Uh oh!
There was an error while loading. Please reload this page.
Fix Type.GetHashCode for RuntimeTypes w/o MethodTable - #97195
Conversation
jkotas
commented
Jan 19, 2024
/azp run runtime-nativeaot-outerloop |
|
Azure Pipelines successfully started running 1 pipeline(s). |
Uh oh!
There was an error while loading. Please reload this page.
jkotas
commented
Jan 19, 2024
/azp run runtime-nativeaot-outerloop |
|
Azure Pipelines successfully started running 1 pipeline(s). |
The hashcode for these was always 0 before this change.
ghost
commented
Jan 19, 2024
Tagging subscribers to this area: @agocke, @MichalStrehovsky, @jkotas Issue Detailsnull
|
jkotas
commented
Jan 19, 2024
@MichalStrehovsky PTLA. It is fixing a minor bug introduced by the RuntimeType refactoring. |
MichalStrehovsky
left a comment
There was a problem hiding this comment.
Thanks!
Do we want to add the test you had here before the force push?
jkotas
commented
Jan 22, 2024
My original assumption was that this bug is doing to lead to crash that is not actually the case. It is not functionality bug; it is a perf bug. The test for this bug would have to generate number of types and verify that their hashcode distibution hits some minimum threshold. We do not have tests like for GetHashCode implementations, so building one for RuntimeType did not seem to be worth it. |
The hashcode for these was always 0 before this change.
No description provided.