- 01 Jan, 2014 4 commits
-
-
Greg Rose authored
Modify the existing Kconfig, Makefile, and MAINTAINERS to add the driver to the kernel. Add a Makefile and a documentation Signed-off-by:
Mitch Williams <mitch.a.williams@intel.com> Signed-off-by:
Greg Rose <gregory.v.rose@intel.com> Tested-by:
Sibai Li <sibai.li@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Greg Rose authored
This patch implements the hardware specific init and management. Signed-off-by:
-
Greg Rose authored
This patch contains the main driver header files, containing structures and data types specific to the linux driver. i40e_osdep.h contains some code that helps us adapt our OS agnostic code to Linux. Signed-off-by:
-
Greg Rose authored
This PCI-E SR-IOV virtual function (VF) driver is dependant upon the physical function (PF) driver (i40e) for nearly all of its hardware configuration. Requests from the VF driver are passed to the PF using the hardware's Admin Queue. This patch contains the functionality for communicating with the PF driver. Because of the delay inherent in this communications channel, most of the replies from the PF driver are handled asynchronously. The exceptions are the "send API version" and "get VF config" messages, which busy-wait because they are done so early during init that interrupts are not yet configured. Signed-off-by:
-
- 31 Dec, 2013 27 commits
-
-
Greg Rose authored
This patch contains the ethtool interface and related functionality. Since the VF driver is mostly unaware of link, much of that functionality is unused. The driver implements ethtool hooks for statistics, driver info, and some basic non-link-related driver settings. Signed-off-by:
-
Greg Rose authored
This file contains the transmit, receive, and NAPI functionality. Some of the functions in this module are extracted from the i40e driver but functions that are not appropriate for virtual function devices have been removed. Signed-off-by:
-
Greg Rose authored
This is the driver for the Intel(R) XL710 X710 Virtual Function. This patch contains the main driver entry points, but does not include transmit and receive or ethtool functionality, which are presented as separate patches in this series. Signed-off-by:
-
Zhi Yong Wu authored
Signed-off-by:
Zhi Yong Wu <wuzhy@linux.vnet.ibm.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Zhi Yong Wu authored
Signed-off-by:
-
Zhi Yong Wu authored
In file included from net/socket.c:99:0: include/net/sock.h: In function ‘sock_rps_record_flow’: include/net/sock.h:849:30: error: ‘const struct sock’ has no member named ‘sk_rxhash’ include/net/sock.h: In function ‘sock_rps_reset_flow’: include/net/sock.h:854:29: error: ‘const struct sock’ has no member named ‘sk_rxhash’ Reported-by:
Fengguang Wu <fengguang.wu@intel.com> Signed-off-by:
-
wangweidong authored
skb_dst_set will use dst, if dst is NULL although is not a problem, then goto the 'no_route' and free nskb, so do the skb_dst_set is pointless. so move the skb_dst_set after dst check. Remove the unnecessary initialization as well. v2: fix the subject line because it would confuse people, as pointed out by Daniel. Signed-off-by:Wang Weidong <wangweidong1@huawei.com> Signed-off-by:
-
Yang Yingliang authored
Add a new attribute to support 64bit rates so that tc can use them to break the 32bit limit. Signed-off-by:
Yang Yingliang <yangyingliang@huawei.com> Acked-by:
Stephen Hemminger <stephen@networkplumber.org> Signed-off-by:
-
Yang Yingliang authored
With TSO/GSO/GRO packets, skb->len doesn't represent a precise amount of bytes on wire. This patch replace skb->len with qdisc_pkt_len(skb) which is more precise. Signed-off-by:
-
Or Gerlitz authored
When the device tunneling offloads mode is vxlan do the following - call SET_PORT with the relevant setting - add DMFS steering vxlan rule for the device self and multicast mac addresses of the form: {<ETH, outer-mac> <VXLAN, ANY vnid> <ETH, ANY mac>} --> RSS QP - set relevant QPC fields in RSS context and RX ring QPs - in TX flow, set WQE fields to generate HW checksum, and handle gso skbs which are marked for encapsulation such that the HW will segment them properly. - in RX flow, read HW offloaded checksum for encapsulated packets from the CQE - advertize hw_enc_features and NETIF_F_GSO_UDP_TUNNEL to the networking stack Signed-off-by:Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by:
-
Or Gerlitz authored
Add the low-level device commands and definitions used for TCP/IP HW offloads of tunneled/vxlan traffic which are supported by the ConnectX3-pro NIC. This is done through the following elements: - read tunneling device caps in QUERY_DEV_CAP - add helper function to do SET_PORT for tunneling - add DMFS VXLAN steering rule definitions - add CQE and WQE checksum offload field definitions Signed-off-by:
-
Daniel Borkmann authored
In order to facilitate development for netlink protocol dissector, fill the unused field skb->pkt_type of the cloned skb with a hint of the address space of the new owner (receiver) socket in the notion of "to kernel" resp. "to user". At the time we invoke __netlink_deliver_tap_skb(), we already have set the new skb owner via netlink_skb_set_owner_r(), so we can use that for netlink_is_kernel() probing. In normal PF_PACKET network traffic, this field denotes if the packet is destined for us (PACKET_HOST), if it's broadcast (PACKET_BROADCAST), etc. As we only have 3 bit reserved, we can use the value (= 6) of PACKET_FASTROUTE as it's _not used_ anywhere in the whole kernel and not supported anywhere, and packets of such type were never exposed to user space, so there are no overlapping users of such kind. Thus, as wished, that seems the only way to make both PACKET_* values non-overlapping and therefore device agnostic. By using those two flags for netlink skbs on nlmon devices, they can be made available and picked up via sll_pkttype (previously unused in netlink context) in struct sockaddr_ll. We now have these two directions: - PACKET_USER (= 6) -> to user space - PACKET_KERNEL (= 7) -> to kernel space Partial `ip a` example strace for sa_family=AF_NETLINK with detected nl msg direction: syscall: direction: sendto(3, ...) = 40 /* to kernel */ recvmsg(3, ...) = 3404 /* to user */ recvmsg(3, ...) = 1120 /* to user */ recvmsg(3, ...) = 20 /* to user */ sendto(3, ...) = 40 /* to kernel */ recvmsg(3, ...) = 168 /* to user */ recvmsg(3, ...) = 144 /* to user */ recvmsg(3, ...) = 20 /* to user */ Signed-off-by:
Daniel Borkmann <dborkman@redhat.com> Signed-off-by:
Jakub Zawadzki <darkjames-ws@darkjames.pl> Signed-off-by:
-
Daniel Borkmann authored
We should also deliver packets to nlmon devices when we are in netlink_unicast_kernel(), and only one of the {src,dst} sockets is user sk and the other one kernel sk. That's e.g. the case in netlink diag, netlink route, etc. Still, forbid to deliver messages from kernel to kernel sks. Signed-off-by: -
David S. Miller authored
Neil Horman says: ==================== sctp: Consolidate and ratelimit deprecation warnings The SCTP protocol has several deprecation warnings in its setsockopt path that can be triggered by unprivlidged users. Since these are not ratelimited, we can spam the logs quite easily here. Since these are all deprecation warnings, and that type of warning isn't uncommon in the rest of the kernel, lets make a common pr_warn_deprecated macro to produce somewhat generalized ratelimited deprecation warnings easily ==================== Signed-off-by:
Neil Horman <nhorman@tuxdriver.com> Signed-off-by:
-
Neil Horman authored
During a recent discussion regarding some sctp socket options, it was noted that we have several points at which we issue log warnings that can be flooded at an unbounded rate by any user. Fix this by converting all the pr_warns in the sctp_setsockopt path to be pr_warn_ratelimited. Note there are several debug level messages as well. I'm leaving those alone, as, if you turn on pr_debug, you likely want lots of verbosity. Signed-off-by:
-
Neil Horman authored
sctp has several points in its setsockopt path in which it issues deprecation warnings. It seems like it might be handy to macrotize such a warning so other subsystems can use it easily Signed-off-by:
-
Yang Yingliang authored
In dsmark_drop(), the function name printed by pr_debug is "dsmark_reset", correct it to "dsmark_drop" by using __func__ . BTW, replace the other function names with __func__ . Signed-off-by:
-
Yang Yingliang authored
Do not use C99 // comments and correct a spelling typo. Signed-off-by:
-
Yang Yingliang authored
Prefer pr_warn(... to pr_warning(... Signed-off-by:
-
Weilong Chen authored
Cleanup checkpatch errors.Specially,the second changed line is exactly 80 columns long. Signed-off-by:
Weilong Chen <chenweilong@huawei.com> Signed-off-by:
-
Li RongQing authored
Since addrconf_get_prefix_route inputs the address prefix to fib6_locate, which does not uses the data which is out of the prefix_len length, so do not need to use ipv6_addr_prefix to get address prefix. Signed-off-by:
Li RongQing <roy.qing.li@gmail.com> Acked-by:
Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by:
-
David S. Miller authored
Zhi Yong Wu says: ==================== tun: add the RFS support Since Tom Herbert's hash related patchset was modified and got merged, his pachset about adding support for RFS on tun flows also need to get adjusted accordingly. I tried to update them, and before i will start to do some perf tests, i hope to get one correct code base, so it's time to post them out now. Any constructive comments are welcome, thanks. ==================== Signed-off-by: -
Tom Herbert authored
This patch adds support so that the rps_flow_tables (RFS) can be programmed using the tun flows which are already set up to track flows for the purposes of queue selection. On the receive path (corresponding to select_queue and tun_net_xmit) the rxhash is saved in the flow_entry. The original code only does flow lookup in select_queue, so this patch adds a flow lookup in tun_net_xmit if num_queues == 1 (select_queue is not called from dev_queue_xmit->netdev_pick_tx in that case). The flow is recorded (processing CPU) in tun_flow_update (TX path), and reset when flow is deleted. Signed-off-by:
Tom Herbert <therbert@google.com> Signed-off-by:
-
Tom Herbert authored
This patch adds sock_rps_record_flow_hash and sock_rps_reset_flow_hash which take a hash value as an argument and sets the sock_flow_table accordingly. This allows the table to be populated in cases where flow is being tracked outside of a sock structure. sock_rps_record_flow and sock_rps_reset_flow call this function where the hash is taken from sk_rxhash. Signed-off-by:
-
dingtianhong authored
Signed-off-by:
Ding Tianhong <dingtianhong@huawei.com> Signed-off-by:
-
dingtianhong authored
The bond 3ad and TLB/ALB has the same check path, so combine them. Signed-off-by:
-
dingtianhong authored
The bond driver could set the lp_interval when loading module. Suggested-by:
Scott Feldman <sfeldma@cumulusnetworks.com> Signed-off-by:
-
- 29 Dec, 2013 5 commits
-
-
stephen hemminger authored
Remove leftover code that is not used anywhere in current tree. Signed-off-by:
-
stephen hemminger authored
bond_xmit_slave_id is only used in main. Signed-off-by:
-
stephen hemminger authored
The following are only used in one file: tcp_connect_init tcp_set_rto Signed-off-by:
-
Geert Uytterhoeven authored
This allows to drop a few casts. Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
-
Geert Uytterhoeven authored
Most of them reported by checkpatch.pl Signed-off-by:
-
- 28 Dec, 2013 4 commits
-
-
Stephen Hemminger authored
Don't export ping_table or ping_v4_sendmsg. Both are only used inside ping code. Signed-off-by:
-
Stephen Hemminger authored
inetpeer_invalidate_family defined but never used Signed-off-by:
-
Stephen Hemminger authored
Don't export arp_invalidate, only used in arp.c Signed-off-by:
-
Stephen Hemminger authored
Make fib_detect_death function static only used in one file. Signed-off-by:
-
