[libcxx-commits] [PATCH] D118029: Introduce branchless sorting functions for sort3, sort4 and sort5.

[Nikolas Klauser via Phabricator via libcxx-commits]

philnik added a comment.

I'm guessing all the ternaries are making the compiler generate `cmov`s?
I noticed that the benchmarks for size 4 are generally pretty bad. Do you know if anything can be done there?
I haven't yet looked at the logic itself very closely.



================
Comment at: libcxx/include/__algorithm/sort.h:152
+inline _LIBCPP_HIDDEN
+    __enable_if_t<_VSTD::is_integral< typename iterator_traits<_RandomAccessIterator>::value_type>::value, void>
+    __sort3_internal(_RandomAccessIterator __x1, _RandomAccessIterator __x2, _RandomAccessIterator __x3, _Compare __c) {
----------------
Ditto below


================
Comment at: libcxx/include/__algorithm/sort.h:160
+inline _LIBCPP_HIDDEN
+    __enable_if_t<!_VSTD::is_integral< typename iterator_traits<_RandomAccessIterator>::value_type>::value, void>
+    __sort3_internal(_RandomAccessIterator __x1, _RandomAccessIterator __x2, _RandomAccessIterator __x3, _Compare __c) {
----------------
Is there a reason you restricted this to integrals? Does it pessimize some code? I guessing at least all primitive types should be allowed. I haven't benchmarked, but looking at code generation it seems floating points should also benefit from this change.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D118029/new/

https://reviews.llvm.org/D118029