neon

50 Questions
Newest Active Unanswered Frequent
Score
View
Card Compact
What is the most efficient way to handle integer multiplication overflow with saturation with ARM Neon intrinsics?
user_105764940
•
2
1
211
neon intrinsics saturation-arithmetic arm simd
aarch64 xtn2 clearing lower half
user_85823830
•
3
1
95
armv8 arm64 neon simd assembly
Why doesn’t Clang use vcnt for __builtin_popcountll on AArch32?
user_18133490
•
3
1
345
neon arm micro-optimization bit-manipulation clang
Loop takes more cycles to execute than expected in an ARM Cortex-A72 CPU
user_5230790
•
6
3
356
neon arm assembly optimization performance
NEON assembly code requires more cycles on Cortex-A72 vs Cortex-A53
user_39833300
•
2
1
324
raspberry-pi arm64 neon arm assembly
Neon equivalent of mm_madd_epi16 and mm_maddubs_epi16
user_2668360
•
2
1
158
neon arm sse c
Why does gcc, with -O3, unnecessarily clear a local ARM NEON array?
user_5230790
•
6
2
196
compiler-bug arm64 neon gcc c
Is there a way to auto-vectorize while using reciprocals and reciprocal square roots?
user_167451020
•
2
0
133
auto-vectorization avx neon sse c++
What doest `vaddhn_high_s16` actually do?
user_29990960
•
3
1
92
arm64 neon intrinsics simd c++
Why does FADDP D-form have higher throughput than FADDP Q-form on the Cortex-A72
user_30182430
•
2
0
188
micro-architecture arm64 neon simd cpu-architecture
Hot Questions