From: kernel test robot <lkp@intel.com>
To: Mina Almasry <almasrymina@google.com>,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
kvm@vger.kernel.org, virtualization@lists.linux.dev
Cc: oe-kbuild-all@lists.linux.dev,
"Mina Almasry" <almasrymina@google.com>,
"Eric Dumazet" <edumazet@google.com>,
"Jakub Kicinski" <kuba@kernel.org>,
"Paolo Abeni" <pabeni@redhat.com>,
"Stefan Hajnoczi" <stefanha@redhat.com>,
"Stefano Garzarella" <sgarzare@redhat.com>,
"David Howells" <dhowells@redhat.com>,
"Jason Gunthorpe" <jgg@nvidia.com>,
"Christian König" <christian.koenig@amd.com>,
"Shakeel Butt" <shakeelb@google.com>,
"Yunsheng Lin" <linyunsheng@huawei.com>,
"Willem de Bruijn" <willemdebruijn.kernel@gmail.com>
Subject: Re: [PATCH net-next v3 3/3] net: add netmem_ref to skb_frag_t
Date: Sat, 23 Dec 2023 07:39:30 +0800 [thread overview]
Message-ID: <202312230739.g0Tfssdt-lkp@intel.com> (raw)
In-Reply-To: <20231220214505.2303297-4-almasrymina@google.com>
Hi Mina,
kernel test robot noticed the following build errors:
[auto build test ERROR on net-next/main]
url: https://github.com/intel-lab-lkp/linux/commits/Mina-Almasry/vsock-virtio-use-skb_frag_-helpers/20231222-164637
base: net-next/main
patch link: https://lore.kernel.org/r/20231220214505.2303297-4-almasrymina%40google.com
patch subject: [PATCH net-next v3 3/3] net: add netmem_ref to skb_frag_t
config: i386-randconfig-141-20231222 (https://download.01.org/0day-ci/archive/20231223/202312230739.g0Tfssdt-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231223/202312230739.g0Tfssdt-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202312230739.g0Tfssdt-lkp@intel.com/
All errors (new ones prefixed by >>):
net/kcm/kcmsock.c: In function 'kcm_write_msgs':
>> net/kcm/kcmsock.c:637:59: error: 'skb_frag_t' {aka 'struct skb_frag'} has no member named 'bv_len'
637 | msize += skb_shinfo(skb)->frags[i].bv_len;
| ^
vim +637 net/kcm/kcmsock.c
cd6e111bf5be5c Tom Herbert 2016-03-07 578
ab7ac4eb9832e3 Tom Herbert 2016-03-07 579 /* Write any messages ready on the kcm socket. Called with kcm sock lock
ab7ac4eb9832e3 Tom Herbert 2016-03-07 580 * held. Return bytes actually sent or error.
ab7ac4eb9832e3 Tom Herbert 2016-03-07 581 */
ab7ac4eb9832e3 Tom Herbert 2016-03-07 582 static int kcm_write_msgs(struct kcm_sock *kcm)
ab7ac4eb9832e3 Tom Herbert 2016-03-07 583 {
c31a25e1db486f David Howells 2023-06-09 584 unsigned int total_sent = 0;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 585 struct sock *sk = &kcm->sk;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 586 struct kcm_psock *psock;
c31a25e1db486f David Howells 2023-06-09 587 struct sk_buff *head;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 588 int ret = 0;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 589
ab7ac4eb9832e3 Tom Herbert 2016-03-07 590 kcm->tx_wait_more = false;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 591 psock = kcm->tx_psock;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 592 if (unlikely(psock && psock->tx_stopped)) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 593 /* A reserved psock was aborted asynchronously. Unreserve
ab7ac4eb9832e3 Tom Herbert 2016-03-07 594 * it and we'll retry the message.
ab7ac4eb9832e3 Tom Herbert 2016-03-07 595 */
ab7ac4eb9832e3 Tom Herbert 2016-03-07 596 unreserve_psock(kcm);
cd6e111bf5be5c Tom Herbert 2016-03-07 597 kcm_report_tx_retry(kcm);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 598 if (skb_queue_empty(&sk->sk_write_queue))
ab7ac4eb9832e3 Tom Herbert 2016-03-07 599 return 0;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 600
c31a25e1db486f David Howells 2023-06-09 601 kcm_tx_msg(skb_peek(&sk->sk_write_queue))->started_tx = false;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 602 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 603
c31a25e1db486f David Howells 2023-06-09 604 retry:
c31a25e1db486f David Howells 2023-06-09 605 while ((head = skb_peek(&sk->sk_write_queue))) {
c31a25e1db486f David Howells 2023-06-09 606 struct msghdr msg = {
c31a25e1db486f David Howells 2023-06-09 607 .msg_flags = MSG_DONTWAIT | MSG_SPLICE_PAGES,
c31a25e1db486f David Howells 2023-06-09 608 };
c31a25e1db486f David Howells 2023-06-09 609 struct kcm_tx_msg *txm = kcm_tx_msg(head);
c31a25e1db486f David Howells 2023-06-09 610 struct sk_buff *skb;
c31a25e1db486f David Howells 2023-06-09 611 unsigned int msize;
c31a25e1db486f David Howells 2023-06-09 612 int i;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 613
c31a25e1db486f David Howells 2023-06-09 614 if (!txm->started_tx) {
c31a25e1db486f David Howells 2023-06-09 615 psock = reserve_psock(kcm);
c31a25e1db486f David Howells 2023-06-09 616 if (!psock)
c31a25e1db486f David Howells 2023-06-09 617 goto out;
c31a25e1db486f David Howells 2023-06-09 618 skb = head;
c31a25e1db486f David Howells 2023-06-09 619 txm->frag_offset = 0;
c31a25e1db486f David Howells 2023-06-09 620 txm->sent = 0;
c31a25e1db486f David Howells 2023-06-09 621 txm->started_tx = true;
c31a25e1db486f David Howells 2023-06-09 622 } else {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 623 if (WARN_ON(!psock)) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 624 ret = -EINVAL;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 625 goto out;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 626 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 627 skb = txm->frag_skb;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 628 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 629
ab7ac4eb9832e3 Tom Herbert 2016-03-07 630 if (WARN_ON(!skb_shinfo(skb)->nr_frags)) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 631 ret = -EINVAL;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 632 goto out;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 633 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 634
c31a25e1db486f David Howells 2023-06-09 635 msize = 0;
c31a25e1db486f David Howells 2023-06-09 636 for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
c31a25e1db486f David Howells 2023-06-09 @637 msize += skb_shinfo(skb)->frags[i].bv_len;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 638
b2e5852793b6eb Mina Almasry 2023-12-20 639 /* The cast to struct bio_vec* here assumes the frags are
b2e5852793b6eb Mina Almasry 2023-12-20 640 * struct page based. WARN if there is no page in this skb.
b2e5852793b6eb Mina Almasry 2023-12-20 641 */
b2e5852793b6eb Mina Almasry 2023-12-20 642 DEBUG_NET_WARN_ON_ONCE(
b2e5852793b6eb Mina Almasry 2023-12-20 643 !skb_frag_page(&skb_shinfo(skb)->frags[0]));
b2e5852793b6eb Mina Almasry 2023-12-20 644
c31a25e1db486f David Howells 2023-06-09 645 iov_iter_bvec(&msg.msg_iter, ITER_SOURCE,
b2e5852793b6eb Mina Almasry 2023-12-20 646 (const struct bio_vec *)skb_shinfo(skb)->frags,
b2e5852793b6eb Mina Almasry 2023-12-20 647 skb_shinfo(skb)->nr_frags, msize);
c31a25e1db486f David Howells 2023-06-09 648 iov_iter_advance(&msg.msg_iter, txm->frag_offset);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 649
c31a25e1db486f David Howells 2023-06-09 650 do {
264ba53fac79b0 David Howells 2023-06-09 651 ret = sock_sendmsg(psock->sk->sk_socket, &msg);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 652 if (ret <= 0) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 653 if (ret == -EAGAIN) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 654 /* Save state to try again when there's
ab7ac4eb9832e3 Tom Herbert 2016-03-07 655 * write space on the socket
ab7ac4eb9832e3 Tom Herbert 2016-03-07 656 */
ab7ac4eb9832e3 Tom Herbert 2016-03-07 657 txm->frag_skb = skb;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 658 ret = 0;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 659 goto out;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 660 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 661
ab7ac4eb9832e3 Tom Herbert 2016-03-07 662 /* Hard failure in sending message, abort this
ab7ac4eb9832e3 Tom Herbert 2016-03-07 663 * psock since it has lost framing
71a2fae50895b3 Bhaskar Chowdhury 2021-03-27 664 * synchronization and retry sending the
ab7ac4eb9832e3 Tom Herbert 2016-03-07 665 * message from the beginning.
ab7ac4eb9832e3 Tom Herbert 2016-03-07 666 */
ab7ac4eb9832e3 Tom Herbert 2016-03-07 667 kcm_abort_tx_psock(psock, ret ? -ret : EPIPE,
ab7ac4eb9832e3 Tom Herbert 2016-03-07 668 true);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 669 unreserve_psock(kcm);
9f8d0dc0ec4a4b David Howells 2023-06-15 670 psock = NULL;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 671
c31a25e1db486f David Howells 2023-06-09 672 txm->started_tx = false;
cd6e111bf5be5c Tom Herbert 2016-03-07 673 kcm_report_tx_retry(kcm);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 674 ret = 0;
c31a25e1db486f David Howells 2023-06-09 675 goto retry;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 676 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 677
c31a25e1db486f David Howells 2023-06-09 678 txm->sent += ret;
c31a25e1db486f David Howells 2023-06-09 679 txm->frag_offset += ret;
cd6e111bf5be5c Tom Herbert 2016-03-07 680 KCM_STATS_ADD(psock->stats.tx_bytes, ret);
c31a25e1db486f David Howells 2023-06-09 681 } while (msg.msg_iter.count > 0);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 682
ab7ac4eb9832e3 Tom Herbert 2016-03-07 683 if (skb == head) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 684 if (skb_has_frag_list(skb)) {
c31a25e1db486f David Howells 2023-06-09 685 txm->frag_skb = skb_shinfo(skb)->frag_list;
c31a25e1db486f David Howells 2023-06-09 686 txm->frag_offset = 0;
c31a25e1db486f David Howells 2023-06-09 687 continue;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 688 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 689 } else if (skb->next) {
c31a25e1db486f David Howells 2023-06-09 690 txm->frag_skb = skb->next;
c31a25e1db486f David Howells 2023-06-09 691 txm->frag_offset = 0;
c31a25e1db486f David Howells 2023-06-09 692 continue;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 693 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 694
ab7ac4eb9832e3 Tom Herbert 2016-03-07 695 /* Successfully sent the whole packet, account for it. */
c31a25e1db486f David Howells 2023-06-09 696 sk->sk_wmem_queued -= txm->sent;
c31a25e1db486f David Howells 2023-06-09 697 total_sent += txm->sent;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 698 skb_dequeue(&sk->sk_write_queue);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 699 kfree_skb(head);
cd6e111bf5be5c Tom Herbert 2016-03-07 700 KCM_STATS_INCR(psock->stats.tx_msgs);
c31a25e1db486f David Howells 2023-06-09 701 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 702 out:
ab7ac4eb9832e3 Tom Herbert 2016-03-07 703 if (!head) {
ab7ac4eb9832e3 Tom Herbert 2016-03-07 704 /* Done with all queued messages. */
ab7ac4eb9832e3 Tom Herbert 2016-03-07 705 WARN_ON(!skb_queue_empty(&sk->sk_write_queue));
9f8d0dc0ec4a4b David Howells 2023-06-15 706 if (psock)
ab7ac4eb9832e3 Tom Herbert 2016-03-07 707 unreserve_psock(kcm);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 708 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 709
ab7ac4eb9832e3 Tom Herbert 2016-03-07 710 /* Check if write space is available */
ab7ac4eb9832e3 Tom Herbert 2016-03-07 711 sk->sk_write_space(sk);
ab7ac4eb9832e3 Tom Herbert 2016-03-07 712
ab7ac4eb9832e3 Tom Herbert 2016-03-07 713 return total_sent ? : ret;
ab7ac4eb9832e3 Tom Herbert 2016-03-07 714 }
ab7ac4eb9832e3 Tom Herbert 2016-03-07 715
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
next prev parent reply other threads:[~2023-12-22 23:41 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-20 21:44 [PATCH net-next v3 0/3] Abstract page from net stack Mina Almasry
2023-12-20 21:45 ` [PATCH net-next v3 1/3] vsock/virtio: use skb_frag_*() helpers Mina Almasry
2023-12-21 17:17 ` Willem de Bruijn
2023-12-21 21:39 ` Shakeel Butt
2024-01-02 10:00 ` Stefano Garzarella
2023-12-20 21:45 ` [PATCH net-next v3 2/3] net: introduce abstraction for network memory Mina Almasry
2023-12-21 23:23 ` Shakeel Butt
2023-12-21 23:44 ` Mina Almasry
2024-01-04 21:44 ` Jakub Kicinski
2024-01-04 22:15 ` Mina Almasry
2024-01-10 17:50 ` Shakeel Butt
2024-01-11 1:35 ` Jakub Kicinski
2023-12-20 21:45 ` [PATCH net-next v3 3/3] net: add netmem_ref to skb_frag_t Mina Almasry
2023-12-21 17:16 ` Simon Horman
2023-12-21 17:18 ` Willem de Bruijn
2023-12-21 23:27 ` Shakeel Butt
2023-12-22 20:10 ` kernel test robot
2023-12-22 23:39 ` kernel test robot [this message]
2023-12-23 11:16 ` kernel test robot
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=202312230739.g0Tfssdt-lkp@intel.com \
--to=lkp@intel.com \
--cc=almasrymina@google.com \
--cc=christian.koenig@amd.com \
--cc=dhowells@redhat.com \
--cc=edumazet@google.com \
--cc=jgg@nvidia.com \
--cc=kuba@kernel.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linyunsheng@huawei.com \
--cc=netdev@vger.kernel.org \
--cc=oe-kbuild-all@lists.linux.dev \
--cc=pabeni@redhat.com \
--cc=sgarzare@redhat.com \
--cc=shakeelb@google.com \
--cc=stefanha@redhat.com \
--cc=virtualization@lists.linux.dev \
--cc=willemdebruijn.kernel@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).