You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In https://crater-reports.s3.amazonaws.com/pr-117905/index.html, the biggest root regression is the bnum crate. This is easily verified by opening the "regressed: dependencies" category, it even explicitly lists "bnum-0.8.0 (249)" as a block of crates that failed because they depend on bnum.
However, "regressed: root results" does not list "bnum".
I was assuming that "root results" is the only category I have to go through to find out which are the root regressions, but it seems like that is not the case. Either there was a bug in looking for root regressions when generating the report, or the results page is labeled in very confusing ways.
The text was updated successfully, but these errors were encountered:
RalfJung
changed the title
"root regression" does not list the biggest root regression
"root results" does not list the biggest root regression
Nov 20, 2023
What I think is going on here, is that bnum version 0.8.0 is regressed, but not the latest version 0.10. Crater only explicitly cargo builds the latest version of a crate. So when building bnum explicitly, Crater didn't encounter any errors, and therefore it is not listed as a root regression. But, in the process of building other crates that depend on old versions of bnum, errors were encountered building said old versions, so bnum is listed as a regressed dependency.
In https://crater-reports.s3.amazonaws.com/pr-117905/index.html, the biggest root regression is the bnum crate. This is easily verified by opening the "regressed: dependencies" category, it even explicitly lists "bnum-0.8.0 (249)" as a block of crates that failed because they depend on bnum.
However, "regressed: root results" does not list "bnum".
I was assuming that "root results" is the only category I have to go through to find out which are the root regressions, but it seems like that is not the case. Either there was a bug in looking for root regressions when generating the report, or the results page is labeled in very confusing ways.
The text was updated successfully, but these errors were encountered: