summaryrefslogtreecommitdiffhomepage
path: root/pkg/tcpip/transport
AgeCommit message (Collapse)Author
2019-10-28Use the user supplied TCP MSS when creating a new active socketGhanan Gowripalan
This change supports using a user supplied TCP MSS for new active TCP connections. Note, the user supplied MSS must be less than or equal to the maximum possible MSS for a TCP connection's route. If it is greater than the maximum possible MSS, the maximum possible MSS will be used as the connection's MSS instead. This change does not use this user supplied MSS for connections accepted from listening sockets - that will come in a later change. Test: Test that outgoing TCP SYN segments contain a TCP MSS option with the user supplied MSS if it is not greater than the maximum possible MSS for the route. PiperOrigin-RevId: 277185125
2019-10-25Validate the checksum for incoming ICMPv6 packetsGhanan Gowripalan
This change validates the ICMPv6 checksum field before further processing an ICMPv6 packet. Tests: Unittests to make sure that only ICMPv6 packets with a valid checksum are accepted/processed. Existing tests using checker.ICMPv6 now also check the ICMPv6 checksum field. PiperOrigin-RevId: 276779148
2019-10-25Convert DelayOption to the newer/faster SockOpt int type.Ian Gudger
DelayOption is set on all new endpoints in gVisor. PiperOrigin-RevId: 276746791
2019-10-24Remove the amss field from tcpip.tcp.handshake as it was unusedGhanan Gowripalan
The amss field in the tcpip.tcp.handshake was not used anywhere. Removed it to not cause confusion with the amss field in the tcpip.tcp.endpoint struct, which was documented to be used (and is actually being used) for the same purpose. PiperOrigin-RevId: 276577088
2019-10-23Merge pull request #641 from tanjianfeng:mastergVisor bot
PiperOrigin-RevId: 276380008
2019-10-22netstack/tcp: software segmentation offloadAndrei Vagin
Right now, we send each tcp packet separately, we call one system call per-packet. This patch allows to generate multiple tcp packets and send them by sendmmsg. The arguable part of this CL is a way how to handle multiple headers. This CL adds the next field to the Prepandable buffer. Nginx test results: Server Software: nginx/1.15.9 Server Hostname: 10.138.0.2 Server Port: 8080 Document Path: /10m.txt Document Length: 10485760 bytes w/o gso: Concurrency Level: 5 Time taken for tests: 5.491 seconds Complete requests: 100 Failed requests: 0 Total transferred: 1048600200 bytes HTML transferred: 1048576000 bytes Requests per second: 18.21 [#/sec] (mean) Time per request: 274.525 [ms] (mean) Time per request: 54.905 [ms] (mean, across all concurrent requests) Transfer rate: 186508.03 [Kbytes/sec] received sw-gso: Concurrency Level: 5 Time taken for tests: 3.852 seconds Complete requests: 100 Failed requests: 0 Total transferred: 1048600200 bytes HTML transferred: 1048576000 bytes Requests per second: 25.96 [#/sec] (mean) Time per request: 192.576 [ms] (mean) Time per request: 38.515 [ms] (mean, across all concurrent requests) Transfer rate: 265874.92 [Kbytes/sec] received w/o gso: $ ./tcp_benchmark --client --duration 15 --ideal [SUM] 0.0-15.1 sec 2.20 GBytes 1.25 Gbits/sec software gso: $ tcp_benchmark --client --duration 15 --ideal --gso $((1<<16)) --swgso [SUM] 0.0-15.1 sec 3.99 GBytes 2.26 Gbits/sec PiperOrigin-RevId: 276112677
2019-10-21AF_PACKET support for netstack (aka epsocket).Kevin Krakauer
Like (AF_INET, SOCK_RAW) sockets, AF_PACKET sockets require CAP_NET_RAW. With runsc, you'll need to pass `--net-raw=true` to enable them. Binding isn't supported yet. PiperOrigin-RevId: 275909366
2019-10-18Fix typo while initializing protocol for UDP endpoints.Mithun Iyer
Fixes #763 PiperOrigin-RevId: 275563222
2019-10-15epsocket: support /proc/net/snmpJianfeng Tan
Netstack has its own stats, we use this to fill /proc/net/snmp. Note that some metrics are not recorded in Netstack, which will be shown as 0 in the proc file. Signed-off-by: Jianfeng Tan <henry.tjf@antfin.com> Change-Id: Ie0089184507d16f49bc0057b4b0482094417ebe1
2019-10-15netstack: add counters for tcp CurrEstab and EstabResetsJianfeng Tan
Signed-off-by: Jianfeng Tan <henry.tjf@antfin.com>
2019-10-14Internal change.gVisor bot
PiperOrigin-RevId: 274700093
2019-10-14Reorder BUILD license and load functions in netstack.Kevin Krakauer
PiperOrigin-RevId: 274672346
2019-10-09Internal change.gVisor bot
PiperOrigin-RevId: 273861936
2019-10-07Implement IP_TTL.Ian Gudger
Also change the default TTL to 64 to match Linux. PiperOrigin-RevId: 273430341
2019-10-03Implement proper local broadcast behaviorChris Kuiper
The behavior for sending and receiving local broadcast (255.255.255.255) traffic is as follows: Outgoing -------- * A broadcast packet sent on a socket that is bound to an interface goes out that interface * A broadcast packet sent on an unbound socket follows the route table to select the outgoing interface + if an explicit route entry exists for 255.255.255.255/32, use that one + else use the default route * Broadcast packets are looped back and delivered following the rules for incoming packets (see next). This is the same behavior as for multicast packets, except that it cannot be disabled via sockopt. Incoming -------- * Sockets wishing to receive broadcast packets must bind to either INADDR_ANY (0.0.0.0) or INADDR_BROADCAST (255.255.255.255). No other socket receives broadcast packets. * Broadcast packets are multiplexed to all sockets matching it. This is the same behavior as for multicast packets. * A socket can bind to 255.255.255.255:<port> and then receive its own broadcast packets sent to 255.255.255.255:<port> In addition, this change implicitly fixes an issue with multicast reception. If two sockets want to receive a given multicast stream and one is bound to ANY while the other is bound to the multicast address, only one of them will receive the traffic. PiperOrigin-RevId: 272792377
2019-09-30Fix bugs in PickEphemeralPort for TCP.Bhasker Hariharan
Netstack always picks a random start point everytime PickEphemeralPort is called. While this is required for UDP so that DNS requests go out through a randomized set of ports it is not required for TCP. Infact Linux explicitly hashes the (srcip, dstip, dstport) and a one time secret initialized at start of the application to get a random offset. But to ensure it doesn't start from the same point on every scan it uses a static hint that is incremented by 2 in every call to pick ephemeral ports. The reason for 2 is Linux seems to split the port ranges where active connects seem to use even ones while odd ones are used by listening sockets. This CL implements a similar strategy where we use a hash + hint to generate the offset to start the search for a free Ephemeral port. This ensures that we cycle through the available port space in order for repeated connects to the same destination and significantly reduces the chance of picking a recently released port. PiperOrigin-RevId: 272058370
2019-09-27Implement SO_BINDTODEVICE sockoptgVisor bot
PiperOrigin-RevId: 271644926
2019-09-25Remove centralized registration of protocols.Kevin Krakauer
Also removes the need for protocol names. PiperOrigin-RevId: 271186030
2019-09-23netstack: convert more socket options to {Set,Get}SockOptIntAndrei Vagin
PiperOrigin-RevId: 270763208
2019-09-12Implement splice methods for pipes and sockets.Adin Scannell
This also allows the tee(2) implementation to be enabled, since dup can now be properly supported via WriteTo. Note that this change necessitated some minor restructoring with the fs.FileOperations splice methods. If the *fs.File is passed through directly, then only public API methods are accessible, which will deadlock immediately since the locking is already done by fs.Splice. Instead, we pass through an abstract io.Reader or io.Writer, which elide locks and use the underlying fs.FileOperations directly. PiperOrigin-RevId: 268805207
2019-09-12Remove go_test from go_stateify and go_marshalMichael Pratt
They are no-ops, so the standard rule works fine. PiperOrigin-RevId: 268776264
2019-09-09Fix ephemeral port leak.Ian Gudger
Fix a bug where udp.(*endpoint).Disconnect [accessible in gVisor via epsocket.(*SocketOperations).Connect with AF_UNSPEC] would leak a port reservation if the socket/endpoint had an ephemeral port assigned to it. glibc's getaddrinfo uses connect with AF_UNSPEC, causing each call of getaddrinfo to leak a port. Call getaddrinfo too many times and you run out of ports (shows up as connect returning EAGAIN and getaddrinfo returning EAI_NONAME "Name or service not known"). PiperOrigin-RevId: 268071160
2019-09-06Remove reundant global tcpip.LinkEndpointID.Ian Gudger
PiperOrigin-RevId: 267709597
2019-09-04Fix RST generation bugs.Bhasker Hariharan
There are a few cases addressed by this change - We no longer generate a RST in response to a RST packet. - When we receive a RST we cleanup and release all reservations immediately as the connection is now aborted. - An ACK received by a listening socket generates a RST when SYN cookies are not in-use. The only reason an ACK should land at the listening socket is if we are using SYN cookies otherwise the goroutine for the handshake in progress should have gotten the packet and it should never have arrived at the listening endpoint. - Also fixes the error returned when a connection times out due to a Keepalive timer expiration from ECONNRESET to a ETIMEDOUT. PiperOrigin-RevId: 267238427
2019-09-03Make UDP traceroute work.Bhasker Hariharan
Adds support to generate Port Unreachable messages for UDP datagrams received on a port for which there is no valid endpoint. Fixes #703 PiperOrigin-RevId: 267034418
2019-08-29Implement /proc/net/udp.Rahat Mahmood
PiperOrigin-RevId: 266229756
2019-08-26netstack/tcp: Add LastAck transition.Rahat Mahmood
Add missing state transition to LastAck, which should happen when the endpoint has already recieved a FIN from the remote side, and is sending its own FIN. PiperOrigin-RevId: 265568314
2019-08-21Support binding to multicast and broadcast addressesChris Kuiper
This fixes the issue of not being able to bind to either a multicast or broadcast address as well as to send and receive data from it. The way to solve this is to treat these addresses similar to the ANY address and register their transport endpoint ID with the global stack's demuxer rather than the NIC's. That way there is no need to require an endpoint with that multicast or broadcast address. The stack's demuxer is in fact the only correct one to use, because neither broadcast- nor multicast-bound sockets care which NIC a packet was received on (for multicast a join is still needed to receive packets on a NIC). I also took the liberty of refactoring udp_test.go to consolidate a lot of duplicate code and make it easier to create repetitive tests that test the same feature for a variety of packet and socket types. For this purpose I created a "flowType" that represents two things: 1) the type of packet being sent or received and 2) the type of socket used for the test. E.g., a "multicastV4in6" flow represents a V4-mapped multicast packet run through a V6-dual socket. This allows writing significantly simpler tests. A nice example is testTTL(). PiperOrigin-RevId: 264766909
2019-08-21Use tcpip.Subnet in tcpip.RouteTamir Duberstein
This is the first step in replacing some of the redundant types with the standard library equivalents. PiperOrigin-RevId: 264706552
2019-08-16netstack: disconnect an unix socket only if the address family is AF_UNSPECAndrei Vagin
Linux allows to call connect for ANY and the zero port. PiperOrigin-RevId: 263892534
2019-08-15Don't dereference errors passed to panic()Tamir Duberstein
These errors are always pointers; there's no sense in dereferencing them in the panic call. Changed one false positive for clarity. PiperOrigin-RevId: 263611579
2019-08-15netstack: move resumption logic into *_state.goTamir Duberstein
13a98df rearranged some of this code in a way that broke compilation of the netstack-only export at github.com/google/netstack because *_state.go files are not included in that export. This commit moves resumption logic back into *_state.go, fixing the compilation breakage. PiperOrigin-RevId: 263601629
2019-08-14Replace uinptr with int64 when returning lengthsTamir Duberstein
This is in accordance with newer parts of the standard library. PiperOrigin-RevId: 263449916
2019-08-14Improve SendMsg performance.Bhasker Hariharan
SendMsg before this change would copy all the data over into a new slice even if the underlying socket could only accept a small amount of data. This is really inefficient with non-blocking sockets and under high throughput where large writes could get ErrWouldBlock or if there was say a timeout associated with the sendmsg() syscall. With this change we delay copying bytes in till they are needed and only copy what can be potentially sent/held in the socket buffer. Reducing the need to repeatedly copy data over. Also a minor fix to change state FIN-WAIT-1 when shutdown(..., SHUT_WR) is called instead of when we transmit the actual FIN. Otherwise the socket could remain in CONNECTED state even though the user has called shutdown() on the socket. Updates #627 PiperOrigin-RevId: 263430505
2019-08-09Add congestion control states to sender.Bhasker Hariharan
This change just introduces different congestion control states and ensures the sender.state is updated to reflect the current state of the connection. It is not used for any decisions yet but this is required before algorithms like Eiffel/PRR can be implemented. Fixes #394 PiperOrigin-RevId: 262638292
2019-08-08netstack: Don't start endpoint goroutines too soon on restore.Rahat Mahmood
Endpoint protocol goroutines were previously started as part of loading the endpoint. This is potentially too soon, as resources used by these goroutine may not have been loaded. Protocol goroutines may perform meaningful work as soon as they're started (ex: incoming connect) which can cause them to indirectly access resources that haven't been loaded yet. This CL defers resuming all protocol goroutines until the end of restore. PiperOrigin-RevId: 262409429
2019-08-06Fix for a panic due to writing to a closed accept channel.Bhasker Hariharan
This can happen because endpoint.Close() closes the accept channel first and then drains/resets any accepted but not delivered connections. But there can be connections that are connected but not delivered to the channel as the channel was full. But closing the channel can cause these writes to fail with a write to a closed channel. The correct solution is to abort any connections in SYN-RCVD state and drain/abort all completed connections before closing the accept channel. PiperOrigin-RevId: 261951132
2019-08-02Plumbing for iptables sockopts.Kevin Krakauer
PiperOrigin-RevId: 261413396
2019-08-02Automated rollback of changelist 261191548Rahat Mahmood
PiperOrigin-RevId: 261373749
2019-08-01Implement getsockopt(TCP_INFO).Rahat Mahmood
Export some readily-available fields for TCP_INFO and stub out the rest. PiperOrigin-RevId: 261191548
2019-07-31Test connecting UDP sockets to the ANY addressTamir Duberstein
This doesn't currently pass on gVisor. While I'm here, fix a bug where connecting to the v6-mapped v4 address doesn't work in gVisor. PiperOrigin-RevId: 260923961
2019-07-24Add support for a subnet prefix length on interface network addressesChris Kuiper
This allows the user code to add a network address with a subnet prefix length. The prefix length value is stored in the network endpoint and provided back to the user in the ProtocolAddress type. PiperOrigin-RevId: 259807693
2019-07-23Deduplicate EndpointState.connected someTamir Duberstein
This fixes a bug introduced in cl/251934850 that caused connect-accept-close-connect races to result in the second connect call failiing when it should have succeeded. PiperOrigin-RevId: 259584525
2019-07-19Handle interfaceAddr and NIC options separately for IP_MULTICAST_IFChris Kuiper
This tweaks the handling code for IP_MULTICAST_IF to ignore the InterfaceAddr if a NICID is given. PiperOrigin-RevId: 258982541
2019-07-18net/tcp/setockopt: impelment setsockopt(fd, SOL_TCP, TCP_INQ)Andrei Vagin
PiperOrigin-RevId: 258859507
2019-07-16Internal change.gVisor bot
PiperOrigin-RevId: 258424489
2019-07-12Add IPPROTO_RAW, which allows raw sockets to write IP headers.Kevin Krakauer
iptables also relies on IPPROTO_RAW in a way. It opens such a socket to manipulate the kernel's tables, but it doesn't actually use any of the functionality. Blegh. PiperOrigin-RevId: 257903078
2019-07-12Stub out support for TCP_MAXSEG.Bhasker Hariharan
Adds support to set/get the TCP_MAXSEG value but does not really change the segment sizes emitted by netstack or alter the MSS advertised by the endpoint. This is currently being added only to unblock iperf3 on gVisor. Plumbing this correctly requires a bit more work which will come in separate CLs. PiperOrigin-RevId: 257859112
2019-07-03netstack/udp: connect with the AF_UNSPEC address family means disconnectAndrei Vagin
PiperOrigin-RevId: 256433283
2019-06-27Fix various spelling issues in the documentationMichael Pratt
Addresses obvious typos, in the documentation only. COPYBARA_INTEGRATE_REVIEW=https://github.com/google/gvisor/pull/443 from Pixep:fix/documentation-spelling 4d0688164eafaf0b3010e5f4824b35d1e7176d65 PiperOrigin-RevId: 255477779