This repository has been archived by the owner on Nov 15, 2023. It is now read-only.
Fill up request slots via launch_parallel_requests
#3681
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
in case waiting for the next response takes too long.
eskimor
added
A0-please_review
eskimor
changed the title
launch_parallel_requestslaunch_parallel_requests
rphmeier
reviewed
Aug 23, 2021
| // We will also stop, if there has not been a response for `MAX_CHUNK_WAIT`, so | ||
| // `launch_parallel_requests` cann fill up slots again. | ||
| while let Some(request_result) = | ||
| self.requesting_chunks.next().timeout(MAX_CHUNK_WAIT).await.flatten() |
There was a problem hiding this comment.
Previously, a timeout would lead to an invocation of self.shuffling.push_front(validator_index). Now, it no longer does. This could also lead to mistaken "unavailable" returns from the process.
There was a problem hiding this comment.
It still does. I did not change that code. The above timeout is a soft timeout, we still wait for the futures to return and they will still return with a timeout, triggering the push_front call. The only difference is that in the meantime we filled up free slots with new futures.
There was a problem hiding this comment.
Ah, I see. The timeout causes the loop to break so we do another loop iteration where we populate chunks.
ordian
approved these changes
Aug 24, 2021
ordian
added a commit
that referenced
this pull request
Aug 24, 2021
* master: backing-availability-audit: Move ErasureChunk Proof to BoundedVec (#3626) Substrate Companion #9575 (#3695) Fill up requests slots via `launch_parallel_requests` (#3681) Bump serde_json from 1.0.64 to 1.0.66 (#3669) substrate #9202 companion: Multiple vesting schedules (#3407) XCM: Introduce versioning to dispatchables' params (#3693) remove dead_code from chain selection test (#3685) Improve MultiLocation conversion functions in xcm-procedural (#3690)
chevdor
pushed a commit
that referenced
this pull request
Sep 6, 2021
in case waiting for the next response takes too long.
chevdor
pushed a commit
that referenced
this pull request
Sep 10, 2021
in case waiting for the next response takes too long.
chevdor
pushed a commit
that referenced
this pull request
Sep 11, 2021
in case waiting for the next response takes too long.
chevdor
pushed a commit
that referenced
this pull request
Sep 13, 2021
in case waiting for the next response takes too long.
s3krit
pushed a commit
that referenced
this pull request
Sep 15, 2021
* Companion for substrate/pull/9442 (#3594) * Fix compact build * Fix again * fix staking miner * fmt * update Substrate Co-authored-by: parity-processbot <> * More standard staking miner deposits (#3621) * XCM v1 (#2815) * MultiAsset TWO * Draft next MultiAsset API. * XCM core builds * XCM Executor builds * XCM Builder builds * API changes making their way throughout * Some TODOs * Further build fixes * Basic compile builds * First test fixed * All executor tests fixed * Typo * Optimize subsume_assets and add test * Optimize checked_sub * XCM Builder first test fixed * Fix builder tests * Fix doc test * fix some doc tests * spelling * named fields for AllOf * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Reformat * Move to XCM version 1 * Spelling * warnings * Replace some more v0->v1s * warnings * format * Add max_assets param * building * test fixes * tests * another test * final test * tests * Rename Null -> Here * Introduce * More ergonomics * More ergonomics * test fix * test fixes * docs * BuyExecution includes * Fix XCM extrinsics * fmt * Make Vec<MultiAsset>/MultiAssets conversions safe * More MultiAssets conversion safety * spelling * fix doc test * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * fmt * Add v0, remove VersionedMultiAsset * Remove VersionedMultiLocation * Update xcm/src/v1/order.rs Co-authored-by: Amar Singh <[email protected]> * Update xcm/src/v1/mod.rs Co-authored-by: Amar Singh <[email protected]> * XCM v0 backwards compatibility * Full compatibility * fmt * Update xcm/pallet-xcm/src/lib.rs * Update xcm/src/v0/order.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Tweaks to versioning system * Fixes * fmt * Update xcm/xcm-executor/src/assets.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Update xcm/xcm-executor/src/assets.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Grumbles * Update xcm/src/v1/multiasset.rs Co-authored-by: Shawn Tabrizi <[email protected]> * fmt * Update xcm/src/v1/multiasset.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Update xcm/src/v1/multiasset.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Fixes * Formatting Co-authored-by: Alexander Popiak <[email protected]> Co-authored-by: Amar Singh <[email protected]> Co-authored-by: Shawn Tabrizi <[email protected]> * XCM v1 (#2815) * MultiAsset TWO * Draft next MultiAsset API. * XCM core builds * XCM Executor builds * XCM Builder builds * API changes making their way throughout * Some TODOs * Further build fixes * Basic compile builds * First test fixed * All executor tests fixed * Typo * Optimize subsume_assets and add test * Optimize checked_sub * XCM Builder first test fixed * Fix builder tests * Fix doc test * fix some doc tests * spelling * named fields for AllOf * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Reformat * Move to XCM version 1 * Spelling * warnings * Replace some more v0->v1s * warnings * format * Add max_assets param * building * test fixes * tests * another test * final test * tests * Rename Null -> Here * Introduce * More ergonomics * More ergonomics * test fix * test fixes * docs * BuyExecution includes * Fix XCM extrinsics * fmt * Make Vec<MultiAsset>/MultiAssets conversions safe * More MultiAssets conversion safety * spelling * fix doc test * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * fmt * Add v0, remove VersionedMultiAsset * Remove VersionedMultiLocation * Update xcm/src/v1/order.rs Co-authored-by: Amar Singh <[email protected]> * Update xcm/src/v1/mod.rs Co-authored-by: Amar Singh <[email protected]> * XCM v0 backwards compatibility * Full compatibility * fmt * Update xcm/pallet-xcm/src/lib.rs * Update xcm/src/v0/order.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Tweaks to versioning system * Fixes * fmt * Update xcm/xcm-executor/src/assets.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Update xcm/xcm-executor/src/assets.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Grumbles * Update xcm/src/v1/multiasset.rs Co-authored-by: Shawn Tabrizi <[email protected]> * fmt * Update xcm/src/v1/multiasset.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Update xcm/src/v1/multiasset.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Fixes * Formatting Co-authored-by: Alexander Popiak <[email protected]> Co-authored-by: Amar Singh <[email protected]> Co-authored-by: Shawn Tabrizi <[email protected]> * Companion to #9514 (Remove Filter and use Contains instead) (#3591) * Remove Filter and use Contains instead * Fixes * Remove patch * Formatting * update Substrate Co-authored-by: parity-processbot <> * Fix Backwards Compatability with v0 Response (#3597) * fix junction + response * Update xcm/src/v1/mod.rs * better comment * Minor fix to encoding for XCM v1 (#3602) * Ensure MultiLocation always has a canonical representation (#3404) * MultiAsset TWO * Ensure MultiLocation always has a canonical representation * Remove v1 module * Draft next MultiAsset API. * Implement custom encoding/decoding scheme for MultiLocation * Properly implement IntoIterator for Junctions * Implement TryFrom<MultiLocation> for Junctions * Fix spelling mistakes * Fix tests in xcm-executor * XCM core builds * XCM Executor builds * XCM Builder builds * Fix xcm-builder tests and compilation * Make pallet-xcm compile * Use MultiLocation::default() * Make polkadot-runtime-common compile * Make rococo-runtime compile * Change return type of parent_count to u8 * Change MAX_MULTILOCATION_LENGTH to 255 * Make kusama-runtime compile * Fix logic in pallet-xcm * Use MultiLocation::empty() * Fix logic in location_conversion * Fix logic in origin_conversion.rs * Make westend-runtime compile * Rename prefixes and suffixes variables * Rename non_parent to interior * Rename non_parent to interior * Add test for encode/decode roundtrip and fix decode algorithm * API changes making their way throughout * Some TODOs * Further build fixes * Rename non_parent/junctions to interior * Basic compile builds * First test fixed * All executor tests fixed * Typo * Optimize subsume_assets and add test * Optimize checked_sub * XCM Builder first test fixed * Fix builder tests * Fix doc test * Make xcm-simulator compile * Make xcm-simulator-example compile * Make spellcheck happy * cargo fmt * fix some doc tests * spelling * named fields for AllOf * Fix subtle bug where Null is treated as an identifier * Add FIXME comment awaiting for const generics eval stabilization * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Update xcm/src/v0/multiasset.rs Co-authored-by: Alexander Popiak <[email protected]> * Reformat * Move to XCM version 1 * Spelling * warnings * Replace some more v0->v1s * warnings * format * Add max_assets param * building * test fixes * tests * another test * final test * Update rustdocs and add debug_assert where sensible * Revert debug_assert in const fn len() * tests * Rename Null -> Here * Introduce * More ergonomics * More ergonomics * test fix * test fixes * docs * BuyExecution includes * Fix XCM extrinsics * fmt * Make Vec<MultiAsset>/MultiAssets conversions safe * More MultiAssets conversion safety * spelling * fix doc test * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * fmt * Add v0, remove VersionedMultiAsset * Remove VersionedMultiLocation * Update xcm/src/v1/order.rs Co-authored-by: Amar Singh <[email protected]> * Update xcm/src/v1/mod.rs Co-authored-by: Amar Singh <[email protected]> * XCM v0 backwards compatibility * Full compatibility * fmt * Update xcm/pallet-xcm/src/lib.rs * Update xcm/src/v0/order.rs Co-authored-by: Shawn Tabrizi <[email protected]> * Tweaks to versioning system * Fixes * fmt * Fix pallet-xcm tests * fix * Substitute with_parent with with_parents_const * Rename argument name from a to m * Rename split_last to split_last_interior * Allow adding multiple parents in MultiLocation * Rename pop_parent to dec_parent * Ensure relay chain XCM sender receives a MultiLocation without any parents * Block only when MultiLocation destination length is 8 * Cargo fmt * Remove reverse iterators, implement DoubleEndedIterator and add tests * Fix iter_rev lifetime requirements * Cargo fmt * Add an into() method for Junctions for conciseness in const context * Ensure parent count is 0 while executing who in RelayedFrom * Appease spellchecker * Use and_then instead of repeated map_err * Remove custom codec indices for v1 Junctions * Add convenience 'contains_parents_only' method to MultiLocation * Fix merge conflict * Use more convenience methods * Remove with_parachain_interior * Prefer matching against tuple instead of using match guards * Match against tuple instead of using more match guards * Update encode/decode test for MultiLocation * Minor tweaks * Fixes * Fixes * Fixes * Fix MultiLocation * Add deprecation note for iter_rev and into_iter_rev * Update some rustdocs * cargo fmt * Fix xcm-executor unit tests * Fix compilation and unit tests in xcm-builder * cargo fmt * Fix tests in xcm-simulator-example * Publicize MultiLocation fields * Match on the MultiLocation struct directly in xcm-builder * Do not dereference undereferenceable types * Add convenience MultiLocation conversions for tuples * Use clearer import paths * Remove unused dependency * fix junction + response * Import from latest opaque xcm module * Update xcm/src/v1/mod.rs * better comment * Fix ownership transfer * Fix merge * Fix merge * cargo fmt * Fix merge * Fix merge * Fix integration test * More readable Parent syntax * cleanup * cleanup * cleanup * cleanup * cleanup * cleanup * cleanup * cleanup * cargo fmt * Fixes * Fix doc test Co-authored-by: Gav Wood <[email protected]> Co-authored-by: Alexander Popiak <[email protected]> Co-authored-by: Amar Singh <[email protected]> Co-authored-by: Shawn Tabrizi <[email protected]> * Make RelayedFrom typesafe (#3617) * Corrent type * Fix build * Fixes * Fixes * Formatting * Remove BuyExecution::orders (#3640) * Remove BuyExecution::orders * Fixes * Fixes * Fixes * Formatting * XCM: Introduce versioning to dispatchables' params (#3693) * Introduce versioning to dispatchables' params * Fixes * Formatting * Bump * Bump deps * Fixes * Fixes * check runtime version in staking miner (#3628) * check runtime version in staking miner * fmt * add short alias for things * fix fee * print length as well * fix build * review comments * add integration tests to xcm-builder (#3537) * add integration tests to xcm-builder * add an integration test for reserve_transfer_assets * add query holding and teleport tests * formatting * add to barrier doc comments and fix doc tests warnings * use more realistic barrier for integration tests * improve imports * adjust base xcm weight and existential deposit to be in line with Kusama * remove AnyNetwork Co-authored-by: Amar Singh <[email protected]> * add more comments and remove unnecessary code * move mock into separate file * reduce imports * update cargo.lock * remove reserve transfer test from xcm builder integration tests * reword barrier doc comment * elaborate on QueryHolding test scenario * add an integration test for reserve based transfers from parachain to parachain * add teleport tests * fix failing teleport filter tests * Update xcm/xcm-builder/src/integration_tests.rs Co-authored-by: Kian Paimani <[email protected]> * Update xcm/xcm-builder/src/integration_tests.rs Co-authored-by: Kian Paimani <[email protected]> * Update xcm/xcm-builder/src/integration_tests.rs Co-authored-by: Kian Paimani <[email protected]> * Move integration tests to tests/ directory * Fix merge * Replace All wildcard with a concrete seed amount * Rename SEED_AMOUNT to REGISTER_AMOUNT * Fix compilation error * Check for teleport destination first before checking out assets * Fix unit test * Do not run tests in integration mock * Add a permissive assets filter for teleportation * Remove check for teleport location in InitiateTeleport XCM * Remove defunct test * Apply suggestions from code review Co-authored-by: Kian Paimani <[email protected]> * Reword comment Co-authored-by: Amar Singh <[email protected]> Co-authored-by: Kian Paimani <[email protected]> Co-authored-by: Keith Yeung <[email protected]> * Impl WeightTrader for tuple (#3601) * Impl WeightTrader for tuple. * fmt * Renaming. * add tracing for buy_weight * Add comment clarifying the default behavior of a WeightTrader tuple Co-authored-by: Alexander Popiak <[email protected]> Co-authored-by: Keith Yeung <[email protected]> * fix compilation (#3657) * Use proc macros to generate conversion functions for MultiLocation (#3635) * Use proc macros to generate conversion functions for MultiLocation * Add compile test and missing conversion cases * Add common derives for Parent and Ancestor * Generate conversion functions for MultiLocation v0 via proc macro * Add type conversion test and fix a bug * cargo fmt * Do not hardcode 8 as the number of max parents * Use map instead of for loops when generating code fragments * Spelling * cargo fmt * More mapping, less for-looping * Improve MultiLocation conversion functions in xcm-procedural (#3690) * Use fully-qualified paths to reference core traits in proc macro * Ensure a canonicalized v0 MultiLocation first before attempting to convert to v1 * Fix failing test * XCM v1 version notification stub (#3766) * Initial draft * v1 subscriptions * Fixes * Fixes * Formatting * Fixes * Formatting * chore: update substrate to branch polkadot-v0.9.10 * chore: update beefy to branch polkadot-v0.9.10 * Provide dummy dispute coordinator by default. (#3661) * Provide dummy dispute coordinator by default. * cargo fmt * Fill up requests slots via `launch_parallel_requests` (#3681) in case waiting for the next response takes too long. * Fix missing overseer feature * Update beefy deps * xcm executor version bump * Remove request multiplexer (#3624) * WIP: Get rid of request multiplexer. * WIP * Receiver for handling of incoming requests. * Get rid of useless `Fault` abstraction. The things the type system let us do are not worth getting abstracted in its own type. Instead error handling is going to be merely a pattern. * Make most things compile again. * Port availability distribution away from request multiplexer. * Formatting. * Port dispute distribution over. * Fixup statement distribution. * Handle request directly in collator protocol. + Only allow fatal errors at top level. * Use direct request channel for availability recovery. * Finally get rid of request multiplexer Fixes #2842 and paves the way for more back pressure possibilities. * Fix overseer and statement distribution tests. * Fix collator protocol and network bridge tests. * Fix tests in availability recovery. * Fix availability distribution tests. * Fix dispute distribution tests. * Add missing dependency * Typos. * Review remarks. * More remarks. * Add words to the dictionnary * sp-io update * Don't drop UMP queue items if weight exhausted (#3784) * Requeue UMP queue items if weight exhausted * Reduce complexity and remove Deque * Formatting * Formatting * Avoid needless storage writes * Test * Formatting * Docs and cleanup * fmt * Remove now irrelevant comment. * Simplify `take_processed` by using `mem::take` * Clean up & fmt: use `upward_message` directly. * Grumbles Co-authored-by: Shawn Tabrizi <[email protected]> Co-authored-by: Sergei Shulepov <[email protected]> * Update beefy deps * Update Substrate and Beefy * fix for #3537 * Update substrate * Fix spelling * Do not expire HRMP open channel requests (#3543) * Do not expire HRMP open channel requests * Fix the build and update the docs * Implement canceling requests and do not remove them automatically * Fix a borked merge * Fix fmt * Please spellchecker * Apply suggestions from code review Co-authored-by: Amar Singh <[email protected]> * Use `mutate_exists` for maintaining request counts * Apply `rustfmt` * Move newly introduced entrypoint to end to preserve ordering Co-authored-by: Amar Singh <[email protected]> * UMP: Support Overweight messages (#3575) * Introduce new config: ump_max_individual_weight * Implement overweight msg stashing * Test * Add migration module. Also introduces a test for migration * Integrate ExecuteOverweightOrigin to runtimes * Fix more stuff * Add `yeet` into dictionary * Use suggested `Error` variant names * typo * Use 20ms as the maximum individual message weight * Update the test value * rustfmt * Clean up * Remove deprecated field from host config * Remove missed _hrmp_open_request_ttl * Apply typo fix suggestion Co-authored-by: Alexander Popiak <[email protected]> * Rename `migration::migrate_to_latest` * Restore `_hrmp_open_request_ttl` in `v0::HostConfiguration` * Apply suggestion for a rustdoc * Apply the suggestion * Test v0 config with the raw production data fetched from Kusama * Update runtime/parachains/src/ump.rs Co-authored-by: Alexander Popiak <[email protected]> * Expose migration functions * Fix spellcheck Co-authored-by: Alexander Popiak <[email protected]> Co-authored-by: Shawn Tabrizi <[email protected]> Co-authored-by: Keith Yeung <[email protected]> * Remove migration done by runtime version 9090 (#3731) * remove 9090 migrations * remove unused * fmt * revert removal of polkadot migration * remove migration for polkadot Co-authored-by: Kian Paimani <[email protected]> Co-authored-by: Gavin Wood <[email protected]> Co-authored-by: Alexander Popiak <[email protected]> Co-authored-by: Amar Singh <[email protected]> Co-authored-by: Shawn Tabrizi <[email protected]> Co-authored-by: Keith Yeung <[email protected]> Co-authored-by: Shaun Wang <[email protected]> Co-authored-by: Guillaume Thiolliere <[email protected]> Co-authored-by: Robert Klotzner <[email protected]> Co-authored-by: Sergei Shulepov <[email protected]>
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
in case waiting for the next response takes too long. I intend to improve this further, but this should already help a lot with the finality lag issues.