| Message ID | 20241107194357.683732-1-tariqt@nvidia.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | mlx5 esw qos refactor and SHAMPO cleanup | expand |
Hello: This series was applied to netdev/net-next.git (main) by Jakub Kicinski <kuba@kernel.org>: On Thu, 7 Nov 2024 21:43:45 +0200 you wrote: > Hi, > > This patchset for the mlx5 core and Eth drivers consists of 3 parts. > > First patch by Patrisious improves the E-switch mode change operation. > > The following 6 patches by Carolina introduce further refactoring for > the QoS handling, to set the foundation for future extensions. > > [...] Here is the summary with links: - [net-next,01/12] net/mlx5: E-switch, refactor eswitch mode change https://git.kernel.org/netdev/net-next/c/ab85ebf43723 - [net-next,02/12] net/mlx5: Simplify QoS normalization by removing error handling https://git.kernel.org/netdev/net-next/c/5a731857656e - [net-next,03/12] net/mlx5: Generalize max_rate and min_rate setting for nodes https://git.kernel.org/netdev/net-next/c/ac778fefed34 - [net-next,04/12] net/mlx5: Refactor scheduling element configuration bitmasks https://git.kernel.org/netdev/net-next/c/cc4bb15ffa84 - [net-next,05/12] net/mlx5: Generalize scheduling element operations https://git.kernel.org/netdev/net-next/c/663bc605d0db - [net-next,06/12] net/mlx5: Integrate esw_qos_vport_enable logic into rate operations https://git.kernel.org/netdev/net-next/c/d67bfd10e668 - [net-next,07/12] net/mlx5: Make vport QoS enablement more flexible for future extensions https://git.kernel.org/netdev/net-next/c/be034baba83e - [net-next,08/12] net/mlx5e: SHAMPO, Simplify UMR allocation for headers https://git.kernel.org/netdev/net-next/c/8a0ee54027b1 - [net-next,09/12] net/mlx5e: SHAMPO, Fix page_index calculation inconsistency https://git.kernel.org/netdev/net-next/c/1a4b58857704 - [net-next,10/12] net/mlx5e: SHAMPO, Change frag page setup order during allocation https://git.kernel.org/netdev/net-next/c/4f56868b7132 - [net-next,11/12] net/mlx5e: SHAMPO, Drop info array https://git.kernel.org/netdev/net-next/c/945ca432bfd0 - [net-next,12/12] net/mlx5e: SHAMPO, Rework header allocation loop https://git.kernel.org/netdev/net-next/c/ab4219db89da You are awesome, thank you!
Hi, This patchset for the mlx5 core and Eth drivers consists of 3 parts. First patch by Patrisious improves the E-switch mode change operation. The following 6 patches by Carolina introduce further refactoring for the QoS handling, to set the foundation for future extensions. In the following 5 patches by Dragos, we enhance the SHAMPO datapath flow by simplifying some logic, and cleaning up the implementation. Series generated against: commit 702c290a1cb1 ("sctp: Avoid enqueuing addr events redundantly") Thanks, Tariq Carolina Jubran (6): net/mlx5: Simplify QoS normalization by removing error handling net/mlx5: Generalize max_rate and min_rate setting for nodes net/mlx5: Refactor scheduling element configuration bitmasks net/mlx5: Generalize scheduling element operations net/mlx5: Integrate esw_qos_vport_enable logic into rate operations net/mlx5: Make vport QoS enablement more flexible for future extensions Dragos Tatulea (5): net/mlx5e: SHAMPO, Simplify UMR allocation for headers net/mlx5e: SHAMPO, Fix page_index calculation inconsistency net/mlx5e: SHAMPO, Change frag page setup order during allocation net/mlx5e: SHAMPO, Drop info array net/mlx5e: SHAMPO, Rework header allocation loop Patrisious Haddad (1): net/mlx5: E-switch, refactor eswitch mode change drivers/net/ethernet/mellanox/mlx5/core/en.h | 5 +- .../net/ethernet/mellanox/mlx5/core/en_main.c | 15 +- .../net/ethernet/mellanox/mlx5/core/en_rx.c | 123 ++-- .../mellanox/mlx5/core/esw/devlink_port.c | 2 +- .../net/ethernet/mellanox/mlx5/core/esw/qos.c | 558 +++++++----------- .../net/ethernet/mellanox/mlx5/core/esw/qos.h | 1 + .../net/ethernet/mellanox/mlx5/core/eswitch.c | 7 +- .../net/ethernet/mellanox/mlx5/core/eswitch.h | 5 +- .../mellanox/mlx5/core/eswitch_offloads.c | 26 +- 9 files changed, 303 insertions(+), 439 deletions(-)