Tate, Hongliang Tian
Tate, Hongliang Tian
@lucassong-mh Is this PR ready for reviewing? The PR is marked as a draft.
@Tangzh33 As I understand it, this PR isn't ready for reviewing. For one thing, it contains a merge commit. And the commit messages do not follow the current convention. So...
This is a big and important PR. I would like to invite several good hands to help me out. @junyang-zh Could you review this PR with a focus on the...
This PR has been modified greatly since last review. I would like to initiate another round of review with help from these good fellows: @junyang-zh Could you review this PR...
I think this is a good idea
@liqinggd Could you review this PR?
@sdww0 Is this PR urgent for enabling more performance benchmarks? If so, @lucassong-mh could you work with @sdww0 to make this PR at least usable for benchmarking purpose? We will...
This PR is abandoned.
> Any user must release lock in the reverse order of acquiring locks, otherwise deadlock may be introduced. The reported bug is true. But I don't think this claim is...
> I'm not sure there is an elegant way to avoid `transmute`. But _if_ we finally decide to use it, how about using `repr(transparent)` instead of `repr(C)`? > > ```rust...