[llvm-dev] PSA: SmallVector<T> Just Works

[Chris Lattner via llvm-dev]

Awesome, thank you Sean!

-Chris

> On Dec 7, 2020, at 1:32 PM, Sean Silva via llvm-dev <llvm-dev at lists.llvm.org> wrote:
> 
> TL;DR: SmallVector now chooses a default "N", and we recommend using that default.
> 
> We hope that this will
> - Avoid semi-arbitrary choices for the "N" parameter.
> - Save you extra edit/compile cycles for forgotten "N" parameters.
> - Avoid some of the pathological cases of SmallVector use (like sizeof(SmallVector) becoming excessively large)
> 
> The programmer's manual has been updated, and now reads
> 
> In the absence of a well-motivated choice for the number of inlined elements N, it is recommended to use SmallVector<T> (that is, omitting the N). This will choose a default number of inlined elements reasonable for allocation on the stack (for example, trying to keep sizeof(SmallVector<T>) around 64 bytes).
> 
> https://llvm.org/docs/ProgrammersManual.html#llvm-adt-smallvector-h <https://llvm.org/docs/ProgrammersManual.html#llvm-adt-smallvector-h>
> 
> We haven't planned any big refactoring to use the new default, so we are expecting it to catch on organically in new code / code reviews / refactorings.
> 
> Enjoy! :)
> 
> For historical reference: This landed in https://reviews.llvm.org/D92522 <https://reviews.llvm.org/D92522> after much discussion on the list <https://groups.google.com/g/llvm-dev/c/Z-VwNCTRGSg/m/fYi0JHhzAwAJ> and an earlier attempt at a patch <https://reviews.llvm.org/D90884> which went back and forth on various ideas, including more general aspirations for SmallVector's evolution. 
> 
> -- Sean Silva
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20201207/735b2027/attachment.html>