[v1] net: mana: Increase the DEF_RX_BUFFERS_PER_QUEUE to 1024

[PATCH net-next RESEND] net: mana: Increase the DEF_RX_BUFFERS_PER_QUEUE to 1024

Posted by Shradha Gupta 1 month, 4 weeks ago

Through some experiments, we found out that increasing the default
RX buffers count from 512 to 1024, gives slightly better throughput
and significantly reduces the no_wqe_rx errs on the receiver side.
Along with these, other parameters like cpu usage, retrans seg etc
also show some improvement with 1024 value.

Following are some snippets from the experiments

ntttcp tests with 512 Rx buffers
---------------------------------------
connections|  throughput|  no_wqe errs|
---------------------------------------
1          |  40.93Gbps | 123,211     |
16         | 180.15Gbps | 190,120     |
128        | 180.20Gbps | 173,508     |
256        | 180.27Gbps | 189,884     |

ntttcp tests with 1024 Rx buffers
---------------------------------------
connections|  throughput|  no_wqe errs|
---------------------------------------
1          |  44.22Gbps | 19,864      |
16         | 180.19Gbps | 4,430       |
128        | 180.21Gbps | 2,560       |
256        | 180.29Gbps | 1,529       |

So, increasing the default RX buffers per queue count to 1024

Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
---
 include/net/mana/mana.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index f2a5200d8a0f..9b0faa24b758 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -43,7 +43,7 @@ enum TRI_STATE {
  * size beyond this value gets rejected by __alloc_page() call.
  */
 #define MAX_RX_BUFFERS_PER_QUEUE 8192
-#define DEF_RX_BUFFERS_PER_QUEUE 512
+#define DEF_RX_BUFFERS_PER_QUEUE 1024
 #define MIN_RX_BUFFERS_PER_QUEUE 128
 
 /* This max value for TX buffers is derived as the maximum allocatable
-- 
2.34.1

Re: [PATCH net-next RESEND] net: mana: Increase the DEF_RX_BUFFERS_PER_QUEUE to 1024

Posted by Pavan Chebbi 1 month, 4 weeks ago

On Mon, Sep 30, 2024 at 9:14 AM Shradha Gupta
<shradhagupta@linux.microsoft.com> wrote:
>
> Through some experiments, we found out that increasing the default
> RX buffers count from 512 to 1024, gives slightly better throughput
> and significantly reduces the no_wqe_rx errs on the receiver side.
> Along with these, other parameters like cpu usage, retrans seg etc
> also show some improvement with 1024 value.
>
> Following are some snippets from the experiments
>
> ntttcp tests with 512 Rx buffers
> ---------------------------------------
> connections|  throughput|  no_wqe errs|
> ---------------------------------------
> 1          |  40.93Gbps | 123,211     |
> 16         | 180.15Gbps | 190,120     |
> 128        | 180.20Gbps | 173,508     |
> 256        | 180.27Gbps | 189,884     |
>
> ntttcp tests with 1024 Rx buffers
> ---------------------------------------
> connections|  throughput|  no_wqe errs|
> ---------------------------------------
> 1          |  44.22Gbps | 19,864      |
> 16         | 180.19Gbps | 4,430       |
> 128        | 180.21Gbps | 2,560       |
> 256        | 180.29Gbps | 1,529       |
>
> So, increasing the default RX buffers per queue count to 1024
>
> Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com>
> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
> ---
>  include/net/mana/mana.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>

Looks good to me.
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>