On 14.09.2023 16:57, Stefano Garzarella wrote:
> On Mon, Sep 11, 2023 at 11:22:34PM +0300, Arseniy Krasnov wrote:
>> This adds handling of MSG_ZEROCOPY flag on transmission path:
>>
>> 1) If this flag is set and zerocopy transmission is possible (enabled
>> in socket options and transport allows zerocopy), then non-linear
>> skb will be created and filled with the pages of user's buffer.
>> Pages of user's buffer are locked in memory by 'get_user_pages()'.
>> 2) Replaces way of skb owning: instead of 'skb_set_owner_sk_safe()' it
>> calls 'skb_set_owner_w()'. Reason of this change is that
>> '__zerocopy_sg_from_iter()' increments 'sk_wmem_alloc' of socket, so
>> to decrease this field correctly, proper skb destructor is needed:
>> 'sock_wfree()'. This destructor is set by 'skb_set_owner_w()'.
>> 3) Adds new callback to 'struct virtio_transport': 'can_msgzerocopy'.
>> If this callback is set, then transport needs extra check to be able
>> to send provided number of buffers in zerocopy mode. Currently, the
>> only transport that needs this callback set is virtio, because this
>> transport adds new buffers to the virtio queue and we need to check,
>> that number of these buffers is less than size of the queue (it is
>> required by virtio spec). vhost and loopback transports don't need
>> this check.
>>
>> Signed-off-by: Arseniy Krasnov <[email protected]>
>> ---
>> Changelog:
>> v5(big patchset) -> v1:
>> * Refactorings of 'if' conditions.
>> * Remove extra blank line.
>> * Remove 'frag_off' field unneeded init.
>> * Add function 'virtio_transport_fill_skb()' which fills both linear
>> and non-linear skb with provided data.
>> v1 -> v2:
>> * Use original order of last four arguments in 'virtio_transport_alloc_skb()'.
>> v2 -> v3:
>> * Add new transport callback: 'msgzerocopy_check_iov'. It checks that
>> provided 'iov_iter' with data could be sent in a zerocopy mode.
>> If this callback is not set in transport - transport allows to send
>> any 'iov_iter' in zerocopy mode. Otherwise - if callback returns 'true'
>> then zerocopy is allowed. Reason of this callback is that in case of
>> G2H transmission we insert whole skb to the tx virtio queue and such
>> skb must fit to the size of the virtio queue to be sent in a single
>> iteration (may be tx logic in 'virtio_transport.c' could be reworked
>> as in vhost to support partial send of current skb). This callback
>> will be enabled only for G2H path. For details pls see comment
>> 'Check that tx queue...' below.
>> v3 -> v4:
>> * 'msgzerocopy_check_iov' moved from 'struct vsock_transport' to
>> 'struct virtio_transport' as it is virtio specific callback and
>> never needed in other transports.
>> v4 -> v5:
>> * 'msgzerocopy_check_iov' renamed to 'can_msgzerocopy' and now it
>> uses number of buffers to send as input argument. I think there is
>> no need to pass iov to this callback (at least today, it is used only
>> by guest side of virtio transport), because the only thing that this
>> callback does is comparison of number of buffers to be inserted to
>> the tx queue and size of this queue.
>> * Remove any checks for type of current 'iov_iter' with payload (is it
>> 'iovec' or 'ubuf'). These checks left from the earlier versions where I
>> didn't use already implemented kernel API which handles every type of
>> 'iov_iter'.
>> v5 -> v6:
>> * Refactor 'virtio_transport_fill_skb()'.
>> * Add 'WARN_ON_ONCE()' and comment on invalid combination of destination
>> socket and payload in 'virtio_transport_alloc_skb()'.
>> v7 -> v8:
>> * Move '+1' addition from 'can_msgzerocopy' callback body to the caller.
>> This addition means packet header.
>> * In 'virtio_transport_can_zcopy()' rename 'max_to_send' argument to
>> 'pkt_len'.
>> * Update commit message by adding details about new 'can_msgzerocopy'
>> callback.
>> * In 'virtio_transport_init_hdr()' move 'len' argument directly after
>> 'info'.
>> * Add comment about processing last skb in tx loop.
>> * Update comment for 'can_msgzerocopy' callback for more details.
>>
>> include/linux/virtio_vsock.h | 9 +
>> net/vmw_vsock/virtio_transport.c | 32 +++
>> net/vmw_vsock/virtio_transport_common.c | 256 ++++++++++++++++++------
>> 3 files changed, 239 insertions(+), 58 deletions(-)
>>
>> diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
>> index a91fbdf233e4..ebb3ce63d64d 100644
>> --- a/include/linux/virtio_vsock.h
>> +++ b/include/linux/virtio_vsock.h
>> @@ -160,6 +160,15 @@ struct virtio_transport {
>>
>> /* Takes ownership of the packet */
>> int (*send_pkt)(struct sk_buff *skb);
>> +
>> + /* Used in MSG_ZEROCOPY mode. Checks, that provided data
>> + * (number of buffers) could be transmitted with zerocopy
>> + * mode. If this callback is not implemented for the current
>> + * transport - this means that this transport doesn't need
>> + * extra checks and can perform zerocopy transmission by
>> + * default.
>> + */
>> + bool (*can_msgzerocopy)(int bufs_num);
>> };
>>
>> ssize_t
>> diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
>> index 73d730156349..09ba3128e759 100644
>> --- a/net/vmw_vsock/virtio_transport.c
>> +++ b/net/vmw_vsock/virtio_transport.c
>> @@ -455,6 +455,37 @@ static void virtio_vsock_rx_done(struct virtqueue *vq)
>> queue_work(virtio_vsock_workqueue, &vsock->rx_work);
>> }
>>
>> +static bool virtio_transport_can_msgzerocopy(int bufs_num)
>> +{
>> + struct virtio_vsock *vsock;
>> + bool res = false;
>> +
>> + rcu_read_lock();
>> +
>> + vsock = rcu_dereference(the_virtio_vsock);
>> + if (vsock) {
>> + struct virtqueue *vq = vsock->vqs[VSOCK_VQ_TX];
>> +
>> + /* Check that tx queue is large enough to keep whole
>> + * data to send. This is needed, because when there is
>> + * not enough free space in the queue, current skb to
>> + * send will be reinserted to the head of tx list of
>> + * the socket to retry transmission later, so if skb
>> + * is bigger than whole queue, it will be reinserted
>> + * again and again, thus blocking other skbs to be sent.
>> + * Each page of the user provided buffer will be added
>> + * as a single buffer to the tx virtqueue, so compare
>> + * number of pages against maximum capacity of the queue.
>> + */
>> + if (bufs_num <= vq->num_max)
>> + res = true;
>> + }
>> +
>> + rcu_read_unlock();
>> +
>> + return res;
>> +}
>> +
>> static bool virtio_transport_seqpacket_allow(u32 remote_cid);
>>
>> static struct virtio_transport virtio_transport = {
>> @@ -504,6 +535,7 @@ static struct virtio_transport virtio_transport = {
>> },
>>
>> .send_pkt = virtio_transport_send_pkt,
>> + .can_msgzerocopy = virtio_transport_can_msgzerocopy,
>> };
>>
>> static bool virtio_transport_seqpacket_allow(u32 remote_cid)
>> diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
>> index 3a48e48a99ac..e358f118b07e 100644
>> --- a/net/vmw_vsock/virtio_transport_common.c
>> +++ b/net/vmw_vsock/virtio_transport_common.c
>> @@ -37,73 +37,110 @@ virtio_transport_get_ops(struct vsock_sock *vsk)
>> return container_of(t, struct virtio_transport, transport);
>> }
>>
>> -/* Returns a new packet on success, otherwise returns NULL.
>> - *
>> - * If NULL is returned, errp is set to a negative errno.
>> - */
>
> Why we are removing this comment?
I think I can return it like:
/* Returns a new sk_buff on success, otherwise returns NULL. */
Line related to 'errp' is not needed i guess.
>
>> -static struct sk_buff *
>> -virtio_transport_alloc_skb(struct virtio_vsock_pkt_info *info,
>> - size_t len,
>> - u32 src_cid,
>> - u32 src_port,
>> - u32 dst_cid,
>> - u32 dst_port)
>> -{
>> - const size_t skb_len = VIRTIO_VSOCK_SKB_HEADROOM + len;
>> - struct virtio_vsock_hdr *hdr;
>> - struct sk_buff *skb;
>> - void *payload;
>> - int err;
>> +static bool virtio_transport_can_zcopy(struct virtio_vsock_pkt_info *info,
>> + size_t pkt_len)
>> +{
>> + const struct virtio_transport *t_ops;
>> + struct iov_iter *iov_iter;
>>
>> - skb = virtio_vsock_alloc_skb(skb_len, GFP_KERNEL);
>> - if (!skb)
>> - return NULL;
>> + if (!info->msg)
>> + return false;
>>
>> - hdr = virtio_vsock_hdr(skb);
>> - hdr->type = cpu_to_le16(info->type);
>> - hdr->op = cpu_to_le16(info->op);
>> - hdr->src_cid = cpu_to_le64(src_cid);
>> - hdr->dst_cid = cpu_to_le64(dst_cid);
>> - hdr->src_port = cpu_to_le32(src_port);
>> - hdr->dst_port = cpu_to_le32(dst_port);
>> - hdr->flags = cpu_to_le32(info->flags);
>> - hdr->len = cpu_to_le32(len);
>> + iov_iter = &info->msg->msg_iter;
>>
>> - if (info->msg && len > 0) {
>> - payload = skb)put(skb, len);
>> - err = memcpy_from_msg(payload, info->msg, len);
>> - if (err)
>> - goto out;
>> + if (iov_iter->iov_offset)
>> + return false;
>>
>> - if (msg_data_left(info->msg) == 0 &&
>> - info->type == VIRTIO_VSOCK_TYPE_SEQPACKET) {
>> - hdr->flags |= cpu_to_le32(VIRTIO_VSOCK_SEQ_EOM);
>> + /* We can't send whole iov. */
>> + if (iov_iter->count > pkt_len)
>> + return false;
>>
>> - if (info->msg->msg_flags & MSG_EOR)
>> - hdr->flags |= cpu_to_le32(VIRTIO_VSOCK_SEQ_EOR);
>> - }
>> + /* Check that transport can send data in zerocopy mode. */
>> + t_ops = virtio_transport_get_ops(info->vsk);
>
> While reviewing I was wondering why here we don't check if `t_ops` is
> NULL.
>
> Then I realized that the only caller of this function
> (virtio_transport_send_pkt_info()) already get the vsk ops calling
> virtio_transport_get_ops() and also checks if it can be null.
>
> So what about passing the ops as function parameter to avoid to call
> virtio_transport_get_ops() again?
Ack
>
>> +
>> + if (t_ops->can_msgzerocopy) {
>> + int pages_in_iov = iov_iter_npages(iov_iter, MAX_SKB_FRAGS);
>> + int pages_to_send = min(pages_in_iov, MAX_SKB_FRAGS);
>> +
>> + /* +1 is for packet header. */
>> + return t_ops->can_msgzerocopy(pages_to_send + 1);
>> }
>>
>> - if (info->reply)
>> - virtio_vsock_skb_set_reply(skb);
>> + return true;
>> +}
>>
>> - trace_virtio_transport_alloc_pkt(src_cid, src_port,
>> - dst_cid, dst_port,
>> - len,
>> - info->type,
>> - info->op,
>> - info->flags);
>> +static int virtio_transport_init_zcopy_skb(struct vsock_sock *vsk,
>> + struct sk_buff *skb,
>> + struct msghdr *msg,
>> + bool zerocopy)
>> +{
>> + struct ubuf_info *uarg;
>>
>> - if (info->vsk && !skb_set_owner_sk_safe(skb, sk_vsock(info->vsk))) {
>> - WARN_ONCE(1, "failed to allocate skb on vsock socket with sk_refcnt == 0\n");
>> - goto out;
>> + if (msg->msg_ubuf) {
>> + uarg = msg->msg_ubuf;
>> + net_zcopy_get(uarg);
>> + } else {
>> + struct iov_iter *iter = &msg->msg_iter;
>> + struct ubuf_info_msgzc *uarg_zc;
>> +
>> + uarg = msg_zerocopy_realloc(sk_vsock(vsk),
>> + iter->count,
>> + NULL);
>> + if (!uarg)
>> + return -1;
>> +
>> + uarg_zc = uarg_to_msgzc(uarg);
>> + uarg_zc->zerocopy = zerocopy ? 1 : 0;
>> }
>>
>> - return skb;
>> + skb_zcopy_init(skb, uarg);
>>
>> -out:
>> - kfree_skb(skb);
>> - return NULL;
>> + return 0;
>> +}
>> +
>> +static int virtio_transport_fill_skb(struct sk_buff *skb,
>> + struct virtio_vsock_pkt_info *info,
>> + size_t len,
>> + bool zcopy)
>> +{
>> + void *payload;
>> + int err;
>> +
>> + if (zcopy)
>> + return __zerocopy_sg_from_iter(info->msg, NULL, skb,
>> + &info->msg->msg_iter,
>> + len);
>> +
>> + payload = skb_put(skb, len);
>> + err = memcpy_from_msg(payload, info->msg, len);
>> + if (err)
>> + return -1;
>> +
>> + if (msg_data_left(info->msg))
>> + return 0;
>
> We are returning 0 in any case, what is the purpose of this check?
Right, just forget to remove it.
>
>> +
>> + return 0;
>> +}
>> +
>> +static void virtio_transport_init_hdr(struct sk_buff *skb,
>> + struct virtio_vsock_pkt_info *info,
>> + size_t payload_len,
>> + u32 src_cid,
>> + u32 src_port,
>> + u32 dst_cid,
>> + u32 dst_port)
>> +{
>> + struct virtio_vsock_hdr *hdr;
>> +
>> + hdr = virtio_vsock_hdr(skb);
>> + hdr->type = cpu_to_le16(info->type);
>> + hdr->op = cpu_to_le16(info->op);
>> + hdr->src_cid = cpu_to_le64(src_cid);
>> + hdr->dst_cid = cpu_to_le64(dst_cid);
>> + hdr->src_port = cpu_to_le32(src_port);
>> + hdr->dst_port = cpu_to_le32(dst_port);
>> + hdr->flags = cpu_to_le32(info->flags);
>> + hdr->len = cpu_to_le32(payload_len);
>> }
>>
>> static void virtio_transport_copy_nonlinear_skb(const struct sk_buff *skb,
>> @@ -214,6 +251,77 @@ static u16 virtio_transport_get_type(struct sock *sk)
>> return VIRTIO_VSOCK_TYPE_SEQPACKET;
>> }
>>
>> +static struct sk_buff *virtio_transport_alloc_skb(struct vsock_sock *vsk,
>
> Before this patch we used `info->vsk` in virtio_transport_alloc_skb().
> Is it now really necessary to add vsk as a parameter? If so, why?
Agree, I think previous version of virtio_transport_alloc_skb() could be
used.
>
>> + struct virtio_vsock_pkt_info *info,
>> + size_t payload_len,
>> + bool zcopy,
>> + u32 src_cid,
>> + u32 src_port,
>> + u32 dst_cid,
>> + u32 dst_port)
>> +{
>> + struct sk_buff *skb;
>> + size_t skb_len;
>> +
>> + skb_len = VIRTIO_VSOCK_SKB_HEADROOM;
>> +
>> + if (!zcopy)
>> + skb_len += payload_len;
>> +
>> + skb = virtio_vsock_alloc_skb(skb_len, GFP_KERNEL);
>> + if (!skb)
>> + return NULL;
>> +
>> + virtio_transport_init_hdr(skb, info, payload_len, src_cid, src_port,
>> + dst_cid, dst_port);
>> +
>> + /* If 'vsk' != NULL then payload is always present, so we
>> + * will never call '__zerocopy_sg_from_iter()' below without
>> + * setting skb owner in 'skb_set_owner_w()'. The only case
>> + * when 'vsk' == NULL is VIRTIO_VSOCK_OP_RST control message
>> + * without payload.
>> + */
>> + WARN_ON_ONCE(!(vsk && (info->msg && payload_len)) && zcopy);
>> +
>> + /* Set owner here, because '__zerocopy_sg_from_iter()' uses
>> + * owner of skb without check to update 'sk_wmem_alloc'.
>> + */
>> + if (vsk)
>> + skb_set_owner_w(skb, sk_vsock(vsk));
>> +
>> + if (info->msg && payload_len > 0) {
>> + int err;
>> +
>> + err = virtio_transport_fill_skb(skb, info, payload_len, zcopy);
>> + if (err)
>> + goto out;
>> +
>> + if (info->type == VIRTIO_VSOCK_TYPE_SEQPACKET) {
>
> Before this patch, we did these steps only if
> `msg_data_left(info->msg) == 0`, why now we do it in any case?
Ack, it is bug
>
>> + struct virtio_vsock_hdr *hdr = virtio_vsock_hdr(skb);
>> +
>> + hdr->flags |= cpu_to_le32(VIRTIO_VSOCK_SEQ_EOM);
>> +
>> + if (info->msg->msg_flags & MSG_EOR)
>> + hdr->flags |= cpu_to_le32(VIRTIO_VSOCK_SEQ_EOR);
>> + }
>> + }
>> +
>> + if (info->reply)
>> + virtio_vsock_skb_set_reply(skb);
>> +
>> + trace_virtio_transport_alloc_pkt(src_cid, src_port,
>> + dst_cid, dst_port,
>> + payload_len,
>> + info->type,
>> + info->op,
>> + info->flags);
>
> Maybe now we should trace also `zcopy`.
Ack
>
>> +
>> + return skb;
>> +out:
>> + kfree_skb(skb);
>> + return NULL;
>> +}
>> +
>> /* This function can only be used on connecting/connected sockets,
>> * since a socket assigned to a transport is required.
>> *
>> @@ -222,10 +330,12 @@ static u16 virtio_transport_get_type(struct sock *sk)
>> static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
>> struct virtio_vsock_pkt_info *info)
>> {
>> + u32 max_skb_len = VIRTIO_VSOCK_MAX_PKT_BUF_SIZE;
>> u32 src_cid, src_port, dst_cid, dst_port;
>> const struct virtio_transport *t_ops;
>> struct virtio_vsock_sock *vvs;
>> u32 pkt_len = info->pkt_len;
>> + bool can_zcopy = false;
>> u32 rest_len;
>> int ret;
>>
>> @@ -254,15 +364,30 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
>> if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
>> return pkt_len;
>>
>> + if (info->msg) {
>> + /* If zerocopy is not enabled by 'setsockopt()', we behave as
>> + * there is no MSG_ZEROCOPY flag set.
>> + */
>> + if (!sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY))
>> + info->msg->msg_flags &= ~MSG_ZEROCOPY;
>> +
>> + if (info->msg->msg_flags & MSG_ZEROCOPY)
>> + can_zcopy = virtio_transport_can_zcopy(info, pkt_len);
>> +
>> + if (can_zcopy)
>> + max_skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE,
>> + (MAX_SKB_FRAGS * PAGE_SIZE));
>> + }
>> +
>> rest_len = pkt_len;
>>
>> do {
>> struct sk_buff *skb;
>> size_t skb_len;
>>
>> - skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE, rest_len);
>> + skb_len = min(max_skb_len, rest_len);
>>
>> - skb = virtio_transport_alloc_skb(info, skb_len,
>> + skb = virtio_transport_alloc_skb(vsk, info, skb_len, can_zcopy,
>> src_cid, src_port,
>> dst_cid, dst_port);
>> if (!skb) {
>> @@ -270,6 +395,21 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
>> break;
>> }
>>
>> + /* We process buffer part by part, allocating skb on
>> + * each iteration. If this is last skb for this buffer
>> + * and MSG_ZEROCOPY mode is in use - we must allocate
>> + * completion for the current syscall.
>> + */
>> + if (info->msg && info->msg->msg_flags & MSG_ZEROCOPY &&
>> + skb_len == rest_len && info->op == VIRTIO_VSOCK_OP_RW) {
>> + if (virtio_transport_init_zcopy_skb(vsk, skb,
>> + info->msg,
>> + can_zcopy)) {
>> + ret = -ENOMEM;
>> + break;
>> + }
>> + }
>> +
>> virtio_transport_inc_tx_pkt(vvs, skb);
>>
>> ret = t_ops->send_pkt(skb);
>> @@ -985,7 +1125,7 @@ static int virtio_transport_reset_no_sock(const struct virtio_transport *t,
>> if (!t)
>> return -ENOTCONN;
>>
>> - reply = virtio_transport_alloc_skb(&info, 0,
>> + reply = virtio_transport_alloc_skb(NULL, &info, 0, false,
>> le64_to_cpu(hdr->dst_cid),
>> le32_to_cpu(hdr->dst_port),
>> le64_to_cpu(hdr->src_cid),
>> --
>> 2.25.1
>>
>