Conversation
ekpyron
left a comment
ekpyron
left a comment
There was a problem hiding this comment.
If all CI runs pass, I don't see a reason not to do this... Unfortunately json-cpp doesn't seem to have proper Changelogs, so we can't really check, if there's something problematic in there...
bshastry
left a comment
bshastry
left a comment
There was a problem hiding this comment.
Looks good to me if tests pass. I checked locally that the oss fuzz issue is no longer reproducible with this PR (jsoncpp 1.9.2) but was reproducible for jsoncpp 1.8.4 that we are currently depending on.
I could reproduce the test failure with clang-6, UBSan enabled, jsoncpp v1.8.4, latest solidity develop commit but not with this PR. |
Hm, so no way to add a test case into our repo which would reproduce this without ubsan? |
Sadly, yes. Unless we add a UBSan build and test CIs. I feel adding a UBSan may be an overkill for this though because as far as I remember this is the only bug that was found (in a dependency and not solidity) by UBSan. |
|
Is it with the test case from the issue? I can't fathom how that could lead into this. Is it due to some gas estimation field? |
Hopefully fix compiler errors on Clang on Apple Silicon. Patches are based on the following upstream PRs: argotorg/solidity#4160 argotorg/solidity#4235 argotorg/solidity#4470 argotorg/solidity#5699 argotorg/solidity#7268 argotorg/solidity#7827 argotorg/solidity#9383 argotorg/solidity#14725
4 participants
Attempt to solve #5869, but couldn't find any reproducible test case with any compiler version. Any ideas @bshastry? The test case in the fuzzer issue is not reproducing it.