Specialization for Extend<&T> for vec #38182
Conversation
|
(rust_highfive has picked a reviewer for you, use r? to override) |
|
r? @alexcrichton This fix goes forward instead of backwards, you could say, by adding more specializations instead of reverting. |
Specialize to use copy_from_slice when extending a Vec with &[T] where T: Copy.
bluss
force-pushed
the
more-vec-extend
branch
from
41b55da
to
02bf1ce
Compare
|
@bors: r+ Seems ok to me, thanks! I'd still ideally prefer to explore the route of adding codegen tests eventually (to ensure we don't regress) but that can happen in a follow-up |
|
📌 Commit 02bf1ce has been approved by |
|
Ok, there is no codegen test, but codegen for extend_with_slice was examined and was fine before it went in, so in that way it's unlikely that it would have been caught that way. It already optimized fine in many situations. |
|
@dtolnay This PR fixes the regression in the sense that the regression I saw here #38021 (comment) was verified to be back at its old performance at 1.5-1.6 ms/iter. I haven't done further tests. |
Specialization for Extend<&T> for vec Specialize to use copy_from_slice when extending a Vec with &[T] where T: Copy. This specialization results in `.clone()` not being called in `extend_from_slice` and `extend` when the element is `Copy`. Fixes #38021
Specialize to use copy_from_slice when extending a Vec with &[T] where
T: Copy.
This specialization results in
.clone()not being called inextend_from_sliceandextendwhen the element isCopy.Fixes #38021