Uh oh!
There was an error while loading. Please reload this page.
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it'd be better for the BitConverter.IsLittleEndian check to be first in all of these cases, e.g.
If
IsStartArrayisn't inlineable, the JIT needs to assume it might have side effects that can't be elided, and so to preserve order of operations, in the big endian case the JIT would need to compile this as:But with the check first, when it's false the second clause would never be evaluated, so the JIT can compile it just as:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I believe the author was just following the pattern we had established on other types like short/int/long where this check is done second. I don't feel strongly about this, but my inclination would be to leave the order as is for 7.0 and we can invert the check for all types in 8.0 where we have more runway. If you feel strongly about this though, I wouldn't object to making the change here.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should fix it in main. I don't feel strongly about it for release/7.0, since to my knowledge we don't actually ship supported S390X bits from that branch. I am curious, though, where we have existing BitConverter.IsLittleEndian checks at the end of a clause; those should be fixed, too.