Skip to content

Conversation

@chfast
Copy link
Member

@chfast chfast commented May 16, 2025

This is guaranteed by the stack space allocation.

This allows the compiler to use aligned vector load instructions. Unfortunately, no performance gains on latest CPUs because aligned/unaligned instructions have the same cost there.

@codecov
Copy link

codecov bot commented May 16, 2025

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.54%. Comparing base (f29c803) to head (cfba1fb).
Report is 1 commits behind head on master.

Additional details and impacted files
@@           Coverage Diff           @@
##           master    #1233   +/-   ##
=======================================
  Coverage   94.54%   94.54%           
=======================================
  Files         175      175           
  Lines       19694    19696    +2     
=======================================
+ Hits        18620    18622    +2     
  Misses       1074     1074           
Flag Coverage Δ
eest_gmp 15.27% <100.00%> (+<0.01%) ⬆️
eof_execution_spec_tests 19.84% <100.00%> (+<0.01%) ⬆️
ethereum_tests 21.43% <100.00%> (+<0.01%) ⬆️
ethereum_tests_silkpre 18.28% <100.00%> (+<0.01%) ⬆️
execution_spec_tests 18.53% <100.00%> (+<0.01%) ⬆️
unittests 91.87% <100.00%> (+<0.01%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines Coverage Δ
lib/evmone/instructions.hpp 100.00% <100.00%> (ø)
🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@chfast chfast requested review from gumb0 and pdobacz May 19, 2025 11:53
Copy link
Member

@pdobacz pdobacz left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Codewise LGTM, but since

Unfortunately, no performance gains on latest CPUs because aligned/unaligned instructions have the same cost there.

Why do this then?

@chfast
Copy link
Member Author

chfast commented May 19, 2025

Why do this then?

It provides some additional information to the compiler. Maybe handy on other architectures.

This is guaranteed by the stack space allocation.
@chfast chfast force-pushed the baseline/aligned_stack_access branch from ecbfda6 to cfba1fb Compare May 20, 2025 11:55
@chfast
Copy link
Member Author

chfast commented May 20, 2025

No changes on ARM64 either 😢

@chfast chfast merged commit 2033593 into master May 20, 2025
26 checks passed
@chfast chfast deleted the baseline/aligned_stack_access branch May 20, 2025 13:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants