Hi Sivaprasad,

Are you going to send v3 (add log) version?

There is another patch about adjust burst size [1], if this patch
don't continue, maybe we should restart that one.

[1] 
https://patchwork.dpdk.org/project/dpdk/patch/[email protected]/

Thanks

On 10/9/2025 10:32 AM, Tummala, Sivaprasad wrote:
> [AMD Official Use Only - AMD Internal Distribution Only]
> 
> Hi Pavan,
> 
> Please find my response inline.
> 
> ________________________________
> From: Pavan Nikhilesh Bhagavatula <[email protected]>
> Sent: Tuesday, June 10, 2025 7:59 PM
> To: Tummala, Sivaprasad <[email protected]>; [email protected] 
> <[email protected]>; [email protected] 
> <[email protected]>; Jerin Jacob <[email protected]>; 
> [email protected] <[email protected]>; Akhil Goyal 
> <[email protected]>; [email protected] 
> <[email protected]>; Yigit, Ferruh <[email protected]>; 
> [email protected] <[email protected]>; 
> [email protected] <[email protected]>
> Cc: [email protected] <[email protected]>; [email protected] <[email protected]>; 
> [email protected] <[email protected]>; Ande, Venkat Kumar 
> <[email protected]>; Dengdui Huang <[email protected]>
> Subject: Re: [EXTERNAL] [PATCH v2] examples/l3fwd: adjust Tx burst size based 
> on Rx burst
> 
> Caution: This message originated from an External Source. Use proper caution 
> when opening attachments, clicking links, or responding.
> 
> 
>>> Previously, the TX burst size was fixed at 256, leading to performance
>>> degradation in certain scenarios.
>>>
>>> This patch introduces logic to set the TX burst size to match the
>>> configured RX burst size (--burst option, default 32, max 512)
>>> for better efficiency.
>>>
>>> Fixes: d5c4897ecfb2 ("examples/l3fwd: add option to set Rx burst size")
>>> Cc: [email protected]
>>> Cc: [email protected]
>>>
>>> Signed-off-by: Sivaprasad Tummala <[email protected]>
>>> Tested-by: Venkat Kumar Ande <[email protected]>
>>> Tested-by: Dengdui Huang <[email protected]>
> 
>> It would be good if the selected burst sizes of Rx and Tx are logged.
> ACK
> 
>> On CN10K platform we see upto 5% improvement, and upto 30% improvement on 
>> CN9K.
> 
>> Tested-by: Pavan Nikhilesh <[email protected]>
> 
>> ---
>> examples/l3fwd/l3fwd.h        |  8 ++------
>> examples/l3fwd/l3fwd_common.h | 11 +++++++----
>> examples/l3fwd/main.c         |  2 ++
>> 3 files changed, 11 insertions(+), 10 deletions(-)
> 
> 

Reply via email to