Summary:
Bug fix. Since we changed the iteration order of ack processing to be
in reverse order of packet numbers, there has been a bug and both BBR and Cubic
has been using the earliest acked packet instead of the largest acked packet as
the largest acked packet.
Reviewed By: mjoras
Differential Revision: D17909490
fbshipit-source-id: 15310aad53b4a9e5190ed65c3b2df496db77b1ae
Summary: Use the custom variant type for errors.
Reviewed By: yangchi
Differential Revision: D17826935
fbshipit-source-id: 2bf0c3e1cc8ca84b504d201fd6c2a4266878b715
Summary:
They are strongly coupled, which indicate this is probably better to do it as one class.
Pull Request resolved: https://github.com/facebookincubator/mvfst/pull/44
Reviewed By: mjoras
Differential Revision: D17590918
Pulled By: yangchi
fbshipit-source-id: 2eaca079fd760107eefd2b74fa612d7a0c8b3001
Summary:
Use the custom variant type for write frames as well, now that
we use them for read frames.
Reviewed By: mjoras
Differential Revision: D17776862
fbshipit-source-id: 47093146d0f1565c22e5393ed012c70e2e23d279
Summary:
We erase packets from outstandingPackets list during ack handling.
Currently we handle acks in ascending order of the acked packet number, then
erase() the acked packet. Each erase() call only erase one packet, and it can
potentally make std::deque to move a few elements following the erased one.
This diff changes it to handle acks in descending order, and also erase acked
packets in continuous groups. This reduces erase() calls, and also reduces the
possibility that std::deque needs to move following elements in the erase().
Reviewed By: mjoras
Differential Revision: D17579683
fbshipit-source-id: 8bc11f6a7875beb70dc46c497857ab694df7b6a5
Summary: Having access to the state when decrypting tickets gives us more control over ticket acceptance policies.
Reviewed By: knekritz
Differential Revision: D17528945
fbshipit-source-id: a3cb3d4c0917f2494f5669f283cda70776b608c6
Summary:
Make a custom variant type for PacketHeader. By not relying on boost::variant
this reduces the code size of the implementation.
This uses a combination of a union type as well as a enum type to emulate a variant
Reviewed By: yangchi
Differential Revision: D17187589
fbshipit-source-id: 00c2b9b8dd3f3e73af766d84888b13b9d867165a
Summary:
Prior to this diff we would clone out an entire flow control's worth of data from the writebuffer and then clone out a smaller portion of that to write into the packet builder. This is extremely wasteful when we have a large flow control window.
Additionally we would always write the stream data length field even when we are going to fill the remainder of the packet with the current stream frame. By first calculating the amount of data that needs to can be written and writing the header, we can now omit the data length field when we can fill the whole packet.
Reviewed By: yangchi
Differential Revision: D15769514
fbshipit-source-id: 95ac74eebcde87dd06de54405d7f69c42362e29c
Summary:
The CryptoFactory is extended with makePacketNumberCipher . In order to support that feature, FizzCryptoFactory now explicitly takes a QuicFizzFactory as argument instead of a generic fizz::Factory, which is the only type that is used in practice anyways.
The cypher argument was removed because:
1/ Only one cypher is used at all. Fizz also supports ChaCha20, but using it in mvfst will throw an exception.
2/ it seems like the factory should know what cypher it is dealing with.
If a choice of cypher needs to be supported going forward, it can be done by adding state to FizzCryptoFactory.
Pull Request resolved: https://github.com/facebookincubator/mvfst/pull/40
Reviewed By: mjoras
Differential Revision: D16785274
Pulled By: yangchi
fbshipit-source-id: a1c490e34c5ddd107e8e068d8b127c1ed00a59ec
Summary:
After https://github.com/facebookincubator/mvfst/issues/35 , the facility to mock fizz::Aead is not required anymore. Either the test depends on fizz's behavior, in which case a mock was never appropriate, or it doesn, in which case a quic::Aead is more appropriate.
Pull Request resolved: https://github.com/facebookincubator/mvfst/pull/39
Reviewed By: mjoras
Differential Revision: D16785265
Pulled By: yangchi
fbshipit-source-id: ba74155ab7e8ed1c261c58569db0690781d85fe1
Summary: Reduce the number of padding frames that are generated, to decrease overall qlog size.
Reviewed By: sharma95
Differential Revision: D16533449
fbshipit-source-id: 68f9c48e265576bff098dfeee87e576e86a32330
Summary:
Draft-22 onwards uses two one byte length fields to encode connection ID length instead of one one byte length field.
To support this without disrupting existing clients we need to make our parsing version dependent. This diff accomplishes it by special casing the existing Facebook client QUIC version (0xfaceb000), and changing the default Facebook client version going forward to 0xfaceb001.
Note that this diff also changes the behavior of the ticket transport parameters. When we changed from draft-18 to draft-19 transport parameters I apparently forgot to update the ticket transport parameters to the new format.
Reviewed By: yangchi
Differential Revision: D16205090
fbshipit-source-id: e74a92fa959d308f4bb43bad76e58d4b58d07322
Summary:
There is already a createPacket util function in this file. Rename the
newly added generate* functions to make this consistent
Reviewed By: sharma95
Differential Revision: D15948954
fbshipit-source-id: 67e7addefc839226b98d81a3808d37519ce001af
Summary:
This is based on top of https://github.com/facebookincubator/mvfst/issues/15 .
Now that the codebase have been refactored to use Aead/fizz:Aead and MockAead/fizz::test::MockAead in proper places, it is time to ensures the aren't aliases of each others anymore.
This introduces FizzAead as a wrapper for fizz::Aead that implements quic::Aead and forward all calls. Most of the codebase now uses quic::Aead, which a significant step toward being able to swap it for another implementation.
Pull Request resolved: https://github.com/facebookincubator/mvfst/pull/16
Reviewed By: yangchi
Differential Revision: D15474068
Pulled By: mjoras
fbshipit-source-id: 7cccc9909a45ea538409645644641e39a58bf81d
Summary:
Interface for QLogger:
- Takes in a Quic packet (`VersionNegotiation`, `RegularQuicWritePacket`, `RegularQuicPacket`)
- Stores it in QLogger
- each packet is an event
- each frame becomes a log
Reviewed By: sharma95
Differential Revision: D15581299
fbshipit-source-id: dedf933f47e2e8769109aa1e45ff38c2bf013714
Summary: Replace hard coded stateless reset token with a token from the stateless reset token generator.
Reviewed By: yangchi
Differential Revision: D15481858
fbshipit-source-id: 30c96843c38c616600466b2fabb6defd5fcc5799
Summary: After a given amount of time, discards PSKs whose handshakes are too old.
Reviewed By: knekritz
Differential Revision: D14174662
fbshipit-source-id: ef505af91732c1030c6ce27835db49cf07d9c506
Summary:
given packet number space has been split to 3 spaces, using a single
PacketNum isn't enough to serve the end of recovery tracking. Instead, use a
TimePoint value.
Reviewed By: sharma95
Differential Revision: D15219240
fbshipit-source-id: baf05828279789eb0aa170b49a177510ea07836f
Summary:
This is based on top of #12 .
It logically split MockAead and fizz::MockAead in preparation for separation of the two.
Pull Request resolved: https://github.com/facebookincubator/mvfst/pull/15
Reviewed By: yangchi
Differential Revision: D15474045
Pulled By: mjoras
fbshipit-source-id: b61a5cb08ddae0add66a6c37e156eddaef118e0c
Summary:
This introduce quic::Aead as a simple typedef to fizz::Aead and update the codebase to use quic::Aead . This should not impact the functionality of the code in any way.
This is a first step toward introducing an interface that is specific for mvfst so that mvfst can swap fizz for something else.
Pull Request resolved: https://github.com/facebookincubator/mvfst/pull/12
Reviewed By: JunqiWang
Differential Revision: D15335324
Pulled By: mjoras
fbshipit-source-id: fef166a9a5c2cbae08ad9511d0abd749f330c221
Summary:
Fixed QuicClientTransportIntegrationTest:
SetUp() shouldn't set maxStreams manually. Instead, client should read serverTransportParams after 1 rtt.
The tests used to call createBidiStream right after client->start(), changed that to only sending data after onTransportReady.
Reviewed By: mjoras
Differential Revision: D15293209
fbshipit-source-id: 88e739d72f1fad76666af003d446d068e6d5fac7
Summary: This is step 1 for removing reset on reset, since the send side may need to transition to waiting for a reset ack while the read side is an any state.
Reviewed By: lnicco
Differential Revision: D15075849
fbshipit-source-id: 1e094942a8a1ca9a01d4161cd6309b4136a9cfbf
