Commit Graph

53 Commits

Author SHA1 Message Date
klzgrad
f214282adf brotli: Fix missing import in BUILD.gn
is_ubsan is not declared when imported from net/BUILD.gn
2024-03-24 20:05:13 +08:00
klzgrad
5f9facbdcd third_party/protobuf: Disable musttail on MIPS
clang reports:

fatal error: error in backend: failed to perform tail call elimination on a call site marked musttail
2024-03-24 20:05:13 +08:00
klzgrad
544ce1b587 Add continuous integration and tests 2024-03-24 20:05:13 +08:00
klzgrad
7984412bbe Add apk build 2024-03-24 20:05:13 +08:00
klzgrad
f5034cd7da Disable CFI icall for linux x64 2024-03-24 20:05:03 +08:00
klzgrad
b386512ec2 Disable ThinLTO on static builds 2024-03-24 20:04:53 +08:00
klzgrad
65ccb88ecb Add build scripts 2024-03-24 13:20:36 +08:00
klzgrad
64c5033d68 Add example config.json 2024-03-24 13:20:36 +08:00
klzgrad
c5e58398e3 Add README 2024-03-24 13:20:36 +08:00
klzgrad
939a6887db Add LICENSE 2024-03-24 13:20:36 +08:00
klzgrad
ba0bca584e Add source import tool 2024-03-24 13:20:36 +08:00
klzgrad
e82da276e2 Add Naive 2024-03-24 13:20:36 +08:00
klzgrad
6655bb0d0f build: Use thinlto max optimization for everything 2024-03-24 13:20:36 +08:00
klzgrad
65935fcb7f base: Fix Android build adding extra -latomic
We don't use use_sysroot=true, but without it -latomic is added
to is_android.
2024-03-24 13:20:36 +08:00
klzgrad
e69a1de462 allocator: Disable PartitionAlloc memory tagging 2024-03-24 13:20:36 +08:00
klzgrad
1bb7948894 allocator: Fix PartitionAlloc mipsel build 2024-03-24 13:20:36 +08:00
klzgrad
b3c647cf54 allocator: Fix recommitted pages not being zeroed when madvise is not available.
Support kernels built without CONFIG_ADVISE_SYSCALLS on small
embedded devices.
2024-03-24 13:20:36 +08:00
klzgrad
194d753917 allocator: Fix __THROW and mallinfo for Musl 2024-03-24 13:20:36 +08:00
klzgrad
385e1a8163 debug: Fix uClibc macro condition 2024-03-24 13:20:36 +08:00
klzgrad
916560b75f third_party: Fix missing sgidefs.h for Musl 2024-03-24 13:20:36 +08:00
klzgrad
d34f4bccd9 process: Remove use of mallinfo for Musl 2024-03-24 13:20:36 +08:00
klzgrad
dcdfe48420 base: Remove use of mallinfo for Musl 2024-03-24 13:20:35 +08:00
klzgrad
b46f06488f dns: Support Musl 2024-03-24 13:20:35 +08:00
klzgrad
cd68b7be05 lss: Avoid naming conflict in fstatat64
Supports OpenWrt builds.
2024-03-24 13:20:35 +08:00
klzgrad
c8c081012f build: Add sysroot creator script 2024-03-24 13:20:35 +08:00
klzgrad
b3984b14b7 build: Add OpenWrt toolchains 2024-03-24 13:20:35 +08:00
klzgrad
72ce35040b build: Work around MPIS floating point ABI passing in LTO 2024-03-24 13:19:57 +08:00
klzgrad
a0319d0b0d build: Support ARM build without FPU 2024-03-24 13:19:57 +08:00
klzgrad
e7f307a72b build: Support -mcpu= on ARM and ARM64 2024-03-24 13:19:57 +08:00
klzgrad
101e5b42ba quic: Add support for HTTP/3 CONNECT Fast Open
SpdyProxyClientSocket uses read_callback_ for both Connect() and
Read(), and its OnIOComplete() calls read_callback_, thus its fast
connect code checks read_callback_. The code was ported to
QuicProxyClientSocket without much change.

But QuicProxyClientSocket uses a separate connect_callback_ apart from
read_callback_, and its OnIOComplete() calls connect_callback_, thus
when headers are received after Connect() it doesn't need to check
read_callback_ and should always avoid calling connect_callback_.
2024-03-24 13:19:57 +08:00
klzgrad
c5133c7450 h2: Pad RST_STREAM frames
Clients sending too many RST_STREAM is an irregular behavior.

Hack in a preceding END_STREAM DATA frame padded towards [48, 72]
before RST_STREAM so that the TLS record looks like a HEADERS frame.

The server often replies to this with a WINDOW_UPDATE because padding
is accounted in flow control. Whether this constitudes a new irregular
behavior is still unclear.
2024-03-24 13:19:57 +08:00
klzgrad
085c4c5993 h2: Add support for HTTP/2 CONNECT Fast Open
SpdyProxyClientSocket waits for 200 OK before returning OK for Connect.

Change that behavior to returning OK immediately after CONNECT header.

This feature is enabled by a "fastopen" header via the proxy delegate.

Design notes:

The current approach is better than the obvious TCP Fast Open style fake
Connect().

Fast Open should not be used for preconnects as preconnects need actual
connections set up. The Naive client does not use preconnects per se
(using "...RawConnect") but the user agent will use preconnects and the
Naive client has to infer that. Hence there is a need to check the
incoming socket for available bytes right before Connect() and configure
whether a socket should be connected with Fast Open. But fake Connect()
make it difficult to check the incoming socket because it immediately
returns and there is not enough time for the first read of the incoming
socket to arrive.

To check for preconnects it is best to push the first read of the
incoming socket to as late as possible. The other (wrong) way of doing
that is to pass in an early read callback and call it immediately after
sending HEADERS and then send the available bytes right there. This way
is wrong because it does not work with late binding, which assumes
Connect() is idempotent and causes sockets opened in this way to be
potentially bound to the wrong socket requests.

The current approach is to return OK in Connect() right after sending
HEADERS before getting the reply, which is to be received later. If the
reply is received during a subsequent Read() and the reply indicates an
error, the error is returned to the callback of the Read(); otherwise
the error is ignored with the connection disconnected and subsequent
Read() and Write() should discover the disconnection.
2024-03-24 13:19:57 +08:00
klzgrad
98b67cf7fa h2: Reduce warnings about RST on invalid streams
Per RFC 7540#6.4:

  However, after sending the RST_STREAM, the sending endpoint MUST be
  prepared to receive and process additional frames sent on the stream
  that might have been sent by the peer prior to the arrival of the
  RST_STREAM.
2024-03-24 13:19:57 +08:00
klzgrad
d7cc4a2374 socket: Force tunneling for all sockets
In the socket system, only WebSocket sockets are allowed to tunnel
through HTTP/1 proxies. "Raw" sockets in the normal socket pool don't
have it, and their CONNECT headers are not sent, instead the raw
payload is sent as-is to the HTTP/1 proxy, breaking the proxying.

The socket system works like this:

- HTTP sockets via HTTP/1 proxies: normal pool, no tunneling.
- HTTPS sockets via HTTP/1 proxies: normal pool, no tunneling,
  but does its own proxy encapsulation.
- WS sockets via HTTP/1 proxies: WS pool, tunneling.

In Naive, we need the normal pool because the WS pool has some extra
restrictions but we also need tunneling to produce a client socket
with proxy tunneling built in.

Therefore force tunneling for all sockets and have them always send
CONNECT headers. This will otherwise break regular HTTP client sockets
via HTTP/1 proxies, but as we don't use this combination, it is ok.
2024-03-24 13:19:57 +08:00
klzgrad
0f24a69450 socket: Allow higher limits for proxies
As an intermediary proxy we should not enforce stricter connection
limits in addition to what the user is already enforcing.
2024-03-24 13:19:57 +08:00
klzgrad
b3461f890f cert: Handle AIA response in PKCS#7 format 2024-03-24 13:19:57 +08:00
klzgrad
8c6aebf3e5 cert: Use builtin verifier on Android and Linux 2024-03-24 13:19:57 +08:00
klzgrad
25931edd2f cert: Add SystemTrustStoreStaticUnix
It reads CA certificates from:

* The file in environment variable SSL_CERT_FILE
* The first available file of

/etc/ssl/certs/ca-certificates.crt (Debian/Ubuntu/Gentoo etc.)
/etc/pki/tls/certs/ca-bundle.crt (Fedora/RHEL 6)
/etc/ssl/ca-bundle.pem (OpenSUSE)
/etc/pki/tls/cacert.pem (OpenELEC)
/etc/pki/ca-trust/extracted/pem/tls-ca-bundle.pem (CentOS/RHEL 7)
/etc/ssl/cert.pem (Alpine Linux)

* Files in the directory of environment variable SSL_CERT_DIR
* Files in the first available directory of

/etc/ssl/certs (SLES10/SLES11, https://golang.org/issue/12139)
/etc/pki/tls/certs (Fedora/RHEL)
/system/etc/security/cacerts (Android)
2024-03-24 13:19:57 +08:00
klzgrad
27b05522d1 libc++: Disable exceptions and RTTI
Except on Mac, where exceptions are required.
And except on Android, where rtti is required.
2024-03-24 13:19:57 +08:00
klzgrad
0a20d5aaf7 base: Disable trace event
This allows builds with enable_base_tracing=false.
2024-03-24 12:43:24 +08:00
klzgrad
ea69f7c633 base: Fix iwyu in file_path.cc 2024-03-24 12:43:24 +08:00
klzgrad
83304fc750 base: Don't fix Y2038 problem with icu 2024-03-24 12:43:24 +08:00
klzgrad
7bc54497b9 net, url: Remove icu 2024-03-24 12:43:24 +08:00
klzgrad
6638573ea9 build: Force determinism in official build
Helps build with ccache.
2024-03-24 12:43:24 +08:00
klzgrad
cdc1adc369 build: Disable Android java templates 2024-03-24 12:43:24 +08:00
klzgrad
ef8cdf29df build: Disable build_with_chromium
The argument build_with_chromium mainly enables various tests,
data bundling, infra integration, and AFDO profiles.

AFDO can be added by other arguments.
2024-03-24 12:43:24 +08:00
klzgrad
c0105ca6e5 url: Remove Android 2024-03-24 12:43:24 +08:00
klzgrad
99f2d6468c base: Remove JNI function on Android 2024-03-24 12:43:24 +08:00
klzgrad
2beb83e776 base: Add Android stubs 2024-03-24 12:43:24 +08:00
klzgrad
bd063a7cac net: Add Android stubs 2024-03-24 09:03:26 +08:00