Skip to content

Conversation

@williamjallen
Copy link
Collaborator

CDash historically used crc32 hashes to determine uniqueness for various record types as part of a complex deduplication scheme. This has caused numerous issues over the years, with frequent bugs and changes to the hashing logic. Instead, we're now in the process of switching to a system where full records are matched against rather than hashes. This PR removes the builderror crc32 column. No performance hits are expected as a result of this change, because the database can still look up records efficiently by build ID.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant