On Thu, Aug 15, 2024 at 09:08:56AM -0700, Jakub Kicinski wrote: > On Wed, 14 Aug 2024 09:59:13 -0700 Erni Sri Satya Vennela wrote: > > Change VMBus channels macro (VRSS_CHANNEL_DEFAULT) in > > Linux netvsc from 8 to 16 to align with Azure Windows VM > > and improve networking throughput. > > > > For VMs having less than 16 vCPUS, the channels depend > > on number of vCPUs. Between 16 to 32 vCPUs, the channels > > default to VRSS_CHANNEL_DEFAULT. For greater than 32 vCPUs, > > set the channels to number of physical cores / 2 as a way > > to optimize CPU resource utilization and scale for high-end > > processors with many cores. > > Maximum number of channels are by default set to 64. > > > > Based on this change the subchannel creation would change as follows: > > > > ------------------------------------------------------------- > > |No. of vCPU |dev_info->num_chn |subchannel created | > > ------------------------------------------------------------- > > | 0-16 | 16 | vCPU | > > | >16 & <=32 | 16 | 16 | > > | >32 & <=128 | vCPU/2 | vCPU/2 | > > | >128 | vCPU/2 | 64 | > > ------------------------------------------------------------- > > > > Performance tests showed significant improvement in throughput: > > - 0.54% for 16 vCPUs > > - 0.83% for 32 vCPUs > > - 1.76% for 48 vCPUs > > - 10.35% for 64 vCPUs > > - 13.47% for 96 vCPUs > > Is there anything that needs clarifying in my feedback on v1? > > https://lore.kernel.org/all/20240807201857.445f9f95@xxxxxxxxxx/ > > Ignoring maintainer feedback is known to result in angry outbursts. I sincerely apologize for the miss on my part. I will make sure this never happens again. As Haiyang mentioned, we were trying to use a similar logic as in netif_get_num_default_rss_queues(), and trying to make sure there are no potential regressions. I will work on Michael's and Haiyang's follow up comments. Please let us know your opinion on the same.
