Question

Achieving More FMA3 Performance Than The Theoretical Maximum

score 52 · Answer 1 · 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 2 · 2024-03-13T11:11:07.133000

47

Views

High Variance In Manual Vectorization Performance

Published on 13 March 2024 at 11:11

score 235 · Answer 3 · 2023-11-26T17:32:32.493000

235

Views

Can we replace XOR with multiply-add?

Published on 26 November 2023 at 17:32

score 149 · Answer 4 · 2023-10-29T13:27:55.727000

149

Views

Why this AVX2 slowdown with FMA x86 MS C Compiler?

Published on 29 October 2023 at 13:27

score 207 · Answer 5 · 2023-03-27T18:18:00.747000

207

Views

Clang fused multiply-add depends on constancy of expression arguments

Published on 27 March 2023 at 18:18

score 153 · Answer 6 · 2022-11-28T03:42:19.347000

153

Views

v4fmaddps instructions for packed 32-bit integers

Published on 28 November 2022 at 03:42

score 206 · Answer 7 · 2022-11-13T18:53:59.803000

206

Views

GCC 12 (minGW 64): how to enable fused multiply add code generation

Published on 13 November 2022 at 18:53

score 345 · Answer 8 · 2022-09-23T13:22:10.000000

345

Views

How should I implement a generic FMA/FMAF instruction in software?

Published on 23 September 2022 at 13:22

score 107 · Answer 9 · 2022-08-16T10:03:24.427000

107

Views

Fast fixed-size polynomial evaluation: MSVC vs GCC

Published on 16 August 2022 at 10:03

score 123 · Answer 10 · 2022-05-30T11:17:12.890000

123

Views

Deleteing initialization leads to avx2 fma performance drop. Why?

Published on 30 May 2022 at 11:17

score 257 · Answer 11 · 2022-04-11T13:44:15.283000

257

Views

Latency and number of FMA units

Published on 11 April 2022 at 13:44

score 1.2k · Answer 12 · 2022-03-26T10:17:15.903000

1.2k

Views

Fastest way to multiply and sum/add two arrays (dot product) - unaligned surprisingly faster than FMA

Published on 26 March 2022 at 10:17

score 135 · Answer 13 · 2022-02-02T12:26:51.183000

135

Views

Terminology: why "floating multiply-add" instead of "fused multiply-add"?

Published on 02 February 2022 at 12:26

score 173 · Answer 14 · 2021-11-19T16:26:40.300000

173

Views

vfmadd231ps Floating Point Exception c0000090

Published on 19 November 2021 at 16:26

score 406 · Answer 15 · 2021-10-08T11:53:17.987000

406

Views

Why Fma code is performing worse than Avx?

Published on 08 October 2021 at 11:53

score 689 · Answer 16 · 2021-08-03T08:24:21.190000

689

Views

Is there any better implemention for integer 'mul and add' with avx?

Published on 03 August 2021 at 08:24

score 322 · Answer 17 · 2021-02-04T22:53:08.950000

322

Views

How to find magic multipliers for divisions by constant on a GPU?

Published on 04 February 2021 at 22:53

score 738 · Answer 18 · 2021-01-07T19:46:39.333000

738

Views

CUDA half float operations without explicit intrinsics

Published on 07 January 2021 at 19:46

score 1.1k · Answer 19 · 2021-01-03T11:20:57.057000

1.1k

Views

incompatible types when assigning to type ‘__m256d’ from type ‘int’

Published on 03 January 2021 at 11:20

score 303 · Answer 20 · 2020-12-24T13:25:39.933000

303

Views

How to refine floating-point division on FMA-capable GPUs?

Published on 24 December 2020 at 13:25

List Question

Achieving More FMA3 Performance Than The Theoretical Maximum

High Variance In Manual Vectorization Performance

Can we replace XOR with multiply-add?

Why this AVX2 slowdown with FMA x86 MS C Compiler?

Clang fused multiply-add depends on constancy of expression arguments

v4fmaddps instructions for packed 32-bit integers

GCC 12 (minGW 64): how to enable fused multiply add code generation

How should I implement a generic FMA/FMAF instruction in software?

Fast fixed-size polynomial evaluation: MSVC vs GCC

Deleteing initialization leads to avx2 fma performance drop. Why?

Latency and number of FMA units

Fastest way to multiply and sum/add two arrays (dot product) - unaligned surprisingly faster than FMA

Terminology: why "floating multiply-add" instead of "fused multiply-add"?

vfmadd231ps Floating Point Exception c0000090

Why Fma code is performing worse than Avx?

Is there any better implemention for integer 'mul and add' with avx?

How to find magic multipliers for divisions by constant on a GPU?

CUDA half float operations without explicit intrinsics

incompatible types when assigning to type ‘__m256d’ from type ‘int’

How to refine floating-point division on FMA-capable GPUs?

Trending Questions

Popular # Hahtags

Popular Questions