summaryrefslogtreecommitdiffhomepage
path: root/pkg/tcpip/transport
AgeCommit message (Collapse)Author
2018-10-10Enforce message size limits and avoid host calls with too many iovecsMichael Pratt
Currently, in the face of FileMem fragmentation and a large sendmsg or recvmsg call, host sockets may pass > 1024 iovecs to the host, which will immediately cause the host to return EMSGSIZE. When we detect this case, use a single intermediate buffer to pass to the kernel, copying to/from the src/dst buffer. To avoid creating unbounded intermediate buffers, enforce message size checks and truncation w.r.t. the send buffer size. The same functionality is added to netstack unix sockets for feature parity. PiperOrigin-RevId: 216590198 Change-Id: I719a32e71c7b1098d5097f35e6daf7dd5190eff7
2018-09-28Block for link address resolutionSepehr Raissian
Previously, if address resolution for UDP or Ping sockets required sending packets using Write in Transport layer, Resolve would return ErrWouldBlock and Write would return ErrNoLinkAddress. Meanwhile startAddressResolution would run in background. Further calls to Write using same address would also return ErrNoLinkAddress until resolution has been completed successfully. Since Write is not allowed to block and System Calls need to be interruptible in System Call layer, the caller to Write is responsible for blocking upon return of ErrWouldBlock. Now, when startAddressResolution is called a notification channel for the completion of the address resolution is returned. The channel will traverse up to the calling function of Write as well as ErrNoLinkAddress. Once address resolution is complete (success or not) the channel is closed. The caller would call Write again to send packets and check if address resolution was compeleted successfully or not. Fixes google/gvisor#5 Change-Id: Idafaf31982bee1915ca084da39ae7bd468cebd93 PiperOrigin-RevId: 214962200
2018-09-21Remove unnecessary deferIan Gudger
PiperOrigin-RevId: 214073949 Change-Id: I8fab916cd77362c13dac2c9dcf2ecc1710d87a5e
2018-09-19Fix data race on tcp.endpoint.hardError in tcp.(*endpoint).ReadIan Gudger
tcp.endpoint.hardError is protected by tcp.endpoint.mu. PiperOrigin-RevId: 213730698 Change-Id: I4e4f322ac272b145b500b1a652fbee0c7b985be2
2018-09-19Fix RTT estimation when timestamp option is enabled.Bhasker Hariharan
From RFC7323#Section-4 The [RFC6298] RTT estimator has weighting factors, alpha and beta, based on an implicit assumption that at most one RTTM will be sampled per RTT. When multiple RTTMs per RTT are available to update the RTT estimator, an implementation SHOULD try to adhere to the spirit of the history specified in [RFC6298]. An implementation suggestion is detailed in Appendix G. From RFC7323#appendix-G Appendix G. RTO Calculation Modification Taking multiple RTT samples per window would shorten the history calculated by the RTO mechanism in [RFC6298], and the below algorithm aims to maintain a similar history as originally intended by [RFC6298]. It is roughly known how many samples a congestion window worth of data will yield, not accounting for ACK compression, and ACK losses. Such events will result in more history of the path being reflected in the final value for RTO, and are uncritical. This modification will ensure that a similar amount of time is taken into account for the RTO estimation, regardless of how many samples are taken per window: ExpectedSamples = ceiling(FlightSize / (SMSS * 2)) alpha' = alpha / ExpectedSamples beta' = beta / ExpectedSamples Note that the factor 2 in ExpectedSamples is due to "Delayed ACKs". Instead of using alpha and beta in the algorithm of [RFC6298], use alpha' and beta' instead: RTTVAR <- (1 - beta') * RTTVAR + beta' * |SRTT - R'| SRTT <- (1 - alpha') * SRTT + alpha' * R' (for each sample R') PiperOrigin-RevId: 213644795 Change-Id: I52278b703540408938a8edb8c38be97b37f4a10e
2018-09-17Prevent TCP connect from picking bound portsTamir Duberstein
PiperOrigin-RevId: 213387851 Change-Id: Icc6850761bc11afd0525f34863acd77584155140
2018-09-14Pass buffer.Prependable by valueTamir Duberstein
PiperOrigin-RevId: 213053370 Change-Id: I60ea89572b4fca53fd126c870fcbde74fcf52562
2018-09-12Always pass buffer.VectorisedView by valueTamir Duberstein
PiperOrigin-RevId: 212757571 Change-Id: I04200df9e45c21eb64951cd2802532fa84afcb1a
2018-09-12Add multicast supportTamir Duberstein
PiperOrigin-RevId: 212750821 Change-Id: I822fd63e48c684b45fd91f9ce057867b7eceb792
2018-09-12Prevent UDP sockets from binding to bound portsTamir Duberstein
PiperOrigin-RevId: 212653818 Change-Id: Ib4e1d754d9cdddeaa428a066cb675e6ec44d91ad
2018-09-05Update {LinkEndpoint,NetworkEndpoint}#WritePacket to take a VectorisedViewBert Muthalaly
Makes it possible to avoid copying or allocating in cases where DeliverNetworkPacket (rx) needs to turn around and call WritePacket (tx) with its VectorisedView. Also removes the restriction on having VectorisedViews with multiple views in the write path. PiperOrigin-RevId: 211728717 Change-Id: Ie03a65ecb4e28bd15ebdb9c69f05eced18fdfcff
2018-09-05Implement TCP keepalivesTamir Duberstein
PiperOrigin-RevId: 211670620 Change-Id: Ia8a3d8ae53a7fece1dee08ee9c74964bd7f71bb7
2018-09-04Automated rollback of changelist 211156845Bhasker Hariharan
PiperOrigin-RevId: 211525182 Change-Id: I462c20328955c77ecc7bfd8ee803ac91f15858e6
2018-09-04Expose TCP RTTTamir Duberstein
PiperOrigin-RevId: 211504634 Change-Id: I9a7bcbbdd40e5036894930f709278725ef477293
2018-09-04Distinguish Element and Linker for ilist.Adin Scannell
Furthermore, allow for the specification of an ElementMapper. This allows a single "Element" type to exist on multiple inline lists, and work without having to embed the entry type. This is a requisite change for supporting a per-Inode list of Dirents. PiperOrigin-RevId: 211467497 Change-Id: If2768999b43e03fdaecf8ed15f435fe37518d163
2018-08-31Automated rollback of changelist 211103930Googler
PiperOrigin-RevId: 211156845 Change-Id: Ie28011d7eb5f45f3a0158dbee2a68c5edf22f6e0
2018-08-31ipv6: ICMP supportTamir Duberstein
This CL does NDP link-address discovery for IPv6. It includes several small changes necessary to get linux to talk to this implementation. In particular, a hop limit of 255 is necessary for ICMPv6. PiperOrigin-RevId: 211103930 Change-Id: If25370ab84c6b1decfb15de917f3b0020f2c4e0e
2018-08-29ping: update comment about UDPIan Gudger
PiperOrigin-RevId: 210788012 Change-Id: I5ebdcf3d02bfab3484a1374fbccba870c9d68954
2018-08-27Add various statisticsTamir Duberstein
PiperOrigin-RevId: 210442599 Change-Id: I9498351f461dc69c77b7f815d526c5693bec8e4a
2018-08-23netstack: make listening tcp socket close state setting and cleanup atomic.Zhaozhong Ni
Otherwise the socket saving logic might find workers still running for closed sockets unexpectedly. PiperOrigin-RevId: 210018905 Change-Id: I443a04d355613f5f9983252cc6863bff6e0eda3a
2018-08-23Encapsulate netstack metricsIan Gudger
PiperOrigin-RevId: 209943212 Change-Id: I96dcbc7c2ab2426e510b94a564436505256c5c79
2018-08-16Internal change.Googler
PiperOrigin-RevId: 209060862 Change-Id: I2cd02f0032b80d0087110095548b1a8ffa696ac2
2018-08-08Fix data race in unix.BoundEndpoint.UnidirectionalConnect.Ian Gudger
Data race is: Read: (*connectionlessEndpoint).UnidirectionalConnect: writeQueue: e.receiver.(*queueReceiver).readQueue, Write: (*connectionlessEndpoint).Close: e.receiver = nil The problem is that (*connectionlessEndpoint).UnidirectionalConnect assumed that baseEndpoint.receiver is immutable which is explicitly not the case. Fixing this required two changes: 1. Add synchronization around access of baseEndpoint.receiver in (*connectionlessEndpoint).UnidirectionalConnect. 2. Check for baseEndpoint.receiver being nil in (*connectionlessEndpoint).UnidirectionalConnect. PiperOrigin-RevId: 207984402 Change-Id: Icddeeb43805e777fa3ef874329fa704891d14181
2018-08-07Adds support to dump out cubic internal state.Bhasker Hariharan
PiperOrigin-RevId: 207754087 Change-Id: I83abce64348ea93f8692da81a881b364dae2158b
2018-08-07Netstack doesn't handle sending after SHUT_WR correctly.Brian Geffon
PiperOrigin-RevId: 207715032 Change-Id: I7b6690074c5be283145192895d706a92e921b22c
2018-08-03Cubic implementation for Netstack.Bhasker Hariharan
This CL implements CUBIC as described in https://tools.ietf.org/html/rfc8312. PiperOrigin-RevId: 207353142 Change-Id: I329cbf3277f91127e99e488f07d906f6779c6603
2018-08-02Automated rollback of changelist 207037226Zhaozhong Ni
PiperOrigin-RevId: 207125440 Change-Id: I6c572afb4d693ee72a0c458a988b0e96d191cd49
2018-08-01Move stack clock to options structIan Gudger
PiperOrigin-RevId: 207039273 Change-Id: Ib8f55a6dc302052ab4a10ccd70b07f0d73b373df
2018-08-01Automated rollback of changelist 207007153Michael Pratt
PiperOrigin-RevId: 207037226 Change-Id: I8b5f1a056d4f3eab17846f2e0193bb737ecb5428
2018-08-01stateify: convert all packages to use explicit mode.Zhaozhong Ni
PiperOrigin-RevId: 207007153 Change-Id: Ifedf1cc3758dc18be16647a4ece9c840c1c636c9
2018-07-30netstack: support disconnect-on-save option per fdbased link.Zhaozhong Ni
PiperOrigin-RevId: 206659972 Change-Id: I5e0e035f97743b6525ad36bed2c802791609beaf
2018-07-27stateify: support explicit annotation mode; convert refs and stack packages.Zhaozhong Ni
We have been unnecessarily creating too many savable types implicitly. PiperOrigin-RevId: 206334201 Change-Id: Idc5a3a14bfb7ee125c4f2bb2b1c53164e46f29a8
2018-07-23Refactor new reno congestion control logic out of sender.Bhasker Hariharan
This CL also puts the congestion control logic behind an interface so that we can easily swap it out for say CUBIC in the future. PiperOrigin-RevId: 205732848 Change-Id: I891cdfd17d4d126b658b5faa0c6bd6083187944b
2018-07-17netstack: update goroutine save / restore safety comments.Zhaozhong Ni
PiperOrigin-RevId: 204930314 Change-Id: Ifc4c41ed28616cd57fafbf7c92e87141a945c41f
2018-07-12netstack: save tcp endpoint accepted channel directly.Zhaozhong Ni
PiperOrigin-RevId: 204356873 Change-Id: I5e2f885f58678e693aae1a69e8bf8084a685af28
2018-07-12netstack: do not defer panicable logic in tcp main loop.Zhaozhong Ni
PiperOrigin-RevId: 204355026 Change-Id: I1a8229879ea3b58aa861a4eb4456fd7aff99863d
2018-07-10netstack: only do connected TCP S/R for loopback connections.Zhaozhong Ni
PiperOrigin-RevId: 204006237 Change-Id: Ica8402ab54d9dd7d11cc41c6d74aacef51d140b7
2018-07-10Internal changeMichael Pratt
PiperOrigin-RevId: 203997995 Change-Id: I8974fe74f1582bc9b2622f18a4bc4ab47ff5d622
2018-07-10netstack: tcp socket connected state S/R support.Zhaozhong Ni
PiperOrigin-RevId: 203958972 Change-Id: Ia6fe16547539296d48e2c6731edacdd96bd6e93c
2018-07-09Notify UDP and Ping endpoints on closeIan Gudger
PiperOrigin-RevId: 203883138 Change-Id: I7500c0a70f5d71c3fb37e2477f7fc466fa92fd3e
2018-07-09Fix two race conditions in tcp stack.Brian Geffon
PiperOrigin-RevId: 203880278 Change-Id: I66b790a616de59142859cc12db4781b57ea626d3
2018-07-09Switch netstack licenses to Apache 2.0.Nicolas Lacasse
Fixes #27 PiperOrigin-RevId: 203825288 Change-Id: Ie9f3a2b2c1e296b026b024f75c07da1a7e118633
2018-06-29Panic in netstack during cleanup where a FIN becomes a RST.Brian Geffon
There is a subtle bug where during cleanup with unread data a FIN can be converted to a RST, at that point the entire connection should be aborted as we're not expecting any ACKs to the RST. PiperOrigin-RevId: 202691271 Change-Id: Idae70800208ca26e07a379bc6b2b8090805d0a22
2018-06-26Fix data races in Unix socketsIan Gudger
PiperOrigin-RevId: 202175558 Change-Id: I0113cb9a90d7a0cd7964bf43eef67f70c92d9589
2018-06-26Automated rollback of changelist 201596247Brian Geffon
PiperOrigin-RevId: 202151720 Change-Id: I0491172c436bbb32b977f557953ba0bc41cfe299
2018-06-21netstack: tcp socket connected state S/R support.Zhaozhong Ni
PiperOrigin-RevId: 201596247 Change-Id: Id22f47b2cdcbe14aa0d930f7807ba75f91a56724
2018-06-15Replace crypto/rand with internal rand packageMichael Pratt
PiperOrigin-RevId: 200784607 Change-Id: I39aa6ee632936dcbb00fc298adccffa606e9f4c0
2018-06-08Add a protocol option to set congestion control algorithm.Bhasker Hariharan
Also adds support to query available congestion control algorithms. PiperOrigin-RevId: 199826897 Change-Id: I2b338b709820ee9cf58bb56d83aa7b1a39f4eab2
2018-06-05netstack: make TCP endpoint closed and error state cleanup work synchronous.Zhaozhong Ni
So that when saving TCP endpoint in these states, there is no pending or background activities. Also lift tcp network save rejection error to tcpip package. PiperOrigin-RevId: 199370748 Change-Id: Ief7b45c2a7338d12414cd7c23db95de6a9c22700
2018-05-29Automated rollback of changelist 196886839Fabricio Voznika
PiperOrigin-RevId: 198457660 Change-Id: I6ea5cf0b4cfe2b5ba455325a7e5299880e5a088a
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-12 01:36:18 +0000