Extend vectorized algorithms to ARM64/ARM64EC

[cbezault]

This work is ongoing thanks to the Arm engineers @hazzlim and @jaykang10, with assistance from the Vector Lord @AlexGuteniev. Dashboard as of 2026-03-24:

STL/stl/inc/xutility

Lines 76 to 101 in 7b9a0f5

	#define _VECTORIZED_ADJACENT_FIND _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_BITSET_FROM_STRING _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_BITSET_TO_STRING _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_COUNT _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_FIND _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_FIND_END _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_FIND_FIRST_OF _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_FIND_LAST _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_FIND_LAST_OF _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_INCLUDES _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_IS_SORTED_UNTIL _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_MINMAX _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_MINMAX_ELEMENT _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_MISMATCH _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_REMOVE _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_REMOVE_COPY _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_REPLACE _VECTORIZED_FOR_X64_X86
	#define _VECTORIZED_REPLACE_COPY _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_REVERSE _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_REVERSE_COPY _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_ROTATE _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_SEARCH _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_SEARCH_N _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_SWAP_RANGES _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_UNIQUE _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC
	#define _VECTORIZED_UNIQUE_COPY _VECTORIZED_FOR_X64_X86_ARM64_ARM64EC

[StephanTLavavej]

C:\Program Files (x86)\Microsoft Visual Studio\2019\Preview\VC\Tools\MSVC\14.26.28801\include\isa_availability.h defines:

• __ISA_AVAILABLE_ARMNT (no NEON support)
• __ISA_AVAILABLE_NEON (128-bit NEON support for 32-bit ARM)
• __ISA_AVAILABLE_NEON_ARM64 (128-bit NEON support for ARM64; there's a comment about how the distinction is "temporary" but I suspect it's permanent)

[sylveon]

We can assume NEON is present in both 32 bits and 64 bits. https://docs.microsoft.com/en-ca/cpp/build/overview-of-arm-abi-conventions specifies WoA requires it.

[StephanTLavavej]

Updating this issue to no longer mention ARM32; at this time we still need to keep it compiling and working, but we no longer care about optimizing for it. Only ARM64 performance matters.

[AlexGuteniev]

Will need to take care of alignment; see #5766

Edit: no, we don't. ARM64 handles misaligned well.