Nate McMaster
9a4a810aa8
Fix race conditions in test event listener
2017-03-15 10:26:09 -07:00
Cesar Blum Silveira
632780dd16
Remove splitting of path and path base ( #1050 ).
2017-03-15 09:57:33 -07:00
Nate McMaster
d4c0d4b81e
Adds an EventSource named 'Microsoft-AspNetCore-Server-Kestrel' with the following event and properties:
...
ConnectionStart:
- connectionId
- scheme
- localEndPoint
- remoteEndPoint
ConnectionStop:
- connectionId
2017-03-14 16:59:11 -07:00
Nate McMaster
5073d4fdaf
Re-add .NET Framework to tests and other cleanup ( #1494 )
2017-03-14 11:02:49 -07:00
Ben Adams
b94912bcb1
InitializeHeaders only at start of parsing/Fix remaining ( #1488 )
...
* Don't reinitialize header collection each loop
* Correct remaining tracking value
* Add tests
2017-03-13 15:32:28 -07:00
David Fowler
5644310811
Clean up benchmarks ( #1487 )
...
- Benchmarks are suffixed with *Benchmarks
- Removed Custom RpsColumn.cs and used Ops/Second column
- Removed params from RequestParsingBenchmark
2017-03-12 22:52:24 -07:00
Cesar Blum Silveira
49f09d5a25
Log rejected request targets.
2017-03-12 18:37:08 -07:00
David Fowler
f479720d5a
Remove our custom bootstrapper and use the BenchmarkSwitcher ( #1480 )
2017-03-11 18:19:13 -08:00
Ben Adams
3c8ee39f1d
Fxi writing perf test ( #1478 )
2017-03-10 12:04:28 -08:00
Nate McMaster
49b328d4c2
Handle absolute, asterisk, and authority-form request targets
...
Improves compliance with RFC 7230 on the expected handling of requests
that have URI or asterisk in the request target.
This means rejecting asterisk requests that are not OPTIONS and rejecting
authority-form requests taht are not CONNECT.
This also means the server will handle the path and query on targets
with absolute URIs as request-targets.
2017-03-09 16:54:12 -08:00
Ben Adams
941d396942
Speed up ParseRequestLine ( #1463 )
2017-03-08 19:27:56 -08:00
Nate McMaster
63e3c9428a
You can't have two benchmarks as the baseline.
2017-03-08 17:28:13 -08:00
Nate McMaster
3a1f3d14f9
Add absolute-uri benchmark and change plaintext, live aspnet, and unicode benchmarks to use origin-form
2017-03-08 17:21:21 -08:00
Cesar Blum Silveira
bb973decb8
Unify header rejection messages.
...
- Log bad headers with escaped non-vchar characters
2017-03-08 15:27:44 -08:00
Cesar Blum Silveira
e25eb418bb
Change non-printable char representation in log messages from <0xXX> to \xXX.
2017-03-08 10:05:53 -08:00
David Fowler
13519b6079
Fix CopyTo implementation in benchmark. ( #1462 )
...
* wip
* ooops
2017-03-07 17:39:56 -08:00
Cesar Blum Silveira
5743d740b4
Convert HTTP parsing FrameTests to IHttpParser tests ( #1416 ).
...
- Also fix an issue in KestrelHttpParser where "Header: \r\n" is
parsed with a value of " " instead of "".
2017-03-07 15:51:11 -08:00
Ben Adams
e2f8c226ef
Simplify TakeSingleHeader and Vectorize ( #1457 )
...
* Sanitize unsafe code
* Vectorize
* Extract Contains and add more tests
2017-03-07 13:30:57 -08:00
Pavel Krymets
1294c00618
Cleanup unused code ( #1458 )
2017-03-07 11:52:45 -08:00
David Fowler
02a4342908
Made changes to TakeSingleHeader ( #1453 )
...
* Made changes to TakeSingleHeader
- Remove state machine and just parse in place
- Inline OnHeader into TakeSingleHeader
- Use IndexOfVectorized instead of custom indexof
- Normalize header whitespace error
- Combine IndexOf and IndexOfAny into a single IndexOfNameEnd call
2017-03-07 08:58:55 -08:00
John Luo
06134bc6e0
Add IPv6 loopback address by default #1434
2017-03-06 18:02:28 -08:00
Cesar Blum Silveira
11c7eb5665
Verify all request rejections are logged ( #1295 ).
2017-03-06 12:05:38 -08:00
John Luo
7d94abd606
Enable default server address test
2017-03-06 10:26:32 -08:00
Cesar Blum Silveira
0404bcc58c
Add more microbenchmarks.
2017-03-06 10:14:22 -08:00
David Fowler
537b06f025
Interleave multispan and single span code path ( #1442 )
...
- Attempt at making the reader better for multispan parsing
- Try tighter inner loop
- Fix boundary case and clean code up
- Update the cursor once instead of after every header
- Fix errors with not updating consumed state on incomplete header payload
- Filled a test hole, removed a condition that should never happen
- Avoid struct copies every iteration when parsing headers
2017-03-06 09:06:28 -08:00
David Fowler
0ce111d9f1
Fix write size in benchmark ( #1449 )
2017-03-06 03:31:45 -08:00
Cesar Blum Silveira
83edc38e72
Use TechEmpower request as baseline for request parsing benchmarks.
2017-03-03 15:55:07 -08:00
Cesar Blum Silveira
de2c3d3608
Merge branch 'KrzysztofCwalina/NewBenchmark' into dev
2017-03-03 15:28:34 -08:00
Stephen Halter
ac60f13312
Modify RequestProcessingAsync to call single parse method ( #1427 )
...
* Modify RequestProcessingAsync to call single parse method
* Fix bad request logging
2017-03-03 14:43:32 -08:00
Krzysztof Cwalina
7cc5c537a4
Added a new benchmark
2017-03-03 13:13:11 -08:00
David Fowler
1d685e195e
Parser cleanup and remove line continuation header error ( #1431 )
...
- Cleaned up some parsing logic (removed locals etc)
- Removing line continuation errors cleaned up code duplication
a little bit
2017-03-03 10:04:44 -08:00
David Fowler
4533383612
React to hosting change
...
- Kestrel binds to ipv4 by default
Fixes #1432
2017-03-02 23:09:13 -08:00
David Fowler
4d7c6ff69f
use rtm appveyor
2017-03-02 21:44:36 -08:00
Stephen Halter
1e0f2b3951
Fix systemd activation tests ( #1429 )
2017-03-02 17:26:41 -08:00
David Fowler
8929b40527
Single span optimizations ( #1421 )
...
- Added a fast path for single span in the start line parsing
- Added a fast path for single span header parsing
- Changed the out header loop to be pointer based (instead of slicing)
2017-03-02 12:17:39 -08:00
Ben Adams
40ee51846c
Add allocations column ( #1422 )
2017-03-02 06:56:05 -08:00
David Fowler
cb6059c143
Make the IHttpParser per frame and add a reset ( #1415 )
...
* Make the IHttpParser per frame and add a reset
- Made the IHttpParser a per frame object so state can be stored
across method calls and parses.
- Added HttpParserFactory to ServiceContext
2017-03-01 13:12:03 -08:00
Pavel Krymets
d3694f085a
Add IHttpParser interface ( #1414 )
2017-03-01 11:55:36 -08:00
Krzysztof Cwalina
c56de066d3
Optimized parsing GET verb and version ( #1399 )
...
* Optimized parsing GET verb
* optimized http version parsing
* Added microbenchmarks for GetKnownMethod and GetKnownVersion
2017-02-28 17:02:52 -08:00
John Luo
6ad9f3e8b0
Reacting to HeaderUtitilities renames
2017-02-28 14:50:02 -08:00
Cesar Blum Silveira
568aaff9c4
Improve HTTP parsing tests ( #1393 ).
...
- Add several more test cases
- Share data between functional and unit tests
2017-02-28 14:23:27 -08:00
Stephen Halter
fde0f6b2fc
Add an option to Kestrel to disable threadpool dispatching
2017-02-28 10:14:58 -08:00
Cesar Blum Silveira
a95743c5f6
Add functional test to verify generated code is up to date ( #1369 ).
2017-02-27 14:19:56 -08:00
David Fowler
c6705d8693
Convert TakeStartLine and TakeMessageHeaders to be state machines ( #1401 )
...
- Less passes over the buffer
- Single pass to find all start line delimiters instead
of calling IndexOf multiple times.
- Made TakeStartLine and TakeMessageHeaders a state machine
- Only check length against remaining bytes once
- Change variable names to match TakeStartLine
- Use ReadableBuffer.First.Span instead of ToSpan()
- Added test for missing path with a querystring
2017-02-27 11:55:30 -08:00
David Fowler
5692f51bf7
Revert "Revert "Use Spans to parse the start line and headers ( #1394 )""
...
This reverts commit 19c3107deb .
2017-02-24 10:22:05 -08:00
David Fowler
19c3107deb
Revert "Use Spans to parse the start line and headers ( #1394 )"
...
This reverts commit 8140b8cdfe .
2017-02-24 10:03:32 -08:00
David Fowler
4544f881a2
Revert "Skipping failing tests to unblock CI"
...
This reverts commit 0860981ada .
2017-02-24 09:48:44 -08:00
David Fowler
0860981ada
Skipping failing tests to unblock CI
2017-02-24 09:31:43 -08:00
David Fowler
8140b8cdfe
Use Spans to parse the start line and headers ( #1394 )
...
* Use Spans to parse the start line and headers
- Use `Span<byte>` to parse the start line and headers
2017-02-23 23:02:29 -08:00
Pavel Krymets
990e2a8dc4
Use pass through connection adapter instead of ssl for max buffer size tests ( #1391 )
2017-02-23 08:04:43 -08:00