Conversation
| BOOST_CHECK(!successCompile("(DUP" + boost::lexical_cast<string>(i) + ")")); | ||
| for (unsigned i = 1; i <= 16; i++) | ||
| BOOST_CHECK(!successCompile("(SWAP" + boost::lexical_cast<string>(i) + ")")); | ||
| BOOST_CHECK(!successCompile("(JUMPDEST)")); |
There was a problem hiding this comment.
It can still be used via assembly: (asm JUMPDEST). See #3066.
There was a problem hiding this comment.
OK, then it's a non-issue.
There was a problem hiding this comment.
Also note it was removed a week ago or so, this PR removes pushnn from assembly and adds tests for the rest too.
There was a problem hiding this comment.
it is an issue. some tests rely on (JUMPDEST)
and we dont have a unit test that actually checks that all lll opcodes are actually supported by lllc version before running the test fillers.
There was a problem hiding this comment.
We build just parser of (opcode) to bytecode.
There was a problem hiding this comment.
the older version of solidity lllc works just like that. if you suggest to write a new parser...
There was a problem hiding this comment.
The older version of lllc is buggy and it produces unreliable output, at least output which we shouldn't rely on in testing the entire ecosystem with.
There was a problem hiding this comment.
@winsvega either fork an older lllc and fix the bugs, or create a new compiler.
|
Yes, I think so when the tests pass. |
|
@pirapira the tests failures are not related to this PR |
|
Restarted the emscripten build. |
|
@pirapira ok to merge? |
3 participants
Found in #3002.