lcnr
lcnr
the impact on diesel seems quite desirable however. I think we can improve the performance of fast reject significantly so that even the ~20% hit rate from typenum is a...
@bors try @rust-timer queue
ah also, the bitmaps regression is quite certainly not caused by slightly more than 1000 uses of deep reject. I expect that the worse performance of the `DeepRejectCtxt` in other...
can you stash the commit adding all new uses (or open a separate PR with these new uses removed) to check the perf impact of the refactor of `DeepRejectCtxt` itself
one option there is to outline all the very rare branches of the match into a `#[cold]` `#[inline(never)]` function
@rustbot author
> Not sure why the `ty::Error` branch was returning true, but after removing it (moving to the `... => false` in the end), the perf got better: > Range Mean...
I'd like to see another rebase + perf run before merging this. Will only get to reviewing it by the end of this week/start of the next
sorry for the delay! nominating for lang, to let them decide whether they mind delegating this to t-types. We simply reuse the vtable of the parent trait when dropping the...
> > This relies on all vtables to have the same header which is used for dropping and size_of_val/align_of_val. I don't know whether that's currently guaranteed anywhere, cc @rust-lang/opsem i...