Closed carterqw2 closed 11 months ago
Coverage from tests in ./e2e_test/...
for ./consensus/istanbul/...
at commit 5c44f5dcfbff4fbbacdcddd4eb25063be47e1fbf
coverage: 63.2% of statements in consensus/istanbul coverage: 42.7% of statements in consensus/istanbul/announce coverage: 55.8% of statements in consensus/istanbul/backend coverage: 0.0% of statements in consensus/istanbul/backend/backendtest coverage: 24.3% of statements in consensus/istanbul/backend/internal/replica coverage: 62.1% of statements in consensus/istanbul/core coverage: 50.0% of statements in consensus/istanbul/db coverage: 0.0% of statements in consensus/istanbul/proxy coverage: 64.2% of statements in consensus/istanbul/uptime coverage: 51.8% of statements in consensus/istanbul/validator coverage: 79.2% of statements in consensus/istanbul/validator/random
Test failures: |
---|
TestTransferCELOPreGingerbread: e2e_test
|
TestSyncNoStorageAndOneCodeCappedPeer: snap
|
TestEthClient: ethclient
|
TestEthClient/TestTxInBlockInterrupted: ethclient
|
This test report was produced by the test-summary action. Made with ❤️ in Cambridge. |
All modified and coverable lines are covered by tests :white_check_mark:
Comparison is base (
8ceea79
) 55.08% compared to head (8780433
) 46.58%. Report is 1 commits behind head on master.:exclamation: Current head 8780433 differs from pull request most recent head 62369f2. Consider uploading reports for the commit 62369f2 to get more accurate results
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Looks good, maybe it's worthwhile to add the benchmarking code as well?
Added.
Description
Avoid decoding and re-encoding of the block body by using available RLP-encoded block body from the database, combine it with the RLP-encoded block hash to avoid unnecessary processing and extra memory allocations.
Benchmarks:
3x more CPU-efficient and 1.5x more memory-efficient. A larger change would be storing RLP-encoded blockBodyWithBlockHash structures instead of just block bodies but it would significantly increase the diff.
Other changes
No.
Tested
Existing tests pass, added unit tests and the benchmark.
Backwards compatibility
Yes.