- 25 Jan, 2017 16 commits
-
-
Sunil Goutham authored
Adds support to set Rx/Tx queue sizes from ethtool. Fixes an issue with retrieving queue size. Also sets SQ's CQ_LIMIT based on configured Tx queue size such that HW doesn't process SQEs when there is no sufficient space in CQ. Signed-off-by:
Sunil Goutham <sgoutham@cavium.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Willy Tarreau authored
Without TFO, any subsequent connect() call after a successful one returns -1 EISCONN. The last API update ensured that __inet_stream_connect() can return -1 EINPROGRESS in response to sendmsg() when TFO is in use to indicate that the connection is now in progress. Unfortunately since this function is used both for connect() and sendmsg(), it has the undesired side effect of making connect() now return -1 EINPROGRESS as well after a successful call, while at the same time poll() returns POLLOUT. This can confuse some applications which happen to call connect() and to check for -1 EISCONN to ensure the connection is usable, and for which EINPROGRESS indicates a need to poll, causing a loop. This problem was encountered in haproxy where a call to connect() is precisely used in certain cases to confirm a connection's readiness. While arguably haproxy's behaviour should be improved here, it seems important to aim at a more robust behaviour when the goal of the new API is to make it easier to implement TFO in existing applications. This patch simply ensures that we preserve the same semantics as in the non-TFO case on the connect() syscall when using TFO, while still returning -1 EINPROGRESS on sendmsg(). For this we simply tell __inet_stream_connect() whether we're doing a regular connect() or in fact connecting for a sendmsg() call. Cc: Wei Wang <weiwan@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Eric Dumazet <edumazet@google.com> Signed-off-by:
Willy Tarreau <w@1wt.eu> Signed-off-by:
-
David S. Miller authored
Wei Wang says: ==================== net/tcp-fastopen: Add new userspace API support The patch series is to add support for new userspace API for TCP fastopen sockets. In the current code, user has to call sendto()/sendmsg() with special flag MSG_FASTOPEN for TCP fastopen sockets. This API is quite different from the normal TCP socket API and can be cumbersome for applications to make use fastopen sockets. So this new patch introduces a new way of using TCP fastopen sockets which is similar to normal TCP sockets with a new sockopt TCP_FASTOPEN_CONNECT. More details about it is described in the third patch. (First 2 patches are preparations for the third patch.) ==================== Signed-off-by: -
Wei Wang authored
This patch adds a new socket option, TCP_FASTOPEN_CONNECT, as an alternative way to perform Fast Open on the active side (client). Prior to this patch, a client needs to replace the connect() call with sendto(MSG_FASTOPEN). This can be cumbersome for applications who want to use Fast Open: these socket operations are often done in lower layer libraries used by many other applications. Changing these libraries and/or the socket call sequences are not trivial. A more convenient approach is to perform Fast Open by simply enabling a socket option when the socket is created w/o changing other socket calls sequence: s = socket() create a new socket setsockopt(s, IPPROTO_TCP, TCP_FASTOPEN_CONNECT …); newly introduced sockopt If set, new functionality described below will be used. Return ENOTSUPP if TFO is not supported or not enabled in the kernel. connect() With cookie present, return 0 immediately. With no cookie, initiate 3WHS with TFO cookie-request option and return -1 with errno = EINPROGRESS. write()/sendmsg() With cookie present, send out SYN with data and return the number of bytes buffered. With no cookie, and 3WHS not yet completed, return -1 with errno = EINPROGRESS. No MSG_FASTOPEN flag is needed. read() Return -1 with errno = EWOULDBLOCK/EAGAIN if connect() is called but write() is not called yet. Return -1 with errno = EWOULDBLOCK/EAGAIN if connection is established but no msg is received yet. Return number of bytes read if socket is established and there is msg received. The new API simplifies life for applications that always perform a write() immediately after a successful connect(). Such applications can now take advantage of Fast Open by merely making one new setsockopt() call at the time of creating the socket. Nothing else about the application's socket call sequence needs to change. Signed-off-by:Wei Wang <weiwan@google.com> Acked-by:
Eric Dumazet <edumazet@google.com> Acked-by:
Yuchung Cheng <ycheng@google.com> Signed-off-by:
-
Wei Wang authored
Remove __sk_dst_reset() in the failure handling because __sk_dst_reset() will eventually get called when sk is released. No need to handle it in the protocol specific connect call. This is also to make the code path consistent with ipv4. Signed-off-by:
-
Wei Wang authored
Refactor the cookie check logic in tcp_send_syn_data() into a function. This function will be called else where in later changes. Signed-off-by:
-
hayeswang authored
Replace rumtime with runtime. Signed-off-by:
Hayes Wang <hayeswang@realtek.com> Signed-off-by:
-
Andrew Lunn authored
Replace spaces with tabs. Fix indentation to be multiples of tabs, not a mixture or tabs and spaces. Signed-off-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
David S. Miller authored
Daniel Borkmann says: ==================== BPF tracepoints This set adds tracepoints to BPF for better introspection and debugging. The first two patches are prerequisite for the actual third patch that adds the tracepoints. I think the first two are small and straight forward enough that they could ideally go via net-next, but I'm also open to other suggestions on how to route them in case that's not applicable (it would reduce potential merge conflicts on BPF side, though). For details, please see individual patches. ==================== Signed-off-by: -
Daniel Borkmann authored
This work adds a number of tracepoints to paths that are either considered slow-path or exception-like states, where monitoring or inspecting them would be desirable. For bpf(2) syscall, tracepoints have been placed for main commands when they succeed. In XDP case, tracepoint is for exceptions, that is, f.e. on abnormal BPF program exit such as unknown or XDP_ABORTED return code, or when error occurs during XDP_TX action and the packet could not be forwarded. Both have been split into separate event headers, and can be further extended. Worst case, if they unexpectedly should get into our way in future, they can also removed [1]. Of course, these tracepoints (like any other) can be analyzed by eBPF itself, etc. Example output: # ./perf record -a -e bpf:* sleep 10 # ./perf script sock_example 6197 [005] 283.980322: bpf:bpf_map_create: map type=ARRAY ufd=4 key=4 val=8 max=256 flags=0 sock_example 6197 [005] 283.980721: bpf:bpf_prog_load: prog=a5ea8fa30ea6849c type=SOCKET_FILTER ufd=5 sock_example 6197 [005] 283.988423: bpf:bpf_prog_get_type: prog=a5ea8fa30ea6849c type=SOCKET_FILTER sock_example 6197 [005] 283.988443: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[06 00 00 00] val=[00 00 00 00 00 00 00 00] [...] sock_example 6197 [005] 288.990868: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[01 00 00 00] val=[14 00 00 00 00 00 00 00] swapper 0 [005] 289.338243: bpf:bpf_prog_put_rcu: prog=a5ea8fa30ea6849c type=SOCKET_FILTER [1] https://lwn.net/Articles/705270/Signed-off-by:Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Daniel Borkmann authored
Add support for the __print_hex_str() macro that was added for tracing, so that user space tools such as perf can understand it as well. Signed-off-by:
-
Daniel Borkmann authored
For upcoming tracepoint support for BPF, we want to dump the program's tag. Format should be similar to __print_hex(), but without spacing. Add a __print_hex_str() variant for exactly that purpose that reuses trace_print_hex_seq(). Signed-off-by:
-
Eric Dumazet authored
tcp_add_backlog() can use skb_condense() helper to get better gains and less SKB_TRUESIZE() magic. This only happens when socket backlog has to be used. Some attacks involve specially crafted out of order tiny TCP packets, clogging the ofo queue of (many) sockets. Then later, expensive collapse happens, trying to copy all these skbs into single ones. This unfortunately does not work if each skb has no neighbor in TCP sequence order. By using skb_condense() if the skb could not be coalesced to a prior one, we defeat these kind of threats, potentially saving 4K per skb (or more, since this is one page fragment). A typical NAPI driver allocates gro packets with GRO_MAX_HEAD bytes in skb->head, meaning the copy done by skb_condense() is limited to about 200 bytes. Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linuxDavid S. Miller authored
Saeed Mahameed says: ==================== mlx5-updates-2017-24-01 The first seven patches from Or Gerlitz in this series further enhances the mlx5 SRIOV switchdev mode to support offloading IPv6 tunnels using the TC tunnel key set (encap) and unset (decap) actions. Or Gerlitz says: ======================== As part of doing this change, few cleanups are done in the IPv4 code, later we move to use the full tunnel key info provided to the driver as the key for our internal hashing which is used to identify cases where the same tunnel is used for encapsulating multiple flows. As done in the IPv4 case, the control path for offloading IPv6 tunnels uses route/neigh lookups and construction of the IPv6 tunnel headers on the encap path and matching on the outer hears in the decap path. The last patch of the series enlarges the HW FDB size for the switchdev mode, so it has now room to contain offloaded flows as many as min(max number of HW flow counters supported, max HW table size supported). ======================== Next to Or's series you can find several patches handling several topics. From Mohamad, add support for SRIOV VF min rate guarantee by using the TSAR BW share weights mechanism. From Or, Two patches to enable Eth VFs to query their min-inline value for user-space. for that we move a mlx5 low level min inline helper function from mlx5 ethernet driver into the core driver and then use it in mlx5_ib to expose the inline mode to rdma applications through libmlx5. From Kamal Heib, Reduce memory consumption on kdump kernel. From Shaker Daibes, code reuse in CQE compression control logic ==================== Signed-off-by: -
Dan Carpenter authored
We shuffled some code around and added some new case statements here and now "res" isn't initialized on all paths. Fixes: 01fd12bb ("tipc: make replicast a user selectable option") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Jamal Hadi Salim authored
Introduce optional 128-bit action cookie. Like all other cookie schemes in the networking world (eg in protocols like http or existing kernel fib protocol field, etc) the idea is to save user state that when retrieved serves as a correlator. The kernel _should not_ intepret it. The user can store whatever they wish in the 128 bits. Sample exercise(showing variable length use of cookie) .. create an accept action with cookie a1b2c3d4 sudo $TC actions add action ok index 1 cookie a1b2c3d4 .. dump all gact actions.. sudo $TC -s actions ls action gact action order 0: gact action pass random type none pass val 0 index 1 ref 1 bind 0 installed 5 sec used 5 sec Action statistics: Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 cookie a1b2c3d4 .. bind the accept action to a filter.. sudo $TC filter add dev lo parent ffff: protocol ip prio 1 \ u32 match ip dst 127.0.0.1/32 flowid 1:1 action gact index 1 ... send some traffic.. $ ping 127.0.0.1 -c 3 PING 127.0.0.1 (127.0.0.1) 56(84) bytes of data. 64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.020 ms 64 bytes from 127.0.0.1: icmp_seq=2 ttl=64 time=0.027 ms 64 bytes from 127.0.0.1: icmp_seq=3 ttl=64 time=0.038 ms Signed-off-by:
-
- 24 Jan, 2017 24 commits
-
-
David S. Miller authored
Stephen Hemminger says: ==================== netvsc driver enhancements for net-next Lots of little things in here. Support for minor more ethtool control, negotiation of offload parameters with host (based on FreeBSD) and several cleanups. ==================== Signed-off-by: -
stephen hemminger authored
To improve performance, netvsc can call network stack directly and avoid the local backlog queue. This is safe since incoming packets are handled in softirq context already because the receive function callback is called from a tasklet. Signed-off-by:
Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by:
-
stephen hemminger authored
Use kernel for_each_clear_bit macro to simplify finding next available send section. Signed-off-by:
-
Simon Xiao authored
Report packets and bytes transferred through a vmbus channel via ethtool. This supersedes need for per-cpu statistics. Example: $ ethtool -S eth0 NIC statistics: ... tx_queue_0_packets: 3523179 tx_queue_0_bytes: 505370920 rx_queue_0_packets: 41430490 rx_queue_0_bytes: 62714661254 tx_queue_1_packets: 0 tx_queue_1_bytes: 0 rx_queue_1_packets: 0 rx_queue_1_bytes: 0 ... Reviewed-by:Long Li <longli@microsoft.com> Reviewed-by:
K. Y. Srinivasan <kys@microsoft.com> Reviewed-by:
Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by:
Simon Xiao <sixiao@microsoft.com> Signed-off-by:
-
stephen hemminger authored
Most drivers do not increment transmit statistics until after the transmit is completed. This will also be necessary for BQL support. Slight additional complexity because the netvsc driver aggregates multiple packets into one transmit. Signed-off-by:
-
stephen hemminger authored
Since now keep track of per-queue outstanding sends, we can avoid one atomic update by removing no longer needed per-device atomic. Signed-off-by:
-
stephen hemminger authored
All caller's already have pointer to netvsc_device so pass it. Signed-off-by:
-
stephen hemminger authored
All the caller's/callee's know that the format of the device_add parameter is a netvsc_device_info struct. Signed-off-by:
-
stephen hemminger authored
Do manual optimizations of receive path: - remove checks for impossible conditions (but keep checks for bad data from host) - pass argument down, rather than having callee recompute what is already known - remove indirection about receive buffer datalength - remove dependence on VLAN_TAG_PRESENCE - use _hot/_cold and likely/unlikely Signed-off-by: -
stephen hemminger authored
Put all the per-channel state together in one data struct. Signed-off-by:
-
stephen hemminger authored
Fixes set but never used warnings Signed-off-by:
-
stephen hemminger authored
The netvsc select queue function was missing many of the flow caching features that exist in default tx queue selection. Add the same logic to remember queue based on socket and implement two level mapping (like RSS). Signed-off-by:
-
stephen hemminger authored
Allow setting receive indirection table. Also uses the system standard for initialization. Signed-off-by:
-
stephen hemminger authored
This allows for number of channels to be managed in a manner similar to existing hardware drivers. It also removes the restriction of maximum 8 channels and allows as many as the host will allow. Signed-off-by:
-
stephen hemminger authored
For some cases it is useful to be able to change RSS key value. For example, replacing RSS key with a symmetric hash. Signed-off-by:
-
stephen hemminger authored
Report current components used in RSS hash. Signed-off-by:
-
stephen hemminger authored
Report actual number of receive queues to ethtool. Signed-off-by:
-
stephen hemminger authored
Redo how Hyper-V network driver negotiates offload features. Query the host to determine offload settings, and use the result. Also: * disable IPv4 header checksum offload (not used by Linux) * enable TSO only if host supports * enable UDP checksum offload if supported * don't advertise support for checksumming of non-IP protocols * adjust GSO maximum segment size * enable HIGHDMA Signed-off-by:
-
stephen hemminger authored
The ring buffer mapping now handles the wraparound case inside get_next_pkt_raw. Therefore it is not necessary to have an additional special receive staging buffer. See commit 1562edaed8c164ca5199 ("Drivers: hv: ring_buffer: count on wrap around mappings") Signed-off-by: -
David S. Miller authored
Andrew Lunn says: ==================== External MDIO support for mv88e6xxx The mv88e6390 family of switches has two MDIO busses, one internal to the switch and a second one for external usage. Older generations of switches have a single MDIO bus, which is used both internally and externally. Refactor the existing MDIO driver code to allow for multiple MDIO busses, and implement the second MDIO bus on mv88e6390. This is a rewrite of a patch previously submitted as part of "Batch 3". It has been broken up into 5 smaller patches. A compatible string is now used in the device tree to indicate the external MDIO bus. ==================== Signed-off-by: -
Andrew Lunn authored
With all the infrastructure in place, implement access to the external MDIO bus on the 6390 family. Signed-off-by:
-
Andrew Lunn authored
The mv88e6390 has multiple MDIO busses. Generalize the parsing of the device tree to support multiple mdio nodes. The external mdio bus has a compatible strings to indicate it is external. Keep a linked list of busses, placing the external mdio bus at the tail of the list. When within the driver an mdio bus is needed, e.g. for EEE or SERDES, use the head of the list which should be the internal bus. Signed-off-by:
-
Andrew Lunn authored
Have the MDIO bus driver code allocate a private structure and make the chip a member of it. This will allow us to add further members in the future. Signed-off-by:
-
Andrew Lunn authored
In preparation for supporting multiple MDIO busses, pass the mii_bus structure to all PHY operations. It will in future then be clear on which MDIO bus the operation should be performed. For reads/write from phylib, the mii_bus is readily available. However some internal code also access the PHY, e.g. for EEE and SERDES. Make this code use the one and only currently available MDIO bus. Signed-off-by:
-
