More generic configuration system is introduced in order to i) unify the
way how modules access the configurations instead of using low-level
get/setenv functions and ii) enable the customization for where configuration
is stored. This could be extended to support flag, file, etc.
Default configuration system uses environment variables as before so
basically this is expected to work just as it did. This behavior can
change by redefining GPR_GLOBAL_CONFIG_DEFINE_*type* macros.
* Migrated configuration
GRPC_CLIENT_CHANNEL_BACKUP_POLL_INTERVAL_MS
GRPC_EXPERIMENTAL_DISABLE_FLOW_CONTROL
GRPC_ABORT_ON_LEAKS
GRPC_NOT_USE_SYSTEM_SSL_ROOTS
take_first(), grpc_undo_first(), ... are very clostly
methods that unnecessarily copy grpc_slice with extra
ref counting requirements.
Introduce alternatives to avoid copies and refs wherever
possible.
This results in 1% improvements in the benchmarks.
alignment options (for cache-alignment).
We shrink by:
1) Removing an unnecessary zone pointer.
2) Replacing gpr_mu (40 bytes when using pthread_mutex_t) with
std::atomic_flag.
We also header-inline the fastpath alloc (ie. when not doing a zone
alloc) and move the malloc() for a zone alloc outside of the mutex
critical zone, which allows us to replace the mutex with a spinlock.
We also cache-align created arenas.
grpc_byte_buffer_reader_next() copies and references the slice. This
is not always necessary since the caller will not use the slice
after destroying the byte buffer.
A prominent example is the protobuf parser, which
calls grpc_byte_buffer_reader_next() and immediately unrefs the slice
after the call. This ref() and unref() calls can be very expensive
in the hot path.
This commit introduces grpc_byte_buffer_reader_peek() which
essentialy return a pointer to the slice in the buffer, i.e.,
no copies, and no refs.
QPS of 1MiB 1 Channel callback benchmark increases by 5%.
More importantly insructions per cycle is increased by 10%.
Also add tests and benchmarks for byte_buffer_reader_peek()
This commit reaplies 509e77a5a3
- Pass extra param to grpc_endpoint_read() as the API has changed.
- Fixed build error seen with Xcode 10.
- Enable pipefail to xcodebuild errors are propagated to the caller.
TCP_INQ is a socket option we added to Linux to report pending bytes
on the socket as a control message.
Using TCP_INQ we can accurately decide whether to continue read or not.
Add an urgent parameter, when we do not want to wait for EPOLLIN.
This commit improves the latency of 1 RPC unary (minimal benchmark)
significantly:
Before:
l_50: 61.3584984733
l_90: 94.8328711277
l_99: 126.211351174
l_999: 158.722406029
After:
l_50: 51.3546011488 (-16%)
l_90: 72.3420731581 (-23%)
l_99: 103.280218974 (-18%)
l_999: 130.905689996 (-17%)