* Re: [PATCH net] vhost: correctly check the iova range when waking virtqueue
From: Peter Xu @ 2018-08-24 9:36 UTC (permalink / raw)
To: Jason Wang; +Cc: netdev, virtualization, linux-kernel, kvm, mst
In-Reply-To: <20180824085313.21798-1-jasowang@redhat.com>
On Fri, Aug 24, 2018 at 04:53:13PM +0800, Jason Wang wrote:
> We don't wakeup the virtqueue if the first byte of pending iova range
> is the last byte of the range we just got updated. This will lead a
> virtqueue to wait for IOTLB updating forever. Fixing by correct the
> check and wake up the virtqueue in this case.
>
> Fixes: 6b1e6cc7855b ("vhost: new device IOTLB API")
> Reported-by: Peter Xu <peterx@redhat.com>
> Signed-off-by: Jason Wang <jasowang@redhat.com>
Without this patch, this command will trigger the IO hang merely every
time from host to guest:
netperf -H 1.2.3.4 -l 5 -t TCP_RR -- -b 100
After applying, I can run it 10 times continuously without a problem.
Reviewed-by: Peter Xu <peterx@redhat.com>
Tested-by: Peter Xu <peterx@redhat.com>
Thanks,
--
Peter Xu
^ permalink raw reply
* [PATCH net] vhost: correctly check the iova range when waking virtqueue
From: Jason Wang @ 2018-08-24 8:53 UTC (permalink / raw)
To: mst, jasowang; +Cc: netdev, linux-kernel, kvm, virtualization
We don't wakeup the virtqueue if the first byte of pending iova range
is the last byte of the range we just got updated. This will lead a
virtqueue to wait for IOTLB updating forever. Fixing by correct the
check and wake up the virtqueue in this case.
Fixes: 6b1e6cc7855b ("vhost: new device IOTLB API")
Reported-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
The patch is needed for -stable.
---
drivers/vhost/vhost.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 96c1d8400822..b13c6b4b2c66 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -952,7 +952,7 @@ static void vhost_iotlb_notify_vq(struct vhost_dev *d,
list_for_each_entry_safe(node, n, &d->pending_list, node) {
struct vhost_iotlb_msg *vq_msg = &node->msg.iotlb;
if (msg->iova <= vq_msg->iova &&
- msg->iova + msg->size - 1 > vq_msg->iova &&
+ msg->iova + msg->size - 1 >= vq_msg->iova &&
vq_msg->type == VHOST_IOTLB_MISS) {
vhost_poll_queue(&node->vq->poll);
list_del(&node->node);
--
2.17.1
^ permalink raw reply related
* [PULL] virtio,vhost: fixes, tweaks
From: Michael S. Tsirkin @ 2018-08-23 17:42 UTC (permalink / raw)
To: Linus Torvalds
Cc: peter.maydell, cdall, kvm, mst, jean-philippe.brucker, netdev,
penguin-kernel, linux-kernel, virtualization, marc.zyngier, akpm,
mhocko, suzuki.poulose
The following changes since commit 94710cac0ef4ee177a63b5227664b38c95bbf703:
Linux 4.18 (2018-08-12 13:41:04 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git tags/for_linus
for you to fetch changes up to 864d39df09b43f9d09d80bc29d8e8888294b3c4b:
vhost/scsi: increase VHOST_SCSI_PREALLOC_PROT_SGLS to 2048 (2018-08-22 01:01:47 +0300)
----------------------------------------------------------------
virtio, vhost: fixes, tweaks
No new features but a bunch of tweaks such as
switching balloon from oom notifier to shrinker.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
----------------------------------------------------------------
Greg Edwards (2):
vhost: allow vhost-scsi driver to be built-in
vhost/scsi: increase VHOST_SCSI_PREALLOC_PROT_SGLS to 2048
Suzuki K Poulose (2):
virtio: mmio-v1: Validate queue PFN
virtio: pci-legacy: Validate queue pfn
Wei Wang (3):
virtio-balloon: remove BUG() in init_vqs
virtio-balloon: kzalloc the vb struct
virtio_balloon: replace oom notifier with shrinker
drivers/vhost/Kconfig | 2 +-
drivers/vhost/scsi.c | 2 +-
drivers/virtio/virtio_balloon.c | 125 ++++++++++++++++++++-----------------
drivers/virtio/virtio_mmio.c | 20 +++++-
drivers/virtio/virtio_pci_legacy.c | 14 ++++-
5 files changed, 99 insertions(+), 64 deletions(-)
^ permalink raw reply
* [PATCH] vhost/scsi: truncate T10 PI iov_iter to prot_bytes
From: Greg Edwards @ 2018-08-22 19:21 UTC (permalink / raw)
To: virtualization; +Cc: Michael S. Tsirkin, Mike Christie, Paolo Bonzini
Commands with protection information included were not truncating the
protection iov_iter to the number of protection bytes in the command.
This resulted in vhost_scsi mis-calculating the size of the protection
SGL in vhost_scsi_calc_sgls(), and including both the protection and
data SG entries in the protection SGL.
Fixes: 09b13fa8c1a1 ("vhost/scsi: Add ANY_LAYOUT support in vhost_scsi_handle_vq")
Signed-off-by: Greg Edwards <gedwards@ddn.com>
---
drivers/vhost/scsi.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 76f8d649147b..cbe0ea26c1ff 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -964,7 +964,8 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
prot_bytes = vhost32_to_cpu(vq, v_req_pi.pi_bytesin);
}
/*
- * Set prot_iter to data_iter, and advance past any
+ * Set prot_iter to data_iter and truncate it to
+ * prot_bytes, and advance data_iter past any
* preceeding prot_bytes that may be present.
*
* Also fix up the exp_data_len to reflect only the
@@ -973,6 +974,7 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
if (prot_bytes) {
exp_data_len -= prot_bytes;
prot_iter = data_iter;
+ iov_iter_truncate(&prot_iter, prot_bytes);
iov_iter_advance(&data_iter, prot_bytes);
}
tag = vhost64_to_cpu(vq, v_req_pi.tag);
--
2.17.1
^ permalink raw reply related
* Re: [PATCH] vhost/scsi: increase VHOST_SCSI_PREALLOC_PROT_SGLS to 2048
From: Greg Edwards @ 2018-08-22 17:25 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: pbonzini, virtualization
In-Reply-To: <20180808233807-mutt-send-email-mst@kernel.org>
On Wed, Aug 08, 2018 at 11:42:22PM +0300, Michael S. Tsirkin wrote:
> On Wed, Aug 08, 2018 at 01:29:55PM -0600, Greg Edwards wrote:
>> The current value of VHOST_SCSI_PREALLOC_PROT_SGLS is too small to
>> accommodate larger I/Os, e.g. 16-32 MiB, when the VIRTIO_SCSI_F_T10_PI
>> feature bit is negotiated and the backing store supports T10 PI.
>>
>> vhost-scsi rejects the command with errors like:
>>
>> [ 59.581317] vhost_scsi_calc_sgls: requested sgl_count: 1820 exceeds pre-allocated max_sgls: 512
>>
>> Signed-off-by: Greg Edwards <gedwards@ddn.com>
>> ---
>> drivers/vhost/scsi.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
>> index 17fcd3b2e686..8c32cf58d6fa 100644
>> --- a/drivers/vhost/scsi.c
>> +++ b/drivers/vhost/scsi.c
>> @@ -56,7 +56,7 @@
>> #define VHOST_SCSI_DEFAULT_TAGS 256
>> #define VHOST_SCSI_PREALLOC_SGLS 2048
>> #define VHOST_SCSI_PREALLOC_UPAGES 2048
>> -#define VHOST_SCSI_PREALLOC_PROT_SGLS 512
>> +#define VHOST_SCSI_PREALLOC_PROT_SGLS 2048
>>
>> struct vhost_scsi_inflight {
>> /* Wait for the flush operation to finish */
>
> I guess it's ok since PREALLOC_SGLS is already 2K ... or
> am I missing something. Paolo, any input on this?
Ignore this patch. I believe I've identified the root cause [1], and
will send out a new patch once I've done some more testing.
Greg
[1] https://www.spinics.net/lists/linux-scsi/msg122825.html
^ permalink raw reply
* IEEE Record # 41985: 2018 3rd International Conference on Contemporary Computing and Informatics (IC3I).
From: Dr. S K Niranjan Aradhya @ 2018-08-22 4:34 UTC (permalink / raw)
To: virtualization
[-- Attachment #1.1: Type: text/plain, Size: 1581 bytes --]
*<< Apologies for cross-postings >><<< Please circulate among your friends,
peers and researchers >>>*
IEEE Conference Record No.: #41985;
2018 3rd International Conference on Contemporary Computing and Informatics
(IC3I).
Conference Date : 10 - 12 October 2018
Submission Deadline: 1 September 2018
*Submission Link:http://cmsweb.com.sg/ic3i18/index.php/ic3i18/ic3i18/login
<http://cmsweb.com.sg/ic3i18/index.php/ic3i18/ic3i18/login>*
IEEE ISBN : 978-1-5386-6894-8
IEEE Part No. : CFP18AWQ-ART
Selected, accepted and extended paper will be published in Scopus Indexed
International Journal of Forensic Software Engineering published by
InderScience
All accepted and presented papers will be submitted to the IEEE for
possible publication in IEEE Xplore Digital Library. Previous edition
indexed in: SCOPUS, ISI Web of Science, Engineering Index, Google, etc.
If you like to join the TPC or propose a special session or symposiums
please write to: secretariat@ic3i.org
General Chair(s)
IC3I 2018 Conference
----------------------
Disclaimer: We have clearly mentioned the subject lines and your email
address won't be misleading in any form. We have found your mail address
through our own efforts on the web search and not through any illegal way.
If you wish to remove your information from our mailing list or no longer
receive future announcements, please email with REMOVE in subject. Your
request to opt-out will be effective within a reasonable amount of time.
ic3i-cfp.pdf
<https://drive.google.com/file/d/1Tg7HlYL-7wwm0RB6BHrU0RqBaU_ol34F/view?usp=drive_web>
[-- Attachment #1.2: Type: text/html, Size: 2604 bytes --]
[-- Attachment #2: Type: text/plain, Size: 183 bytes --]
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* Re: [PATCH net-next v8 3/7] net: vhost: factor out busy polling logic to vhost_net_busy_poll()
From: Jason Wang @ 2018-08-21 3:15 UTC (permalink / raw)
To: xiangxia.m.yue, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-4-git-send-email-xiangxia.m.yue@gmail.com>
On 2018年08月19日 20:11, xiangxia.m.yue@gmail.com wrote:
> From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
>
> Factor out generic busy polling logic and will be
> used for in tx path in the next patch. And with the patch,
> qemu can set differently the busyloop_timeout for rx queue.
>
> To avoid duplicate codes, introduce the helper functions:
> * sock_has_rx_data(changed from sk_has_rx_data)
> * vhost_net_busy_poll_try_queue
>
> Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
> ---
> drivers/vhost/net.c | 111 +++++++++++++++++++++++++++++++++-------------------
> 1 file changed, 71 insertions(+), 40 deletions(-)
>
> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
> index 32c1b52..453c061 100644
> --- a/drivers/vhost/net.c
> +++ b/drivers/vhost/net.c
> @@ -440,6 +440,75 @@ static void vhost_net_signal_used(struct vhost_net_virtqueue *nvq)
> nvq->done_idx = 0;
> }
>
> +static int sock_has_rx_data(struct socket *sock)
> +{
> + if (unlikely(!sock))
> + return 0;
> +
> + if (sock->ops->peek_len)
> + return sock->ops->peek_len(sock);
> +
> + return skb_queue_empty(&sock->sk->sk_receive_queue);
> +}
> +
> +static void vhost_net_busy_poll_try_queue(struct vhost_net *net,
> + struct vhost_virtqueue *vq)
> +{
> + if (!vhost_vq_avail_empty(&net->dev, vq)) {
> + vhost_poll_queue(&vq->poll);
> + } else if (unlikely(vhost_enable_notify(&net->dev, vq))) {
> + vhost_disable_notify(&net->dev, vq);
> + vhost_poll_queue(&vq->poll);
> + }
> +}
> +
> +static void vhost_net_busy_poll(struct vhost_net *net,
> + struct vhost_virtqueue *rvq,
> + struct vhost_virtqueue *tvq,
> + bool *busyloop_intr,
> + bool poll_rx)
> +{
> + unsigned long busyloop_timeout;
> + unsigned long endtime;
> + struct socket *sock;
> + struct vhost_virtqueue *vq = poll_rx ? tvq : rvq;
> +
> + mutex_lock_nested(&vq->mutex, poll_rx ? VHOST_NET_VQ_TX: VHOST_NET_VQ_RX);
> + vhost_disable_notify(&net->dev, vq);
> + sock = rvq->private_data;
> +
> + busyloop_timeout = poll_rx ? rvq->busyloop_timeout:
> + tvq->busyloop_timeout;
> +
> + preempt_disable();
> + endtime = busy_clock() + busyloop_timeout;
> +
> + while (vhost_can_busy_poll(endtime)) {
> + if (vhost_has_work(&net->dev)) {
> + *busyloop_intr = true;
> + break;
> + }
> +
> + if ((sock_has_rx_data(sock) &&
> + !vhost_vq_avail_empty(&net->dev, rvq)) ||
> + !vhost_vq_avail_empty(&net->dev, tvq))
> + break;
> +
> + cpu_relax();
> + }
> +
> + preempt_enable();
> +
> + if (poll_rx)
> + vhost_net_busy_poll_try_queue(net, tvq);
> + else if (sock_has_rx_data(sock))
> + vhost_net_busy_poll_try_queue(net, rvq);
This could be simplified like:
if (poll_rx || sock_has_rx_data(sock))
vhost_net_busy_poll_try_queue(net, vq);
Thanks
> + else /* On tx here, sock has no rx data. */
> + vhost_enable_notify(&net->dev, rvq);
> +
> + mutex_unlock(&vq->mutex);
> +}
> +
> static int vhost_net_tx_get_vq_desc(struct vhost_net *net,
> struct vhost_net_virtqueue *nvq,
> unsigned int *out_num, unsigned int *in_num,
> @@ -753,16 +822,6 @@ static int peek_head_len(struct vhost_net_virtqueue *rvq, struct sock *sk)
> return len;
> }
>
> -static int sk_has_rx_data(struct sock *sk)
> -{
> - struct socket *sock = sk->sk_socket;
> -
> - if (sock->ops->peek_len)
> - return sock->ops->peek_len(sock);
> -
> - return skb_queue_empty(&sk->sk_receive_queue);
> -}
> -
> static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk,
> bool *busyloop_intr)
> {
> @@ -770,41 +829,13 @@ static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk,
> struct vhost_net_virtqueue *tnvq = &net->vqs[VHOST_NET_VQ_TX];
> struct vhost_virtqueue *rvq = &rnvq->vq;
> struct vhost_virtqueue *tvq = &tnvq->vq;
> - unsigned long uninitialized_var(endtime);
> int len = peek_head_len(rnvq, sk);
>
> - if (!len && tvq->busyloop_timeout) {
> + if (!len && rvq->busyloop_timeout) {
> /* Flush batched heads first */
> vhost_net_signal_used(rnvq);
> /* Both tx vq and rx socket were polled here */
> - mutex_lock_nested(&tvq->mutex, VHOST_NET_VQ_TX);
> - vhost_disable_notify(&net->dev, tvq);
> -
> - preempt_disable();
> - endtime = busy_clock() + tvq->busyloop_timeout;
> -
> - while (vhost_can_busy_poll(endtime)) {
> - if (vhost_has_work(&net->dev)) {
> - *busyloop_intr = true;
> - break;
> - }
> - if ((sk_has_rx_data(sk) &&
> - !vhost_vq_avail_empty(&net->dev, rvq)) ||
> - !vhost_vq_avail_empty(&net->dev, tvq))
> - break;
> - cpu_relax();
> - }
> -
> - preempt_enable();
> -
> - if (!vhost_vq_avail_empty(&net->dev, tvq)) {
> - vhost_poll_queue(&tvq->poll);
> - } else if (unlikely(vhost_enable_notify(&net->dev, tvq))) {
> - vhost_disable_notify(&net->dev, tvq);
> - vhost_poll_queue(&tvq->poll);
> - }
> -
> - mutex_unlock(&tvq->mutex);
> + vhost_net_busy_poll(net, rvq, tvq, busyloop_intr, true);
>
> len = peek_head_len(rnvq, sk);
> }
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* Re: [PATCH net-next v8 7/7] net: vhost: make busyloop_intr more accurate
From: Jason Wang @ 2018-08-21 0:47 UTC (permalink / raw)
To: xiangxia.m.yue, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <f85bfa97-ab9c-2d51-2053-1fe6bb3d45bc@redhat.com>
On 2018年08月21日 08:33, Jason Wang wrote:
>
>
> On 2018年08月19日 20:11, xiangxia.m.yue@gmail.com wrote:
>> From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
>>
>> The patch uses vhost_has_work_pending() to check if
>> the specified handler is scheduled, because in the most case,
>> vhost_has_work() return true when other side handler is added
>> to worker list. Use the vhost_has_work_pending() insead of
>> vhost_has_work().
>>
>> Topology:
>> [Host] ->linux bridge -> tap vhost-net ->[Guest]
>>
>> TCP_STREAM (netperf):
>> * Without the patch: 38035.39 Mbps, 3.37 us mean latency
>> * With the patch: 38409.44 Mbps, 3.34 us mean latency
>
> The improvement is not obvious as last version. Do you imply there's
> some recent changes of vhost that make it faster?
>
I misunderstood the numbers, please ignore this.
It shows less than 1% improvement. I'm not sure it's worth to do so. Can
you try bi-directional pktgen to see if it has more obvious effect?
Thanks
> Thanks
>
>>
>> Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
>> ---
>> drivers/vhost/net.c | 9 ++++++---
>> 1 file changed, 6 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
>> index db63ae2..b6939ef 100644
>> --- a/drivers/vhost/net.c
>> +++ b/drivers/vhost/net.c
>> @@ -487,10 +487,8 @@ 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)) {
>> - *busyloop_intr = true;
>> + if (vhost_has_work(&net->dev))
>> break;
>> - }
>> if ((sock_has_rx_data(sock) &&
>> !vhost_vq_avail_empty(&net->dev, rvq)) ||
>> @@ -513,6 +511,11 @@ static void vhost_net_busy_poll(struct vhost_net
>> *net,
>> !vhost_has_work_pending(&net->dev, VHOST_NET_VQ_RX))
>> vhost_net_enable_vq(net, rvq);
>> + if (vhost_has_work_pending(&net->dev,
>> + poll_rx ?
>> + VHOST_NET_VQ_RX: VHOST_NET_VQ_TX))
>> + *busyloop_intr = true;
>> +
>> mutex_unlock(&vq->mutex);
>> }
>
> _______________________________________________
> Virtualization mailing list
> Virtualization@lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/virtualization
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* Re: [PATCH net-next v8 5/7] net: vhost: introduce bitmap for vhost_poll
From: Jason Wang @ 2018-08-21 0:45 UTC (permalink / raw)
To: xiangxia.m.yue, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-6-git-send-email-xiangxia.m.yue@gmail.com>
On 2018年08月19日 20:11, xiangxia.m.yue@gmail.com wrote:
> From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
>
> The bitmap of vhost_dev can help us to check if the
> specified poll is scheduled. This patch will be used
> for next two patches.
>
> Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
> ---
> drivers/vhost/net.c | 11 +++++++++--
> drivers/vhost/vhost.c | 17 +++++++++++++++--
> drivers/vhost/vhost.h | 7 ++++++-
> 3 files changed, 30 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
> index 1eff72d..23d7ffc 100644
> --- a/drivers/vhost/net.c
> +++ b/drivers/vhost/net.c
> @@ -1135,8 +1135,15 @@ static int vhost_net_open(struct inode *inode, struct file *f)
> }
> vhost_dev_init(dev, vqs, VHOST_NET_VQ_MAX);
>
> - 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,
> + VHOST_NET_VQ_TX,
> + EPOLLOUT, dev);
> +
> + vhost_poll_init(n->poll + VHOST_NET_VQ_RX,
> + handle_rx_net,
> + VHOST_NET_VQ_RX,
> + EPOLLIN, dev);
>
> f->private_data = n;
>
> diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
> index a1c06e7..dc88a60 100644
> --- a/drivers/vhost/vhost.c
> +++ b/drivers/vhost/vhost.c
> @@ -186,7 +186,7 @@ void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn)
>
> /* Init poll structure */
> void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
> - __poll_t mask, struct vhost_dev *dev)
> + __u8 poll_id, __poll_t mask, struct vhost_dev *dev)
> {
> init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
> init_poll_funcptr(&poll->table, vhost_poll_func);
> @@ -194,6 +194,7 @@ void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
> poll->dev = dev;
> poll->wqh = NULL;
>
> + poll->poll_id = poll_id;
> vhost_work_init(&poll->work, fn);
> }
> EXPORT_SYMBOL_GPL(vhost_poll_init);
> @@ -276,8 +277,16 @@ bool vhost_has_work(struct vhost_dev *dev)
> }
> EXPORT_SYMBOL_GPL(vhost_has_work);
>
> +bool vhost_has_work_pending(struct vhost_dev *dev, int poll_id)
> +{
> + return !llist_empty(&dev->work_list) &&
> + test_bit(poll_id, dev->work_pending);
I think we've already had something similar. E.g can we test
VHOST_WORK_QUEUED instead?
Thanks
> +}
> +EXPORT_SYMBOL_GPL(vhost_has_work_pending);
> +
> void vhost_poll_queue(struct vhost_poll *poll)
> {
> + set_bit(poll->poll_id, poll->dev->work_pending);
> vhost_work_queue(poll->dev, &poll->work);
> }
> EXPORT_SYMBOL_GPL(vhost_poll_queue);
> @@ -354,6 +363,7 @@ static int vhost_worker(void *data)
> if (!node)
> schedule();
>
> + bitmap_zero(dev->work_pending, VHOST_DEV_MAX_VQ);
> node = llist_reverse_order(node);
> /* make sure flag is seen after deletion */
> smp_wmb();
> @@ -420,6 +430,8 @@ void vhost_dev_init(struct vhost_dev *dev,
> struct vhost_virtqueue *vq;
> int i;
>
> + BUG_ON(nvqs > VHOST_DEV_MAX_VQ);
> +
> dev->vqs = vqs;
> dev->nvqs = nvqs;
> mutex_init(&dev->mutex);
> @@ -428,6 +440,7 @@ void vhost_dev_init(struct vhost_dev *dev,
> dev->iotlb = NULL;
> dev->mm = NULL;
> dev->worker = NULL;
> + bitmap_zero(dev->work_pending, VHOST_DEV_MAX_VQ);
> init_llist_head(&dev->work_list);
> init_waitqueue_head(&dev->wait);
> INIT_LIST_HEAD(&dev->read_list);
> @@ -445,7 +458,7 @@ 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);
> + i, EPOLLIN, dev);
> }
> }
> EXPORT_SYMBOL_GPL(vhost_dev_init);
> diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
> index 6c844b9..60b6f6d 100644
> --- a/drivers/vhost/vhost.h
> +++ b/drivers/vhost/vhost.h
> @@ -30,6 +30,7 @@ struct vhost_poll {
> wait_queue_head_t *wqh;
> wait_queue_entry_t wait;
> struct vhost_work work;
> + __u8 poll_id;
> __poll_t mask;
> struct vhost_dev *dev;
> };
> @@ -37,9 +38,10 @@ 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);
> +bool vhost_has_work_pending(struct vhost_dev *dev, int poll_id);
>
> void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
> - __poll_t mask, struct vhost_dev *dev);
> + __u8 id, __poll_t mask, struct vhost_dev *dev);
> int vhost_poll_start(struct vhost_poll *poll, struct file *file);
> void vhost_poll_stop(struct vhost_poll *poll);
> void vhost_poll_flush(struct vhost_poll *poll);
> @@ -152,6 +154,8 @@ struct vhost_msg_node {
> struct list_head node;
> };
>
> +#define VHOST_DEV_MAX_VQ 128
> +
> struct vhost_dev {
> struct mm_struct *mm;
> struct mutex mutex;
> @@ -159,6 +163,7 @@ struct vhost_dev {
> int nvqs;
> struct eventfd_ctx *log_ctx;
> struct llist_head work_list;
> + DECLARE_BITMAP(work_pending, VHOST_DEV_MAX_VQ);
> struct task_struct *worker;
> struct vhost_umem *umem;
> struct vhost_umem *iotlb;
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* Re: [PATCH net-next v8 7/7] net: vhost: make busyloop_intr more accurate
From: Jason Wang @ 2018-08-21 0:33 UTC (permalink / raw)
To: xiangxia.m.yue, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-8-git-send-email-xiangxia.m.yue@gmail.com>
On 2018年08月19日 20:11, xiangxia.m.yue@gmail.com wrote:
> From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
>
> The patch uses vhost_has_work_pending() to check if
> the specified handler is scheduled, because in the most case,
> vhost_has_work() return true when other side handler is added
> to worker list. Use the vhost_has_work_pending() insead of
> vhost_has_work().
>
> Topology:
> [Host] ->linux bridge -> tap vhost-net ->[Guest]
>
> TCP_STREAM (netperf):
> * Without the patch: 38035.39 Mbps, 3.37 us mean latency
> * With the patch: 38409.44 Mbps, 3.34 us mean latency
The improvement is not obvious as last version. Do you imply there's
some recent changes of vhost that make it faster?
Thanks
>
> Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
> ---
> drivers/vhost/net.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
> index db63ae2..b6939ef 100644
> --- a/drivers/vhost/net.c
> +++ b/drivers/vhost/net.c
> @@ -487,10 +487,8 @@ 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)) {
> - *busyloop_intr = true;
> + if (vhost_has_work(&net->dev))
> break;
> - }
>
> if ((sock_has_rx_data(sock) &&
> !vhost_vq_avail_empty(&net->dev, rvq)) ||
> @@ -513,6 +511,11 @@ static void vhost_net_busy_poll(struct vhost_net *net,
> !vhost_has_work_pending(&net->dev, VHOST_NET_VQ_RX))
> vhost_net_enable_vq(net, rvq);
>
> + if (vhost_has_work_pending(&net->dev,
> + poll_rx ?
> + VHOST_NET_VQ_RX: VHOST_NET_VQ_TX))
> + *busyloop_intr = true;
> +
> mutex_unlock(&vq->mutex);
> }
>
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* Re: [PATCH net-next v8 0/7] net: vhost: improve performance when enable busyloop
From: Michael S. Tsirkin @ 2018-08-20 20:34 UTC (permalink / raw)
To: xiangxia.m.yue; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
On Sun, Aug 19, 2018 at 05:11:19AM -0700, xiangxia.m.yue@gmail.com wrote:
> From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
>
> This patches improve the guest receive performance.
> On the handle_tx side, we poll the sock receive queue
> at the same time. handle_rx do that in the same way.
>
> For more performance report, see patch 4, 6, 7
Thanks for the patches. I'm traveling this week,
will do my best to review next week.
> Tonghao Zhang (7):
> net: vhost: lock the vqs one by one
> net: vhost: replace magic number of lock annotation
> net: vhost: factor out busy polling logic to vhost_net_busy_poll()
> net: vhost: add rx busy polling in tx path
> net: vhost: introduce bitmap for vhost_poll
> net: vhost: disable rx wakeup during tx busypoll
> net: vhost: make busyloop_intr more accurate
>
> drivers/vhost/net.c | 169 +++++++++++++++++++++++++++++++-------------------
> drivers/vhost/vhost.c | 41 ++++++------
> drivers/vhost/vhost.h | 7 ++-
> 3 files changed, 133 insertions(+), 84 deletions(-)
>
> --
> 1.8.3.1
^ permalink raw reply
* RE: [PATCH V2 net-next 2/6] sctp: Handle sctp packets with CHECKSUM_PARTIAL
From: David Laight @ 2018-08-20 15:39 UTC (permalink / raw)
To: 'Marcelo Ricardo Leitner', Vladislav Yasevich
Cc: virtio-dev@lists.oasis-open.org, nhorman@tuxdriver.com,
mst@redhat.com, netdev@vger.kernel.org,
virtualization@lists.linux-foundation.org,
linux-sctp@vger.kernel.org
In-Reply-To: <20180820145415.GA5310@localhost.localdomain>
From: Marcelo Ricardo Leitner
> Sent: 20 August 2018 15:54
> On Wed, May 02, 2018 at 11:38:24AM -0300, Marcelo Ricardo Leitner wrote:
> > On Tue, May 01, 2018 at 10:07:35PM -0400, Vladislav Yasevich wrote:
> > > With SCTP checksum offload available in virtio, it is now
> > > possible for virtio to receive a sctp packet with CHECKSUM_PARTIAL
> > > set (guest-to-guest traffic). SCTP doesn't really have a
> > > partial checksum like TCP does, because CRC32c can't do partial
> > > additive checksumming.
...
Actually that isn't entirely true.
For all crc, crc(a) ^ crc(b) == crc(a^b).
Since crc(0) == 0 you can xor together two separately calculated crc
provided they both end at the same point.
The slight problem is that you are more likely to be appending
one buffer to another - which requires appending lots of zero
bytes to one of the crcs.
This could be speeded up by using lookup tables that add moderate
sized blocks of zero bytes to a crc instead of adding the zero
bytes one at a time.
Doing it without large const data and/or data cache trashing
is left as an exercise to the implementer.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
^ permalink raw reply
* Re: [PATCH RESEND] x86-64: use RIP-relative calls for paravirt indirect ones
From: Thomas Gleixner @ 2018-08-20 14:54 UTC (permalink / raw)
To: Jan Beulich
Cc: Juergen Gross, the arch/x86 maintainers, linux-kernel,
Linux Virtualization, hpa, Alok Kataria
In-Reply-To: <5B73DFB602000078001DE35E@prv1-mh.provo.novell.com>
On Wed, 15 Aug 2018, Jan Beulich wrote:
> This saves one insn byte per instance, summing up to a savings of over
> 4k in my (stripped down) configuration. No variant of to be patched in
> replacement code relies on the one byte larger size.
>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
> Reviewed-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
^ permalink raw reply
* Call for Workshops Proposals - WorldCIST'19, La Toja Island, Spain
From: Maria Lemos @ 2018-08-19 22:03 UTC (permalink / raw)
To: virtualization
[-- Attachment #1.1: Type: text/plain, Size: 4421 bytes --]
----------------- CALL FOR WORKSHOPS PROPOSALS --------------------
WorldCIST'19 - 7th World Conference on Information Systems and Technologies
16th-19th of April 2019, La Toja Island, Galicia, Spain
http://www.worldcist.org/ <http://www.worldcist.org/>
-----------------------------------------------------------------------------------
The Information Systems and Technologies research and industrial community is invited to submit proposals for the organization of Workshops at WorldCist'19 - 7th World Conference on Information Systems and Technologies, to be held at La Toja Island, Galicia, Spain, 16 - 19 April 2019. WorldCist is a global forum for researchers and practitioners to present and discuss the most recent innovations, trends, results, experiences and concerns in the several perspectives of Information Systems and Technologies.
###############
WORKSHOP FORMAT
###############
Workshops should focus on a specific scientific subject on the scope of WorldCist'19 but not directly included on the main conference areas. Each workshop will be coordinated by an Organizing Committee composed of, at least, two researchers in the field, preferably from different institutions and different countries. The organizers should create an international Program Committee for the Workshop, with recognized researchers within the specific Workshop scientific area. Each workshop should have at least ten submissions and five accepted papers in order to be conducted at WorldCist'19.
The selection of Workshops will be performed by WorldCist'19 Conference/Workshop Chairs. Workshops full and short papers will be published in the conference main proceedings in specific Workshop chapters published by Springer in a book of the AISC series. Proceedings will be submitted for indexation by ISI Thomson, SCOPUS, DBLP, EI-Compendex among several other scientific databases. Extended versions of best selected papers will be published in journals indexed by ISI/SCI, SCOPUS and DBLP. Detailed and up-to-date information may be found at WorldCist'19 website: http://www.worldcist.org/ <http://www.worldcist.org/>
#####################
WORKSHOP ORGANIZATION
#####################
The Organizing Committee of each Workshop will be responsible for:
- Producing and distributing the Workshop Call for Papers (CFP);
- Coordinating the review and selection process for the papers submitted to the Workshop, as Workshop chairs (on the paper submission system to be installed);
- Delivering the final versions of the papers accepted for the Workshop in accordance with the guidelines and deadlines defined by WorldCist'19 organizers;
- Coordinating and chairing the Workshop sessions at the conference.
WorldCist'19 organizers reserve the right to cancel any Workshop if deadlines are missed or if the number of registered attendees is too low to support the costs associated with the Workshop.
################
PROPOSAL CONTENT
################
Workshop proposals should contain the following information:
- Workshop title;
- Brief description of the specific scientific scope of the Workshop;
- List of topics of interest (max 15 topics);
- Reasons the Workshop should be held within WorldCist’19;
- Name, postal address, phone and email of all the members of the Workshop Organizing Committee;
- Preliminary proposal for the Workshop Program Committee (Names and affiliations).
Proposals should be submitted at https://easychair.org/conferences/?conf=worldcist-workshops2019 <https://easychair.org/conferences/?conf=worldcist-workshops2019> in PDF (in English), by September 10, 2018.
###############
IMPORTANT DATES
###############
- Deadline for Workshop proposals: September 10, 2018
- Notification of Workshop acceptance: September 20, 2018
- Workshop Final Information and Program Committee: October 10, 2018
- Deadline for paper submission: November 30, 2018
- Notification of paper acceptance: January 6, 2019
- Deadline for final versions and conference registration: January 20, 2019
- Conference dates: April 16-19, 2019
#####
CHAIR
#####
Luis Paulo Reis, AISTI, IEEE & University of Porto, Portugal
WorldCIST'19 Website: http://www.worldcist.org/ <http://www.worldcist.org/>
---
This email has been checked for viruses by AVG.
https://www.avg.com
[-- Attachment #1.2: Type: text/html, Size: 6068 bytes --]
[-- Attachment #2: Type: text/plain, Size: 183 bytes --]
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* [PATCH net-next v8 7/7] net: vhost: make busyloop_intr more accurate
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
The patch uses vhost_has_work_pending() to check if
the specified handler is scheduled, because in the most case,
vhost_has_work() return true when other side handler is added
to worker list. Use the vhost_has_work_pending() insead of
vhost_has_work().
Topology:
[Host] ->linux bridge -> tap vhost-net ->[Guest]
TCP_STREAM (netperf):
* Without the patch: 38035.39 Mbps, 3.37 us mean latency
* With the patch: 38409.44 Mbps, 3.34 us mean latency
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
---
drivers/vhost/net.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index db63ae2..b6939ef 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -487,10 +487,8 @@ 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)) {
- *busyloop_intr = true;
+ if (vhost_has_work(&net->dev))
break;
- }
if ((sock_has_rx_data(sock) &&
!vhost_vq_avail_empty(&net->dev, rvq)) ||
@@ -513,6 +511,11 @@ static void vhost_net_busy_poll(struct vhost_net *net,
!vhost_has_work_pending(&net->dev, VHOST_NET_VQ_RX))
vhost_net_enable_vq(net, rvq);
+ if (vhost_has_work_pending(&net->dev,
+ poll_rx ?
+ VHOST_NET_VQ_RX: VHOST_NET_VQ_TX))
+ *busyloop_intr = true;
+
mutex_unlock(&vq->mutex);
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 6/7] net: vhost: disable rx wakeup during tx busypoll
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
In the handle_tx, the busypoll will vhost_net_disable/enable_vq
because we have poll the sock. This can improve performance.
This is suggested by Toshiaki Makita and Jason Wang.
If the rx handle is scheduled, we will not enable vq, because it's
not necessary. We do it not in last 'else' because if we receive
the data, but can't queue the rx handle(rx vring is full), then we
enable the vq to avoid case: guest receives the data, vring is not
full then guest can get more data, but vq is disabled, rx vq can't
be wakeup to receive more data.
Topology:
[Host] ->linux bridge -> tap vhost-net ->[Guest]
TCP_STREAM (netperf):
* Without the patch: 37598.20 Mbps, 3.43 us mean latency
* With the patch: 38035.39 Mbps, 3.37 us mean latency
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
---
drivers/vhost/net.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 23d7ffc..db63ae2 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -480,6 +480,9 @@ static void vhost_net_busy_poll(struct vhost_net *net,
busyloop_timeout = poll_rx ? rvq->busyloop_timeout:
tvq->busyloop_timeout;
+ if (!poll_rx)
+ vhost_net_disable_vq(net, rvq);
+
preempt_disable();
endtime = busy_clock() + busyloop_timeout;
@@ -506,6 +509,10 @@ static void vhost_net_busy_poll(struct vhost_net *net,
else /* On tx here, sock has no rx data. */
vhost_enable_notify(&net->dev, rvq);
+ if (!poll_rx &&
+ !vhost_has_work_pending(&net->dev, VHOST_NET_VQ_RX))
+ vhost_net_enable_vq(net, rvq);
+
mutex_unlock(&vq->mutex);
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 5/7] net: vhost: introduce bitmap for vhost_poll
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
The bitmap of vhost_dev can help us to check if the
specified poll is scheduled. This patch will be used
for next two patches.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
---
drivers/vhost/net.c | 11 +++++++++--
drivers/vhost/vhost.c | 17 +++++++++++++++--
drivers/vhost/vhost.h | 7 ++++++-
3 files changed, 30 insertions(+), 5 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 1eff72d..23d7ffc 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -1135,8 +1135,15 @@ static int vhost_net_open(struct inode *inode, struct file *f)
}
vhost_dev_init(dev, vqs, VHOST_NET_VQ_MAX);
- 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,
+ VHOST_NET_VQ_TX,
+ EPOLLOUT, dev);
+
+ vhost_poll_init(n->poll + VHOST_NET_VQ_RX,
+ handle_rx_net,
+ VHOST_NET_VQ_RX,
+ EPOLLIN, dev);
f->private_data = n;
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index a1c06e7..dc88a60 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -186,7 +186,7 @@ void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn)
/* Init poll structure */
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
- __poll_t mask, struct vhost_dev *dev)
+ __u8 poll_id, __poll_t mask, struct vhost_dev *dev)
{
init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
init_poll_funcptr(&poll->table, vhost_poll_func);
@@ -194,6 +194,7 @@ void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
poll->dev = dev;
poll->wqh = NULL;
+ poll->poll_id = poll_id;
vhost_work_init(&poll->work, fn);
}
EXPORT_SYMBOL_GPL(vhost_poll_init);
@@ -276,8 +277,16 @@ bool vhost_has_work(struct vhost_dev *dev)
}
EXPORT_SYMBOL_GPL(vhost_has_work);
+bool vhost_has_work_pending(struct vhost_dev *dev, int poll_id)
+{
+ return !llist_empty(&dev->work_list) &&
+ test_bit(poll_id, dev->work_pending);
+}
+EXPORT_SYMBOL_GPL(vhost_has_work_pending);
+
void vhost_poll_queue(struct vhost_poll *poll)
{
+ set_bit(poll->poll_id, poll->dev->work_pending);
vhost_work_queue(poll->dev, &poll->work);
}
EXPORT_SYMBOL_GPL(vhost_poll_queue);
@@ -354,6 +363,7 @@ static int vhost_worker(void *data)
if (!node)
schedule();
+ bitmap_zero(dev->work_pending, VHOST_DEV_MAX_VQ);
node = llist_reverse_order(node);
/* make sure flag is seen after deletion */
smp_wmb();
@@ -420,6 +430,8 @@ void vhost_dev_init(struct vhost_dev *dev,
struct vhost_virtqueue *vq;
int i;
+ BUG_ON(nvqs > VHOST_DEV_MAX_VQ);
+
dev->vqs = vqs;
dev->nvqs = nvqs;
mutex_init(&dev->mutex);
@@ -428,6 +440,7 @@ void vhost_dev_init(struct vhost_dev *dev,
dev->iotlb = NULL;
dev->mm = NULL;
dev->worker = NULL;
+ bitmap_zero(dev->work_pending, VHOST_DEV_MAX_VQ);
init_llist_head(&dev->work_list);
init_waitqueue_head(&dev->wait);
INIT_LIST_HEAD(&dev->read_list);
@@ -445,7 +458,7 @@ 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);
+ i, EPOLLIN, dev);
}
}
EXPORT_SYMBOL_GPL(vhost_dev_init);
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 6c844b9..60b6f6d 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -30,6 +30,7 @@ struct vhost_poll {
wait_queue_head_t *wqh;
wait_queue_entry_t wait;
struct vhost_work work;
+ __u8 poll_id;
__poll_t mask;
struct vhost_dev *dev;
};
@@ -37,9 +38,10 @@ 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);
+bool vhost_has_work_pending(struct vhost_dev *dev, int poll_id);
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
- __poll_t mask, struct vhost_dev *dev);
+ __u8 id, __poll_t mask, struct vhost_dev *dev);
int vhost_poll_start(struct vhost_poll *poll, struct file *file);
void vhost_poll_stop(struct vhost_poll *poll);
void vhost_poll_flush(struct vhost_poll *poll);
@@ -152,6 +154,8 @@ struct vhost_msg_node {
struct list_head node;
};
+#define VHOST_DEV_MAX_VQ 128
+
struct vhost_dev {
struct mm_struct *mm;
struct mutex mutex;
@@ -159,6 +163,7 @@ struct vhost_dev {
int nvqs;
struct eventfd_ctx *log_ctx;
struct llist_head work_list;
+ DECLARE_BITMAP(work_pending, VHOST_DEV_MAX_VQ);
struct task_struct *worker;
struct vhost_umem *umem;
struct vhost_umem *iotlb;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 4/7] net: vhost: add rx busy polling in tx path
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
This patch improves the guest receive performance.
On the handle_tx side, we poll the sock receive queue at the
same time. handle_rx do that in the same way.
We set the poll-us=100us and use the netperf to test throughput
and mean latency. When running the tests, the vhost-net kthread
of that VM, is alway 100% CPU. The commands are shown as below.
Rx performance is greatly improved by this patch. There is not
notable performance change on tx with this series though. This
patch is useful for bi-directional traffic.
netperf -H IP -t TCP_STREAM -l 20 -- -O "THROUGHPUT, THROUGHPUT_UNITS, MEAN_LATENCY"
Topology:
[Host] ->linux bridge -> tap vhost-net ->[Guest]
TCP_STREAM:
* Without the patch: 19842.95 Mbps, 6.50 us mean latency
* With the patch: 37598.20 Mbps, 3.43 us mean latency
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
---
drivers/vhost/net.c | 33 +++++++++++++--------------------
1 file changed, 13 insertions(+), 20 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 453c061..1eff72d 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -510,31 +510,24 @@ static void vhost_net_busy_poll(struct vhost_net *net,
}
static int vhost_net_tx_get_vq_desc(struct vhost_net *net,
- struct vhost_net_virtqueue *nvq,
+ struct vhost_net_virtqueue *tnvq,
unsigned int *out_num, unsigned int *in_num,
bool *busyloop_intr)
{
- struct vhost_virtqueue *vq = &nvq->vq;
- unsigned long uninitialized_var(endtime);
- int r = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
+ struct vhost_net_virtqueue *rnvq = &net->vqs[VHOST_NET_VQ_RX];
+ struct vhost_virtqueue *rvq = &rnvq->vq;
+ struct vhost_virtqueue *tvq = &tnvq->vq;
+
+ int r = vhost_get_vq_desc(tvq, tvq->iov, ARRAY_SIZE(tvq->iov),
out_num, in_num, NULL, NULL);
- if (r == vq->num && vq->busyloop_timeout) {
- if (!vhost_sock_zcopy(vq->private_data))
- vhost_net_signal_used(nvq);
- preempt_disable();
- endtime = busy_clock() + vq->busyloop_timeout;
- while (vhost_can_busy_poll(endtime)) {
- if (vhost_has_work(vq->dev)) {
- *busyloop_intr = true;
- break;
- }
- if (!vhost_vq_avail_empty(vq->dev, vq))
- break;
- cpu_relax();
- }
- preempt_enable();
- r = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
+ if (r == tvq->num && tvq->busyloop_timeout) {
+ if (!vhost_sock_zcopy(tvq->private_data))
+ vhost_net_signal_used(tnvq);
+
+ vhost_net_busy_poll(net, rvq, tvq, busyloop_intr, false);
+
+ r = vhost_get_vq_desc(tvq, tvq->iov, ARRAY_SIZE(tvq->iov),
out_num, in_num, NULL, NULL);
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 3/7] net: vhost: factor out busy polling logic to vhost_net_busy_poll()
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Factor out generic busy polling logic and will be
used for in tx path in the next patch. And with the patch,
qemu can set differently the busyloop_timeout for rx queue.
To avoid duplicate codes, introduce the helper functions:
* sock_has_rx_data(changed from sk_has_rx_data)
* vhost_net_busy_poll_try_queue
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
---
drivers/vhost/net.c | 111 +++++++++++++++++++++++++++++++++-------------------
1 file changed, 71 insertions(+), 40 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 32c1b52..453c061 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -440,6 +440,75 @@ static void vhost_net_signal_used(struct vhost_net_virtqueue *nvq)
nvq->done_idx = 0;
}
+static int sock_has_rx_data(struct socket *sock)
+{
+ if (unlikely(!sock))
+ return 0;
+
+ if (sock->ops->peek_len)
+ return sock->ops->peek_len(sock);
+
+ return skb_queue_empty(&sock->sk->sk_receive_queue);
+}
+
+static void vhost_net_busy_poll_try_queue(struct vhost_net *net,
+ struct vhost_virtqueue *vq)
+{
+ if (!vhost_vq_avail_empty(&net->dev, vq)) {
+ vhost_poll_queue(&vq->poll);
+ } else if (unlikely(vhost_enable_notify(&net->dev, vq))) {
+ vhost_disable_notify(&net->dev, vq);
+ vhost_poll_queue(&vq->poll);
+ }
+}
+
+static void vhost_net_busy_poll(struct vhost_net *net,
+ struct vhost_virtqueue *rvq,
+ struct vhost_virtqueue *tvq,
+ bool *busyloop_intr,
+ bool poll_rx)
+{
+ unsigned long busyloop_timeout;
+ unsigned long endtime;
+ struct socket *sock;
+ struct vhost_virtqueue *vq = poll_rx ? tvq : rvq;
+
+ mutex_lock_nested(&vq->mutex, poll_rx ? VHOST_NET_VQ_TX: VHOST_NET_VQ_RX);
+ vhost_disable_notify(&net->dev, vq);
+ sock = rvq->private_data;
+
+ busyloop_timeout = poll_rx ? rvq->busyloop_timeout:
+ tvq->busyloop_timeout;
+
+ preempt_disable();
+ endtime = busy_clock() + busyloop_timeout;
+
+ while (vhost_can_busy_poll(endtime)) {
+ if (vhost_has_work(&net->dev)) {
+ *busyloop_intr = true;
+ break;
+ }
+
+ if ((sock_has_rx_data(sock) &&
+ !vhost_vq_avail_empty(&net->dev, rvq)) ||
+ !vhost_vq_avail_empty(&net->dev, tvq))
+ break;
+
+ cpu_relax();
+ }
+
+ preempt_enable();
+
+ if (poll_rx)
+ vhost_net_busy_poll_try_queue(net, tvq);
+ else if (sock_has_rx_data(sock))
+ vhost_net_busy_poll_try_queue(net, rvq);
+ else /* On tx here, sock has no rx data. */
+ vhost_enable_notify(&net->dev, rvq);
+
+ mutex_unlock(&vq->mutex);
+}
+
static int vhost_net_tx_get_vq_desc(struct vhost_net *net,
struct vhost_net_virtqueue *nvq,
unsigned int *out_num, unsigned int *in_num,
@@ -753,16 +822,6 @@ static int peek_head_len(struct vhost_net_virtqueue *rvq, struct sock *sk)
return len;
}
-static int sk_has_rx_data(struct sock *sk)
-{
- struct socket *sock = sk->sk_socket;
-
- if (sock->ops->peek_len)
- return sock->ops->peek_len(sock);
-
- return skb_queue_empty(&sk->sk_receive_queue);
-}
-
static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk,
bool *busyloop_intr)
{
@@ -770,41 +829,13 @@ static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk,
struct vhost_net_virtqueue *tnvq = &net->vqs[VHOST_NET_VQ_TX];
struct vhost_virtqueue *rvq = &rnvq->vq;
struct vhost_virtqueue *tvq = &tnvq->vq;
- unsigned long uninitialized_var(endtime);
int len = peek_head_len(rnvq, sk);
- if (!len && tvq->busyloop_timeout) {
+ if (!len && rvq->busyloop_timeout) {
/* Flush batched heads first */
vhost_net_signal_used(rnvq);
/* Both tx vq and rx socket were polled here */
- mutex_lock_nested(&tvq->mutex, VHOST_NET_VQ_TX);
- vhost_disable_notify(&net->dev, tvq);
-
- preempt_disable();
- endtime = busy_clock() + tvq->busyloop_timeout;
-
- while (vhost_can_busy_poll(endtime)) {
- if (vhost_has_work(&net->dev)) {
- *busyloop_intr = true;
- break;
- }
- if ((sk_has_rx_data(sk) &&
- !vhost_vq_avail_empty(&net->dev, rvq)) ||
- !vhost_vq_avail_empty(&net->dev, tvq))
- break;
- cpu_relax();
- }
-
- preempt_enable();
-
- if (!vhost_vq_avail_empty(&net->dev, tvq)) {
- vhost_poll_queue(&tvq->poll);
- } else if (unlikely(vhost_enable_notify(&net->dev, tvq))) {
- vhost_disable_notify(&net->dev, tvq);
- vhost_poll_queue(&tvq->poll);
- }
-
- mutex_unlock(&tvq->mutex);
+ vhost_net_busy_poll(net, rvq, tvq, busyloop_intr, true);
len = peek_head_len(rnvq, sk);
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 2/7] net: vhost: replace magic number of lock annotation
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Use the VHOST_NET_VQ_XXX as a subclass for mutex_lock_nested.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
---
drivers/vhost/net.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 367d802..32c1b52 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -712,7 +712,7 @@ static void handle_tx(struct vhost_net *net)
struct vhost_virtqueue *vq = &nvq->vq;
struct socket *sock;
- mutex_lock(&vq->mutex);
+ mutex_lock_nested(&vq->mutex, VHOST_NET_VQ_TX);
sock = vq->private_data;
if (!sock)
goto out;
@@ -777,7 +777,7 @@ static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk,
/* Flush batched heads first */
vhost_net_signal_used(rnvq);
/* Both tx vq and rx socket were polled here */
- mutex_lock_nested(&tvq->mutex, 1);
+ mutex_lock_nested(&tvq->mutex, VHOST_NET_VQ_TX);
vhost_disable_notify(&net->dev, tvq);
preempt_disable();
@@ -919,7 +919,7 @@ static void handle_rx(struct vhost_net *net)
__virtio16 num_buffers;
int recv_pkts = 0;
- mutex_lock_nested(&vq->mutex, 0);
+ mutex_lock_nested(&vq->mutex, VHOST_NET_VQ_RX);
sock = vq->private_data;
if (!sock)
goto out;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 1/7] net: vhost: lock the vqs one by one
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
In-Reply-To: <1534680686-3108-1-git-send-email-xiangxia.m.yue@gmail.com>
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
This patch changes the way that lock all vqs
at the same, to lock them one by one. It will
be used for next patch to avoid the deadlock.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
drivers/vhost/vhost.c | 24 +++++++-----------------
1 file changed, 7 insertions(+), 17 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index a502f1a..a1c06e7 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -294,8 +294,11 @@ static void vhost_vq_meta_reset(struct vhost_dev *d)
{
int i;
- for (i = 0; i < d->nvqs; ++i)
+ for (i = 0; i < d->nvqs; ++i) {
+ mutex_lock(&d->vqs[i]->mutex);
__vhost_vq_meta_reset(d->vqs[i]);
+ mutex_unlock(&d->vqs[i]->mutex);
+ }
}
static void vhost_vq_reset(struct vhost_dev *dev,
@@ -890,20 +893,6 @@ static inline void __user *__vhost_get_user(struct vhost_virtqueue *vq,
#define vhost_get_used(vq, x, ptr) \
vhost_get_user(vq, x, ptr, VHOST_ADDR_USED)
-static void vhost_dev_lock_vqs(struct vhost_dev *d)
-{
- int i = 0;
- for (i = 0; i < d->nvqs; ++i)
- mutex_lock_nested(&d->vqs[i]->mutex, i);
-}
-
-static void vhost_dev_unlock_vqs(struct vhost_dev *d)
-{
- int i = 0;
- for (i = 0; i < d->nvqs; ++i)
- mutex_unlock(&d->vqs[i]->mutex);
-}
-
static int vhost_new_umem_range(struct vhost_umem *umem,
u64 start, u64 size, u64 end,
u64 userspace_addr, int perm)
@@ -953,7 +942,10 @@ static void vhost_iotlb_notify_vq(struct vhost_dev *d,
if (msg->iova <= vq_msg->iova &&
msg->iova + msg->size - 1 > vq_msg->iova &&
vq_msg->type == VHOST_IOTLB_MISS) {
+ mutex_lock(&node->vq->mutex);
vhost_poll_queue(&node->vq->poll);
+ mutex_unlock(&node->vq->mutex);
+
list_del(&node->node);
kfree(node);
}
@@ -985,7 +977,6 @@ static int vhost_process_iotlb_msg(struct vhost_dev *dev,
int ret = 0;
mutex_lock(&dev->mutex);
- vhost_dev_lock_vqs(dev);
switch (msg->type) {
case VHOST_IOTLB_UPDATE:
if (!dev->iotlb) {
@@ -1019,7 +1010,6 @@ static int vhost_process_iotlb_msg(struct vhost_dev *dev,
break;
}
- vhost_dev_unlock_vqs(dev);
mutex_unlock(&dev->mutex);
return ret;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v8 0/7] net: vhost: improve performance when enable busyloop
From: xiangxia.m.yue @ 2018-08-19 12:11 UTC (permalink / raw)
To: jasowang, mst, makita.toshiaki; +Cc: netdev, virtualization
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
This patches improve the guest receive performance.
On the handle_tx side, we poll the sock receive queue
at the same time. handle_rx do that in the same way.
For more performance report, see patch 4, 6, 7
Tonghao Zhang (7):
net: vhost: lock the vqs one by one
net: vhost: replace magic number of lock annotation
net: vhost: factor out busy polling logic to vhost_net_busy_poll()
net: vhost: add rx busy polling in tx path
net: vhost: introduce bitmap for vhost_poll
net: vhost: disable rx wakeup during tx busypoll
net: vhost: make busyloop_intr more accurate
drivers/vhost/net.c | 169 +++++++++++++++++++++++++++++++-------------------
drivers/vhost/vhost.c | 41 ++++++------
drivers/vhost/vhost.h | 7 ++-
3 files changed, 133 insertions(+), 84 deletions(-)
--
1.8.3.1
^ permalink raw reply
* Call for Papers - ICITS'19 - Quito, Ecuador
From: Maria Lemos @ 2018-08-18 11:02 UTC (permalink / raw)
To: virtualization
[-- Attachment #1.1: Type: text/plain, Size: 6043 bytes --]
***** Proceedings by Springer. Indexed by Scopus, ISI, etc.
------------
ICITS'19 - The 2019 International Conference on Information Technology & Systems
6 - 8 February 2019, Quito, Ecuador
http://www.icits.me/ <http://www.icits.me/>
------------------------------ ------------------------------ ------------------------------ ------------------------
ICITS'19 - The 2019 International Conference on Information Technology & Systems, to be held at Quito, Ecuador, 6 - 8 February 2019, is an international forum for researchers and practitioners to present and discuss the most recent innovations, trends, results, experiences and concerns in the several perspectives of Information Technology & Systems.
We are pleased to invite you to submit your papers to ICITS'19. They can be written in English, Spanish or Portuguese. All submissions will be reviewed on the basis of relevance, originality, importance and clarity.
Topics
Submitted papers should be related with one or more of the main themes proposed for the Conference:
A) Information and Knowledge Management (IKM);
B) Organizational Models and Information Systems (OMIS);
C) Software and Systems Modeling (SSM);
D) Software Systems, Architectures, Applications and Tools (SSAAT);
E) Multimedia Systems and Applications (MSA);
F) Computer Networks, Mobility and Pervasive Systems (CNMPS);
G) Intelligent and Decision Support Systems (IDSS);
H) Big Data Analytics and Applications (BDAA);
I) Human-Computer Interaction (HCI);
J) Ethics, Computers and Security (ECS)
K) Health Informatics (HIS);
L) Information Technologies in Education (ITE);
M) Cybersecurity and Cyber-defense;
N) Electromagnetics, Sensors and Antennas for Security.
Submission and Decision
Submitted papers written in English (until 10-page limit) must comply with the format of Advances in Intelligent Systems and Computing series (see Instructions for Authors at Springer Website <http://www.springer.com/series/11156> or download a DOC example <http://www.icits.me/springerformat.doc>), must not have been published before, not be under review for any other conference or publication and not include any information leading to the authors’ identification. Therefore, the authors’ names, affiliations and bibliographic references should not be included in the version for evaluation by the Scientific Committee. This information should only be included in the camera-ready version, saved in Word or Latex format and also in PDF format. These files must be accompanied by the Consent to Publish form <http://www.icits.me/copyright.pdf> filled out, in a ZIP file, and uploaded at the conference management system.
Submitted papers written in Spanish or Portuguese (until 15-page limit) must comply with the format of RISTI <http://www.risti.xyz/> - Revista Ibérica de Sistemas e Tecnologias de Informação (download instructions/template for authors in Spanish <http://www.risti.xyz/formato-es.doc> or Portuguese <http://www.risti.xyz/formato-pt.doc>), must not have been published before, not be under review for any other conference or publication and not include any information leading to the authors’ identification. Therefore, the authors’ names, affiliations and bibliographic references should not be included in the version for evaluation by the Scientific Committee. This information should only be included in the camera-ready version, saved in Word. These file must be uploaded at the conference management system in a ZIP file.
All papers will be subjected to a “double-blind review” by at least two members of the Scientific Committee.
Based on Scientific Committee evaluation, a paper can be rejected or accepted by the Conference Chairs. In the later case, it can be accepted as paper or poster.
The authors of papers accepted as posters must build and print a poster to be exhibited during the Conference. This poster must follow an A1 or A2 vertical format. The Conference can includes Work Sessions where these posters are presented and orally discussed, with a 7 minute limit per poster.
The authors of accepted papers will have 15 minutes to present their work in a Conference Work Session; approximately 5 minutes of discussion will follow each presentation.
Publication and Indexing
To ensure that an accepted paper is published, at least one of the authors must be fully registered by the 9th of October 2018, and the paper must comply with the suggested layout and page-limit. Additionally, all recommended changes must be addressed by the authors before they submit the camera-ready version.
No more than one paper per registration will be published. An extra fee must be paid for publication of additional papers, with a maximum of one additional paper per registration. One registration permits only the participation of one author in the conference.
Papers written in English and accepted and registered will be published in Proceedings by Springer, in a book of the Advances in Intelligent Systems and Computing <http://www.springer.com/series/11156>series, will be submitted for indexation by ISI, EI-Compendex, SCOPUS and DBLP, among others, and will be available in the SpringerLink Digital Library <http://link.springer.com/>.
Papers written in Spanish or Portuguese and accepted and registered will be published in a Special Issue of RISTI <http://www.risti.xyz/index.php?option=com_content&view=article&id=3&Itemid=104&lang=es> and will be submitted for indexation by SCOPUS, among others.
Important Dates
Paper Submission: September 16, 2018
Notification of Acceptance: October 28, 2018
Payment of Registration, to ensure the inclusion of an accepted paper in the conference proceedings: November 9, 2018.
Camera-ready Submission: November 9, 2018
Website of ICITS'19: http://www.icits.me/ <http://www.icits.me/>
---
This email has been checked for viruses by AVG.
https://www.avg.com
[-- Attachment #1.2: Type: text/html, Size: 10805 bytes --]
[-- Attachment #2: Type: text/plain, Size: 183 bytes --]
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
^ permalink raw reply
* [PATCH v4 3/3] virtio_balloon: replace oom notifier with shrinker
From: Wei Wang @ 2018-08-16 7:50 UTC (permalink / raw)
To: virtio-dev, linux-kernel, virtualization, linux-mm, mst, mhocko,
akpm, penguin-kernel
In-Reply-To: <1534405858-27085-1-git-send-email-wei.w.wang@intel.com>
The OOM notifier is getting deprecated to use for the reasons:
- As a callout from the oom context, it is too subtle and easy to
generate bugs and corner cases which are hard to track;
- It is called too late (after the reclaiming has been performed).
Drivers with large amuont of reclaimable memory is expected to
release them at an early stage of memory pressure;
- The notifier callback isn't aware of oom contrains;
Link: https://lkml.org/lkml/2018/7/12/314
This patch replaces the virtio-balloon oom notifier with a shrinker
to release balloon pages on memory pressure. The balloon pages are
given back to mm adaptively by returning the number of pages that the
reclaimer is asking for (i.e. sc->nr_to_scan).
Currently the max possible value of sc->nr_to_scan passed to the balloon
shrinker is SHRINK_BATCH, which is 128. This is smaller than the
limitation that only VIRTIO_BALLOON_ARRAY_PFNS_MAX (256) pages can be
returned via one invocation of leak_balloon. But this patch still
considers the case that SHRINK_BATCH or shrinker->batch could be changed
to a value larger than VIRTIO_BALLOON_ARRAY_PFNS_MAX, which will need to
do multiple invocations of leak_balloon.
Historically, the feature VIRTIO_BALLOON_F_DEFLATE_ON_OOM has been used
to release balloon pages on OOM. We continue to use this feature bit for
the shrinker, so the shrinker is only registered when this feature bit
has been negotiated with host.
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
---
drivers/virtio/virtio_balloon.c | 110 +++++++++++++++++++++-------------------
1 file changed, 59 insertions(+), 51 deletions(-)
diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
index d97d73c..d1c1f62 100644
--- a/drivers/virtio/virtio_balloon.c
+++ b/drivers/virtio/virtio_balloon.c
@@ -27,7 +27,6 @@
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/balloon_compaction.h>
-#include <linux/oom.h>
#include <linux/wait.h>
#include <linux/mm.h>
#include <linux/mount.h>
@@ -40,13 +39,8 @@
*/
#define VIRTIO_BALLOON_PAGES_PER_PAGE (unsigned)(PAGE_SIZE >> VIRTIO_BALLOON_PFN_SHIFT)
#define VIRTIO_BALLOON_ARRAY_PFNS_MAX 256
-#define OOM_VBALLOON_DEFAULT_PAGES 256
#define VIRTBALLOON_OOM_NOTIFY_PRIORITY 80
-static int oom_pages = OOM_VBALLOON_DEFAULT_PAGES;
-module_param(oom_pages, int, S_IRUSR | S_IWUSR);
-MODULE_PARM_DESC(oom_pages, "pages to free on OOM");
-
#ifdef CONFIG_BALLOON_COMPACTION
static struct vfsmount *balloon_mnt;
#endif
@@ -86,8 +80,8 @@ struct virtio_balloon {
/* Memory statistics */
struct virtio_balloon_stat stats[VIRTIO_BALLOON_S_NR];
- /* To register callback in oom notifier call chain */
- struct notifier_block nb;
+ /* To register a shrinker to shrink memory upon memory pressure */
+ struct shrinker shrinker;
};
static struct virtio_device_id id_table[] = {
@@ -365,38 +359,6 @@ static void update_balloon_size(struct virtio_balloon *vb)
&actual);
}
-/*
- * virtballoon_oom_notify - release pages when system is under severe
- * memory pressure (called from out_of_memory())
- * @self : notifier block struct
- * @dummy: not used
- * @parm : returned - number of freed pages
- *
- * The balancing of memory by use of the virtio balloon should not cause
- * the termination of processes while there are pages in the balloon.
- * If virtio balloon manages to release some memory, it will make the
- * system return and retry the allocation that forced the OOM killer
- * to run.
- */
-static int virtballoon_oom_notify(struct notifier_block *self,
- unsigned long dummy, void *parm)
-{
- struct virtio_balloon *vb;
- unsigned long *freed;
- unsigned num_freed_pages;
-
- vb = container_of(self, struct virtio_balloon, nb);
- if (!virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_DEFLATE_ON_OOM))
- return NOTIFY_OK;
-
- freed = parm;
- num_freed_pages = leak_balloon(vb, oom_pages);
- update_balloon_size(vb);
- *freed += num_freed_pages;
-
- return NOTIFY_OK;
-}
-
static void update_balloon_stats_func(struct work_struct *work)
{
struct virtio_balloon *vb;
@@ -550,6 +512,52 @@ static struct file_system_type balloon_fs = {
#endif /* CONFIG_BALLOON_COMPACTION */
+static unsigned long virtio_balloon_shrinker_scan(struct shrinker *shrinker,
+ struct shrink_control *sc)
+{
+ unsigned long pages_to_free, pages_freed = 0;
+ struct virtio_balloon *vb = container_of(shrinker,
+ struct virtio_balloon, shrinker);
+
+ pages_to_free = sc->nr_to_scan * VIRTIO_BALLOON_PAGES_PER_PAGE;
+
+ /*
+ * One invocation of leak_balloon can deflate at most
+ * VIRTIO_BALLOON_ARRAY_PFNS_MAX balloon pages, so we call it
+ * multiple times to deflate pages till reaching pages_to_free.
+ */
+ while (vb->num_pages && pages_to_free) {
+ pages_to_free -= pages_freed;
+ pages_freed += leak_balloon(vb, pages_to_free);
+ }
+ update_balloon_size(vb);
+
+ return pages_freed / VIRTIO_BALLOON_PAGES_PER_PAGE;
+}
+
+static unsigned long virtio_balloon_shrinker_count(struct shrinker *shrinker,
+ struct shrink_control *sc)
+{
+ struct virtio_balloon *vb = container_of(shrinker,
+ struct virtio_balloon, shrinker);
+
+ return vb->num_pages / VIRTIO_BALLOON_PAGES_PER_PAGE;
+}
+
+static void virtio_balloon_unregister_shrinker(struct virtio_balloon *vb)
+{
+ unregister_shrinker(&vb->shrinker);
+}
+
+static int virtio_balloon_register_shrinker(struct virtio_balloon *vb)
+{
+ vb->shrinker.scan_objects = virtio_balloon_shrinker_scan;
+ vb->shrinker.count_objects = virtio_balloon_shrinker_count;
+ vb->shrinker.seeks = DEFAULT_SEEKS;
+
+ return register_shrinker(&vb->shrinker);
+}
+
static int virtballoon_probe(struct virtio_device *vdev)
{
struct virtio_balloon *vb;
@@ -580,17 +588,10 @@ static int virtballoon_probe(struct virtio_device *vdev)
if (err)
goto out_free_vb;
- vb->nb.notifier_call = virtballoon_oom_notify;
- vb->nb.priority = VIRTBALLOON_OOM_NOTIFY_PRIORITY;
- err = register_oom_notifier(&vb->nb);
- if (err < 0)
- goto out_del_vqs;
-
#ifdef CONFIG_BALLOON_COMPACTION
balloon_mnt = kern_mount(&balloon_fs);
if (IS_ERR(balloon_mnt)) {
err = PTR_ERR(balloon_mnt);
- unregister_oom_notifier(&vb->nb);
goto out_del_vqs;
}
@@ -599,12 +600,19 @@ static int virtballoon_probe(struct virtio_device *vdev)
if (IS_ERR(vb->vb_dev_info.inode)) {
err = PTR_ERR(vb->vb_dev_info.inode);
kern_unmount(balloon_mnt);
- unregister_oom_notifier(&vb->nb);
goto out_del_vqs;
}
vb->vb_dev_info.inode->i_mapping->a_ops = &balloon_aops;
#endif
-
+ /*
+ * We continue to use VIRTIO_BALLOON_F_DEFLATE_ON_OOM to decide if a
+ * shrinker needs to be registered to relieve memory pressure.
+ */
+ if (virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_DEFLATE_ON_OOM)) {
+ err = virtio_balloon_register_shrinker(vb);
+ if (err)
+ goto out_del_vqs;
+ }
virtio_device_ready(vdev);
if (towards_target(vb))
@@ -636,8 +644,8 @@ static void virtballoon_remove(struct virtio_device *vdev)
{
struct virtio_balloon *vb = vdev->priv;
- unregister_oom_notifier(&vb->nb);
-
+ if (virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_DEFLATE_ON_OOM))
+ virtio_balloon_unregister_shrinker(vb);
spin_lock_irq(&vb->stop_update_lock);
vb->stop_update = true;
spin_unlock_irq(&vb->stop_update_lock);
--
2.7.4
^ permalink raw reply related
* [PATCH v4 2/3] virtio-balloon: kzalloc the vb struct
From: Wei Wang @ 2018-08-16 7:50 UTC (permalink / raw)
To: virtio-dev, linux-kernel, virtualization, linux-mm, mst, mhocko,
akpm, penguin-kernel
In-Reply-To: <1534405858-27085-1-git-send-email-wei.w.wang@intel.com>
Zero all the vb fields at alloaction, so that we don't need to
zero-initialize each field one by one later.
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
---
drivers/virtio/virtio_balloon.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
index 8100e77..d97d73c 100644
--- a/drivers/virtio/virtio_balloon.c
+++ b/drivers/virtio/virtio_balloon.c
@@ -561,7 +561,7 @@ static int virtballoon_probe(struct virtio_device *vdev)
return -EINVAL;
}
- vdev->priv = vb = kmalloc(sizeof(*vb), GFP_KERNEL);
+ vdev->priv = vb = kzalloc(sizeof(*vb), GFP_KERNEL);
if (!vb) {
err = -ENOMEM;
goto out;
@@ -570,8 +570,6 @@ static int virtballoon_probe(struct virtio_device *vdev)
INIT_WORK(&vb->update_balloon_stats_work, update_balloon_stats_func);
INIT_WORK(&vb->update_balloon_size_work, update_balloon_size_func);
spin_lock_init(&vb->stop_update_lock);
- vb->stop_update = false;
- vb->num_pages = 0;
mutex_init(&vb->balloon_lock);
init_waitqueue_head(&vb->acked);
vb->vdev = vdev;
@@ -602,7 +600,6 @@ static int virtballoon_probe(struct virtio_device *vdev)
err = PTR_ERR(vb->vb_dev_info.inode);
kern_unmount(balloon_mnt);
unregister_oom_notifier(&vb->nb);
- vb->vb_dev_info.inode = NULL;
goto out_del_vqs;
}
vb->vb_dev_info.inode->i_mapping->a_ops = &balloon_aops;
--
2.7.4
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox