Question

Using `static` on a AVX2 counter function increases performance ~10x in MT environment without any change in Compiler optimizations

score 76 · Answer 1 · 2024-03-30T14:27:49.290000

76

Views

Using `static` on a AVX2 counter function increases performance ~10x in MT environment without any change in Compiler optimizations

Published on 30 March 2024 at 14:27

score 73 · Answer 2 · 2024-03-21T21:01:25.743000

73

Views

Convert Variable Width Bitstream (2-bit or 4-bit symbols) into Fixed Width

Published on 21 March 2024 at 21:01

score 52 · Answer 3 · 2024-03-14T13:58:49.053000

52

Views

Achieving More FMA3 Performance Than The Theoretical Maximum

Published on 14 March 2024 at 13:58

score 47 · Answer 4 · 2024-03-13T11:11:07.133000

47

Views

High Variance In Manual Vectorization Performance

Published on 13 March 2024 at 11:11

score 84 · Answer 5 · 2024-03-05T04:37:31.340000

84

Views

AVX2 vectorization for code similar to prefix sum (decrement by count of preceding matches in short fixed-length arrays)

Published on 05 March 2024 at 04:37

score 166 · Answer 6 · 2024-02-22T18:02:30.420000

166

Views

Multiplying packed 32-bit integers by a 32-bit float with AVX2

Published on 22 February 2024 at 18:02

score 193 · Answer 7 · 2024-01-28T15:16:37.683000

193

Views

Are there processors on which VPMASKMOVD generates faults for the masked-out elements?

Published on 28 January 2024 at 15:16

score 123 · Answer 8 · 2024-01-22T17:14:51.247000

123

Views

Nan problem with Intel 2022 compiler using AVX2 & /fp:fast

Published on 22 January 2024 at 17:14

score 94 · Answer 9 · 2024-01-13T22:18:17.940000

94

Views

_mm256_insert_epi32() has no effect

Published on 13 January 2024 at 22:18

score 34 · Answer 10 · 2024-01-09T15:36:13.050000

34

Views

Find common minimum CPU features to expect when targeting a certain macOS deployment target

Published on 09 January 2024 at 15:36

score 137 · Answer 11 · 2024-01-08T09:26:18.293000

137

Views

AVX2 narrowing conversion, from uint16_t to uint8_t

Published on 08 January 2024 at 09:26

score 1.4k · Answer 12 · 2023-12-25T07:56:10.160000

1.4k

Views

Why performance for this index-of-max function over many arrays of 256 bytes is so slow on Intel i3-N305 compared to AMD Ryzen 7 3800X?

Published on 25 December 2023 at 07:56

score 61 · Answer 13 · 2023-12-19T13:14:48.373000

61

Views

dst[i] eqaul src[i] multiply by dst[i-1] in avx or sse

Published on 19 December 2023 at 13:14

score 137 · Answer 14 · 2023-12-12T03:46:49.027000

137

Views

Why can't Oracle Linux automatically detect CPUs with AVX?

Published on 12 December 2023 at 03:46

score 99 · Answer 15 · 2023-12-12T02:50:29.573000

99

Views

No Speedup in Float Multiply with Rust SSE Intrinsics

Published on 12 December 2023 at 02:50

score 117 · Answer 16 · 2023-12-11T15:25:08.057000

117

Views

Fast int32_t dot product of two C++ integer vectors using AVX is not faster

Published on 11 December 2023 at 15:25

score 84 · Answer 17 · 2023-12-08T10:44:36.087000

84

Views

Adding slightly shifted vectors

Published on 08 December 2023 at 10:44

score 184 · Answer 18 · 2023-11-18T05:48:30.413000

184

Views

How to Improve XORing of large uint64 arrays?

Published on 18 November 2023 at 05:48

score 128 · Answer 19 · 2023-11-03T14:30:43.103000

128

Views

Setting/getting 1-bits of __m256i vector from integer array of bit positions

Published on 03 November 2023 at 14:30

score 98 · Answer 20 · 2023-11-01T17:47:52.063000

98

Views

how to calculate a parallel product in ispc

Published on 01 November 2023 at 17:47

List Question

Using `static` on a AVX2 counter function increases performance ~10x in MT environment without any change in Compiler optimizations

Convert Variable Width Bitstream (2-bit or 4-bit symbols) into Fixed Width

Achieving More FMA3 Performance Than The Theoretical Maximum

High Variance In Manual Vectorization Performance

AVX2 vectorization for code similar to prefix sum (decrement by count of preceding matches in short fixed-length arrays)

Multiplying packed 32-bit integers by a 32-bit float with AVX2

Are there processors on which VPMASKMOVD generates faults for the masked-out elements?

Nan problem with Intel 2022 compiler using AVX2 & /fp:fast

_mm256_insert_epi32() has no effect

Find common minimum CPU features to expect when targeting a certain macOS deployment target

AVX2 narrowing conversion, from uint16_t to uint8_t

Why performance for this index-of-max function over many arrays of 256 bytes is so slow on Intel i3-N305 compared to AMD Ryzen 7 3800X?

dst[i] eqaul src[i] multiply by dst[i-1] in avx or sse

Why can't Oracle Linux automatically detect CPUs with AVX?

No Speedup in Float Multiply with Rust SSE Intrinsics

Fast int32_t dot product of two C++ integer vectors using AVX is not faster

Adding slightly shifted vectors

How to Improve XORing of large uint64 arrays?

Setting/getting 1-bits of __m256i vector from integer array of bit positions

how to calculate a parallel product in ispc

Trending Questions

Popular # Hahtags

Popular Questions