feat(tests): add EIP-7954 jumpdest test past the old code-size limit

[danceratopz]

🗒️ Description

EIP-7954 raises MAX_CODE_SIZE to 64 KiB and the initcode limit to 128 KiB. Before this fork, deployed runtime code could never exceed 24 KiB, so no existing test ever executes a contract at a program counter past the old limit. The EIP-7954 max-size contracts are either deploy-only or run a short prefix that STOPs before their JUMPDEST padding, and EXTCODECOPY/CODECOPY only read the code rather than run it.

A client whose jumpdest analysis or code execution is still bounded by the old limit would pass the existing suite yet break on a real 64 KiB contract.

test_max_code_size_high_jumpdest deploys a MAX_CODE_SIZE contract and jumps near the new limit, far beyond the old 24 KiB code and 48 KiB initcode limits:

• Valid JUMPDEST: The jump succeeds and the code runs at the high offset, storing a sentinel.
• Non-JUMPDEST byte: The jump is rejected and nothing is stored.

The contract is called through a caller that records the call result, so a correct client is told apart from one that truncated its jumpdest analysis (the valid jump fails) or accepts any high offset (the invalid jump wrongly succeeds).

Final version

🔗 Related Issues or PRs

• EIP-7954 tracker: EIP-7954 Implementation Tracker: Increase Maximum Contract Size #2028
• Test release tracker: Test Release Tracker: glamsterdam-devnet@v6.0.0 #2915 (glamsterdam-devnet@v6.0.0)
• Follow-up to feat(tests, spec-specs): raise EIP-7954 max code size to 64KiB #2987, which raised the EIP-7954 limits to 64 KiB code / 128 KiB initcode.

✅ Checklist

• All: Ran fast static checks to avoid unnecessary CI fails, see also Code Standards and Enabling Pre-commit Checks:

  just static

• All: PR title adheres to the repo standard - it will be used as the squash commit message and should start type(scope):.
• All: Considered updating the online docs in the ./docs/ directory.
• All: Set appropriate labels for the changes (only maintainers can apply labels).

Cute Animal Picture

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.20%. Comparing base (e35d103) to head (a9b7ed7).
⚠️ Report is 5 commits behind head on forks/amsterdam.

Additional details and impacted files

@@                 Coverage Diff                  @@
##           forks/amsterdam    #2993       +/-   ##
====================================================
+ Coverage            81.28%   93.20%   +11.92%     
====================================================
  Files                  620      620               
  Lines                36641    38759     +2118     
  Branches              3311     3341       +30     
====================================================
+ Hits                 29784    36127     +6343     
+ Misses                6335     1773     -4562     
- Partials               522      859      +337     

Flag	Coverage Δ
unittests	93.20% <ø> (+11.92%)	⬆️

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

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[marioevz]

Looks good, just a couple of suggestions, feel free to apply.

[LouisTsai-Csie]

Good points from Mario, LGTM once those are addressed. No other comments from my side.

Follow-up PR idea on JUMPDEST analysis: clients run it to ensure a jump target doesn't land on an immediate byte. We could add a scenario where the jump targets a 0x5b immediate byte of a PUSH, SWAPN, DUPN, or EXCHANGE instruction, asserting it's rejected as an invalid jump destination. The contract size should be at the new max code size, so the analysis is done well past the increase.

[danceratopz]

Follow-up PR idea on JUMPDEST analysis: clients run it to ensure a jump target doesn't land on an immediate byte. We could add a scenario where the jump targets a 0x5b immediate byte of a PUSH, SWAPN, DUPN, or EXCHANGE instruction, asserting it's rejected as an invalid jump destination. The contract size should be at the new max code size, so the analysis is done well past the increase.

Thanks for the review @LouisTsai-Csie! About the idea above, I think it has low additional value, but happy to add if you disagree. I think it would only catch a client that had hard-coded the wrong contract sizes in EIP-8024 context?

Behavior (the jump target byte)	Whose rule	Already tested in forks/amsterdam?
0x5B buried in PUSH data → rejected	classic JUMPDEST analysis (pre-EIP)	Rule: runtime.py#L62-L67. No dedicated test in the 8024/7954 suites (ancient behavior).
0x5B after DUPN → accepted	EIP-8024	✅ test_dupn_jump_to_immediate_byte_0x5b_succeeds
0x5B after SWAPN → accepted	EIP-8024	✅ test_swapn_jump_to_immediate_byte_0x5b_succeeds
0x5B after EXCHANGE → accepted	EIP-8024	✅ test_exchange_jump_to_immediate_byte (immediate sweep, incl. 0x5B)
valid immediate skipped → jump fails	EIP-8024	✅ test_dupn_jump_to_valid_immediate_fails, test_swapn_jump_to_valid_immediate_fails
any of the above at 64 KiB (MAX_CODE_SIZE)	EIP-7954 (size only)	❌ Not covered. The 8024 tests above use tiny contracts; the basic max-size jump case is in #2993.

[danceratopz]

@LouisTsai-Csie I added the test case here #2998