There is no reason for this function, tokio already blocks automatically until all tasks are ended.
Another reason to remove this feature is `mpsc_background_tasks` unbounded channel. Recently this
channel was reporting too many unprocessed elements. We assume that this was a result of a lot of
very shot lived tasks that somehow flooded this channel.
* refractored group param to enum
* refractored group param to enum
* changed group param to GroupName enum for other methods too such as spawn_inner
updated docs
* revert on task type
* put back ticks in GroupName
* Update client/service/src/task_manager/mod.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* document group name
change specific to actual in enum declaration
* change documentation
* Update client/service/src/task_manager/mod.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* changed Actual to Specific
* Update client/service/src/task_manager/mod.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* Update client/service/src/task_manager/mod.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Update client/service/src/task_manager/mod.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Update client/service/src/task_manager/mod.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
Co-authored-by: Damilare <dakinlose@teamapt.com>
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
Co-authored-by: Andronik Ordian <write@reusable.software>
* state machine proofs.
* initial implementation
* Remove todo.
* Extend test and fix import.
* fix no proof, with proof ko.
* fix start at logic.
* Restore response size.
* Rework comments.
* Add explicit ref
* Use compact proof.
* ref change
* elaborato on empty change set condition.
* KeyValueState renaming.
* Do not add two time child trie with same root to sync reply.
* rust format
* Fix merge.
* fix warnings and fmt
* fmt
* update protocol id to V2
* Removal of light client from substrate
* add missing import
* These tests relate to there being light and non light clients.
* removing lightnodes from test
* cargo fmt
* not needed
* LightDataChecker not needed any longer
* cargo fmt
* Update client/service/test/src/lib.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Update client/service/test/src/lib.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* cargo fmt
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Disable log reloading by default
This disables the log reloading that was enabled by default. The problem
is that the log reload implementation of `tracing` is using a lock to
make the layer replaceable. This lock needs to be locked every time we
need to check if a particular target is enabled (assuming the log level
is high enough). This kills the performance when for example
`sometarget=trace` logging is enabled.
* 🤦
* Remove unused parameter
* Fix test
* Fix
* Prepare UI tests for rust 1.54
* Delete wrong_page.stderr
* CI: run with a staging CI image
* Revert "CI: run with a staging CI image"
This reverts commit 66f5b00d14b50fd9d8fbf773f7e884f380697591.
* CI: debug, again
* LOG_TARGET is only used in std
* Remove unnecessary unsafe
* Fixes
* Use correct rustc locally
* FMT
* Compile with benchmarking
* Review feedback
* Some ui tests
* I know...
* Fix wasm tests
Co-authored-by: Denis P <denis.pisarev@parity.io>
Co-authored-by: thiolliere <gui.thiolliere@gmail.com>
* Use tokio runtime handle instead of TaskExecutor abstraction
Before this pr we had the `TaskExecutor` abstraction which theoretically
allowed that any futures executor could have been used. However, this
was never tested and is currently not really required. Anyone running a
node currently only used tokio and nothing else (because this was hard
coded in CLI). So, this pr removes the `TaskExecutor` abstraction and
relies directly on the tokio runtime handle.
Besides this changes, this pr also makes sure that the http and ws rpc
server use the same tokio runtime. This fixes a panic that occurred when
you drop the rpc servers inside an async function (tokio doesn't like
that a tokio runtime is dropped in the async context of another tokio
runtime).
As we don't use any custom runtime in the http rpc server anymore, this
pr also removes the `rpc-http-threads` cli argument. If external parties
complain that there aren't enough threads for the rpc server, we could
bring support for increasing the thread count of the tokio runtime.
* FMT
* Fix try runtime
* Fix integration tests and some other optimizations
* Remove warnings
* sp-utils => sc-utils
* cargo fmt
* These files are now in the client so should be licensed as GPL3
* Apply suggestions from code review
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
Before we fetched the runtime code from the `TrieBackend` and this lead
to not using the storage cache. Thus, we recalculated the storage hash
for the runtime code on every call into the runtime and this killed the
performance on parachains block authoring. The solution is to fetch the
runtime code from the storage cache, to make sure we use the cached
storage cache.
* Split native executor stuff from wasm executor stuff
* Remove `native_runtime_version` in places
* Fix warning
* Fix test warning
* Remove redundant NativeRuntimeInfo trait
* Add a warning for use_native
* Run cargo fmt
* Revert "Add a warning for use_native"
This reverts commit 9494f765a06037e991dd60524f2ed1b14649bfd6.
* Make choosing an executor (native/wasm) an explicit part of service construction
* Add Cargo.lock
* Rename Executor to ExecutorDispatch
* Update bin/node/executor/src/lib.rs
Co-authored-by: Squirrel <gilescope@gmail.com>
* Fix tests
* Fix minor node-executor error
* Fix node cli command thing
Co-authored-by: Squirrel <gilescope@gmail.com>
* Comment out browser stuff
* Remove browser stuff
* Remove more wasm transport code
* Remove ExtTransport and rework how telemetry initialises.
* Change (most) wasm-timer using code to use std::time
* Rename CI-job
* Aura does not compile for wasm
* Remove testing in the browser on CI
* Update README
* Leave `StreamSink` be
* fmt
* implement "auto" database backend in client/db, in progress, #9201
* move fn supports_ref_counting from DatabaseSource enum to Database trait to make it work correctly for all types of dbs
* update kvdb_rocksdb to 0.13 and use it's new config feature to properly auto start existing database
* tests for auto database reopening
* introduce OpenDbError to cleanup opening database error handling and handle case when database is not enabled at the compile time
* cargo fmt strings again
* cargo fmt strings again
* rename DataSettingsSrc to fix test compilation
* fix the call to the new kvdb-rocksdb interdace in tests to fix compilation
* simplify OpenDbError and make it compile even when paritydb and rocksdb are disabled
* cargo fmt
* fix compilation without flag with-parity-db
* fix unused var compilation warning
* support different paths for rocksdb and paritydb in DatabaseSouce::Auto
* support "auto" database option in substrate cli
* enable Lz4 compression for some of the parity-db colums as per review suggestion
* applied review suggestions
* Decouples light-sync state from chain spec
This decouples the light-sync state from chain spec. First, the
light-sync state currently only works with BABE+Grandpa, so not
all *Substrate* based chains can use this feature. The next problem was
also that this pulled the `sc-consensus-babe` and `sc-finality-grandpa`
crate into `sc-chain-spec`.
If a chain now wants to support the light-sync state, it needs to add
the `LightSyncStateExtension` to the chain spec as an extension. This is
documented in the crate level docs of `sc-sync-state-rpc`. If this
extension is not available, `SyncStateRpc` fails at initialization.
* Fix compilation for browser
* Fmt
* Improve `state` related logs to use a more uniform format
The logging before wasn't that uniform and not that great to read/parse.
Now we are using a uniform format for all the logs. Besides these
changes, there are some minor changes around the code that calls the
state machine.
* Make CI happy
* Use HexDisplay for `ext_id`
* Run cargo fmt on the whole code base
* Second run
* Add CI check
* Fix compilation
* More unnecessary braces
* Handle weights
* Use --all
* Use correct attributes...
* Fix UI tests
* AHHHHHHHHH
* 🤦
* Docs
* Fix compilation
* 🤷
* Please stop
* 🤦 x 2
* More
* make rustfmt.toml consistent with polkadot
Co-authored-by: André Silva <andrerfosilva@gmail.com>
* Do not call `initialize_block` before any runtime api
Before this change we always called `initialize_block` before calling
into the runtime. There was already support with `skip_initialize` to skip
the initialization. Almost no runtime_api requires that
`initialize_block` is called before. Actually this only leads to higher
execution times most of the time, because all runtime modules are
initialized and this is especially expensive when the block contained a
runtime upgrade.
TLDR: Do not call `initialize_block` before calling a runtime api.
* Change `validate_transaction` interface
* Fix rpc test
* Fixes and comments
* Some docs
* Make it possible to override maximum payload of RPC
* Finish it.
* remove todo.
* Update client/cli/src/commands/run_cmd.rs
* Apply suggestions from code review
Co-authored-by: David <dvdplm@gmail.com>
* Apply suggestions from code review
Co-authored-by: David <dvdplm@gmail.com>
* Incorporate suggestions
* Thread rpc_max_payload from configuration to trace_block
* Try obey line gitlab/check_line_width.sh
* update state rpc tests
* Improve readbility
* Apply suggestions from code review
* Apply suggestions from code review
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
Co-authored-by: Zeke Mostov <32168567+emostov@users.noreply.github.com>
Co-authored-by: David <dvdplm@gmail.com>
* Reexport ExecutionStrategies and ExecutionStrategy
* Reexport more of the network
* Reexport the ExecutionStrategy as it's used within ExecutionStrategies
* Removes unnecessary blank impl for Backend
This commit removes a from my perspective unneccessary implementation
for &T which implement Backend.
The current implementation exists (again from my perspective) solely
to satisfy a methods &mut self parameters (i.e. allows to satisfy
this for an & reference via using &mut &Backend).
As all implementors use a RefCell with borrow_mut() where actually
calling the mentioned &mut self method and then forwad to the
{} implementation of either TrieBackend or ProvingBackend, the
current &mut self seems to be not needed.
* Fixed tests client
This introduces a new field `code_substitute` into the chain spec. This
can be used to substitute the on-chain wasm starting from a given block
until there is another wasm on chain (determined through the
`spec_version`). This can be used to fix broken on chain wasm runtimes.
Bastian Köcher