Skip to content
Snippets Groups Projects
Commit bb041753 authored by Branislav Kontur's avatar Branislav Kontur Committed by GitHub
Browse files

Run all benchmarks "as tests" in short setup on every PR (#2846)


* Run all benchmarks "as tests" in short setup on every PR

* Fix

* Update scripts/ci/gitlab/pipeline/short-benchmarks.yml

Co-authored-by: default avatarAlexander Samusev <41779041+alvicsam@users.noreply.github.com>

* Updated TODO

* Removed `--execution wasm`

* [asset-conv] Unused Balances type

* [asset-conv] native asset change

* Dedicated `AssetBalance` type for `pallet_assets` instances

* Improved local vs foreign asset handling + test for not allowing pool_assets in pool

* Removed `into_multiasset_id`

* Fix

* Refactor

* Fixed create_pool for benchmark with LocalAndForeignAssets (finally found it)

* Revert

* fmt

* Migrates pools with `MultiLocation { parents: 0, interior: Here }` to `MultiLocation { parents: 1, interior: Here }`

* Allow `set_storage` for `AllowMultiAssetPools` / `LiquidityWithdrawalFee`

* Benchmarks work

* Removed comment + more defensive migration

* `T::Currency::transfer` -> `Balances::transfer_all` in migration

*...
parent 8171890f
Branches
No related merge requests found
......@@ -7,11 +7,14 @@
stages:
- test
- build
# used for manual job run for regenerate weights for release-* branches (not needed anymore, just leave it here for a while as PlanB)
- benchmarks-build
# used for manual job run for regenerate weights for release-* branches (not needed anymore, just leave it here for a while as PlanB)
- benchmarks-run
- publish
- integration-tests
- zombienet
- short-benchmarks
default:
interruptible: true
......@@ -142,7 +145,10 @@ include:
- scripts/ci/gitlab/pipeline/test.yml
# # build jobs
- scripts/ci/gitlab/pipeline/build.yml
# short-benchmarks jobs
- scripts/ci/gitlab/pipeline/short-benchmarks.yml
# # benchmarks jobs
# # used for manual job run for regenerate weights for release-* branches (not needed anymore, just leave it here for a while as PlanB)
- scripts/ci/gitlab/pipeline/benchmarks.yml
# # publish jobs
- scripts/ci/gitlab/pipeline/publish.yml
......
......@@ -121,3 +121,18 @@ build-runtime-testing:
needs:
- job: build-runtime-starters
artifacts: false
build-short-benchmark:
stage: build
extends:
- .docker-env
- .common-refs
- .collect-artifacts
# this is an artificial job dependency, for pipeline optimization using GitLab's DAGs
needs:
- job: check-rustdoc
artifacts: false
script:
- cargo build --profile release --locked --features=runtime-benchmarks
- mkdir -p ./artifacts
- cp ./target/release/polkadot-parachain ./artifacts/
# This file is part of .gitlab-ci.yml
# Here are all jobs that are executed during "short-benchmarks" stage
# Run all pallet benchmarks only once to check if there are any errors
.short-benchmark-template: &short-bench
stage: short-benchmarks
extends:
- .common-refs
- .docker-env
needs:
- job: build-short-benchmark
artifacts: true
variables:
RUNTIME_CHAIN: benchmarked-runtime-chain
script:
- ./artifacts/polkadot-parachain benchmark pallet --wasm-execution compiled --chain $RUNTIME_CHAIN --pallet "*" --extrinsic "*" --steps 2 --repeat 1
short-benchmark-asset-hub-polkadot:
<<: *short-bench
variables:
RUNTIME_CHAIN: asset-hub-polkadot-dev
short-benchmark-asset-hub-kusama:
<<: *short-bench
variables:
RUNTIME_CHAIN: asset-hub-kusama-dev
short-benchmark-asset-hub-westend:
<<: *short-bench
variables:
RUNTIME_CHAIN: asset-hub-westend-dev
short-benchmark-bridge-hub-polkadot:
<<: *short-bench
variables:
RUNTIME_CHAIN: bridge-hub-polkadot-dev
short-benchmark-bridge-hub-kusama:
<<: *short-bench
variables:
RUNTIME_CHAIN: bridge-hub-kusama-dev
short-benchmark-bridge-hub-rococo:
<<: *short-bench
variables:
RUNTIME_CHAIN: bridge-hub-rococo-dev
short-benchmark-collectives-polkadot :
<<: *short-bench
variables:
RUNTIME_CHAIN: collectives-polkadot-dev
short-benchmark-glutton-kusama :
<<: *short-bench
variables:
RUNTIME_CHAIN: glutton-kusama-dev-1300
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment