Ixgbe internals
8 likes1,997 views
This document discusses the internals of the ixgbe driver, which is the Intel 10 Gigabit Ethernet driver for Linux. It describes how the driver handles transmission and reception of packets using ring buffers and NAPI. It also discusses how the driver supports eXpress Data Path (XDP) by using a separate ring for XDP packets and adjusting page reference counts to support XDP operations like redirection. Upcoming features that will improve XDP support for this driver are also mentioned.
![References
Linux kernel v4.15
https://github.com/torvalds/linux/tree/v4.15/drivers/net/ethernet/intel/ixgb
e
[0/5] Enable XDP for ixgbevf
http://patchwork.ozlabs.org/cover/887197/
[net-next V11 PATCH 00/17] XDP redirect memory return API
https://www.spinics.net/lists/netdev/msg495995.html
ixgbe: tweak page counting for XDP_REDIRECT
https://patchwork.ozlabs.org/patch/889261/
Monitoring and Tuning the Linux Networking Stack: Sending Data
https://blog.packagecloud.io/eng/2017/02/06/monitoring-tuning-linux-net
working-stack-sending-data/
Monitoring and Tuning the Linux Networking Stack: Receiving Data
https://blog.packagecloud.io/eng/2016/06/22/monitoring-tuning-linux-net
working-stack-receiving-data/](https://image.slidesharecdn.com/ixgbe-internals-180426034356/85/Ixgbe-internals-42-320.jpg)
More Related Content
Ixgbe internals
- 1. ixgbe Internals SUSE Labs Taipei Technology Sharing Day 2018 Gary Lin Software Engineer, SUSE Labs [email protected]
- 2. ixgbe?
- 3. Intel 10 Gigabit PCI Express Linux Driver
- 4. Why ixgbe? ● The first driver supporting XDP_REDIRECT in Linux kernel mainline ● Using a different memory model for XDP ● Just For Fun™
- 5. Overview
- 6. NIC driver Application Network Stack QDISC From Driver to AP TX RX NAPI
- 7. TX – QDISC
- 8. ... CPU 1 CPU 2 CPU n TX Q 1 TX Q 2 TX Q n-1 TX Q n CPU n-1 Ring Buffer ... QDISC netdev_pick_tx() + ndo_start_xmit()
- 9. RX – NAPI
- 10. RAM Network Card DMA poll poll poll poll IRQ RX Queue
- 11. ... CPU 1 CPU 2 CPU n napi_poll() CPU Q 1 CPU Q 2 CPU Q n ... enqueue_to_backlog() RX Q 1 RX Q 2 RX Q n-1 RX Q n CPU n-1 CPU Q n-1 CPU i Ring Buffer
- 12. Ring Buffers in ixgbe
- 13. TX Ring RX Ring XDP Ring
- 14. RX Q 2 IRQ q_vector RX Ring 2 TX Q 1 TX Ring 1 Legacy Interrupt RX Q 1 TX Q 2 RX Ring 1 TX Ring 2
- 15. RX Q IRQ q_vector RX Ring TX Q IRQ q_vector TX Ring RX Q IRQ q_vector RX Ring TX Q TX Ring MSI-X
- 17. TX Ring ixgbe_tx_buffer sk_buff QDISCstruct sk_buff *skb unsigned int bytescount
- 18. XDP Ring ixgbe_tx_buffer void *data xdp_buff XDP unsigned int bytescount void *data
- 19. TX Ring + XDP Ring ixgbe_q_vector struct ixgbe_ring_container tx TX Ring 1 TX Ring 2 TX Ring n XDP Ring 1 XDP Ring 2 XDP Ring n
- 20. RX Ring ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page dev_alloc_pages()
- 21. RX Ring ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page DMA
- 22. RX Ring ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page SKB
- 23. RX Ring ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page Flip Flip: page_offset ^= page_size / 2
- 24. RX Ring ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page
- 25. RX Ring – Recycle ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page Flip
- 26. RX Ring – Replace ixgbe_rx_buffer struct page *page __u32 *page_offset __u16 pagecnt_bias page page
- 27. Page Ref Count ● Tracking the number of users of the page ● The possible page ref count (ideally): – 1: The whole page is available. – 2: One half of the page is in use. – 3: The whole page is in use.
- 28. Page Count Operations static inline void set_page_count(struct page *page, int v) { atomic_set(&page->_refcount, v); if (page_ref_tracepoint_active(__tracepoint_page_ref_set)) __page_ref_set(page, v); } static inline void page_ref_add(struct page *page, int nr) { atomic_add(nr, &page->_refcount); if (page_ref_tracepoint_active(__tracepoint_page_ref_mod)) __page_ref_mod(page, nr); } static inline void page_ref_sub(struct page *page, int nr) { atomic_sub(nr, &page->_refcount); if (page_ref_tracepoint_active(__tracepoint_page_ref_mod)) __page_ref_mod(page, -nr); }
- 30. Adjusted Ref Count (1/3) ● A locally maintained pagecnt_bias for the RX page ● Initial value of pagecnt_bias: 1 ● adj_pagecnt = pagecnt – pagecnt_bias – 0: The whole page is available. – 1: One half of the page is in use. – 2: The whole page is in use.
- 31. Adjusted Ref Count (2/3) ● Harvesting a packet: pagecnt_bias-- ● Recycling the page: pagecnt_bias++ – The XDP program returns XDP_DROP or an error. – The packet is small enough to be copied into the allocated skb. – The packet is failed to be packed into skb. ● If pagecnt_bias == 0, set pagecnt_bias to USHRT_MAX and add USHRT_MAX to pagecnt.
- 32. Adjusted Ref Count (3/3) ● Packet consumed: pagecnt-- – The packet is consumed by the network stack. – XDP_TX or XDP_REDIRECT is completed. ● Releasing the page: pagecnt -= pagecnt_bias – With the help of __page_frag_cache_drain()
- 33. XDP
- 34. TX Q RX Q Network Stack PASS DROP eBPF Program TX NIC Driver eXpress Data Path REDIRECT Other CPU or NIC
- 35. Page 1 Page 2 Page 3 Conventional RX Buffer Page 1 Page 2 Page 3 One Packet Per Page Memory Model Switch XDP NOTE: The driver other than ixgbe
- 36. For ixgbe, memory model switch is not necessary!
- 37. XDP in ixgbe eBPF Program XDP_PASS Network Stack RX Ring XDP Ring XDP_DROP XDP_TX XDP_REDIRECT Other CPU or NICXDP_REDIRECT ixgbe_xdp_xmit() xdp_do_redirect() ixgbe_xmit_xdp_ring() pagecnt_bias++
- 38. Incoming New Features ● XDP for ixgbevf (linux-next) – ixgbe blocks XDP if SR-IOV is enabled. ● XDP redirect memory return API (net-next) – Managing pages across drivers – Adopted by ixgbe, i40e, mlx5, tuntap, and virtio_net – Preparing for the AF_XDP zero-copy patch set – ixgbe tweaked the page ref counting scheme for the new API.
- 39. Question?
- 40. Thank You!
- 42. References Linux kernel v4.15 https://github.com/torvalds/linux/tree/v4.15/drivers/net/ethernet/intel/ixgb e [0/5] Enable XDP for ixgbevf http://patchwork.ozlabs.org/cover/887197/ [net-next V11 PATCH 00/17] XDP redirect memory return API https://www.spinics.net/lists/netdev/msg495995.html ixgbe: tweak page counting for XDP_REDIRECT https://patchwork.ozlabs.org/patch/889261/ Monitoring and Tuning the Linux Networking Stack: Sending Data https://blog.packagecloud.io/eng/2017/02/06/monitoring-tuning-linux-net working-stack-sending-data/ Monitoring and Tuning the Linux Networking Stack: Receiving Data https://blog.packagecloud.io/eng/2016/06/22/monitoring-tuning-linux-net working-stack-receiving-data/