From: Yuanhan Liu <yuanhan.liu@linux.intel.com>
To: dev@dpdk.org
Cc: huawei.xie@intel.com, Yuanhan Liu <yuanhan.liu@linux.intel.com>,
"Michael S. Tsirkin" <mst@redhat.com>
Subject: [dpdk-dev] [PATCH 4/6] vhost: workaround stale vring base
Date: Fri, 6 May 2016 23:40:22 -0700 [thread overview]
Message-ID: <1462603224-29510-5-git-send-email-yuanhan.liu@linux.intel.com> (raw)
In-Reply-To: <1462603224-29510-1-git-send-email-yuanhan.liu@linux.intel.com>
When DPDK app crashes (or quits, or gets killed), and when QEMU supports
reconnecting (patches have been sent, not merged yet), a restart of DPDK
app would get stale vring base from QEMU. That would break the kernel
virtio net completely, making it non-work any more, unless a driver reset
is done.
So, instead of getting the stale vring base from QEMU, Huawei suggested
we could get a proper one from used->idx.
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Suggested-by: "Xie, Huawei" <huawei.xie@intel.com>
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
---
Note that a right way to handle reconnect from abnormal quit would be
let the guest driver to initiate a reset when reconnect is detected
from QEMU. As a reset from the virtio net driver would resets all virtio
related meta datas, and trigger the vhost-user re-initiation again,
therefore, it would make the reconnect work as expected.
What's unforunate is that driver reset on reconnect, as the term "driver"
implies, need driver hackings, which could be a linux kernel virtio net
driver, DPDK pmd driver, or even, the windows driver. Apparently, it will
not work so far, or even for a long while, until we are adapted to the
new kernel.
The fix (or more precisely, the workaround) from this patch would make
reconnect work in most case, including the following two hypothesis that
might corrupt virtio memory:
- vring_used_elem might be in stale state when we are in the middle of
updating used vring. Say, we might have updated the "id" field, but
leaving "len" untouched.
- vring desc buff might also be in stale state, when we are copying
data into it.
The reason it still works is that we haven't updated used->idx yet,
which means guest driver will not start processing those buggy used
entries. Therefore, no issues.
However, Michael claims some concerns: he made a good point: a crash
is happening means some memory is corrupted, and it could be the virtio
memory being corrupted. In such case, nothing will work without the
reset.
But I would say, that an app in product use would rare crash, and even
if it crashes, the chance that virtio memory being corrupted would be
relatively smaller then. Besides that, it would work, say when the app
is killed by ctrl-c or kill command. So, it works in the most cases.
But still, I would say it's just a workaround so far.
On the other hand, without this patch, it would be 100% not working
from abnormal quit. But with this patch, it works in most cases, just
don't work in a rare case that when virtio memory is corrupted. Therefore,
I'd say, it is still good to have, until we have a perfect solution.
---
lib/librte_vhost/virtio-net.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/lib/librte_vhost/virtio-net.c b/lib/librte_vhost/virtio-net.c
index c88aaa3..df103aa 100644
--- a/lib/librte_vhost/virtio-net.c
+++ b/lib/librte_vhost/virtio-net.c
@@ -560,6 +560,14 @@ vhost_set_vring_addr(int vid, struct vhost_vring_addr *addr)
return -1;
}
+ if (vq->last_used_idx != vq->used->idx) {
+ RTE_LOG(WARNING, VHOST_CONFIG,
+ "last_used_idx (%u) and vq->used->idx (%u) mismatch\n",
+ vq->last_used_idx, vq->used->idx);
+ vq->last_used_idx = vq->used->idx;
+ vq->last_used_idx_res = vq->used->idx;
+ }
+
vq->log_guest_addr = addr->log_guest_addr;
LOG_DEBUG(VHOST_CONFIG, "(%d) mapped address desc: %p\n",
--
1.9.0
next prev parent reply other threads:[~2016-05-07 6:36 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-07 6:40 [dpdk-dev] [PATCH 0/6] vhost: add vhost-user client mode and reconnect ability Yuanhan Liu
2016-05-07 6:40 ` [dpdk-dev] [PATCH 1/6] vhost: rename structs for enabling client mode Yuanhan Liu
2016-05-07 6:40 ` [dpdk-dev] [PATCH 2/6] vhost: add vhost-user " Yuanhan Liu
2016-05-09 10:33 ` Victor Kaplansky
2016-05-09 20:33 ` Yuanhan Liu
2016-05-09 20:30 ` Michael S. Tsirkin
2016-05-07 6:40 ` [dpdk-dev] [PATCH 3/6] vhost: add reconnect ability Yuanhan Liu
2016-05-09 16:47 ` Xie, Huawei
2016-05-09 18:12 ` Yuanhan Liu
2016-05-10 7:24 ` Xie, Huawei
2016-05-10 7:54 ` Michael S. Tsirkin
2016-05-10 8:07 ` Xie, Huawei
2016-05-10 8:42 ` Michael S. Tsirkin
2016-05-10 9:00 ` Xie, Huawei
2016-05-10 9:17 ` Michael S. Tsirkin
2016-05-10 17:17 ` Loftus, Ciara
2016-05-11 21:46 ` Michael S. Tsirkin
2016-05-07 6:40 ` Yuanhan Liu [this message]
2016-05-09 10:45 ` [dpdk-dev] [PATCH 4/6] vhost: workaround stale vring base Victor Kaplansky
2016-05-09 13:39 ` Xie, Huawei
2016-05-09 18:23 ` Yuanhan Liu
2016-05-09 12:19 ` Michael S. Tsirkin
2016-05-09 16:25 ` Xie, Huawei
2016-05-09 18:22 ` Yuanhan Liu
2016-06-13 20:47 ` Michael S. Tsirkin
2016-05-10 8:21 ` Xie, Huawei
2016-05-07 6:40 ` [dpdk-dev] [PATCH 5/6] examples/vhost: add client and reconnect option Yuanhan Liu
2016-05-09 10:47 ` Victor Kaplansky
2016-05-07 6:40 ` [dpdk-dev] [PATCH 6/6] vhost: add pmd " Yuanhan Liu
2016-05-09 10:54 ` Victor Kaplansky
2016-05-09 18:26 ` Yuanhan Liu
2016-05-10 3:23 ` [dpdk-dev] [PATCH 0/6] vhost: add vhost-user client mode and reconnect ability Xu, Qian Q
2016-05-10 17:41 ` Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 " Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 1/6] vhost: rename structs for enabling client mode Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 2/6] vhost: add vhost-user " Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 3/6] vhost: add reconnect ability Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 4/6] vhost: workaround stale vring base Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 5/6] examples/vhost: add client and reconnect option Yuanhan Liu
2016-05-13 6:16 ` [dpdk-dev] [PATCH v2 6/6] vhost: add pmd " Yuanhan Liu
2016-05-25 17:45 ` Rich Lane
2016-05-26 8:01 ` Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 0/6] vhost: add vhost-user client mode and reconnect ability Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 1/6] vhost: rename structs for enabling client mode Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 2/6] vhost: add vhost-user " Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 3/6] vhost: add reconnect ability Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 4/6] vhost: workaround stale vring base Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 5/6] examples/vhost: add client option Yuanhan Liu
2016-06-07 4:05 ` [dpdk-dev] [PATCH v3 6/6] vhost: add pmd " Yuanhan Liu
2016-06-14 12:00 ` [dpdk-dev] [PATCH v3 0/6] vhost: add vhost-user client mode and reconnect ability 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=1462603224-29510-5-git-send-email-yuanhan.liu@linux.intel.com \
--to=yuanhan.liu@linux.intel.com \
--cc=dev@dpdk.org \
--cc=huawei.xie@intel.com \
--cc=mst@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).