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Outdated suggestions cannot be applied.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.
Language nitpick:
would be leaked
-->leaks
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.
Maybe the leak can be avoided by reordering the code a bit?
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.
@the8472 In this case yes, but I wanted this to translate to how one would initialize a large struct field by field.
If the struct had been a heap allocated struct with a large array field instead of a
Vec
, it wouldn't be possible to construct the array on the stack before moving it to the field in debug builds, since it would overflow the stack.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.
Well, afaik currently there is no guarantee that passing a large array directly to function arguments gets optimized any better than first assigning it to a named variable. You're at the mercy of the optimizer in either case since there's no placement API. rust-lang/rfcs#2884 is still under discussion.
I think to initialize a large array without blowing the stack in debug mode you would have to loop over each item.