On Tue, Feb 04, 2025 at 02:42:52PM -0800, Jakub Kicinski wrote: > On Mon, 3 Feb 2025 13:09:29 -0800 Tony Nguyen wrote: > > Now changing queues using ethtool is also changing MSI-X. If there is > > enough MSI-X it is always one to one. When there is not enough there > > will be more queues than MSI-X. There is a lack of ability to set how > > many queues should be used per MSI-X. Maybe we should introduce another > > ethtool param for it? Sth like queues_per_vector? > > It's probably okay for today. AFAIU ethtool channels basically > correspond to IRQs. As the queue API matures we'll have > the ability to allocate more queues for "channel" == IRQ / event > queue. > Ok, thanks for explanation. > > The following are changes since commit c2933b2befe25309f4c5cfbea0ca80909735fd76: > > Merge tag 'net-6.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net > > and are available in the git repository at: > > git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue 100GbE > > Tony, the patches in your tree are missing your SoB, and I suspect > you may need the same PR to get pulled into RDMA, so I'm not applying > from the list... Please respin. > -- > pw-bot: cr
