DPDK patches and discussions
 help / color / mirror / Atom feed
From: Olivier Matz <olivier.matz@6wind.com>
To: dev@dpdk.org, Yuanhan Liu <yliu@fridaylinux.org>,
	Maxime Coquelin <maxime.coquelin@redhat.com>,
	Tiwei Bie <tiwei.bie@intel.com>
Subject: [dpdk-dev] [PATCH v3 0/4] net/virtio: fix memory leak when reinitializing device
Date: Tue, 23 Jan 2018 16:54:39 +0100	[thread overview]
Message-ID: <20180123155443.8883-1-olivier.matz@6wind.com> (raw)
In-Reply-To: <20180119155556.32597-1-olivier.matz@6wind.com>

When devops->configure() is called, the queues may be reallocated
if the features flag changed, but the previous are not freed. This
patchset fixes this issue.

To really point out the issue, I instrumented rte_malloc,
rte_memzone_reserve, rte_mbuf_alloc to track the allocations and frees.
For reference, the patch is available here:
https://www.droids-corp.org/~zer0/hidden/0001-debug-malloc.patch
Iit conflicts a bit with the patchset, but it is quite easy to solve.

To reproduce the issue:

  cd dpdk
  make config T=x86_64-native-linuxapp-gcc
  sed -i 's,CONFIG_RTE_LIBRTE_VIRTIO_DEBUG_INIT=n,CONFIG_RTE_LIBRTE_VIRTIO_DEBUG_INIT=y,' build/.config
  make -j4
  mkdir -p /mnt/huge
  mount -t hugetlbfs nodev /mnt/huge
  echo 256 > /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
  modprobe uio_pci_generic
  python usertools/dpdk-devbind.py -b uio_pci_generic 0000:00:02.0
  ./build/app/testpmd -l 0,1 --log-level 8 -- --total-num-mbufs=16384 \
     -i --port-topology=chained --disable-hw-vlan-filter \
     --disable-hw-vlan-strip --enable-rx-cksum --enable-lro \
     --txqflags=0 --no-lsc-interrupt

  # testpmd commands
  port stop 0
  port config all rx-cksum off
  port start 0
  quit

Without the patchset, the debug patch displays a leak of mbufs allocated
from virtio_dev_rx_queue_setup_finish() and a leak of rte_mallocs allocated
from virtio_init_queue().

After the patchset, only one malloc is remaining after the device close,
but this is normal, because this area is freed in dev_uninit().

v2 -> v3:
- fix queue flushing with vector Rx enabled (Tiwei).
  The patch is slightly changed compared to what was proposed by mail,
  due to a checkpatch warning.

v1 -> v2:
- move the vq != NULL check from the refactor patch to the memory
  leak fix patch (Yuanhan)
- new patch to fix queue flushing with vector Rx enabled (Tiwei)
- move the backportable fixes at the beginning of the patchset (Yuanhan)
- remove Cc stable on the typo fix (Yuanhan)
  note: I finally kept it in the patchset


Olivier Matz (4):
  net/virtio: fix queue flushing with vector Rx enabled
  net/virtio: fix memory leak when reinitializing device
  net/virtio: rationalize queue flushing
  net/virtio: fix typo in function name

 drivers/net/virtio/virtio_ethdev.c | 71 +++++++++++++++++---------------------
 drivers/net/virtio/virtqueue.c     | 30 +++++++++++++---
 drivers/net/virtio/virtqueue.h     | 13 ++++++-
 3 files changed, 69 insertions(+), 45 deletions(-)

-- 
2.11.0

  parent reply	other threads:[~2018-01-23 15:55 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-18  9:07 [dpdk-dev] [PATCH 0/3] " Olivier Matz
2018-01-18  9:07 ` [dpdk-dev] [PATCH 1/3] net/virtio: fix typo in function name Olivier Matz
2018-01-18 13:27   ` Yuanhan Liu
2018-01-18 13:45     ` Olivier Matz
2018-01-18 14:06       ` Yuanhan Liu
2018-01-18  9:07 ` [dpdk-dev] [PATCH 2/3] net/virtio: rationalize queue flushing Olivier Matz
2018-01-18 13:26   ` Yuanhan Liu
2018-01-18 13:55     ` Olivier Matz
2018-01-18 14:04       ` Yuanhan Liu
2018-01-18 14:05   ` Tiwei Bie
2018-01-18 14:55     ` Olivier Matz
2018-01-18 15:48       ` Tiwei Bie
2018-01-18 15:56         ` Olivier Matz
2018-01-18  9:07 ` [dpdk-dev] [PATCH 3/3] net/virtio: fix memory leak when reinitializing device Olivier Matz
2018-01-19 15:55 ` [dpdk-dev] [PATCH v2 0/4] " Olivier Matz
2018-01-19 15:55   ` [dpdk-dev] [PATCH v2 1/4] net/virtio: fix queue flushing with vector Rx enabled Olivier Matz
2018-01-22  2:56     ` Tiwei Bie
2018-01-22 10:38       ` Olivier Matz
2018-01-23  2:05         ` Tiwei Bie
2018-01-19 15:55   ` [dpdk-dev] [PATCH v2 2/4] net/virtio: fix memory leak when reinitializing device Olivier Matz
2018-01-19 15:55   ` [dpdk-dev] [PATCH v2 3/4] net/virtio: rationalize queue flushing Olivier Matz
2018-01-19 15:55   ` [dpdk-dev] [PATCH v2 4/4] net/virtio: fix typo in function name Olivier Matz
2018-01-23 15:54   ` Olivier Matz [this message]
2018-01-23 15:54     ` [dpdk-dev] [PATCH v3 1/4] net/virtio: fix queue flushing with vector Rx enabled Olivier Matz
2018-01-23 15:54     ` [dpdk-dev] [PATCH v3 2/4] net/virtio: fix memory leak when reinitializing device Olivier Matz
2018-01-23 15:54     ` [dpdk-dev] [PATCH v3 3/4] net/virtio: rationalize queue flushing Olivier Matz
2018-01-23 15:54     ` [dpdk-dev] [PATCH v3 4/4] net/virtio: fix typo in function name Olivier Matz
2018-01-26 15:34     ` [dpdk-dev] [PATCH v3 0/4] net/virtio: fix memory leak when reinitializing device 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=20180123155443.8883-1-olivier.matz@6wind.com \
    --to=olivier.matz@6wind.com \
    --cc=dev@dpdk.org \
    --cc=maxime.coquelin@redhat.com \
    --cc=tiwei.bie@intel.com \
    --cc=yliu@fridaylinux.org \
    /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).