From: Yuanhan Liu <yuanhan.liu@linux.intel.com>
To: "Tan, Jianfeng" <jianfeng.tan@intel.com>
Cc: dev@dpdk.org, Victor Kaplansky <vkaplans@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: [dpdk-dev] [PATCH v3 6/8] vhost: handle VHOST_USER_SEND_RARP request
Date: Fri, 19 Feb 2016 15:03:26 +0800 [thread overview]
Message-ID: <20160219070326.GR21426@yliu-dev.sh.intel.com> (raw)
In-Reply-To: <56C6B218.6080501@intel.com>
On Fri, Feb 19, 2016 at 02:11:36PM +0800, Tan, Jianfeng wrote:
> Hi Yuanhan,
>
> On 1/29/2016 12:58 PM, Yuanhan Liu wrote:
> >While in former patch we enabled GUEST_ANNOUNCE feature, so that the
> >guest OS will broadcast a GARP message after migration to notify the
> >switch about the new location of migrated VM, the thing is that
> >GUEST_ANNOUNCE is enabled since kernel v3.5 only. For older kernel,
> >VHOST_USER_SEND_RARP request comes to rescue.
> >
> >The payload of this new request is the mac address of the migrated VM,
> >with that, we could construct a RARP message, and then broadcast it
> >to host interfaces.
> >
> >That's how this patch works:
> >
> >- list all interfaces, with the help of SIOCGIFCONF ioctl command
> >
> >- construct an RARP message and broadcast it
> >
> >Cc: Thibaut Collet <thibaut.collet@6wind.com>
> >Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
> >---
> ...
> >+
> >+/*
> >+ * Broadcast a RARP message to all interfaces, to update
> >+ * switch's mac table
> >+ */
> >+int
> >+user_send_rarp(struct VhostUserMsg *msg)
> >+{
> >+ uint8_t *mac = (uint8_t *)&msg->payload.u64;
> >+ uint8_t rarp[RARP_BUF_SIZE];
> >+ struct ifconf ifc = {0, };
> >+ struct ifreq *ifr;
> >+ int nr = 16;
> >+ int fd;
> >+ uint32_t i;
> >+
> >+ RTE_LOG(DEBUG, VHOST_CONFIG,
> >+ ":: mac: %02x:%02x:%02x:%02x:%02x:%02x\n",
> >+ mac[0], mac[1], mac[2], mac[3], mac[4], mac[5]);
> >+
> >+ make_rarp_packet(rarp, mac);
> >+
> >+ /*
> >+ * Get all interfaces
> >+ */
> >+ fd = socket(AF_INET, SOCK_DGRAM, 0);
> >+ if (fd < 0) {
> >+ perror("failed to create AF_INET socket");
> >+ return -1;
> >+ }
> >+
> >+again:
> >+ ifc.ifc_len = sizeof(*ifr) * nr;
> >+ ifc.ifc_buf = realloc(ifc.ifc_buf, ifc.ifc_len);
> >+
> >+ if (ioctl(fd, SIOCGIFCONF, &ifc) < 0) {
> >+ perror("failed at SIOCGIFCONF");
> >+ close(fd);
> >+ return -1;
> >+ }
> >+
> >+ if (ifc.ifc_len == (int)sizeof(struct ifreq) * nr) {
> >+ /*
> >+ * current ifc_buf is not big enough to hold
> >+ * all interfaces; double it and try again.
> >+ */
> >+ nr *= 2;
> >+ goto again;
> >+ }
> >+
> >+ ifr = (struct ifreq *)ifc.ifc_buf;
> >+ for (i = 0; i < ifc.ifc_len / sizeof(struct ifreq); i++)
> >+ send_rarp(ifr[i].ifr_name, rarp);
> >+
> >+ close(fd);
> >+
> >+ return 0;
> >+}
>
> From how you implement user_send_rarp(), if I understand it correctly, it
> broadcasts this ARP packets to all host interfaces, which I don't think it's
> appropriate. This ARP packets should be sent to it's own L2 networking. You
> should not make the hypothesis that all interfaces maintained in the kernel
> are in the same L2 networking. Even worse, this could bring problems when
> used in overlay networking, in which two VM in two different overlay
> networking, can have same MAC address.
>
> What I suggest here is to move user_send_rarp() to rte_vhost_dequeue_burst()
> using a flag to control, so that this arp packet can be broadcasted in its
> own L2 network.
I have thought of that, too. It was given up because SEND_RARP request was
handled in different thread from rte_vhost_dequeue_burst(), leading to the
fact that the RARP packet will not be broadcasted immediately after migration
is done: it will be broadcasted only when rte_vhost_dequeue_burst() is invoked.
I was thinking the delay might be a problem. While thinking it twice, it
doesn't look like one then. As GUEST_ANNOUNCE is also broadcasted by
rte_vhost_dequeue_burst(); it's enqueued by guest kernel though. And
judging that we are polling mode driver, it won't be an issue then.
So, thanks. I will give it a quick try; it should work.
--yliu
next prev parent reply other threads:[~2016-02-19 7:02 UTC|newest]
Thread overview: 98+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-02 3:43 [dpdk-dev] [PATCH 0/4 for 2.3] vhost-user live migration support Yuanhan Liu
2015-12-02 3:43 ` [dpdk-dev] [PATCH 1/4] vhost: handle VHOST_USER_SET_LOG_BASE request Yuanhan Liu
2015-12-02 13:53 ` Panu Matilainen
2015-12-02 14:31 ` Yuanhan Liu
2015-12-02 14:48 ` Panu Matilainen
2015-12-02 15:09 ` Yuanhan Liu
2015-12-02 16:58 ` Panu Matilainen
2015-12-02 17:24 ` Michael S. Tsirkin
2015-12-02 16:38 ` Thomas Monjalon
2015-12-03 1:49 ` Yuanhan Liu
2015-12-06 23:07 ` Thomas Monjalon
2015-12-07 2:00 ` Yuanhan Liu
2015-12-07 2:03 ` Thomas Monjalon
2015-12-07 2:18 ` Yuanhan Liu
2015-12-07 2:49 ` Thomas Monjalon
2015-12-07 6:29 ` Panu Matilainen
2015-12-07 11:28 ` Thomas Monjalon
2015-12-07 11:41 ` Panu Matilainen
2015-12-07 13:55 ` Thomas Monjalon
2015-12-07 16:48 ` Panu Matilainen
2015-12-07 17:47 ` Thomas Monjalon
2015-12-08 5:57 ` Xie, Huawei
2015-12-08 7:25 ` Yuanhan Liu
2015-12-02 3:43 ` [dpdk-dev] [PATCH 2/4] vhost: introduce vhost_log_write Yuanhan Liu
2015-12-02 13:53 ` Victor Kaplansky
2015-12-02 14:39 ` Yuanhan Liu
2015-12-09 3:33 ` Xie, Huawei
2015-12-09 3:42 ` Yuanhan Liu
2015-12-09 5:44 ` Xie, Huawei
2015-12-09 8:41 ` Yuanhan Liu
2015-12-02 3:43 ` [dpdk-dev] [PATCH 3/4] vhost: log vring changes Yuanhan Liu
2015-12-02 14:07 ` Victor Kaplansky
2015-12-02 14:38 ` Yuanhan Liu
2015-12-02 15:58 ` Victor Kaplansky
2015-12-02 16:26 ` Michael S. Tsirkin
2015-12-03 2:31 ` Yuanhan Liu
2015-12-09 2:45 ` Xie, Huawei
2015-12-02 3:43 ` [dpdk-dev] [PATCH 4/4] vhost: enable log_shmfd protocol feature Yuanhan Liu
2015-12-02 14:10 ` [dpdk-dev] [PATCH 0/4 for 2.3] vhost-user live migration support Victor Kaplansky
2015-12-02 14:33 ` Yuanhan Liu
2015-12-09 3:41 ` Xie, Huawei
2015-12-17 3:11 ` [dpdk-dev] [PATCH v2 0/6] " Yuanhan Liu
2015-12-17 3:11 ` [dpdk-dev] [PATCH v2 1/6] vhost: handle VHOST_USER_SET_LOG_BASE request Yuanhan Liu
2015-12-21 15:32 ` Xie, Huawei
2015-12-22 2:25 ` Yuanhan Liu
2015-12-22 2:41 ` Xie, Huawei
2015-12-22 2:55 ` Yuanhan Liu
2015-12-17 3:11 ` [dpdk-dev] [PATCH v2 2/6] vhost: introduce vhost_log_write Yuanhan Liu
2015-12-21 15:06 ` Xie, Huawei
2015-12-22 2:40 ` Yuanhan Liu
2015-12-22 2:45 ` Xie, Huawei
2015-12-22 3:04 ` Yuanhan Liu
2015-12-22 7:02 ` Xie, Huawei
2015-12-22 5:11 ` Peter Xu
2015-12-22 6:09 ` Yuanhan Liu
2015-12-17 3:11 ` [dpdk-dev] [PATCH v2 3/6] vhost: log used vring changes Yuanhan Liu
2015-12-22 6:55 ` Peter Xu
2015-12-22 7:07 ` Xie, Huawei
2015-12-22 7:59 ` Peter Xu
2015-12-22 7:13 ` Yuanhan Liu
2015-12-22 8:01 ` Peter Xu
2015-12-17 3:11 ` [dpdk-dev] [PATCH v2 4/6] vhost: log vring desc buffer changes Yuanhan Liu
2015-12-17 3:12 ` [dpdk-dev] [PATCH v2 5/6] vhost: claim that we support GUEST_ANNOUNCE feature Yuanhan Liu
2015-12-22 8:11 ` Peter Xu
2015-12-22 8:21 ` Yuanhan Liu
2015-12-22 8:24 ` Pavel Fedin
2015-12-17 3:12 ` [dpdk-dev] [PATCH v2 6/6] vhost: enable log_shmfd protocol feature Yuanhan Liu
2015-12-17 12:08 ` [dpdk-dev] [PATCH v2 0/6] vhost-user live migration support Iremonger, Bernard
2015-12-17 12:45 ` Yuanhan Liu
2015-12-21 8:17 ` Pavel Fedin
2016-01-29 4:57 ` [dpdk-dev] [PATCH v3 0/8] " Yuanhan Liu
2016-01-29 4:57 ` [dpdk-dev] [PATCH v3 1/8] vhost: handle VHOST_USER_SET_LOG_BASE request Yuanhan Liu
2016-01-29 4:57 ` [dpdk-dev] [PATCH v3 2/8] vhost: introduce vhost_log_write Yuanhan Liu
2016-02-19 14:26 ` Thomas Monjalon
2016-02-22 6:59 ` Yuanhan Liu
2016-01-29 4:57 ` [dpdk-dev] [PATCH v3 3/8] vhost: log used vring changes Yuanhan Liu
2016-01-29 4:57 ` [dpdk-dev] [PATCH v3 4/8] vhost: log vring desc buffer changes Yuanhan Liu
2016-01-29 4:58 ` [dpdk-dev] [PATCH v3 5/8] vhost: claim that we support GUEST_ANNOUNCE feature Yuanhan Liu
2016-03-11 12:39 ` Olivier MATZ
2016-03-11 13:16 ` Thomas Monjalon
2016-03-11 13:22 ` Olivier MATZ
2016-01-29 4:58 ` [dpdk-dev] [PATCH v3 6/8] vhost: handle VHOST_USER_SEND_RARP request Yuanhan Liu
2016-02-19 6:11 ` Tan, Jianfeng
2016-02-19 7:03 ` Yuanhan Liu [this message]
2016-02-19 8:55 ` Yuanhan Liu
2016-02-22 14:36 ` [dpdk-dev] [PATCH] vhost: broadcast RARP pkt by injecting it to receiving mbuf array Yuanhan Liu
2016-02-24 8:15 ` Qiu, Michael
2016-02-24 8:28 ` Yuanhan Liu
2016-02-25 7:55 ` Qiu, Michael
2016-02-29 15:56 ` Thomas Monjalon
2016-01-29 4:58 ` [dpdk-dev] [PATCH v3 7/8] vhost: enable log_shmfd protocol feature Yuanhan Liu
2016-01-29 4:58 ` [dpdk-dev] [PATCH v3 8/8] vhost: remove duplicate header include Yuanhan Liu
2016-02-01 15:54 ` [dpdk-dev] [PATCH v3 0/8] vhost-user live migration support Iremonger, Bernard
2016-02-02 1:53 ` Yuanhan Liu
2016-02-19 15:01 ` Thomas Monjalon
2016-02-22 7:08 ` Yuanhan Liu
2016-02-22 9:56 ` Thomas Monjalon
2016-02-22 14:24 ` Yuanhan Liu
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=20160219070326.GR21426@yliu-dev.sh.intel.com \
--to=yuanhan.liu@linux.intel.com \
--cc=dev@dpdk.org \
--cc=jianfeng.tan@intel.com \
--cc=mst@redhat.com \
--cc=vkaplans@redhat.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).