From: bugzilla@dpdk.org
To: dev@dpdk.org
Subject: [DPDK/ethdev Bug 1591] MLX5 Windows : Issue with Packet Loss When Setting Descriptors Above 1<<14 on ConnectX6-DX
Date: Tue, 03 Dec 2024 13:16:48 +0000 [thread overview]
Message-ID: <bug-1591-3@http.bugs.dpdk.org/> (raw)
[-- Attachment #1: Type: text/plain, Size: 4520 bytes --]
https://bugs.dpdk.org/show_bug.cgi?id=1591
Bug ID: 1591
Summary: MLX5 Windows : Issue with Packet Loss When Setting
Descriptors Above 1<<14 on ConnectX6-DX
Product: DPDK
Version: 24.11
Hardware: x86
OS: Windows
Status: UNCONFIRMED
Severity: normal
Priority: Normal
Component: ethdev
Assignee: dev@dpdk.org
Reporter: a.pollenus@deltacast.tv
Target Milestone: ---
I am encountering an issue with the ConnectX6-DX on Windows. When I set the
number of descriptors to a value greater than 1<<14, all my packets are dropped
(imissed), except for the first one. The root cause is unclear, but I observed
that the maximum number of descriptors reported by rte_eth_dev_info_get() is
32768 (rx_desc_lim.nb_max), which I believe indicates that the number of
descriptors should be set to this value or lower.
Here are some test results using testpmd that demonstrate the issue:
TEST 1 : 4096 descriptors
./dpdk-testpmd -l 2-3 -n 4 -a 0000:03:00.0 --log-level=8
--log-level=pmd.common.mlx5:8 --log-level=pmd.net.mlx5:8 -- --socket-num=0
--burst=64 --txd=4096 --rxd=4096 --mbcache=512 --rxq=4 --txq=4 --nb-cores=1
--txpkts=1500 -i --forward-mode=rxonly --flow-isolate-all
testpmd> show port stats 0
######################## NIC statistics for port 0 ########################
RX-packets: 1626632 RX-missed: 0 RX-bytes: 2152490218
RX-errors: 0
RX-nombuf: 0
TX-packets: 0 TX-errors: 0 TX-bytes: 0
Throughput (since last show)
Rx-pps: 246876 Rx-bps: 2613496560
Tx-pps: 0 Tx-bps: 0
############################################################################
TEST 2 : 16384 descriptors
./dpdk-testpmd -l 2-3 -n 4 -a 0000:03:00.0 --log-level=8
--log-level=pmd.common.mlx5:8 --log-level=pmd.net.mlx5:8 -- --socket-num=0
--burst=64 --txd=4096 --rxd=16384 --mbcache=512 --rxq=4 --txq=4 --nb-cores=1
--txpkts=1500 -i --forward-mode=rxonly --flow-isolate-all
testpmd> show port stats 0
######################## NIC statistics for port 0 ########################
RX-packets: 2923021 RX-missed: 0 RX-bytes: 3867975188
RX-errors: 0
RX-nombuf: 0
TX-packets: 0 TX-errors: 0 TX-bytes: 0
Throughput (since last show)
Rx-pps: 246881 Rx-bps: 2613540240
Tx-pps: 0 Tx-bps: 0
############################################################################
TEST 3 : 20480 descriptors
./dpdk-testpmd -l 2-3 -n 4 -a 0000:03:00.0 --log-level=8
--log-level=pmd.common.mlx5:8 --log-level=pmd.net.mlx5:8 -- --socket-num=0
--burst=64 --txd=4096 --rxd=20480 --mbcache=512 --rxq=4 --txq=4 --nb-cores=1
--txpkts=1500 -i --forward-mode=rxonly --flow-isolate-all
testpmd> show port stats 0
######################## NIC statistics for port 0 ########################
RX-packets: 1 RX-missed: 2732098 RX-bytes: 1328
RX-errors: 0
RX-nombuf: 0
TX-packets: 0 TX-errors: 0 TX-bytes: 0
Throughput (since last show)
Rx-pps: 0 Rx-bps: 0
Tx-pps: 0 Tx-bps: 0
############################################################################
TEST 4 : 32768 descriptors
./dpdk-testpmd -l 2-3 -n 4 -a 0000:03:00.0 --log-level=8
--log-level=pmd.common.mlx5:8 --log-level=pmd.net.mlx5:8 -- --socket-num=0
--burst=64 --txd=4096 --rxd=32768 --mbcache=512 --rxq=4 --txq=4 --nb-cores=1
--txpkts=1500 -i --forward-mode=rxonly --flow-isolate-all
testpmd> show port stats 0
######################## NIC statistics for port 0 ########################
RX-packets: 1 RX-missed: 1129806 RX-bytes: 1328
RX-errors: 0
RX-nombuf: 0
TX-packets: 0 TX-errors: 0 TX-bytes: 0
Throughput (since last show)
Rx-pps: 0 Rx-bps: 0
Tx-pps: 0 Tx-bps: 0
############################################################################
I was able to reproduce this issue on versions 24.11 and 23.11, using DevX
version 24.10.26603.
Thank you in advance for your help,
Please let me know if you need further information.
Thank you.
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #2: Type: text/html, Size: 6447 bytes --]
reply other threads:[~2024-12-03 13:16 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=bug-1591-3@http.bugs.dpdk.org/ \
--to=bugzilla@dpdk.org \
--cc=dev@dpdk.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).