issues
search
mklarqvist
/
positional-popcount
Fast C functions for the computing the positional popcount (pospopcnt).
Apache License 2.0
52
stars
5
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
AVX512BW 8-bit pospopcnt using vpsadbw
#40
WojciechMula
closed
4 years ago
0
A bit faster Harley-Seal-based implementations
#39
WojciechMula
closed
4 years ago
0
Another variant of 8-bit pospopcnt (SSE4 and AVX2 implementations)
#38
WojciechMula
closed
4 years ago
0
Faster 32-bit pospopcnt
#37
WojciechMula
closed
4 years ago
2
Detailed
#36
mklarqvist
closed
4 years ago
0
Positional popcount for 32-bit entities
#35
WojciechMula
closed
4 years ago
0
Proper incrementation order in scalar hist1x4
#34
WojciechMula
closed
4 years ago
0
Pospopcnt for 8-bit data
#33
WojciechMula
closed
4 years ago
0
Add pospopcnt 8-bit procedures
#32
WojciechMula
closed
4 years ago
0
Eliminate warnings
#31
WojciechMula
closed
4 years ago
0
MinGW problems
#30
WojciechMula
opened
4 years ago
1
Fix Windows build
#29
WojciechMula
closed
4 years ago
0
Refactor benchmark program
#28
WojciechMula
closed
4 years ago
0
Add 8-bit positional popcount procedures
#27
WojciechMula
closed
4 years ago
1
Daniel/add get alignment
#26
lemire
closed
5 years ago
0
Discussion
#25
aqrit
closed
4 years ago
10
update scalar_umul128 & sse2_sad
#24
aqrit
closed
5 years ago
1
Use ternary logic instruction
#23
WojciechMula
closed
5 years ago
2
Measuring the counter overhead.
#22
lemire
closed
5 years ago
8
Optionally use aligned input data
#21
WojciechMula
closed
5 years ago
7
Cleanup
#20
mklarqvist
closed
5 years ago
0
Accumulate counters using "horizontal reduction"
#19
WojciechMula
closed
5 years ago
8
Added pospopcnt_u16_scalar_umul128 subroutine.
#18
mklarqvist
closed
5 years ago
0
new scalar umul128 method
#17
aqrit
closed
5 years ago
11
Intel perf
#16
mklarqvist
closed
5 years ago
0
Using ref. cycles too.
#15
lemire
closed
5 years ago
0
Let us display the array size (useful for comparison with caching).
#14
lemire
closed
5 years ago
0
Dev2
#13
mklarqvist
closed
5 years ago
0
We want "n" to be an uint32_t.
#12
lemire
closed
5 years ago
0
Horizontal sum of 16-bit counters
#11
WojciechMula
closed
5 years ago
15
AVX512: scatter update loop
#10
WojciechMula
closed
5 years ago
4
Minor optimization (improving avx512_csa)
#9
lemire
closed
5 years ago
0
Update machine measurments (from CNL)
#8
WojciechMula
closed
5 years ago
0
new scalar tail for sse2_sad method
#7
aqrit
closed
5 years ago
2
New scheme
#6
lemire
closed
5 years ago
11
add population count (e.g., AVX-512)
#5
lemire
closed
5 years ago
0
Some fixes + instrumented tests
#4
lemire
closed
5 years ago
0
avx512 popcnt
#3
mklarqvist
closed
5 years ago
0
avx512
#2
mklarqvist
closed
5 years ago
0
avx512 functions
#1
mklarqvist
closed
5 years ago
0