From: "Lin, Xueqin" <xueqin.lin@intel.com>
To: "Li, WenjieX A" <wenjiex.a.li@intel.com>, "dts@dpdk.org" <dts@dpdk.org>
Cc: "Li, WenjieX A" <wenjiex.a.li@intel.com>
Subject: Re: [dts] [PATCH V1] tests/dynamic_queue: halve the queue number for fvl_40G nic
Date: Wed, 12 Dec 2018 11:02:27 +0000 [thread overview]
Message-ID: <0D300480287911409D9FF92C1FA2A3355B45F03A@SHSMSX104.ccr.corp.intel.com> (raw)
In-Reply-To: <1544609775-27422-1-git-send-email-wenjiex.a.li@intel.com>
Hi Wenjie,
Suggest not decreasing the queue number.
If chang as your patch, queue number will decrease for all FVL 40G, it is not reasonable.
I can't reproduce it for FVL 40G on my environment, only see it on your environment.
My concern as below:
1. Basic TX with multi-queues also has the issue, it is not dynamic queue feature special issue.
2. Your CPU is v2, suggest changing other 40G on this sever if has the issue, if yes, means not NIC issue, maybe CPU is too old.
3. Double check if 10G on this server has the issue.
Best regards,
Xueqin
> -----Original Message-----
> From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of Wenjie Li
> Sent: Wednesday, December 12, 2018 6:16 PM
> To: dts@dpdk.org
> Cc: Li, WenjieX A <wenjiex.a.li@intel.com>
> Subject: [dts] [PATCH V1] tests/dynamic_queue: halve the queue number for
> fvl_40G nic
>
> For fvl_40G nic, txq cases will fail if the queue number is 64 and could pass
> with less queues, so reduce the number.
>
> Signed-off-by: Wenjie Li <wenjiex.a.li@intel.com>
>
> ---
> tests/TestSuite_dynamic_queue.py | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/tests/TestSuite_dynamic_queue.py
> b/tests/TestSuite_dynamic_queue.py
> index f155686..8fc2def 100644
> --- a/tests/TestSuite_dynamic_queue.py
> +++ b/tests/TestSuite_dynamic_queue.py
> @@ -27,6 +27,8 @@ class TestDynamicQueue(TestCase):
> self.dut_testpmd = PmdOutput(self.dut)
>
> def set_up(self):
> + if self.nic in ['fortville_spirit']:
> + self.PF_QUEUE = self.PF_QUEUE / 2
> self.dut_testpmd.start_testpmd(
> "Default", "--port-topology=chained --txq=%s --rxq=%s"
> % (self.PF_QUEUE, self.PF_QUEUE))
> --
> 2.17.2
prev parent reply other threads:[~2018-12-12 11:02 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-12 10:16 Wenjie Li
2018-12-12 11:02 ` Lin, Xueqin [this message]
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=0D300480287911409D9FF92C1FA2A3355B45F03A@SHSMSX104.ccr.corp.intel.com \
--to=xueqin.lin@intel.com \
--cc=dts@dpdk.org \
--cc=wenjiex.a.li@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).