhost.h | 2 +-
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 07181cd8d52e..8ed63651b9eb 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -546,7 +546,7 @@ static void vhost_net_busy_poll(struct vhost_net *net,
endtime = busy_clock() + busyloop_timeout;
while (vhost_can_busy_poll(endtime)) {
- if (vhost_has_work(&net->dev)) {
+ if (vhost_vq_has_work(vq)) {
*busyloop_intr = true;
break;
}
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index e041e116afee..6567aed69ebb 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -262,11 +262,11 @@ void vhost_work_queue(struct vhost_dev *dev, struct
vhost_work *work)
EXPORT_SYMBOL_GPL(vhost_work_queue);
/* A lockless hint for busy polling code to exit the loop */
-bool vhost_has_work(struct vhost_dev *dev)
+bool vhost_vq_has_work(struct vhost_virtqueue *vq)
{
- return dev->worker &&
!llist_empty(&dev->worker->work_list);
+ return vq->worker &&
!llist_empty(&vq->worker->work_list);
}
-EXPORT_SYMBOL_GPL(vhost_has_work);
+EXPORT_SYMBOL_GPL(vhost_vq_has_work);
void vhost_poll_queue(struct vhost_poll *poll)
{
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index e72b665ba3a5..0dde119fb0ee 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -45,7 +45,6 @@ struct vhost_poll {
void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn);
void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work);
-bool vhost_has_work(struct vhost_dev *dev);
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
__poll_t mask, struct vhost_dev *dev);
@@ -195,6 +194,7 @@ int vhost_get_vq_desc(struct vhost_virtqueue *,
struct vhost_log *log, unsigned int *log_num);
void vhost_discard_vq_desc(struct vhost_virtqueue *, int n);
+bool vhost_vq_has_work(struct vhost_virtqueue *vq);
bool vhost_vq_is_setup(struct vhost_virtqueue *vq);
int vhost_vq_init_access(struct vhost_virtqueue *);
int vhost_add_used(struct vhost_virtqueue *, unsigned int head, int len);
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 03/11] vhost: take worker or vq instead of dev for queueing
This patch has the core work queueing function take a worker for when we
support multiple workers. It also adds a helper that takes a vq during
queueing so modules can control which vq/worker to queue work on.
This temp leaves vhost_work_queue. It will be removed when the drivers
are converted in the next patches.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/vhost.c | 44 +++++++++++++++++++++++++++----------------
drivers/vhost/vhost.h | 1 +
2 files changed, 29 insertions(+), 16 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 6567aed69ebb..cc2628ba9a77 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -231,6 +231,34 @@ void vhost_poll_stop(struct vhost_poll *poll)
}
EXPORT_SYMBOL_GPL(vhost_poll_stop);
+static void vhost_work_queue_on(struct vhost_worker *worker,
+ struct vhost_work *work)
+{
+ if (!worker)
+ return;
+
+ if (!test_and_set_bit(VHOST_WORK_QUEUED, &work->flags)) {
+ /* We can only add the work to the list after we're
+ * sure it was not in the list.
+ * test_and_set_bit() implies a memory barrier.
+ */
+ llist_add(&work->node, &worker->work_list);
+ wake_up_process(worker->vtsk->task);
+ }
+}
+
+void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
+{
+ vhost_work_queue_on(dev->worker, work);
+}
+EXPORT_SYMBOL_GPL(vhost_work_queue);
+
+void vhost_vq_work_queue(struct vhost_virtqueue *vq, struct vhost_work *work)
+{
+ vhost_work_queue_on(vq->worker, work);
+}
+EXPORT_SYMBOL_GPL(vhost_vq_work_queue);
+
void vhost_dev_flush(struct vhost_dev *dev)
{
struct vhost_flush_struct flush;
@@ -245,22 +273,6 @@ void vhost_dev_flush(struct vhost_dev *dev)
}
EXPORT_SYMBOL_GPL(vhost_dev_flush);
-void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
-{
- if (!dev->worker)
- return;
-
- if (!test_and_set_bit(VHOST_WORK_QUEUED, &work->flags)) {
- /* We can only add the work to the list after we're
- * sure it was not in the list.
- * test_and_set_bit() implies a memory barrier.
- */
- llist_add(&work->node, &dev->worker->work_list);
- wake_up_process(dev->worker->vtsk->task);
- }
-}
-EXPORT_SYMBOL_GPL(vhost_work_queue);
-
/* A lockless hint for busy polling code to exit the loop */
bool vhost_vq_has_work(struct vhost_virtqueue *vq)
{
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 0dde119fb0ee..b64ee4ef387d 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -194,6 +194,7 @@ int vhost_get_vq_desc(struct vhost_virtqueue *,
struct vhost_log *log, unsigned int *log_num);
void vhost_discard_vq_desc(struct vhost_virtqueue *, int n);
+void vhost_vq_work_queue(struct vhost_virtqueue *vq, struct vhost_work *work);
bool vhost_vq_has_work(struct vhost_virtqueue *vq);
bool vhost_vq_is_setup(struct vhost_virtqueue *vq);
int vhost_vq_init_access(struct vhost_virtqueue *);
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 04/11] vhost: take worker or vq instead of dev for flushing
This patch has the core work flush function take a worker. When we
support multiple workers we can then flush each worker during device
removal, stoppage, etc.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/vhost.c | 24 +++++++++++++++---------
1 file changed, 15 insertions(+), 9 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index cc2628ba9a77..6160aa1cc922 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -247,6 +247,20 @@ static void vhost_work_queue_on(struct vhost_worker
*worker,
}
}
+static void vhost_work_flush_on(struct vhost_worker *worker)
+{
+ struct vhost_flush_struct flush;
+
+ if (!worker)
+ return;
+
+ init_completion(&flush.wait_event);
+ vhost_work_init(&flush.work, vhost_flush_work);
+
+ vhost_work_queue_on(worker, &flush.work);
+ wait_for_completion(&flush.wait_event);
+}
+
void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
{
vhost_work_queue_on(dev->worker, work);
@@ -261,15 +275,7 @@ EXPORT_SYMBOL_GPL(vhost_vq_work_queue);
void vhost_dev_flush(struct vhost_dev *dev)
{
- struct vhost_flush_struct flush;
-
- if (dev->worker) {
- init_completion(&flush.wait_event);
- vhost_work_init(&flush.work, vhost_flush_work);
-
- vhost_work_queue(dev, &flush.work);
- wait_for_completion(&flush.wait_event);
- }
+ vhost_work_flush_on(dev->worker);
}
EXPORT_SYMBOL_GPL(vhost_dev_flush);
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 05/11] vhost: convert poll work to be vq based
This has the drivers pass in their poll to vq mapping and then converts
the core poll code to use the vq based helpers. In the next patches we
will allow vqs to be handled by different workers, so to allow drivers
to execute operations like queue, stop, flush, etc on specific polls/vqs
we need to know the mappings.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/net.c | 6 ++++--
drivers/vhost/vhost.c | 8 +++++---
drivers/vhost/vhost.h | 4 +++-
3 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 8ed63651b9eb..4a9b757071a2 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -1342,8 +1342,10 @@ static int vhost_net_open(struct inode *inode, struct
file *f)
VHOST_NET_PKT_WEIGHT, VHOST_NET_WEIGHT, true,
NULL);
- vhost_poll_init(n->poll + VHOST_NET_VQ_TX, handle_tx_net, EPOLLOUT, dev);
- vhost_poll_init(n->poll + VHOST_NET_VQ_RX, handle_rx_net, EPOLLIN, dev);
+ vhost_poll_init(n->poll + VHOST_NET_VQ_TX, handle_tx_net, EPOLLOUT, dev,
+ vqs[VHOST_NET_VQ_TX]);
+ vhost_poll_init(n->poll + VHOST_NET_VQ_RX, handle_rx_net, EPOLLIN, dev,
+ vqs[VHOST_NET_VQ_RX]);
f->private_data = n;
n->page_frag.page = NULL;
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 6160aa1cc922..6968f8fc17e8 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -187,13 +187,15 @@ EXPORT_SYMBOL_GPL(vhost_work_init);
/* Init poll structure */
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
- __poll_t mask, struct vhost_dev *dev)
+ __poll_t mask, struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
{
init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
init_poll_funcptr(&poll->table, vhost_poll_func);
poll->mask = mask;
poll->dev = dev;
poll->wqh = NULL;
+ poll->vq = vq;
vhost_work_init(&poll->work, fn);
}
@@ -288,7 +290,7 @@ EXPORT_SYMBOL_GPL(vhost_vq_has_work);
void vhost_poll_queue(struct vhost_poll *poll)
{
- vhost_work_queue(poll->dev, &poll->work);
+ vhost_vq_work_queue(poll->vq, &poll->work);
}
EXPORT_SYMBOL_GPL(vhost_poll_queue);
@@ -510,7 +512,7 @@ void vhost_dev_init(struct vhost_dev *dev,
vhost_vq_reset(dev, vq);
if (vq->handle_kick)
vhost_poll_init(&vq->poll, vq->handle_kick,
- EPOLLIN, dev);
+ EPOLLIN, dev, vq);
}
}
EXPORT_SYMBOL_GPL(vhost_dev_init);
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index b64ee4ef387d..d9b8abbe3a26 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -41,13 +41,15 @@ struct vhost_poll {
struct vhost_work work;
__poll_t mask;
struct vhost_dev *dev;
+ struct vhost_virtqueue *vq;
};
void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn);
void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work);
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
- __poll_t mask, struct vhost_dev *dev);
+ __poll_t mask, struct vhost_dev *dev,
+ struct vhost_virtqueue *vq);
int vhost_poll_start(struct vhost_poll *poll, struct file *file);
void vhost_poll_stop(struct vhost_poll *poll);
void vhost_poll_queue(struct vhost_poll *poll);
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 06/11] vhost-sock: convert to vhost_vq_work_queue
Convert from vhost_work_queue to vhost_vq_work_queue. Signed-off-by: Mike Christie <michael.christie at oracle.com> --- drivers/vhost/vsock.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c index c8e6087769a1..1dcbc8669f95 100644 --- a/drivers/vhost/vsock.c +++ b/drivers/vhost/vsock.c @@ -285,7 +285,7 @@ vhost_transport_send_pkt(struct sk_buff *skb) atomic_inc(&vsock->queued_replies); virtio_vsock_skb_queue_tail(&vsock->send_pkt_queue, skb); - vhost_work_queue(&vsock->dev, &vsock->send_pkt_work); + vhost_vq_work_queue(&vsock->vqs[VSOCK_VQ_RX], &vsock->send_pkt_work); rcu_read_unlock(); return len; @@ -582,7 +582,7 @@ static int vhost_vsock_start(struct vhost_vsock *vsock) /* Some packets may have been queued before the device was started, * let's kick the send worker to send them. */ - vhost_work_queue(&vsock->dev, &vsock->send_pkt_work); + vhost_vq_work_queue(&vsock->vqs[VSOCK_VQ_RX], &vsock->send_pkt_work); mutex_unlock(&vsock->dev.mutex); return 0; -- 2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 07/11] vhost-scsi: make SCSI cmd completion per vq
This patch separates the scsi cmd completion code paths so we can complete
cmds based on their vq instead of having all cmds complete on the same
worker/CPU. This will be useful with the next patches that allow us to
create mulitple worker threads and bind them to different vqs, so we can
have completions running on different threads/CPUs.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
Reviewed-by: Stefan Hajnoczi <stefanha at redhat.com>
---
drivers/vhost/scsi.c | 56 ++++++++++++++++++++------------------------
1 file changed, 26 insertions(+), 30 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 3b0b556c57ef..ecb5cd7450b8 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -167,6 +167,7 @@ MODULE_PARM_DESC(max_io_vqs, "Set the max number of IO
virtqueues a vhost scsi d
struct vhost_scsi_virtqueue {
struct vhost_virtqueue vq;
+ struct vhost_scsi *vs;
/*
* Reference counting for inflight reqs, used for flush operation. At
* each time, one reference tracks new commands submitted, while we
@@ -181,6 +182,9 @@ struct vhost_scsi_virtqueue {
struct vhost_scsi_cmd *scsi_cmds;
struct sbitmap scsi_tags;
int max_cmds;
+
+ struct vhost_work completion_work;
+ struct llist_head completion_list;
};
struct vhost_scsi {
@@ -190,12 +194,8 @@ struct vhost_scsi {
struct vhost_dev dev;
struct vhost_scsi_virtqueue *vqs;
- unsigned long *compl_bitmap;
struct vhost_scsi_inflight **old_inflight;
- struct vhost_work vs_completion_work; /* cmd completion work item */
- struct llist_head vs_completion_list; /* cmd completion queue */
-
struct vhost_work vs_event_work; /* evt injection work item */
struct llist_head vs_event_list; /* evt injection queue */
@@ -368,10 +368,11 @@ static void vhost_scsi_release_cmd(struct se_cmd *se_cmd)
} else {
struct vhost_scsi_cmd *cmd = container_of(se_cmd,
struct vhost_scsi_cmd, tvc_se_cmd);
- struct vhost_scsi *vs = cmd->tvc_vhost;
+ struct vhost_scsi_virtqueue *svq = container_of(cmd->tvc_vq,
+ struct vhost_scsi_virtqueue, vq);
- llist_add(&cmd->tvc_completion_list, &vs->vs_completion_list);
- vhost_work_queue(&vs->dev, &vs->vs_completion_work);
+ llist_add(&cmd->tvc_completion_list, &svq->completion_list);
+ vhost_vq_work_queue(&svq->vq, &svq->completion_work);
}
}
@@ -534,17 +535,17 @@ static void vhost_scsi_evt_work(struct vhost_work *work)
*/
static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
{
- struct vhost_scsi *vs = container_of(work, struct vhost_scsi,
- vs_completion_work);
+ struct vhost_scsi_virtqueue *svq = container_of(work,
+ struct vhost_scsi_virtqueue, completion_work);
struct virtio_scsi_cmd_resp v_rsp;
struct vhost_scsi_cmd *cmd, *t;
struct llist_node *llnode;
struct se_cmd *se_cmd;
struct iov_iter iov_iter;
- int ret, vq;
+ bool signal = false;
+ int ret;
- bitmap_zero(vs->compl_bitmap, vs->dev.nvqs);
- llnode = llist_del_all(&vs->vs_completion_list);
+ llnode = llist_del_all(&svq->completion_list);
llist_for_each_entry_safe(cmd, t, llnode, tvc_completion_list) {
se_cmd = &cmd->tvc_se_cmd;
@@ -564,21 +565,17 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work
*work)
cmd->tvc_in_iovs, sizeof(v_rsp));
ret = copy_to_iter(&v_rsp, sizeof(v_rsp), &iov_iter);
if (likely(ret == sizeof(v_rsp))) {
- struct vhost_scsi_virtqueue *q;
+ signal = true;
+
vhost_add_used(cmd->tvc_vq, cmd->tvc_vq_desc, 0);
- q = container_of(cmd->tvc_vq, struct vhost_scsi_virtqueue, vq);
- vq = q - vs->vqs;
- __set_bit(vq, vs->compl_bitmap);
} else
pr_err("Faulted on virtio_scsi_cmd_resp\n");
vhost_scsi_release_cmd_res(se_cmd);
}
- vq = -1;
- while ((vq = find_next_bit(vs->compl_bitmap, vs->dev.nvqs, vq + 1))
- < vs->dev.nvqs)
- vhost_signal(&vs->dev, &vs->vqs[vq].vq);
+ if (signal)
+ vhost_signal(&svq->vs->dev, &svq->vq);
}
static struct vhost_scsi_cmd *
@@ -1795,6 +1792,7 @@ static int vhost_scsi_set_features(struct vhost_scsi *vs,
u64 features)
static int vhost_scsi_open(struct inode *inode, struct file *f)
{
+ struct vhost_scsi_virtqueue *svq;
struct vhost_scsi *vs;
struct vhost_virtqueue **vqs;
int r = -ENOMEM, i, nvqs = vhost_scsi_max_io_vqs;
@@ -1813,10 +1811,6 @@ static int vhost_scsi_open(struct inode *inode, struct
file *f)
}
nvqs += VHOST_SCSI_VQ_IO;
- vs->compl_bitmap = bitmap_alloc(nvqs, GFP_KERNEL);
- if (!vs->compl_bitmap)
- goto err_compl_bitmap;
-
vs->old_inflight = kmalloc_array(nvqs, sizeof(*vs->old_inflight),
GFP_KERNEL | __GFP_ZERO);
if (!vs->old_inflight)
@@ -1831,7 +1825,6 @@ static int vhost_scsi_open(struct inode *inode, struct
file *f)
if (!vqs)
goto err_local_vqs;
- vhost_work_init(&vs->vs_completion_work, vhost_scsi_complete_cmd_work);
vhost_work_init(&vs->vs_event_work, vhost_scsi_evt_work);
vs->vs_events_nr = 0;
@@ -1842,8 +1835,14 @@ static int vhost_scsi_open(struct inode *inode, struct
file *f)
vs->vqs[VHOST_SCSI_VQ_CTL].vq.handle_kick = vhost_scsi_ctl_handle_kick;
vs->vqs[VHOST_SCSI_VQ_EVT].vq.handle_kick = vhost_scsi_evt_handle_kick;
for (i = VHOST_SCSI_VQ_IO; i < nvqs; i++) {
- vqs[i] = &vs->vqs[i].vq;
- vs->vqs[i].vq.handle_kick = vhost_scsi_handle_kick;
+ svq = &vs->vqs[i];
+
+ vqs[i] = &svq->vq;
+ svq->vs = vs;
+ init_llist_head(&svq->completion_list);
+ vhost_work_init(&svq->completion_work,
+ vhost_scsi_complete_cmd_work);
+ svq->vq.handle_kick = vhost_scsi_handle_kick;
}
vhost_dev_init(&vs->dev, vqs, nvqs, UIO_MAXIOV,
VHOST_SCSI_WEIGHT, 0, true, NULL);
@@ -1858,8 +1857,6 @@ static int vhost_scsi_open(struct inode *inode, struct
file *f)
err_vqs:
kfree(vs->old_inflight);
err_inflight:
- bitmap_free(vs->compl_bitmap);
-err_compl_bitmap:
kvfree(vs);
err_vs:
return r;
@@ -1879,7 +1876,6 @@ static int vhost_scsi_release(struct inode *inode, struct
file *f)
kfree(vs->dev.vqs);
kfree(vs->vqs);
kfree(vs->old_inflight);
- bitmap_free(vs->compl_bitmap);
kvfree(vs);
return 0;
}
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 08/11] vhost-scsi: convert to vhost_vq_work_queue
Convert from vhost_work_queue to vhost_vq_work_queue.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/scsi.c | 18 +++++++++---------
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index ecb5cd7450b8..3e86b5fbeca6 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -363,8 +363,9 @@ static void vhost_scsi_release_cmd(struct se_cmd *se_cmd)
if (se_cmd->se_cmd_flags & SCF_SCSI_TMR_CDB) {
struct vhost_scsi_tmf *tmf = container_of(se_cmd,
struct vhost_scsi_tmf, se_cmd);
+ struct vhost_virtqueue *vq = &tmf->svq->vq;
- vhost_work_queue(&tmf->vhost->dev, &tmf->vwork);
+ vhost_vq_work_queue(vq, &tmf->vwork);
} else {
struct vhost_scsi_cmd *cmd = container_of(se_cmd,
struct vhost_scsi_cmd, tvc_se_cmd);
@@ -1357,11 +1358,9 @@ static void vhost_scsi_ctl_handle_kick(struct vhost_work
*work)
}
static void
-vhost_scsi_send_evt(struct vhost_scsi *vs,
- struct vhost_scsi_tpg *tpg,
- struct se_lun *lun,
- u32 event,
- u32 reason)
+vhost_scsi_send_evt(struct vhost_scsi *vs, struct vhost_virtqueue *vq,
+ struct vhost_scsi_tpg *tpg, struct se_lun *lun,
+ u32 event, u32 reason)
{
struct vhost_scsi_evt *evt;
@@ -1383,7 +1382,7 @@ vhost_scsi_send_evt(struct vhost_scsi *vs,
}
llist_add(&evt->list, &vs->vs_event_list);
- vhost_work_queue(&vs->dev, &vs->vs_event_work);
+ vhost_vq_work_queue(vq, &vs->vs_event_work);
}
static void vhost_scsi_evt_handle_kick(struct vhost_work *work)
@@ -1397,7 +1396,8 @@ static void vhost_scsi_evt_handle_kick(struct vhost_work
*work)
goto out;
if (vs->vs_events_missed)
- vhost_scsi_send_evt(vs, NULL, NULL, VIRTIO_SCSI_T_NO_EVENT, 0);
+ vhost_scsi_send_evt(vs, vq, NULL, NULL, VIRTIO_SCSI_T_NO_EVENT,
+ 0);
out:
mutex_unlock(&vq->mutex);
}
@@ -2016,7 +2016,7 @@ vhost_scsi_do_plug(struct vhost_scsi_tpg *tpg,
goto unlock;
if (vhost_has_feature(vq, VIRTIO_SCSI_F_HOTPLUG))
- vhost_scsi_send_evt(vs, tpg, lun,
+ vhost_scsi_send_evt(vs, vq, tpg, lun,
VIRTIO_SCSI_T_TRANSPORT_RESET, reason);
unlock:
mutex_unlock(&vq->mutex);
--
2.25.1
vhost_work_queue is no longer used. Each driver is using the poll or vq
based queueing, so remove vhost_work_queue.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/vhost.c | 6 ------
drivers/vhost/vhost.h | 1 -
2 files changed, 7 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 6968f8fc17e8..f812daf25648 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -263,12 +263,6 @@ static void vhost_work_flush_on(struct vhost_worker
*worker)
wait_for_completion(&flush.wait_event);
}
-void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
-{
- vhost_work_queue_on(dev->worker, work);
-}
-EXPORT_SYMBOL_GPL(vhost_work_queue);
-
void vhost_vq_work_queue(struct vhost_virtqueue *vq, struct vhost_work *work)
{
vhost_work_queue_on(vq->worker, work);
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index d9b8abbe3a26..ef55fae2517c 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -45,7 +45,6 @@ struct vhost_poll {
};
void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn);
-void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work);
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
__poll_t mask, struct vhost_dev *dev,
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 10/11] vhost-scsi: flush IO vqs then send TMF rsp
With one worker we will always send the scsi cmd responses then send the
TMF rsp, because LIO will always complete the scsi cmds first then call
into us to send the TMF response.
With multiple workers, the IO vq workers could be running while the
TMF/ctl vq worker is so this has us do a flush before completing the TMF
to make sure cmds are completed when it's work is later queued and run.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/scsi.c | 22 +++++++++++++++++++---
drivers/vhost/vhost.c | 6 ++++++
drivers/vhost/vhost.h | 1 +
3 files changed, 26 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 3e86b5fbeca6..48dba4fe2dac 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -1158,12 +1158,28 @@ static void vhost_scsi_tmf_resp_work(struct vhost_work
*work)
{
struct vhost_scsi_tmf *tmf = container_of(work, struct vhost_scsi_tmf,
vwork);
- int resp_code;
+ struct vhost_virtqueue *ctl_vq, *vq;
+ int resp_code, i;
+
+ if (tmf->scsi_resp == TMR_FUNCTION_COMPLETE) {
+ /*
+ * Flush IO vqs that don't share a worker with the ctl to make
+ * sure they have sent their responses before us.
+ */
+ ctl_vq = &tmf->vhost->vqs[VHOST_SCSI_VQ_CTL].vq;
+ for (i = VHOST_SCSI_VQ_IO; i < tmf->vhost->dev.nvqs; i++) {
+ vq = &tmf->vhost->vqs[i].vq;
+
+ if (vhost_vq_is_setup(vq) &&
+ vq->worker != ctl_vq->worker) {
+ vhost_vq_flush(vq);
+ }
+ }
- if (tmf->scsi_resp == TMR_FUNCTION_COMPLETE)
resp_code = VIRTIO_SCSI_S_FUNCTION_SUCCEEDED;
- else
+ } else {
resp_code = VIRTIO_SCSI_S_FUNCTION_REJECTED;
+ }
vhost_scsi_send_tmf_resp(tmf->vhost, &tmf->svq->vq,
tmf->in_iovs,
tmf->vq_desc, &tmf->resp_iov, resp_code);
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index f812daf25648..1fa5e9a49092 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -275,6 +275,12 @@ void vhost_dev_flush(struct vhost_dev *dev)
}
EXPORT_SYMBOL_GPL(vhost_dev_flush);
+void vhost_vq_flush(struct vhost_virtqueue *vq)
+{
+ vhost_work_flush_on(vq->worker);
+}
+EXPORT_SYMBOL_GPL(vhost_vq_flush);
+
/* A lockless hint for busy polling code to exit the loop */
bool vhost_vq_has_work(struct vhost_virtqueue *vq)
{
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index ef55fae2517c..395707c680e5 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -53,6 +53,7 @@ int vhost_poll_start(struct vhost_poll *poll, struct file
*file);
void vhost_poll_stop(struct vhost_poll *poll);
void vhost_poll_queue(struct vhost_poll *poll);
void vhost_dev_flush(struct vhost_dev *dev);
+void vhost_vq_flush(struct vhost_virtqueue *vq);
struct vhost_log {
u64 addr;
--
2.25.1
Mike Christie
2023-Mar-28 02:17 UTC
[PATCH v6 11/11] vhost: allow userspace to create workers
For vhost-scsi with 3 vqs and a workload like that tries to use those vqs
like:
fio --filename=/dev/sdb --direct=1 --rw=randrw --bs=4k \
--ioengine=libaio --iodepth=128 --numjobs=3
the single vhost worker thread will become a bottlneck and we are stuck
at around 500K IOPs no matter how many jobs, virtqueues, and CPUs are
used.
To better utilize virtqueues and available CPUs, this patch allows
userspace to create workers and bind them to vqs. You can have N workers
per dev and also share N workers with M vqs.
With the patches and doing a worker per vq, we can scale to at least
16 vCPUs/vqs (that's my system limit) with the same command fio command
above with numjobs=16:
fio --filename=/dev/sdb --direct=1 --rw=randrw --bs=4k \
--ioengine=libaio --iodepth=64 --numjobs=16
which gives around 2326K IOPs.
Note that for testing I dropped depth to 64 above because the vhost/virt
layer supports only 1024 total commands per device. And the only tuning I
did was set LIO's emulate_pr to 0 to avoid LIO's PR lock in the main IO
path which becomes an issue at around 12 jobs/virtqueues.
Signed-off-by: Mike Christie <michael.christie at oracle.com>
---
drivers/vhost/vhost.c | 177 ++++++++++++++++++++++++++++---
drivers/vhost/vhost.h | 4 +-
include/uapi/linux/vhost.h | 22 ++++
include/uapi/linux/vhost_types.h | 15 +++
4 files changed, 204 insertions(+), 14 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 1fa5e9a49092..e40699e83c6d 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -271,7 +271,11 @@ EXPORT_SYMBOL_GPL(vhost_vq_work_queue);
void vhost_dev_flush(struct vhost_dev *dev)
{
- vhost_work_flush_on(dev->worker);
+ struct vhost_worker *worker;
+ unsigned long i;
+
+ xa_for_each(&dev->worker_xa, i, worker)
+ vhost_work_flush_on(worker);
}
EXPORT_SYMBOL_GPL(vhost_dev_flush);
@@ -489,7 +493,6 @@ void vhost_dev_init(struct vhost_dev *dev,
dev->umem = NULL;
dev->iotlb = NULL;
dev->mm = NULL;
- dev->worker = NULL;
dev->iov_limit = iov_limit;
dev->weight = weight;
dev->byte_weight = byte_weight;
@@ -499,7 +502,7 @@ void vhost_dev_init(struct vhost_dev *dev,
INIT_LIST_HEAD(&dev->read_list);
INIT_LIST_HEAD(&dev->pending_list);
spin_lock_init(&dev->iotlb_lock);
-
+ xa_init_flags(&dev->worker_xa, XA_FLAGS_ALLOC);
for (i = 0; i < dev->nvqs; ++i) {
vq = dev->vqs[i];
@@ -562,32 +565,67 @@ static void vhost_detach_mm(struct vhost_dev *dev)
dev->mm = NULL;
}
-static void vhost_worker_free(struct vhost_dev *dev)
+static void vhost_worker_put(struct vhost_dev *dev, struct vhost_worker
*worker)
{
- struct vhost_worker *worker = dev->worker;
-
if (!worker)
return;
- dev->worker = NULL;
+ if (!refcount_dec_and_test(&worker->refcount))
+ return;
+
WARN_ON(!llist_empty(&worker->work_list));
vhost_task_stop(worker->vtsk);
kfree(worker);
}
+static void vhost_vq_detach_worker(struct vhost_virtqueue *vq)
+{
+ if (vq->worker)
+ vhost_worker_put(vq->dev, vq->worker);
+ vq->worker = NULL;
+}
+
+static void vhost_workers_free(struct vhost_dev *dev)
+{
+ struct vhost_worker *worker;
+ unsigned long i;
+
+ if (!dev->use_worker)
+ return;
+
+ for (i = 0; i < dev->nvqs; i++)
+ vhost_vq_detach_worker(dev->vqs[i]);
+ /*
+ * Drop the refcount taken during allocation, and handle the default
+ * worker and the cases where userspace might have crashed or was lazy
+ * and did a VHOST_NEW_WORKER but not a VHOST_FREE_WORKER.
+ */
+ xa_for_each(&dev->worker_xa, i, worker) {
+ xa_erase(&dev->worker_xa, worker->id);
+ vhost_worker_put(dev, worker);
+ }
+ xa_destroy(&dev->worker_xa);
+}
+
static struct vhost_worker *vhost_worker_create(struct vhost_dev *dev)
{
struct vhost_worker *worker;
struct vhost_task *vtsk;
char name[TASK_COMM_LEN];
+ int ret;
+ u32 id;
worker = kzalloc(sizeof(*worker), GFP_KERNEL_ACCOUNT);
if (!worker)
return NULL;
- dev->worker = worker;
worker->kcov_handle = kcov_common_handle();
init_llist_head(&worker->work_list);
+ /*
+ * We increase the refcount for the initial creation and then
+ * later each time it's attached to a virtqueue.
+ */
+ refcount_set(&worker->refcount, 1);
snprintf(name, sizeof(name), "vhost-%d", current->pid);
vtsk = vhost_task_create(vhost_worker, worker, name);
@@ -596,14 +634,104 @@ static struct vhost_worker *vhost_worker_create(struct
vhost_dev *dev)
worker->vtsk = vtsk;
vhost_task_start(vtsk);
+
+ ret = xa_alloc(&dev->worker_xa, &id, worker, xa_limit_32b,
GFP_KERNEL);
+ if (ret < 0)
+ goto stop_worker;
+ worker->id = id;
+
return worker;
+stop_worker:
+ vhost_task_stop(vtsk);
free_worker:
kfree(worker);
- dev->worker = NULL;
return NULL;
}
+/* Caller must have device and virtqueue mutex */
+static void __vhost_vq_attach_worker(struct vhost_virtqueue *vq,
+ struct vhost_worker *worker)
+{
+ refcount_inc(&worker->refcount);
+ vhost_vq_detach_worker(vq);
+ vq->worker = worker;
+}
+
+/* Caller must have device and virtqueue mutex */
+static int vhost_vq_attach_worker(struct vhost_virtqueue *vq,
+ struct vhost_vring_worker *info)
+{
+ unsigned long index = info->worker_id;
+ struct vhost_dev *dev = vq->dev;
+ struct vhost_worker *worker;
+
+ if (!dev->use_worker)
+ return -EINVAL;
+
+ /*
+ * We don't support setting a worker on an active vq to make flushing
+ * and removal simple.
+ */
+ if (vhost_vq_get_backend(vq))
+ return -EBUSY;
+
+ worker = xa_find(&dev->worker_xa, &index, UINT_MAX, XA_PRESENT);
+ if (!worker || worker->id != info->worker_id)
+ return -ENODEV;
+
+ __vhost_vq_attach_worker(vq, worker);
+ return 0;
+}
+
+/* Caller must have device mutex */
+static int vhost_new_worker(struct vhost_dev *dev,
+ struct vhost_worker_state *info)
+{
+ struct vhost_worker *worker;
+
+ if (!dev->use_worker)
+ return -EINVAL;
+
+ worker = vhost_worker_create(dev);
+ if (!worker)
+ return -ENOMEM;
+
+ info->worker_id = worker->id;
+ return 0;
+}
+
+/* Caller must have device mutex */
+static int vhost_free_worker(struct vhost_dev *dev,
+ struct vhost_worker_state *info)
+{
+ unsigned long index = info->worker_id;
+ struct vhost_worker *worker;
+
+ if (!dev->use_worker)
+ return -EINVAL;
+
+ worker = xa_find(&dev->worker_xa, &index, UINT_MAX, XA_PRESENT);
+ if (!worker || worker->id != info->worker_id)
+ return -ENODEV;
+
+ /*
+ * We can free the worker if it's not attached to any virtqueues.
+ */
+ if (refcount_read(&worker->refcount) != 1)
+ return -EBUSY;
+
+ xa_erase(&dev->worker_xa, worker->id);
+ /*
+ * Make sure if there was a flush that saw the worker in the XA that
+ * it has completed.
+ */
+ vhost_work_flush_on(worker);
+
+ vhost_worker_put(dev, worker);
+ return 0;
+}
+
/* Caller should have device mutex */
long vhost_dev_set_owner(struct vhost_dev *dev)
{
@@ -624,7 +752,7 @@ long vhost_dev_set_owner(struct vhost_dev *dev)
goto err_worker;
for (i = 0; i < dev->nvqs; i++)
- dev->vqs[i]->worker = worker;
+ __vhost_vq_attach_worker(dev->vqs[i], worker);
}
err = vhost_dev_alloc_iovecs(dev);
@@ -633,7 +761,7 @@ long vhost_dev_set_owner(struct vhost_dev *dev)
return 0;
err_iovecs:
- vhost_worker_free(dev);
+ vhost_workers_free(dev);
err_worker:
vhost_detach_mm(dev);
err_mm:
@@ -726,7 +854,7 @@ void vhost_dev_cleanup(struct vhost_dev *dev)
dev->iotlb = NULL;
vhost_clear_msg(dev);
wake_up_interruptible_poll(&dev->wait, EPOLLIN | EPOLLRDNORM);
- vhost_worker_free(dev);
+ vhost_workers_free(dev);
vhost_detach_mm(dev);
}
EXPORT_SYMBOL_GPL(vhost_dev_cleanup);
@@ -1616,6 +1744,7 @@ long vhost_vring_ioctl(struct vhost_dev *d, unsigned int
ioctl, void __user *arg
struct eventfd_ctx *ctx = NULL;
u32 __user *idxp = argp;
struct vhost_virtqueue *vq;
+ struct vhost_vring_worker w;
struct vhost_vring_state s;
struct vhost_vring_file f;
u32 idx;
@@ -1723,7 +1852,16 @@ long vhost_vring_ioctl(struct vhost_dev *d, unsigned int
ioctl, void __user *arg
if (copy_to_user(argp, &s, sizeof(s)))
r = -EFAULT;
break;
- default:
+ case VHOST_ATTACH_VRING_WORKER:
+ if (copy_from_user(&w, argp, sizeof(w))) {
+ r = -EFAULT;
+ break;
+ }
+ r = vhost_vq_attach_worker(vq, &w);
+ if (!r && copy_to_user(argp, &w, sizeof(w)))
+ r = -EFAULT;
+ break;
+default:
r = -ENOIOCTLCMD;
}
@@ -1776,6 +1914,7 @@ EXPORT_SYMBOL_GPL(vhost_init_device_iotlb);
/* Caller must have device mutex */
long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user
*argp)
{
+ struct vhost_worker_state w;
struct eventfd_ctx *ctx;
u64 p;
long r;
@@ -1836,6 +1975,18 @@ long vhost_dev_ioctl(struct vhost_dev *d, unsigned int
ioctl, void __user *argp)
if (ctx)
eventfd_ctx_put(ctx);
break;
+ case VHOST_NEW_WORKER:
+ r = vhost_new_worker(d, &w);
+ if (!r && copy_to_user(argp, &w, sizeof(w)))
+ r = -EFAULT;
+ break;
+ case VHOST_FREE_WORKER:
+ if (copy_from_user(&w, argp, sizeof(w))) {
+ r = -EFAULT;
+ break;
+ }
+ r = vhost_free_worker(d, &w);
+ break;
default:
r = -ENOIOCTLCMD;
break;
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 395707c680e5..a67ae8293c38 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -30,6 +30,8 @@ struct vhost_worker {
struct vhost_task *vtsk;
struct llist_head work_list;
u64 kcov_handle;
+ refcount_t refcount;
+ u32 id;
};
/* Poll a file (eventfd or socket) */
@@ -156,7 +158,6 @@ struct vhost_dev {
struct vhost_virtqueue **vqs;
int nvqs;
struct eventfd_ctx *log_ctx;
- struct vhost_worker *worker;
struct vhost_iotlb *umem;
struct vhost_iotlb *iotlb;
spinlock_t iotlb_lock;
@@ -166,6 +167,7 @@ struct vhost_dev {
int iov_limit;
int weight;
int byte_weight;
+ struct xarray worker_xa;
bool use_worker;
int (*msg_handler)(struct vhost_dev *dev, u32 asid,
struct vhost_iotlb_msg *msg);
diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h
index 92e1b700b51c..7329e7f349dd 100644
--- a/include/uapi/linux/vhost.h
+++ b/include/uapi/linux/vhost.h
@@ -45,6 +45,23 @@
#define VHOST_SET_LOG_BASE _IOW(VHOST_VIRTIO, 0x04, __u64)
/* Specify an eventfd file descriptor to signal on log write. */
#define VHOST_SET_LOG_FD _IOW(VHOST_VIRTIO, 0x07, int)
+/* By default, a device gets one vhost_worker that its virtqueues share. This
+ * command allows the owner of the device to create an additional vhost_worker
+ * for the device. It can later be bound to 1 or more of its virtqueues using
+ * the VHOST_ATTACH_VRING_WORKER command.
+ *
+ * This must be called after VHOST_SET_OWNER and the caller must be the owner
+ * of the device. The new thread will inherit caller's cgroups and
namespaces,
+ * and will share the caller's memory space. The new thread will also be
+ * counted against the caller's RLIMIT_NPROC value.
+ */
+#define VHOST_NEW_WORKER _IOW(VHOST_VIRTIO, 0x8, struct vhost_worker_state)
+/* Free a worker created with VHOST_NEW_WORKER if it's not attached to any
+ * virtqueue. If userspace is not able to call this for workers its created,
+ * the kernel will free all the device's workers when the device is closed
and
+ * the last reference to the device has been released.
+ */
+#define VHOST_FREE_WORKER _IOR(VHOST_VIRTIO, 0x9, struct vhost_worker_state)
/* Ring setup. */
/* Set number of descriptors in ring. This parameter can not
@@ -70,6 +87,11 @@
#define VHOST_VRING_BIG_ENDIAN 1
#define VHOST_SET_VRING_ENDIAN _IOW(VHOST_VIRTIO, 0x13, struct
vhost_vring_state)
#define VHOST_GET_VRING_ENDIAN _IOW(VHOST_VIRTIO, 0x14, struct
vhost_vring_state)
+/* Attach a vhost_worker created with VHOST_NEW_WORKER to one of the
device's
+ * virtqueues. This must be done before the virtqueue is active.
+ */
+#define VHOST_ATTACH_VRING_WORKER _IOR(VHOST_VIRTIO, 0x15, \
+ struct vhost_vring_worker)
/* The following ioctls use eventfd file descriptors to signal and poll
* for events. */
diff --git a/include/uapi/linux/vhost_types.h b/include/uapi/linux/vhost_types.h
index c5690a8992d8..ad0fe2e721be 100644
--- a/include/uapi/linux/vhost_types.h
+++ b/include/uapi/linux/vhost_types.h
@@ -47,6 +47,21 @@ struct vhost_vring_addr {
__u64 log_guest_addr;
};
+struct vhost_worker_state {
+ /*
+ * For VHOST_NEW_WORKER the kernel will return the new vhost_worker id.
+ * For VHOST_FREE_WORKER this must be set to the id of the vhost_worker
+ * to free.
+ */
+ int worker_id;
+};
+
+struct vhost_vring_worker {
+ unsigned int index;
+ /* The id of the vhost_worker returned from VHOST_NEW_WORKER */
+ int worker_id;
+};
+
/* no alignment requirement */
struct vhost_iotlb_msg {
__u64 iova;
--
2.25.1
On Mon, Mar 27, 2023 at 09:17:06PM -0500, Mike Christie wrote:> The following patches were built over linux-next which contains various > vhost patches in mst's tree and the vhost_task patchset in Christian > Brauner's tree: > > git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git > > kernel.user_worker branch: > > https://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git/log/?h=kernel.user_workerLooks like it's going to miss this merge cycle. Hopefully in the next one.> The latter patchset handles the review comment for the patches in thread > to make sure that worker threads we create are accounted for in the parent > process's NPROC limit. The patches are scheduled to be sent to Linus for > 6.4. > > The patches in this patchset allow us to support multiple vhost workers > per device. The design is a modified version of Stefan's original idea > where userspace has the kernel create a worker and we pass back the pid. > In this version instead of passing the pid between user/kernel space we > use a worker_id which is just an integer managed by the vhost driver and > we allow userspace to create and free workers and then attach them to > virtqueues at setup time. > > All review comments from the past reviews should be handled. If I didn't > reply to a review comment, I agreed with the comment and should have > handled it in this posting. Let me know if I missed one. > > Results: > -------- > > fio jobs 1 2 4 8 12 16 > ---------------------------------------------------------- > 1 worker 160k 488k - - - - > worker per vq 160k 310k 620k 1300k 1836k 2326k > > Notes: > 0. This used a simple fio command: > > fio --filename=/dev/sdb --direct=1 --rw=randrw --bs=4k \ > --ioengine=libaio --iodepth=128 --numjobs=$JOBS_ABOVE > > and I used a VM with 16 vCPUs and 16 virtqueues. > > 1. The patches were tested with LIO's emulate_pr=0 which drops the > LIO PR lock use. This was a bottleneck at around 12 vqs/jobs. > > 2. Because we have a hard limit of 1024 cmds, if the num jobs * iodepth > was greater than 1024, I would decrease iodepth. So 12 jobs used 85 cmds, > and 16 used 64. > > 3. The perf issue above at 2 jobs is because when we only have 1 worker > we execute more cmds per vhost_work due to all vqs funneling to one worker. > This results in less context switches and better batching without having to > tweak any settings. I'm working on patches to add back batching during lio > completion and do polling on the submission side. > > We will still want the threading patches, because if we batch at the fio > level plus use the vhost theading patches, we can see a big boost like > below. So hopefully doing it at the kernel will allow apps to just work > without having to be smart like fio. > > fio using io_uring and batching with the iodepth_batch* settings: > > fio jobs 1 2 4 8 12 16 > ------------------------------------------------------------- > 1 worker 494k 520k - - - - > worker per vq 496k 878k 1542k 2436k 2304k 2590k > > > V6: > - Rebase against vhost_task patchset. > - Used xa instead of idr. > V5: > - Rebase against user_worker patchset. > - Rebase against flush patchset. > - Redo vhost-scsi tmf flush handling so it doesn't access vq->worker. > V4: > - fix vhost-sock VSOCK_VQ_RX use. > - name functions called directly by ioctl cmd's to match the ioctl cmd. > - break up VHOST_SET_VRING_WORKER into a new, free and attach cmd. > - document worker lifetime, and cgroup, namespace, mm, rlimit > inheritance, make it clear we currently only support sharing within the > device. > - add support to attach workers while IO is running. > - instead of passing a pid_t of the kernel thread, pass a int allocated > by the vhost layer with an idr. > > V3: > - fully convert vhost code to use vq based APIs instead of leaving it > half per dev and half per vq. > - rebase against kernel worker API. > - Drop delayed worker creation. We always create the default worker at > VHOST_SET_OWNER time. Userspace can create and bind workers after that. > > V2: > - change loop that we take a refcount to the worker in > - replaced pid == -1 with define. > - fixed tabbing/spacing coding style issue > - use hash instead of list to lookup workers. > - I dropped the patch that added an ioctl cmd to get a vq's worker's > pid. I saw we might do a generic netlink interface instead. >
Possibly Parallel Threads
- [PATCH v6 11/11] vhost: allow userspace to create workers
- [PATCH v6 04/11] vhost: take worker or vq instead of dev for flushing
- [PATCH v6 11/11] vhost: allow userspace to create workers
- [PATCH 1/1] vhost: Fix crash during early vhost_transport_send_pkt calls
- [PATCH 1/1] vhost: Fix crash during early vhost_transport_send_pkt calls