David Benjamin
0e7dbd579b
Chromium's renegotiation handling currently relies on reads being the only thing that can discover a renegotiation. However, for a number of reasons, we would like to eagerly drive the read loop after a handshake: - 0-RTT + HTTP/1.1 will otherwise not pick up ServerHellos until after we send a request. In particular, if we preconnect a 0-RTT socket sufficiently in advance, such that the ServerHello comes in by the time we use it, we should send 1-RTT data rather than 0-RTT. - In TLS 1.2 False Start, if HTTP/1.1 or preconnect, we will not pick up the server Finished and NewSessionTicket until later. This way we pick it up sooner. - If the server does not implement https://boringssl-review.googlesource.com/c/boringssl/+/34948, this plugs the theoretical deadlock on the client end. The False Start and 0-RTT scenarios above also have theoretical deadlocks and cannot be mitigated on the server. - TLS 1.3 client certificate alerts interact badly with TCP reset. Eagerly reading from the socket makes it behave slightly better, though it's still not reliable unless the server defers closing the socket. So we can SSL_peek without triggering a renegotiation, add an ssl_renegotiate_explicit mode to defer processing the renegotiation. Bug: chromium:950706, chromium:958638 Change-Id: I78242d93d651b7a32a5c4c24ea9032ef63a027cf Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/37944 Commit-Queue: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <agl@google.com>
BoringSSL SSL Tests
This directory contains BoringSSL's protocol-level test suite.
Testing a TLS implementation can be difficult. We need to produce invalid but sufficiently correct handshakes to get our implementation close to its edge cases. TLS's cryptographic steps mean we cannot use a transcript and effectively need a TLS implementation on the other end. But we do not wish to litter BoringSSL with options for bugs to test against.
Instead, we use a fork of the Go crypto/tls package, heavily patched with
configurable bugs. This code, along with a test suite and harness written in Go,
lives in the runner directory. The harness runs BoringSSL via a C/C++ shim
binary which lives in this directory. All communication with the shim binary
occurs with command-line flags, sockets, and standard I/O.
This strategy also ensures we always test against a second implementation. All features should be implemented twice, once in C for BoringSSL and once in Go for testing. If possible, the Go code should be suitable for potentially upstreaming. However, sometimes test code has different needs. For example, our test DTLS code enforces strict ordering on sequence numbers and has controlled packet drop simulation.
To run the tests manually, run go test from the runner directory. It takes
command-line flags found at the top of runner/runner.go. The -help option
also works after using go test -c to make a runner.test binary first.
If adding a new test, these files may be a good starting point:
runner/runner.go: the test harness and all the individual tests.runner/common.go: contains theConfigandProtocolBugsstruct which control the Go TLS implementation's behavior.test_config.h,test_config.cc: the command-line flags which control the shim's behavior.bssl_shim.cc: the shim binary itself.
For porting the test suite to a different implementation see PORTING.md.