This requires two changes:
1) Support for more than one socket to join a given multicast group.
2) Duplicate delivery of incoming multicast packets to all sockets listening
for it.
In addition, I tweaked the code (and added a test) to disallow duplicates
IP_ADD_MEMBERSHIP calls for the same group and NIC. This is how Linux does
it.
PiperOrigin-RevId: 246437315
Change-Id: Icad8300b4a8c3f501d9b4cd283bd3beabef88b72
Based on the guidelines at
https://opensource.google.com/docs/releasing/authors/.
1. $ rg -l "Google LLC" | xargs sed -i 's/Google LLC.*/The gVisor Authors./'
2. Manual fixup of "Google Inc" references.
3. Add AUTHORS file. Authors may request to be added to this file.
4. Point netstack AUTHORS to gVisor AUTHORS. Drop CONTRIBUTORS.
Fixes#209
PiperOrigin-RevId: 245823212
Change-Id: I64530b24ad021a7d683137459cafc510f5ee1de9
Add the CloseRead & CloseWrite methods that performs shutdown on the
corresponding Read & Write sides of a connection.
Change-Id: I3996a2abdc7cd68a2becba44dc4bd9f0919d2ce1
PiperOrigin-RevId: 245537950
This CL fixes the following bugs:
- Uses atomic to set/read status instead of binary.LittleEndian.PutUint32 etc
which are not atomic.
- Increments ringOffsets for frames that are truncated (i.e status is
tpStatusCopy)
- Does not ignore frames with tpStatusLost bit set as they are valid frames and
only indicate that there some frames were lost before this one and metrics can
be retrieved with a getsockopt call.
- Adds checks to make sure blockSize is a multiple of page size. This is
required as the kernel allocates in pages per block and rejects sizes that are
not page aligned with an EINVAL.
Updates #210
PiperOrigin-RevId: 244959464
Change-Id: I5d61337b7e4c0f8a3063dcfc07791d4c4521ba1f
Support shutdown on only the read side of an endpoint. Reads performed
after a call to Shutdown with only the ShutdownRead flag will return
ErrClosedForReceive without data.
Break out the shutdown(2) with SHUT_RD syscall test into to two tests.
The first tests that no packets are sent when shutting down the read
side of a socket. The second tests that, after shutting down the read
side of a socket, unread data can still be read, or an EOF if there is
no more data to read.
Change-Id: I9d7c0a06937909cbb466b7591544a4bcaebb11ce
PiperOrigin-RevId: 244459430
It is possible to create a listening socket which will accept
IPv4 and IPv6 connections. In this case, we set IPv6ProtocolNumber
for all accepted endpoints, even if they handle IPv4 connections.
This means that we can't use endpoint.netProto to set gso.L3HdrLen.
PiperOrigin-RevId: 244227948
Change-Id: I5e1863596cb9f3d216febacdb7dc75651882eef1
Having raw socket code together will make it easier to add support for other raw
network protocols. Currently, only ICMP uses the raw endpoint. However, adding
support for other protocols such as UDP shouldn't be much more difficult than
adding a few switch cases.
PiperOrigin-RevId: 241564875
Change-Id: I77e03adafe4ce0fd29ba2d5dfdc547d2ae8f25bf
The panic was caused by modifying the tree while iterating which invalidated the
iterator.
Also fixes another bug in SACKScoreboard.Insert() which was causing blocks to be
merged incorrectly.
PiperOrigin-RevId: 240895053
Change-Id: Ia72b8244297962df5c04283346da5226434740af
The linux packet socket can handle GSO packets, so we can segment packets to
64K instead of the MTU which is usually 1500.
Here are numbers for the nginx-1m test:
runsc: 579330.01 [Kbytes/sec] received
runsc-gso: 1794121.66 [Kbytes/sec] received
runc: 2122139.06 [Kbytes/sec] received
and for tcp_benchmark:
$ tcp_benchmark --duration 15 --ideal
[ 4] 0.0-15.0 sec 86647 MBytes 48456 Mbits/sec
$ tcp_benchmark --client --duration 15 --ideal
[ 4] 0.0-15.0 sec 2173 MBytes 1214 Mbits/sec
$ tcp_benchmark --client --duration 15 --ideal --gso 65536
[ 4] 0.0-15.0 sec 19357 MBytes 10825 Mbits/sec
PiperOrigin-RevId: 240809103
Change-Id: I2637f104db28b5d4c64e1e766c610162a195775a
This is a preparation for GSO changes (cl/234508902).
RELNOTES[gofers]: Refactor checksum code to include length, which
it already did, but in a convoluted way. Should be a no-op.
PiperOrigin-RevId: 240460794
Change-Id: I537381bc670b5a9f5d70a87aa3eb7252e8f5ace2
Previous memory allocation was excessive (80 MB). Changed
it to use 2 MB instead. There is no drop in perfomance due
to this change:
ab -n 100 -c 10 http://server/latin10m.txt ==> 10 MB file
80 MB: 178 MB/s
2 MB: 181 MB/s
PiperOrigin-RevId: 238321594
Change-Id: I1c8aed13cad5d75f4506d2b406b305117055fbe5
gonet.PacketConn now implements net.Conn, allowing it to be returned from
net.Dial.Dialer functions.
PiperOrigin-RevId: 238111980
Change-Id: I174884385ff4d9b8e9918fac7bbb5b93ca366ba7
HandleLocal is very similar conceptually to MULTICAST_LOOP, so we can unify
the implementations. This has the benefit of making HandleLocal apply even when
the fdbased link endpoint isn't in use.
In addition, move looping logic to route creation so that it doesn't need to be
run for each packet. This should improve performance.
PiperOrigin-RevId: 238099480
Change-Id: I72839f16f25310471453bc9d3fb8544815b25c23
IP_MULTICAST_LOOP controls whether or not multicast packets sent on the default
route are looped back. In order to implement this switch, support for sending
and looping back multicast packets on the default route had to be implemented.
For now we only support IPv4 multicast.
PiperOrigin-RevId: 237534603
Change-Id: I490ac7ff8e8ebef417c7eb049a919c29d156ac1c
...in accordance with RFCs 1112 and 2464.
Fixes IPv4 multicast when IP_MULTICAST_IF is specified.
Don't return ErrNoRoute when no route is needed.
Don't set Route.NextHop when no route is needed.
PiperOrigin-RevId: 236199813
Change-Id: I48ed33e1b7f760deaa37e18ad7f1b8b62819ab43
Broadly, this change:
* Enables sockets to be created via `socket(AF_INET, SOCK_RAW, IPPROTO_ICMP)`.
* Passes the network-layer (IP) header up the stack to the transport endpoint,
which can pass it up to the socket layer. This allows a raw socket to return
the entire IP packet to users.
* Adds functions to stack.TransportProtocol, stack.Stack, stack.transportDemuxer
that enable incoming packets to be delivered to raw endpoints. New raw sockets
of other protocols (not ICMP) just need to register with the stack.
* Enables ping.endpoint to return IP headers when created via SOCK_RAW.
PiperOrigin-RevId: 235993280
Change-Id: I60ed994f5ff18b2cbd79f063a7fdf15d093d845a
Also exposes ipv4.MaxTotalSize since it is a generally useful constant.
PiperOrigin-RevId: 235799755
Change-Id: I1fa8d5294bf355acf5527cfdf274b3687d3c8b13
An earlier CL excessively minimizes the period in which it
holds a lock on NIC. This earlier CL had done this out of
the mistaken impression it fixed a broken test, when in
fact it just reduced the rate of failure of a flaky test
in tcp_test.go. This new change holds the lock on NIC
for the duration of the loop over n.endpoints.
PiperOrigin-RevId: 235732487
Change-Id: I53ee6df264f093ddc4d29e9acdcba6b4838cb112
This change does not make use of SACK information but adds support to track
SACK information and store it in the endpoint.
The actual SACK based recovery will be in a separate CL.
Part of commits to add RFC 6675 support to Netstack.
PiperOrigin-RevId: 235612264
Change-Id: I261f94844d7bad5abda803152ce6cc6125a467ff
cl/234850781 introduced a race condition in NIC.DeliverNetworkPacket
by failing to hold a lock. This change fixes this regressesion by acquiring
a read lock before iterating through n.endpoints, and then releasing the lock
once iteration is complete.
PiperOrigin-RevId: 235549770
Change-Id: Ib0133288be512d478cf759c3314dc95ec3205d4b
This change adds support for the SO_BROADCAST socket option in gVisor Netstack.
This support includes getsockopt()/setsockopt() functionality for both UDP and
TCP endpoints (the latter being a NOOP), dispatching broadcast messages up and
down the stack, and route finding/creation for broadcast packets. Finally, a
suite of tests have been implemented, exercising this functionality through the
Linux syscall API.
PiperOrigin-RevId: 234850781
Change-Id: If3e666666917d39f55083741c78314a06defb26c
tcp_proxy now uses an AF_PACKET socket as the FD for netstack link layer
endpoint instead of a tap device. It also changes the link layer endpoint to use
PacketMMap dispatch instead of Readv. This reduces overall cpu and reflects the
current runsc setup which uses PacketMMap and also uses veth devices to receive
packets.
Also fixed a bug in gonet where Read() was not doing coalescing read and would
read small amounts at a time.
PiperOrigin-RevId: 234714768
Change-Id: Idabf8e600e4512489d3ba441c4096dc74deba5d7
This allows setting a default send interface for IPv4 multicast. IPv6 support
will come later.
PiperOrigin-RevId: 234251379
Change-Id: I65922341cd8b8880f690fae3eeb7ddfa47c8c173
SO_TIMESTAMP is reimplemented in ping and UDP sockets (and needs to be added for
TCP), but can just be implemented in epsocket for simplicity. This will also
make SIOCGSTAMP easier to implement.
PiperOrigin-RevId: 234179300
Change-Id: Ib5ea0b1261dc218c1a8b15a65775de0050fe3230
PACKET_RX_RING allows the use of an mmapped buffer to receive packets from the
kernel. This should cut down the number of host syscalls that need to be made
to receive packets when the underlying fd is a socket of the AF_PACKET type.
PiperOrigin-RevId: 233834998
Change-Id: I8060025c6ced206986e94cc46b8f382b81bfa47f
RFC7323 recommends that if the timestamp option was negotiated
then all packets should carry a TCP Timestamp and any packets that
do not should be dropped.
Netstack implemented this behaviour. Linux OTOH does not and will
accept such packets. This change makes Netstack behaviour compatible
with Linux.
Also now that we allow such packets, we do need to update RTO calculations
based on these packets even if timestamp option is enabled.
PiperOrigin-RevId: 233432268
Change-Id: I9f4742ae6b63930ac3b5e37d8c238761e6a4b29f
Also includes a few fixes for IPv4 multicast support. IPv6 support is coming in
a followup CL.
PiperOrigin-RevId: 233008638
Change-Id: If7dae6222fef43fda48033f0292af77832d95e82
Nothing reads them and they can simply get stale.
Generated with:
$ sed -i "s/licenses(\(.*\)).*/licenses(\1)/" **/BUILD
PiperOrigin-RevId: 231818945
Change-Id: Ibc3f9838546b7e94f13f217060d31f4ada9d4bf0
quoting what "rscheff@gmx.at" pointed out over email.
"IsLost in RFC3517 is defined as >= (DupThresh * SMSS) while
RFC6675 improves upon this, and defines IsLost as >
((DupThresh - 1) * SMSS + 1).
The latter addresses situations where partial segments (size < MSS)
are sent (eg. last segment of a http protocol message sent with PSH
being less than MSS is common)."
PiperOrigin-RevId: 231512331
Change-Id: I1addd4a92e3e7baeb0bdda46463ebfae435da958
This should reduce the number of syscalls required to process packets
significantly and improve throughputs.
PiperOrigin-RevId: 231366886
Change-Id: I8b38077262bf9c53176bc4a94b530188d3d7c0ca
This option allows multiple sockets to be bound to the same port.
Incoming packets are distributed to sockets using a hash based on source and
destination addresses. This means that all packets from one sender will be
received by the same server socket.
PiperOrigin-RevId: 227153413
Change-Id: I59b6edda9c2209d5b8968671e9129adb675920cf
We don't explicitly support out-of-band data and treat it like normal in-band
data. This is equilivent to SO_OOBINLINE being enabled, so always report that
it is enabled.
PiperOrigin-RevId: 226572742
Change-Id: I4c30ccb83265e76c30dea631cbf86822e6ee1c1b
Within gVisor, plumb new socket options to netstack.
Within netstack, fix GetSockOpt and SetSockOpt return value logic.
PiperOrigin-RevId: 226532229
Change-Id: If40734e119eed633335f40b4c26facbebc791c74
Same as with broadcast packets, sending of a multicast packet shouldn't require
accessing the route table. The same applies to IPv6 link-local addresses, which
aren't routable at all (they don't belong to any subnet by definition).
PiperOrigin-RevId: 225775870
Change-Id: Ic53e6560c125a83be2be9c3d112e66b36e8dfe7b
Currently sending a broadcast packet (for DHCP, e.g.) requires a "default
route" of the format "0.0.0.0/0 via 0.0.0.0 <intf>". There is no good reason
for this and on devices with several ports this creates a rather akward route
table with lots of such default routes (which defeats the purpose of a default
route).
PiperOrigin-RevId: 224378769
Change-Id: Icd7ec8a206eb08083cff9a837f6f9ab231c73a19
This removes code that should have never made it in in the first place, but did so due to incomplete testing. With the new tests the original code fails, the new code passes.
PiperOrigin-RevId: 224086966
Change-Id: I646fef76977f4528f3705f497b95fad6b3ec32bc
The signature of time.now has remained unchanged:
c2412a7681/src/time/time.go (L1072)
PiperOrigin-RevId: 224061160
Change-Id: Ic84bd6ee8fb9952cd9ab580bcb0892444ce7c2da
...to (remote, local), reflecting the (correct) names in the implementation of
DeliverNetworkPacket (see tcpip/stack/nic.go).
Also trim the names in DeliverNetworkPacket and elsewhere to avoid stuttering;
since the type is tcpip.LinkAddress, there's no need to include "LinkAddr" in
the parameter names.
Note that every callsite passes arguments in the order (src, dst).
PiperOrigin-RevId: 221514396
Change-Id: I3637454ad0d6e62a19e4dcbc2a16493798bd0f09
Previously, TCP_NODELAY was always enabled and we would lie about it being
configurable. TCP_NODELAY is now disabled by default (to match Linux) in the
socket layer so that non-gVisor users don't automatically start using this
questionable optimization.
PiperOrigin-RevId: 221368472
Change-Id: Ib0240f66d94455081f4e0ca94f09d9338b2c1356
Increase timeout to prevent the entry from being
found when there is delay on the address resolution
goroutine that doesn't mark the request as failed.
PiperOrigin-RevId: 220504789
Change-Id: I7e44fd95d8624bd69962f862fbf5517a81395f2a
This field was added in the intial implementation, before Route existed
to pass the local and remote addresses to the packet-writing path.
Today, the Route's members should be respected. A similar bug was
previously fixed in 214650822.
PiperOrigin-RevId: 219474095
Change-Id: Id2a8ee4421d2841c8d88ccb3c193c455086350ee
The channels {cancel,resCh} have roughly the same lifetime and are used for
roughly the same purpose as an entry's waiters; we can unify the state
management of the two mechanisms, while also reducing unncessary mutex locking
and unlocking.
Made some cosmetic changes while I'm here.
PiperOrigin-RevId: 218343915
Change-Id: Ic69546a2b7b390162b2231f07f335dd6199472d7
This change also adds extensive testing to the p9 package via mocks. The sanity
checks and type checks are moved from the gofer into the core package, where
they can be more easily validated.
PiperOrigin-RevId: 218296768
Change-Id: I4fc3c326e7bf1e0e140a454cbacbcc6fd617ab55
* Integrate recvMsg and sendMsg functions into Recv and Send respectively as
they are no longer shared.
* Clean up partial read/write error handling code.
* Re-order code to make sense given that there is no longer a host.endpoint
type.
PiperOrigin-RevId: 217255072
Change-Id: Ib43fe9286452f813b8309d969be11f5fa40694cd
host.endpoint contained duplicated logic from the sockerpair implementation and
host.ConnectedEndpoint. Remove host.endpoint in favor of a
host.ConnectedEndpoint wrapped in a socketpair end.
PiperOrigin-RevId: 217240096
Change-Id: I4a3d51e3fe82bdf30e2d0152458b8499ab4c987c
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
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.
Fixesgoogle/gvisor#5
Change-Id: Idafaf31982bee1915ca084da39ae7bd468cebd93
PiperOrigin-RevId: 214962200
There is a subtle bug that is the result of two changes made when upstreaming
ICMPv6 support from Fuchsia:
1) ipv6.endpoint.WritePacket writes the local address it was initialized with,
rather than the provided route's local address
2) ipv6.endpoint.handleICMP doesn't set its route's local address to the ICMP
target address before writing the response
The result is that the ICMP response erroneously uses the target ipv6 address
(rather than icmp) as its source address in the response. When trying to debug
this by fixing (2), we ran into problems with bad ipv6 checksums because (1)
didn't respect the local address of the route being passed to it.
This fixes both problems.
PiperOrigin-RevId: 214650822
Change-Id: Ib6148bf432e6428d760ef9da35faef8e4b610d69
...by increasing the allotted timeout and using direct comparison rather than
reflect.DeepEqual (which should be faster).
PiperOrigin-RevId: 214027024
Change-Id: I0a2690e65c7e14b4cc118c7312dbbf5267dc78bc
This allows a NetworkDispatcher to implement transparent bridging,
assuming all implementations of LinkEndpoint.WritePacket call eth.Encode
with header.EthernetFields.SrcAddr set to the passed
Route.LocalLinkAddress, if it is provided.
PiperOrigin-RevId: 213686651
Change-Id: I446a4ac070970202f0724ef796ff1056ae4dd72a
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
This was previously broken in 212917409, resulting in "missing function body"
compilation errors.
PiperOrigin-RevId: 213323695
Change-Id: I32a95b76a1c73fd731f223062ec022318b979bd4
Netstack needs to be portable, so this seems to be preferable to using raw
system calls.
PiperOrigin-RevId: 212917409
Change-Id: I7b2073e7db4b4bf75300717ca23aea4c15be944c
Now that it's possible to remove subnets, we must iterate over them with locks
held.
Also do the removal more efficiently while I'm here.
PiperOrigin-RevId: 211737416
Change-Id: I29025ec8b0c3ad11f22d4447e8ad473f1c785463
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
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
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
The previous use of non-blocking writes could result in corrupt PCAP files if a
partial write occurs. Using (*os.File).Write solves this problem by not
allowing partial writes. This change does not increase allocations (in one path
it actually reduces them), but does add additional copying.
PiperOrigin-RevId: 209652974
Change-Id: I4b1cf2eda4cfd7f237a4245aceb7391b3055a66c
Add option to redirect packet back to netstack if it's destined to itself.
This fixes the problem where connecting to the local NIC address would
not work, e.g.:
echo bar | nc -l -p 8080 &
echo foo | nc 192.168.0.2 8080
PiperOrigin-RevId: 207995083
Change-Id: I17adc2a04df48bfea711011a5df206326a1fb8ef
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
This CL implements CUBIC as described in https://tools.ietf.org/html/rfc8312.
PiperOrigin-RevId: 207353142
Change-Id: I329cbf3277f91127e99e488f07d906f6779c6603
We have been unnecessarily creating too many savable types implicitly.
PiperOrigin-RevId: 206334201
Change-Id: Idc5a3a14bfb7ee125c4f2bb2b1c53164e46f29a8