This still assumes that the client did not start any transactions
before calling into runtime. This is the case for benchmarking
as long as either NativeWhenPossible or AlwaysWasm exection
strategy is chosen. Using any other will result in a panic.
* Add transactional storage functionality to OverlayChanges
A collection already has a natural None state. No need to
wrap it with an option.
* Add storage transactions runtime interface
* Add frame support for transactions
* Fix committed typo
* Rename 'changes' variable to 'overlay'
* Fix renaming change
* Fixed strange line break
* Rename clear to clear_where
* Add comment regarding delete value on mutation
* Add comment which changes are covered by a transaction
* Do force the arg to with_transaction return a Result
* Use rust doc comments on every documentable place
* Fix wording of insert_diry doc
* Improve doc on start_transaction
* Rename value to overlayed in close_transaction
* Inline negation
* Improve wording of close_transaction comments
* Get rid of an expect by using get_or_insert_with
* Remove trailing whitespace
* Rename should to expected in tests
* Rolling back a transaction must mark the overlay as dirty
* Protect client initiated storage tx from being droped by runtime
* Review nits
* Return Err when entering or exiting runtime fails
* Documentation fixup
* Remove close type
* Move enter/exit runtime to excute_aux in the state-machine
* Rename Discard -> Rollback
* Move child changeset creation to constructor
* Move child spawning into the closure
* Apply suggestions from code review
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Fixup for code suggestion
* Unify re-exports
* Rename overlay_changes to mod.rs and move into subdir
* Change proof wording
* Adapt a new test from master to storage-tx
* Suggestions from the latest round of review
* Fix warning message
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Don't clone values when calculating storage root
Instead of cloning all the keys and values of the overlay when
calculating the storage root, we pass all the values by reference. This
should probably bring some performance improvements when calculating the
storage root.
* no cow version (#6113)
Co-authored-by: cheme <emericchevalier.pro@gmail.com>
* Hide internal structure of OverlayChanges
* Fix tests for OverlayChanges refactor
* Do not clone pending changes
Discarding prospective changes should be equivalent as a state machine
is not to be called with peding changes.
This will be replaced by a storage transaction that is rolled back before
executing the call the second time removing this constraint.
* Doc fixes
* Remove overlong line
* Revert "Do not clone pending changes"
This reverts commit 4799491f4ac16f8517287a0fcf4a3f84ad56f46e.
* Deduplicate chield tries returned from child_infos()
* Remove redundant type annotation
* Avoid changing the storage root in tests
* Preserve extrinsic indices in trie build test
* Swap order of comitted and prospective in fn child_infos
This is only for consistency and does not impact the result.
* Rename set_pending to replace_pending for clearity
* Avoid changing overlay committed layer.
* basic test
* Add some tx in the test.
* only update from backend value on missing entry in both layer.
deleted entry is replace by empty vec.
* test and review changes
* additional test and review change
* remove test on changing existing value, it does not always panic
depending on existing content
* Update primitives/state-machine/src/overlayed_changes.rs
* Update primitives/state-machine/src/overlayed_changes.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Update to latest staking
* generate tests for benchmarking
* add tests, fix warnings
* starting on democracy
* impl_benchmark_tests
* Way more readable
* add test feature flag (does this work?)
* Fix `successful_origin` impl
* democracry benchmark tests
* Fix example benchmarks, add tests
* identity benchmark tests
* Update im-online benchmark tests
* try to add session benchmarking tests (problem with mock)
* staking and timestamp
* add test for treasury, issue with dynamic contains
* utility
* Vesting
* test instead of check
* hide until we figure out what is wrong
* add docs
* close code
* Create custom mock for session-pallet-benchmarking
* Use refcell pattern
* make un-pub
* test-linux-stable includes `runtime-benchmarks` feature
* Revert "test-linux-stable includes `runtime-benchmarks` feature"
This reverts commit a2dab38abd18ac3eb8a6220e4a00e687740bd38c.
* run tests in `--release`
* undo balance change
* build wasm
* Don't include `:code` by default in storage proofs (#5060)
* Adds test to verify that the runtime currently is always contained in
the proof
* Start passing the runtime wasm code from the outside
* Fix compilation
* More build fixes
* Make the test work as expected now :)
* Last fixes
* Fixes benchmarks
* Review feedback
* Apply suggestions from code review
Co-Authored-By: Sergei Pepyakin <sergei@parity.io>
* Review feedback
* Fix compilation
Co-authored-by: Sergei Pepyakin <s.pepyakin@gmail.com>
* Fix compilation and change the way `RuntimeCode` works
* Fix tests
* Switch to `Cow`
Co-authored-by: Benjamin Kampmann <ben@gnunicorn.org>
Co-authored-by: Sergei Pepyakin <s.pepyakin@gmail.com>
This reduces the usage of `Blake2Hasher` in the code base and replaces
it with `BlakeTwo256`. The most important change is the removal of the
custom extern function for `Blake2Hasher`. The runtime `Hash` trait is
now also simplified and directly requires that the implementing type
implements `Hashable`.
* Adds test to verify that the runtime currently is always contained in
the proof
* Start passing the runtime wasm code from the outside
* Fix compilation
* More build fixes
* Make the test work as expected now :)
* Last fixes
* Fixes benchmarks
* Review feedback
* Apply suggestions from code review
Co-Authored-By: Sergei Pepyakin <sergei@parity.io>
* Review feedback
* Fix compilation
Co-authored-by: Sergei Pepyakin <s.pepyakin@gmail.com>
* Adds documentation for `wipe` and `commit`
This adds documentation to `wipe` and `commit` of `Externalities`.
Besides that it removes the default implementation that would just panic
and requires that all implementers of the trait implement the functions.
* Update primitives/externalities/src/lib.rs
Co-Authored-By: joe petrowski <25483142+joepetrowski@users.noreply.github.com>
Co-authored-by: joe petrowski <25483142+joepetrowski@users.noreply.github.com>
* Implement runtime version checks in `set_code`
Check that the new runtime code given to `set_code` fullfills some
requirements:
- `spec_name` matches
- `spec_version` does not decreases
- `impl_version` does not decreases
- Either `spec_version` and `impl_version` increase
* Make tests almost work
* Some fixes after master merge
* Fix tests
* Add missed file
* Make depedency check happy?
* Remove leftover `sc-executor`
* AHHHHH
* Reset debug stuff
* Remove some 'static
* More 'static
* Some docs
* Update `Cargo.lock`
* Extend `Proposer` to optionally generate a proof of the proposal
* Something
* Refactor sr-api to not depend on client anymore
* Fix benches
* Apply suggestions from code review
Co-Authored-By: Tomasz Drwięga <tomusdrw@users.noreply.github.com>
* Apply suggestions from code review
* Introduce new `into_storage_changes` function
* Switch to runtime api for `execute_block` and don't require `H256`
anywhere in the code
* Put the `StorageChanges` into the `Proposal`
* Move the runtime api error to its own trait
* Adds `StorageTransactionCache` to the runtime api
This requires that we add `type NodeBlock = ` to the
`impl_runtime_apis!` macro to work around some bugs in rustc :(
* Remove `type NodeBlock` and switch to a "better" hack
* Start using the transaction cache from the runtime api
* Make it compile
* Move `InMemory` to its own file
* Make all tests work again
* Return block, storage_changes and proof from Blockbuilder::bake()
* Make sure that we use/set `storage_changes` when possible
* Add test
* Fix deadlock
* Remove accidentally added folders
* Introduce `RecordProof` as argument type to be more explicit
* Update client/src/client.rs
Co-Authored-By: Tomasz Drwięga <tomusdrw@users.noreply.github.com>
* Update primitives/state-machine/src/ext.rs
Co-Authored-By: Tomasz Drwięga <tomusdrw@users.noreply.github.com>
* Integrates review feedback
* Remove `unsafe` usage
* Update client/block-builder/src/lib.rs
Co-Authored-By: Benjamin Kampmann <ben@gnunicorn.org>
* Update client/src/call_executor.rs
* Bump versions
Co-authored-by: Tomasz Drwięga <tomusdrw@users.noreply.github.com>
Co-authored-by: Benjamin Kampmann <ben.kampmann@googlemail.com>
* In progress, runtime io must switch to future proof root +
child_specific (unique id) + u32 type.
* Switch interface, sr-io seems ok, rpc could use similar interface to
sr-io, genesis json broken if there is child trie in existing encoding
genesis.
* test from previous implementation.
* fix proving test.
* Restore Keyspacedb from other branch, only apply to child trie.
* Removing unneeded child_info from child root (child info are stored
if things changed, otherwhise the root does not change).
* Switch rpc to use same format as ext: more future proof.
* use root from child info for trie backend essence.
* Breaking long lines.
* Update doc and clean pr a bit.
* fix error type
* Restore removed doc on merge and update sr-io doc.
* Switch child storage api to use directly unique id, if managed id
where to be put in place, the api will change at this time.
* Clean deprecated host interface from child.
* Removing assertion on child info (can fail depending on root
memoization).
* merging child info in the overlay when possible.
* child iteration by prefix using child_info.
* Using ChainInfo in frame support. ChainInfo gets redesign to avoid
buffers allocation on every calls.
* Add length of root to the data of child info.
* comments
* Encode compact.
* Remove child info with root.
* Fix try_update condition.
* Comment Ext child root caching.
* Replace tuples by struct with field
* remove StorageTuple alias.
* Fix doc tests, and remove StorageOverlay and ChildStorageOverlay
aliases.
* The storage runtime interface should not enforce a hash type
Currently the runtime interface enforces `H256` as hash type, but in the
future people could use whatever they want as hash type. The hash type
always needs to match between the runtime and the node, but that is
already required.
* Update primitives/externalities/src/lib.rs
Co-Authored-By: thiolliere <gui.thiolliere@gmail.com>
Alexander Theißen