Revert "Fixes GetEntryScriptHash and GetCallingScriptHash (2.x) (#138)"

[erikzhang]

This reverts commit 38c5dc0.

[codecov-io]

Codecov Report

Merging #157 into master-2.x will increase coverage by 11.1%.
The diff coverage is 100%.

@@              Coverage Diff               @@
##           master-2.x     #157      +/-   ##
==============================================
+ Coverage       80.38%   91.49%   +11.1%     
==============================================
  Files              45       17      -28     
  Lines            4548     1986    -2562     
==============================================
- Hits             3656     1817    -1839     
+ Misses            892      169     -723

Impacted Files	Coverage Δ
src/neo-vm/ExecutionContext.cs	87.5% <100%> (+2.88%)	⬆️
src/neo-vm/ExecutionEngine.cs	99.56% <100%> (-0.01%)	⬇️
tests/neo-vm.Tests/Converters/ScriptConverter.cs
tests/neo-vm.Tests/UtDebugger.cs
tests/neo-vm.Tests/Extensions/JsonExtensions.cs
tests/neo-vm.Tests/Cryptography/ECC/ECCurve.cs
tests/neo-vm.Tests/Types/VMUT.cs
tests/neo-vm.Tests/Types/VMUTEntry.cs
...ts/neo-vm.Tests/Cryptography/ECC/ECFieldElement.cs
tests/neo-vm.Tests/Types/ScriptTable.cs
... and 19 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update cd5c3d0...fe4296e. Read the comment docs.

[ixje]

Can you describe why this is needed?
I had this ported to neo-python so need to know if I need to rollback as well.

[erikzhang]

#138 may not be compatible with previous contracts and is being tested.

[ixje]

Ok, I'll keep watching this issue then

[vncoelho]

This is strange, because we synced Testnet and no issues were reported.
And @superboyiii synced the Mainnet and also did not found inconsistencies.

Let's double check this before reverting.

Check by NGD team

[ixje]

😞

[vncoelho]

What happened, @ixje?

I am not sure if it really affected storage of Mainnet since I only compared for Testnet.
Did NGD confirmed inconsistencies, @erikzhang?
If true, let's keep 2.x like it is now and focus on the future.

[erikzhang]

I have not tested it myself, but the NGD team has confirmed this problem.

[superboyiii]

@vncoelho Could be seen here for the detail

[ixje]

@superboyiii there seems to be some doubt in whether this is the solution

Should be, I think. Or if there might be a better solution...

Did you try a comparison with a version that had this PR merged to validate this is the actual solution to the problem?

[superboyiii]

Yes, I just completed the test of revert#138, it got the same correct storage with v2.10.0 and v2.9.4. There actually has contradiction between storage compatibility and TAIL CALL fixing. What a pity.....

[ixje]

Great! Thanks for the input. Now I can confidently revert this change as well