Use __ldg on Scalar Read-only Data Stores #1326

Job	Run time
perftests (gcc-8, release)	1m 32s
perftests (gcc-9, release)	1m 28s
perftests (gcc-10, release)	1m 44s
perftests (gcc-11, release)	1m 37s
perftests (gcc-12, debug)	2m 11s
perftests (gcc-12, release)	1m 42s
perftests (gcc-13, debug)	2m 4s
perftests (gcc-13, release)	1m 40s
perftests (clang-11, debug)	1m 49s
perftests (clang-11, release)	1m 52s
perftests (clang-12, debug)	1m 59s
perftests (clang-12, release)	1m 55s
perftests (clang-13, release)	1m 57s
perftests (clang-14, release)	2m 6s
perftests (clang-15, debug)	1m 47s
perftests (clang-15, release)	1m 54s
perftests (clang-16, debug)	1m 55s
perftests (clang-16, release)	1m 55s
perftests (clang-14-cuda-11, debug)	5m 50s
perftests (clang-14-cuda-11, release)	4m 56s
perftests (gcc-10-cuda-11.8, debug)	5m 48s
perftests (gcc-10-cuda-11.8, release)	5m 7s
perftests (gcc-11-cuda-12.0, debug)	5m 36s
perftests (gcc-11-cuda-12.0, release)	5m 6s
perftests (gcc-12-cuda-12.3, debug)	5m 54s
perftests (gcc-12-cuda-12.3, release)	5m 18s
perftests (gcc-12-cuda-12.4, debug)	5m 47s
perftests (gcc-12-cuda-12.4, release)	5m 3s
perftests (base-hip, debug)	7m 14s
perftests (base-hip, release)	6m 14s
perftests (gcc-10-hpx, release)	3m 24s
perftests (nvhpc-23.3, debug)	9m 27s
perftests (nvhpc-23.3, release)	9m 58s
perftests (nvhpc-23.9, debug)	12m 29s
perftests (nvhpc-23.9, release)	10m 33s
	2h 26m 51s

Provide feedback