patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Kevin Traynor <ktraynor@redhat.com>
To: Maxime Coquelin <maxime.coquelin@redhat.com>
Cc: Tiwei Bie <tiwei.bie@intel.com>, dpdk stable <stable@dpdk.org>
Subject: [dpdk-stable] patch 'vhost: prevent disabled rings to be processed with zero-copy' has been queued to LTS release 18.11.2
Date: Tue, 16 Apr 2019 15:36:47 +0100	[thread overview]
Message-ID: <20190416143719.21601-29-ktraynor@redhat.com> (raw)
In-Reply-To: <20190416143719.21601-1-ktraynor@redhat.com>

Hi,

FYI, your patch has been queued to LTS release 18.11.2

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 04/24/19. So please
shout if anyone has objections.

Also note that after the patch there's a diff of the upstream commit vs the
patch applied to the branch. This will indicate if there was any rebasing
needed to apply to the stable branch. If there were code changes for rebasing
(ie: not only metadata diffs), please double check that the rebase was
correctly done.

Thanks.

Kevin Traynor

---
From 7dadea51de8e770dad51940d6a5efc347fd9b30c Mon Sep 17 00:00:00 2001
From: Maxime Coquelin <maxime.coquelin@redhat.com>
Date: Thu, 28 Feb 2019 18:57:04 +0100
Subject: [PATCH] vhost: prevent disabled rings to be processed with zero-copy

[ upstream commit 11d5253a3e93a38be14da65328dd19960fd8cb14 ]

The vhost-user spec says that once the vring is disabled, the
client has to stop processing it. But it can happen when
dequeue zero-copy is enabled if outstanding descriptors buffers
are still being processed by an external NIC or another guest.

The fix consists in draining the zmbufs list to ensure no more
descriptors buffers are in the wild.

Note that this fix is only working in the case REPLY_ACK
protocol feature is enabled, which is not the case by default
for now (it is only enabled when IOMMU feature is enabled in
the vhost library).

Fixes: b0a985d1f340 ("vhost: add dequeue zero copy")

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
---
 lib/librte_vhost/vhost_user.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index be4f3c6c8..2d1123e60 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -1346,4 +1346,8 @@ vhost_user_set_vring_enable(struct virtio_net **pdev,
 				index, enable);
 
+	/* On disable, rings have to be stopped being processed. */
+	if (!enable && dev->dequeue_zero_copy)
+		drain_zmbuf_list(dev->virtqueue[index]);
+
 	dev->virtqueue[index]->enabled = enable;
 
-- 
2.20.1

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2019-04-16 15:34:26.487909570 +0100
+++ 0029-vhost-prevent-disabled-rings-to-be-processed-with-ze.patch	2019-04-16 15:34:25.174180083 +0100
@@ -1,8 +1,10 @@
-From 11d5253a3e93a38be14da65328dd19960fd8cb14 Mon Sep 17 00:00:00 2001
+From 7dadea51de8e770dad51940d6a5efc347fd9b30c Mon Sep 17 00:00:00 2001
 From: Maxime Coquelin <maxime.coquelin@redhat.com>
 Date: Thu, 28 Feb 2019 18:57:04 +0100
 Subject: [PATCH] vhost: prevent disabled rings to be processed with zero-copy
 
+[ upstream commit 11d5253a3e93a38be14da65328dd19960fd8cb14 ]
+
 The vhost-user spec says that once the vring is disabled, the
 client has to stop processing it. But it can happen when
 dequeue zero-copy is enabled if outstanding descriptors buffers
@@ -17,7 +19,6 @@
 the vhost library).
 
 Fixes: b0a985d1f340 ("vhost: add dequeue zero copy")
-Cc: stable@dpdk.org
 
 Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
 Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
@@ -26,7 +27,7 @@
  1 file changed, 4 insertions(+)
 
 diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
-index 36c0c676d..555d09ad9 100644
+index be4f3c6c8..2d1123e60 100644
 --- a/lib/librte_vhost/vhost_user.c
 +++ b/lib/librte_vhost/vhost_user.c
 @@ -1346,4 +1346,8 @@ vhost_user_set_vring_enable(struct virtio_net **pdev,

  parent reply	other threads:[~2019-04-16 14:38 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-16 14:36 [dpdk-stable] patch 'eal: support strlcat function' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'common/cpt: fix null auth only' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'crypto/openssl: fix big numbers after computations' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'crypto/openssl: fix modexp' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'test/crypto: fix duplicate id used by CCP device' " Kevin Traynor
2019-04-17  6:55   ` [dpdk-stable] [EXT] " Hemant Agrawal
2019-04-16 14:36 ` [dpdk-stable] patch 'event/opdl: replace sprintf with snprintf' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/nfp: fix setting MAC address' " Kevin Traynor
2019-04-23 10:24   ` Pablo Cascón
2019-04-23 13:03     ` Kevin Traynor
2019-04-23 13:35       ` Pablo Cascón
2019-04-25 16:02         ` Kevin Traynor
2019-04-26  9:26           ` Pablo Cascón
2019-04-26  9:32             ` Kevin Traynor
2019-04-26 10:00               ` Pablo Cascón
2019-04-16 14:36 ` [dpdk-stable] patch 'net/i40e: fix time sync for 25G' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/qede: support IOVA VA mode' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/mlx5: fix packet inline on Tx queue wraparound' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/bnxt: silence IOVA warnings' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/bnxt: suppress spurious error log' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/nfp: fix RSS query' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/ixgbe: restore VLAN filter for VF' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'app/testpmd: remove unused field from port struct' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'app/testpmd: fix a typo in log message' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'ethdev: fix method name in doxygen comment' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/qede: fix Rx packet drop' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: fix negative error codes' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: remove unused variable' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: remove extra checks for error codes' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: fix buffer overflow' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: fix EEPROM get for small and uneven lengths' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: fix link configuration' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: fix missing VLAN filter offload' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/atlantic: fix xstats return' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/enic: fix max MTU calculation' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/octeontx: fix vdev name' " Kevin Traynor
2019-04-16 14:36 ` Kevin Traynor [this message]
2019-04-16 14:36 ` [dpdk-stable] patch 'net/virtio-user: fix multiqueue with vhost kernel' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'vhost: fix interrupt suppression for the split ring' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/virtio: add barrier in interrupt enable' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'app/testpmd: fix stdout flush after printing stats' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/nfp: fix possible buffer overflow' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/tap: fix getting max iovec' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/sfc: fix speed capabilities reported in device info' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/bonding: fix LACP negotiation' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/cxgbe: fix missing checksum flags and packet type' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'doc: fix examples in bonding guide' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/bonding: fix port id types' " Kevin Traynor
2019-04-16 14:36 ` [dpdk-stable] patch 'net/bonding: fix slave " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'net/bonding: fix packet count type for LACP' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'net/bonding: fix queue index types' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'drivers/net: fix possible overflow using strlcat' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'compress/qat: fix setup inter buffers' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'examples/ipsec-secgw: fix AES-CTR block size' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'examples/ipsec-secgw: fix debug logs' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'cryptodev: fix driver name comparison' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'test/crypto: fix possible overflow using strlcat' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'build: remove meson warning for Arm' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'doc: update cross Arm toolchain in Linux guide' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'acl: fix compiler flags with meson and AVX2 runtime' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'malloc: fix documentation of realloc function' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'eal/linux: fix log levels for pagemap reading failure' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'ring: enforce reading tail before slots' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'test/spinlock: remove delay for correct benchmarking' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'test/spinlock: amortize the cost of getting time' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'spinlock: reimplement with atomic one-way barrier' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'rwlock: reimplement with atomic builtins' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'eal/ppc: fix global memory barrier' " Kevin Traynor
2019-04-16 14:37 ` [dpdk-stable] patch 'vfio: document multiprocess limitation for container API' " Kevin Traynor

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=20190416143719.21601-29-ktraynor@redhat.com \
    --to=ktraynor@redhat.com \
    --cc=maxime.coquelin@redhat.com \
    --cc=stable@dpdk.org \
    --cc=tiwei.bie@intel.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).