Craig Tiller
21dd43a1cd
Merge github.com:grpc/grpc into bm_call_create
2017-02-21 16:06:13 -08:00
Craig Tiller
8b01e8f5e6
Merge pull request #9708 from ctiller/bm_cq
...
Add microbenchmark of single-threaded CQ operations
2017-02-21 16:05:50 -08:00
Craig Tiller
7f4d30a032
Separate CAS/ADD RMWs
2017-02-21 10:24:00 -08:00
Craig Tiller
f3dec9c995
Add counters for rmw atomic operations to microbenchmarks
2017-02-21 10:02:31 -08:00
Craig Tiller
c92290329f
Merge github.com:grpc/grpc into we-have-better-methods-now
2017-02-18 15:13:59 -08:00
Yang Gao
36ffc6e0f7
Merge pull request #9205 from yang-g/health
...
C++ default health checking service.
2017-02-17 21:42:55 -08:00
Craig Tiller
221edc1e5b
Remove dead code
2017-02-17 13:58:19 -08:00
Craig Tiller
05fbe3c518
Get gtest building with Bazel, add first test in test/cpp/common/BUILD
2017-02-17 13:27:21 -08:00
Nicolas "Pixel" Noble
abd887012f
Initial test/cpp/util/BUILD file.
2017-02-17 21:56:10 +01:00
yang-g
55c6ebfd7a
Merge branch 'master' into health
2017-02-17 09:55:03 -08:00
Craig Tiller
2bc4cc4237
Merge github.com:grpc/grpc into bm_call_create
2017-02-17 06:55:55 -08:00
Craig Tiller
f422cc77ba
Merge github.com:grpc/grpc into bm_cq
2017-02-17 06:51:47 -08:00
Paul Marks
69c0437e75
Merge pull request #9634 from pmarks-net/master
...
Connect to "localhost" from interop clients.
2017-02-16 16:34:42 -08:00
Craig Tiller
985736358c
Merge github.com:grpc/grpc into bm_cq
2017-02-16 09:45:17 -08:00
Craig Tiller
be42e1cb6a
Fix missing header
2017-02-16 09:25:23 -08:00
Craig Tiller
d3b8cd6617
Merge github.com:grpc/grpc into bm_closure
2017-02-15 23:55:31 -08:00
Craig Tiller
03eec51790
Merge pull request #9697 from ctiller/bm_trickle
...
Add a test that measures flow control stalls in a deterministic way
2017-02-15 23:52:46 -08:00
Craig Tiller
4f60fbf104
Add initial metadata benchmarks
2017-02-15 23:50:36 -08:00
Craig Tiller
3a04debe74
Add load reporting
2017-02-15 23:01:47 -08:00
Craig Tiller
c52ba3a03f
Hoist constant setting out of loop - and shave some ns from call creation
2017-02-15 22:57:43 -08:00
Craig Tiller
a25c9b4aa0
Call out no-filter case
2017-02-15 17:19:22 -08:00
Craig Tiller
4bba37c228
Add a no-filter stack for comparison
2017-02-15 17:16:44 -08:00
Craig Tiller
571b7fffb3
Add benchmarks for a good selection of filters
2017-02-15 17:10:53 -08:00
Craig Tiller
fd044b8b22
Call creation benchmark
2017-02-15 16:15:48 -08:00
Craig Tiller
872e932b09
Merge github.com:grpc/grpc into bm_call_create
2017-02-15 13:08:43 -08:00
Craig Tiller
1f8c55835d
Merge branch 'bm_closure' into bm_cq
2017-02-15 08:34:49 -08:00
Craig Tiller
2b085f9923
Merge branch 'bm_trickle' into bm_closure
2017-02-15 08:26:37 -08:00
Craig Tiller
22482a41b6
Merge branch 'bm_trickle' of github.com:ctiller/grpc into bm_trickle
2017-02-15 08:25:09 -08:00
Craig Tiller
d9bc2107e5
Update microbenchmarking framework for new benchmark
2017-02-15 08:24:55 -08:00
Craig Tiller
7f9184f314
Review feedback
2017-02-14 22:39:47 -08:00
Craig Tiller
5752570b02
Fix memory leak
2017-02-14 21:58:46 -08:00
Craig Tiller
14ae5381af
Add microbenchmark of single-threaded CQ operations
2017-02-14 12:43:50 -08:00
Craig Tiller
ef6d744c23
Include faster networks
2017-02-13 19:23:09 -08:00
Craig Tiller
4cb0a7a5a1
Better test
2017-02-13 17:14:20 -08:00
Craig Tiller
fe5f497f77
Add a test that measures flow control stalls in a deterministic way
2017-02-13 16:26:27 -08:00
Craig Tiller
4dde8b277d
Fix stack corruption
2017-02-13 12:44:14 -08:00
Craig Tiller
96f8728b06
Add a comparison benchmark of acquiring a mutex
2017-02-13 12:39:12 -08:00
Craig Tiller
b3f34b6c2b
Add benchmarks of closures, combiners, exec_ctx primitives
2017-02-13 10:29:33 -08:00
Craig Tiller
e7c60e6282
Make latency traces more useful
2017-02-13 08:56:36 -08:00
Craig Tiller
7dd42bf3d8
Merge pull request #9522 from ctiller/bm_perf
...
Update latency profiler to use (more appropriate) microbenchmarks
2017-02-08 13:57:09 -08:00
David G. Quintas
5ef95a96ab
Merge pull request #9194 from dgquintas/lb_secure_naming_3
...
Secure naming support for gRPCLB.
2017-02-07 21:00:59 -08:00
David Garcia Quintas
012915045f
Secure naming support for gRPCLB
2017-02-07 17:41:57 -08:00
Paul Marks
3a5bba0a20
Connect to "localhost" from interop clients.
...
This is for compatibility with IPv6-only environments; see the Google
bug b/32531998.
Note that InteropClient.cs and http2_client.cc don't appear to be tested
by our internal branch, but I've included them for consistency.
2017-02-07 16:28:09 -08:00
Craig Tiller
ea2c59420c
Merge pull request #9619 from ctiller/bm_counters
...
Add counters for important metrics to bm_fullstack
2017-02-07 15:16:40 -08:00
yang-g
076bac017b
Remove async
2017-02-07 13:50:36 -08:00
yang-g
13a537bcae
merge with head
2017-02-07 13:13:20 -08:00
Harvey Tuch
5f3cfe960f
Fix read from uninitialized memory bug in GrpcBufferWriter.
...
This commit fixes an issue in which the following sequence of operations
leads to use of uninitialized memory:
1. Caller invokes GrpcBufferWriter::Next(), and then makes use of 8191
bytes in the returned buffer (which is 8192 bytes in size).
2. Caller then returns the unused single byte via
GrpcBufferWriter::BackUp(). This method invokes
g_core_codegen_interface->grpc_slice_split_tail(), which causes
backup_slice_ to be a grpc_slice with one byte.
3. At the next invocation of GrpcBufferWriter::Next(), a reference to
the single byte grpc_slice is returned to the caller.
The problem here is that the returned reference is to the inlined buffer
in the grpc_slice, which is resident in slice_, not the location of the
buffer inside slice_buffer_ after
g_core_codegen_interface->grpc_slice_buffer_add() in
GrpcBufferWriter::Next(). As a result, any data the caller writes to the
returned void* data is lost.
The solution is to avoid inlined backup slices.
2017-02-07 15:45:44 -05:00
Craig Tiller
d52b76de44
Merge github.com:grpc/grpc into bm_perf
2017-02-07 10:30:58 -08:00
Sree Kuchibhotla
3d4f443527
Merge pull request #9590 from sreecha/micro_bm
...
Add Streaming ping-pong micro-benchmarks
2017-02-07 10:21:44 -08:00
Craig Tiller
01d7d9b2e5
Add allocations/iteration counters to bm_fullstack
2017-02-07 08:39:05 -08:00