DPDK patches and discussions
 help / color / mirror / Atom feed
From: Olivier Matz <olivier.matz@6wind.com>
To: dev@dpdk.org, Adrien Mazarguil <adrien.mazarguil@6wind.com>,
	Nelio Laranjeiro <nelio.laranjeiro@6wind.com>,
	Yongseok Koh <yskoh@mellanox.com>
Cc: stable@dpdk.org
Subject: Re: [dpdk-dev] [PATCH 1/2] net/mlx5: fix return value of start operation
Date: Fri, 19 Jan 2018 14:42:08 +0100	[thread overview]
Message-ID: <20180119134208.ztdnqlnv7viptw6w@platinum> (raw)
In-Reply-To: <20180118130043.31773-1-olivier.matz@6wind.com>

On Thu, Jan 18, 2018 at 02:00:42PM +0100, Olivier Matz wrote:
> On error, mlx5_dev_start() does not return a negative value
> as it is supposed to do. The consequence is that the application
> (ex: testpmd) does not notice that the port is not started
> and begins the rxtx on an uninitialized port, which crashes.
> 
> Fixes: e1016cb73383 ("net/mlx5: fix Rx interrupts management")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>

For reference, here is how to reproduce the problem.


The topology of the target:

   socket 0                 socket 1
  +---------------------+  +---------------------+
  |        c0        c1 |  |        c0        c1 |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  | |  0| 16| |  1| 17| |  | |  8| 24| |  9| 25| |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  |        c2        c3 |  |        c2        c3 |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  | |  2| 18| |  3| 19| |  | | 10| 26| | 11| 27| |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  |        c4        c5 |  |        c4        c5 |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  | |  4| 20| |  5| 21| |  | | 12| 28| | 13| 29| |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  |        c6        c7 |  |        c6        c7 |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  | |  6| 22| |  7| 23| |  | | 14| 30| | 15| 31| |
  | +-------+ +-------+ |  | +-------+ +-------+ |
  +---------------------+  +---------------------+

The cx4 devices are on socket 1, but I use cores and memory from
socket 0. I know it is not optimal, but it should work.



root@dut-cx4:~# cd dpdk.org
root@dut-cx4:~/dpdk.org# make config T=x86_64-native-linuxapp-gcc
root@dut-cx4:~/dpdk.org# make -j32
root@dut-cx4:~/dpdk.org# mkdir -p /mnt/huge
root@dut-cx4:~/dpdk.org# mount -t hugetlbfs nodev /mnt/huge
root@dut-cx4:~/dpdk.org# echo 256 > /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
root@dut-cx4:~/dpdk.org# testpmd -l 0,2 -- --total-num-mbufs=16384 -i --port-topology=chained --no-numa
EAL: Detected 32 lcore(s)
EAL: No free hugepages reported in hugepages-1048576kB
EAL: Probing VFIO support...
EAL: PCI device 0000:02:00.0 on NUMA socket 0
EAL:   probe driver: 8086:1572 net_i40e
EAL: PCI device 0000:02:00.1 on NUMA socket 0
EAL:   probe driver: 8086:1572 net_i40e
EAL: PCI device 0000:02:00.2 on NUMA socket 0
EAL:   probe driver: 8086:1572 net_i40e
EAL: PCI device 0000:02:00.3 on NUMA socket 0
EAL:   probe driver: 8086:1572 net_i40e
EAL: PCI device 0000:04:00.0 on NUMA socket 0
EAL:   probe driver: 14e4:16d7 net_bnxt
EAL: PCI device 0000:04:00.1 on NUMA socket 0
EAL:   probe driver: 14e4:16d7 net_bnxt
EAL: PCI device 0000:06:00.0 on NUMA socket 0
EAL:   probe driver: 8086:1521 net_e1000_igb
EAL: PCI device 0000:83:00.0 on NUMA socket 1
EAL:   probe driver: 8086:10fb net_ixgbe
EAL: PCI device 0000:83:00.1 on NUMA socket 1
EAL:   probe driver: 8086:10fb net_ixgbe
EAL: PCI device 0000:86:00.0 on NUMA socket 1
EAL:   probe driver: 15b3:1013 net_mlx5
PMD: net_mlx5: PCI information matches, using device "mlx5_0" (SR-IOV: false)
PMD: net_mlx5: 1 port(s) detected
PMD: net_mlx5: MPS is disabled
PMD: net_mlx5: port 1 MAC address is e4:1d:2d:e7:0d:06
EAL: PCI device 0000:86:00.1 on NUMA socket 1
EAL:   probe driver: 15b3:1013 net_mlx5
PMD: net_mlx5: PCI information matches, using device "mlx5_1" (SR-IOV: false)
PMD: net_mlx5: 1 port(s) detected
PMD: net_mlx5: MPS is disabled
PMD: net_mlx5: port 1 MAC address is e4:1d:2d:e7:0d:07
Interactive-mode selected
USER1: create a new mbuf pool <mbuf_pool_socket_0>: n=16384, size=2176, socket=0
Configuring Port 0 (socket 0)
PMD: net_mlx5: 0x1459e40: TX queues number update: 0 -> 1
PMD: net_mlx5: 0x1459e40: RX queues number update: 0 -> 1
PMD: net_mlx5: cannot allocate CQ for drop queue
PMD: net_mlx5: 0x1459e40: Drop queue allocation failed: Unknown error -1
Port 0: E4:1D:2D:E7:0D:06
Configuring Port 1 (socket 0)
PMD: net_mlx5: 0x145dec0: TX queues number update: 0 -> 1
PMD: net_mlx5: 0x145dec0: RX queues number update: 0 -> 1
PMD: net_mlx5: cannot allocate CQ for drop queue
PMD: net_mlx5: 0x145dec0: Drop queue allocation failed: Unknown error -1
Port 1: E4:1D:2D:E7:0D:07
Checking link statuses...
Done
testpmd> start
Segmentation fault (core dumped)

  parent reply	other threads:[~2018-01-19 13:42 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-18 13:00 Olivier Matz
2018-01-18 13:00 ` [dpdk-dev] [PATCH 2/2] net/mlx5: fix allocation when no memory on device NUMA node Olivier Matz
2018-01-18 16:19   ` Nélio Laranjeiro
2018-01-18 16:04 ` [dpdk-dev] [PATCH 1/2] net/mlx5: fix return value of start operation Nélio Laranjeiro
2018-01-18 16:13   ` Olivier Matz
2018-01-19  6:28     ` Yongseok Koh
2018-01-19  8:35     ` Nélio Laranjeiro
2018-01-19  8:43       ` Olivier Matz
2018-01-19 13:30         ` Olivier Matz
2018-01-19 13:43           ` Nélio Laranjeiro
2018-01-19 14:18           ` Nélio Laranjeiro
2018-01-19 13:42 ` Olivier Matz [this message]
2018-01-19 16:25 ` [dpdk-dev] [PATCH v2 " Olivier Matz
2018-01-19 16:25   ` [dpdk-dev] [PATCH v2 2/2] net/mlx5: fix allocation when no memory on device NUMA node Olivier Matz
2018-01-21  6:58     ` [dpdk-dev] [dpdk-stable] " Shahaf Shuler
2018-01-22  8:20       ` Olivier Matz
2018-01-22 12:33   ` [dpdk-dev] [PATCH v3 1/2] net/mlx5: fix return value of start operation Olivier Matz
2018-01-22 12:33     ` [dpdk-dev] [PATCH v3 2/2] net/mlx5: fix allocation when no memory on device NUMA node Olivier Matz
2018-01-22 20:27     ` [dpdk-dev] [dpdk-stable] [PATCH v3 1/2] net/mlx5: fix return value of start operation Shahaf Shuler

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=20180119134208.ztdnqlnv7viptw6w@platinum \
    --to=olivier.matz@6wind.com \
    --cc=adrien.mazarguil@6wind.com \
    --cc=dev@dpdk.org \
    --cc=nelio.laranjeiro@6wind.com \
    --cc=stable@dpdk.org \
    --cc=yskoh@mellanox.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).