From: Ferruh Yigit <ferruh.yigit@amd.com>
To: "Varghese, Vipin" <Vipin.Varghese@amd.com>,
Thomas Monjalon <thomas@monjalon.net>,
Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Cc: "david.marchand@redhat.com" <david.marchand@redhat.com>,
"Tummala, Sivaprasad" <Sivaprasad.Tummala@amd.com>,
"dev@dpdk.org" <dev@dpdk.org>,
Narcisa Ana Maria Vasile <navasile@linux.microsoft.com>,
Dmitry Malloy <dmitrym@microsoft.com>,
Pallavi Kadam <pallavi.kadam@intel.com>
Subject: Re: [PATCH] usertools: enhance logic to display NUMA
Date: Wed, 14 Jun 2023 11:48:02 +0100 [thread overview]
Message-ID: <4bdbc7a6-ae16-898c-a9e3-5b196396b662@amd.com> (raw)
In-Reply-To: <MN2PR12MB3085DB6773CBB668CCB0308E821D9@MN2PR12MB3085.namprd12.prod.outlook.com>
On 3/28/2022 3:56 AM, Varghese, Vipin wrote:
> Hi Thomas,
>
> <snipp>
>
> 26/03/2022 08:32, Vipin Varghese:
>>> enhance python logic to accomadate NUMA information. Current logic
>>> considers physical socket with CPU threads to core map. With new AMD
>>> SKU vairant NUMA is no longer same as SOCKET. Single physical socket
>>> can be partitioned to variant of 1,2 and 4.
>>>
>>> The changes address the new mapping with Socket-NUMA to CPU cores.
>>>
>>> Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
>>> ---
>>> usertools/cpu_layout.py | 76
>>> +++++++++++++++++++++++++----------------
>>> 1 file changed, 47 insertions(+), 29 deletions(-)
>> Honestly, I'm not sure it is a good idea to keep this script in the DPDK repo.
>> Can it be replaced with hwloc usage?
>
> thanks for the suggestion, it is genuine and useful. Following is my observations
>
> - It takes some effort to identify the NUMA with `Group`
> - One needs to install ` lstopo-no-graphics` on distro and manually build and add on custom Linux.
>
>
>> What is the output on the new AMD SKU for this command?
>> lstopo-no-graphics --merge
>
> I have tried ` lstopo-no-graphics --merge` on a ` 2 Socket AMD EPYC 7713 64-Core Processor` with
> possible NUMA configuration such as 1, 2 and 4.
>
Hi Thomas,
Should we document preferred tool and usage in DPDK, if so where is good
location for it?
Also @Dmitry, is there solution for Windows for this issue (a tool to
replace cpu_layout.py)?
Thanks,
ferruh
next prev parent reply other threads:[~2023-06-14 10:48 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-26 7:32 Vipin Varghese
2022-03-26 9:46 ` Thomas Monjalon
2022-03-28 2:56 ` Varghese, Vipin
2022-03-28 7:44 ` Tummala, Sivaprasad
2023-06-14 10:48 ` Ferruh Yigit [this message]
2023-06-14 11:25 ` Dmitry Kozlyuk
2023-06-14 14:30 ` Thomas Monjalon
2023-07-11 6:38 ` Varghese, Vipin
2023-07-11 15:42 ` Stephen Hemminger
2023-07-11 16:26 ` Thomas Monjalon
2023-07-14 9:14 ` Varghese, Vipin
2023-07-17 15:07 ` Thomas Monjalon
2023-07-18 10:37 ` Varghese, Vipin
2023-07-18 14:45 ` Thomas Monjalon
2023-08-12 11:39 ` Varghese, Vipin
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=4bdbc7a6-ae16-898c-a9e3-5b196396b662@amd.com \
--to=ferruh.yigit@amd.com \
--cc=Sivaprasad.Tummala@amd.com \
--cc=Vipin.Varghese@amd.com \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=dmitry.kozliuk@gmail.com \
--cc=dmitrym@microsoft.com \
--cc=navasile@linux.microsoft.com \
--cc=pallavi.kadam@intel.com \
--cc=thomas@monjalon.net \
/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).