opensource/grpc-go
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,298 Commits 1 Branch 0 Tags
Commit Graph

1298 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Xiang Li
11ef22ebfb make comment on invoke/sendmsg more clear 2016-04-01 10:47:25 -07:00
Tamir Duberstein
2a118f74a2 minor cleanup: simplify looping 2016-03-31 22:05:49 -04:00
Menghan Li
68dadb4d00 Merge pull request #620 from iamqizhao/master 
fix a test error msg
 2016-03-31 09:36:46 -07:00
iamqizhao
b0a5402a2a fix a test error msg 2016-03-30 11:10:27 -07:00
Qi Zhao
33966f5391 Merge pull request #602 from menghanl/metadata_doc 
Add metadata documentation
 2016-03-29 14:16:34 -07:00
Menghan Li
3187ce79ff Merge pull request #617 from iamqizhao/master 
Added the unmapped http2 error and fixed a related bug in RST_STREAM …
 2016-03-29 13:58:50 -07:00
iamqizhao
2f0d12ea11 Added the unmapped http2 error and fixed a related bug in RST_STREAM handling. 2016-03-29 11:07:22 -07:00
Menghan Li
b60d4e72a1 Add metadata documentation 2016-03-29 11:00:44 -07:00
Tamir Duberstein
9e8752e7a2 transport: allocate bigger slabs and reuse 
This provides a substantial speed improvement with large payloads.

Benchmarked using https://github.com/cockroachdb/rpc-bench:

```
name                 old time/op    new time/op    delta
GRPCServeHTTP_1K-4      178µs ± 2%     177µs ± 1%     ~     (p=1.000 n=5+5)
GRPCServeHTTP_64K-4    1.30ms ± 3%    1.09ms ± 4%  -16.82%  (p=0.008 n=5+5)

name                 old speed      new speed      delta
GRPCServeHTTP_1K-4   11.5MB/s ± 2%  11.6MB/s ± 1%     ~     (p=1.000 n=5+5)
GRPCServeHTTP_64K-4   100MB/s ± 2%   121MB/s ± 4%  +20.26%  (p=0.008 n=5+5)

name                 old alloc/op   new alloc/op   delta
GRPCServeHTTP_1K-4     88.5kB ± 0%    93.9kB ± 0%   +6.16%  (p=0.008 n=5+5)
GRPCServeHTTP_64K-4     801kB ± 0%     791kB ± 0%   -1.16%  (p=0.008 n=5+5)

name                 old allocs/op  new allocs/op  delta
GRPCServeHTTP_1K-4        162 ± 0%       156 ± 0%   -3.70%  (p=0.008 n=5+5)
GRPCServeHTTP_64K-4       645 ± 0%       284 ± 0%  -55.96%  (p=0.016 n=5+4)
```
 2016-03-29 05:29:55 -04:00
Qi Zhao
d1f7152c46 Merge pull request #612 from peter-edge/makefile_updates 
Update Makefile
 2016-03-28 23:31:13 -07:00
Qi Zhao
877b524786 Merge pull request #588 from tamird/perf-less 
transport: performance improvements
 2016-03-28 23:30:09 -07:00
Qi Zhao
3c688e321a Merge pull request #611 from iamqizhao/master 
Bump the required Go version
 2016-03-28 23:15:44 -07:00
Qi Zhao
8e8c6a1b40 Merge pull request #604 from broady/travis 
travis: build against go1.5.3, go1.6
 2016-03-28 23:15:11 -07:00
Tamir Duberstein
54ac7579a7 transport: skip the backlog if it's empty 
This provides a modest reduction in allocation count.

Benchmarked using https://github.com/cockroachdb/rpc-bench:

```
name                 old time/op    new time/op    delta
GRPCServeHTTP_1K-4      181µs ± 3%     178µs ± 2%    ~     (p=0.310 n=5+5)
GRPCServeHTTP_64K-4    1.31ms ± 4%    1.30ms ± 3%    ~     (p=0.548 n=5+5)

name                 old speed      new speed      delta
GRPCServeHTTP_1K-4   11.3MB/s ± 3%  11.5MB/s ± 2%    ~     (p=0.246 n=5+5)
GRPCServeHTTP_64K-4   100MB/s ± 4%   100MB/s ± 2%    ~     (p=0.548 n=5+5)

name                 old alloc/op   new alloc/op   delta
GRPCServeHTTP_1K-4     88.5kB ± 0%    88.5kB ± 0%  -0.09%  (p=0.008 n=5+5)
GRPCServeHTTP_64K-4     801kB ± 0%     801kB ± 0%  -0.05%  (p=0.008 n=5+5)

name                 old allocs/op  new allocs/op  delta
GRPCServeHTTP_1K-4        167 ± 0%       162 ± 0%  -2.99%  (p=0.008 n=5+5)
GRPCServeHTTP_64K-4       672 ± 0%       645 ± 0%  -4.08%  (p=0.008 n=5+5)
```
 2016-03-28 21:26:36 -04:00
iamqizhao
58c3636a27 revise wording 2016-03-28 15:44:29 -07:00
Chris Broadfoot
785723ef9c travis: build against go1.5.3, go1.6 2016-03-28 15:06:19 -07:00
Peter Edge
bc6951eb0b Update Makefile. 
Signed-off-by: Peter Edge <peter.edge@gmail.com>
 2016-03-29 00:05:04 +02:00
iamqizhao
a6e9e042c5 Bump the required Go version 2016-03-28 15:02:37 -07:00
Qi Zhao
5309b8314a Merge pull request #608 from menghanl/better_http2_header_newline_error 
Use ErrorDetail() to provide better error msg for malformed HTTP2
 2016-03-28 13:17:04 -07:00
Menghan Li
71e3dbec56 Use ErrorDetail() to provide better error msg for malformed HTTP2 2016-03-28 11:34:29 -07:00
Chris Broadfoot
85c387ba1c Add instructions to CONTRIBUTING.md 2016-03-24 13:55:06 -07:00
Qi Zhao
9e3a674ceb Merge pull request #599 from iamqizhao/master 
lowercase credential keys
 2016-03-22 14:41:09 -07:00
iamqizhao
e5f60381dd lowercase credential keys 2016-03-22 11:33:19 -07:00
Qi Zhao
2dc70c28da Merge pull request #596 from enisoc/errorf 
Don't treat StatusDesc() as a format string.
 2016-03-22 11:11:59 -07:00
Qi Zhao
896247136f Merge pull request #598 from iamqizhao/master 
Specialize http2.StreamError when the server reads frames
 2016-03-22 11:07:08 -07:00
iamqizhao
a31af4bde2 Add some comments to the new test. 2016-03-22 10:56:29 -07:00
iamqizhao
a098b2c182 Specialize http2.StreamError when the server reads frames 2016-03-21 18:23:30 -07:00
Anthony Yeh
a294a45ba0 Don't treat StatusDesc() as a format string. 
For example, if the error message contains "%v", then passing
StatusDesc() as the format string to Errorf() will replace "%v" with
"%!v(MISSING)".
 2016-03-19 17:15:21 -07:00
Qi Zhao
d3ddb4469d Merge pull request #595 from vieux/typos 
fix a few typos
 2016-03-17 10:50:43 -07:00
Victor Vieux
047af78181 fix a few typos 2016-03-16 16:40:16 -07:00
Qi Zhao
75cc451428 Merge pull request #593 from dsymonds/prbot-gofmt 
gofmt everything
 2016-03-16 10:40:56 -07:00
David Symonds
a0594a1683 Run gofmt over Go source files. 2016-03-16 15:11:58 +11:00
Qi Zhao
986912c907 Merge pull request #590 from petermattis/pmattis/context-with-stream 
transport: remove unnecessary allocation
 2016-03-09 11:37:48 -08:00
Qi Zhao
ef489cba5c Merge pull request #591 from petermattis/pmattis/buffered-reader 
Use a bufio.Reader for input.
 2016-03-09 11:35:03 -08:00
Peter Mattis
a1d076a617 Use a bufio.Reader for input. 
This reduces the number of system calls for reading network data,
providing a nice speedup when there are concurrent RPCs.
 2016-03-09 12:07:58 -05:00
Peter Mattis
8b936e927a transport: remove unnecessary allocation 
Storing an integer in an interface requires an allocation. Storing an
empty struct does not. This matches the pattern used for peer.NewContext
and metadata.NewContext.
 2016-03-09 10:27:53 -05:00
Qi Zhao
b1439a2743 Merge pull request #522 from dsymonds/master 
Prepare for adding a compatibility marker to proto generated code.
 2016-03-07 12:19:07 -08:00
David Symonds
710dbc764f Prepare for adding a compatibility marker to proto generated code. 
This will assert at build time that the generated code is compatible
with the grpc package that it is linked to. A future change in the
grpc plugin for protoc-gen-go will generate a line like
	const _ = grpc.SupportPackageIsVersion1

This constant will be renamed in the future when incompatible changes
are made that require synchronised updates to grpc and protoc-gen-go.
 2016-03-07 10:12:28 +11:00
Qi Zhao
fea7689493 Merge pull request #584 from bradfitz/h2 
Simplify server setup when testing the http.Handler-based server transport
 2016-03-02 10:59:35 -08:00
Brad Fitzpatrick
18b2da6250 Simplify server setup when testing the http.Handler-based server transport 
This simplifies the code that's used for wiring up the http.Handler-based
transport for testing. It's not used outside of tests.

http2.Server.ServeConn was added somewhat recently for other reasons and
I just realized it would simplify this code as well.
 2016-03-02 18:49:23 +00:00
Qi Zhao
0e69ad1ebd Merge pull request #580 from iamqizhao/master 
Fix err handling of malformed http2
 2016-03-01 11:42:18 -08:00
iamqizhao
f3c6dc545f Fix err handling of malformed http2 
polish the commit
 2016-03-01 10:52:13 -08:00
Qi Zhao
b4b62c5f29 Merge pull request #571 from bradfitz/internal 
Remove test-only methods from grpc package.
 2016-02-26 14:33:54 -08:00
Qi Zhao
b88c12e7ca Merge pull request #570 from bradfitz/meta 
transport: use http2.Framer's MetaHeadersFrame functionality
 2016-02-26 13:37:15 -08:00
Qi Zhao
7459645a33 Merge pull request #574 from iamqizhao/master 
Skip all goroutines in C code for the leak detection
 2016-02-26 10:49:11 -08:00
Qi Zhao
b4adf7d359 Merge pull request #573 from bradfitz/no_cover 
Disable the code coverage checker.
 2016-02-25 18:35:59 -08:00
iamqizhao
a5e4790bfe coalesce runtime_mcall cases 2016-02-25 17:36:29 -08:00
iamqizhao
d5c0571ebc skip all goroutines in C code for the leak detector 2016-02-25 16:55:22 -08:00
Qi Zhao
8a7d92a735 Merge pull request #572 from iamqizhao/master 
Add more uninteresting goroutines for the leak detector
 2016-02-25 16:09:25 -08:00
Brad Fitzpatrick
3ecc13ec96 Disable the code coverage checker. 
It's too noisy and not worth it. I don't like encouraging the code
coverage game, wherein people write bad tests to get points.
It's annoying to get a red X in the PR status when coverage dropped
by 1 line or 0.4%.

Good code review should be sufficient to verify we have both tests
and good tests.
 2016-02-26 00:02:08 +00:00