Question

CMake: Enable '/arch:AVX2' in Visual Studio 2013 projects

score 4.2k · Answer 1 · 2015-06-10 07:36:37

4.2k

Views

CMake: Enable '/arch:AVX2' in Visual Studio 2013 projects

Published on 10 June 2015 at 07:36

score 217 · Answer 2 · 2015-06-11 19:04:04

217

Views

For some reason serial code runs faster than SIMD code

Published on 11 June 2015 at 19:04

score 1.3k · Answer 3 · 2015-06-15 19:48:54

1.3k

Views

SSE - AVX conversion from double to char

Published on 15 June 2015 at 19:48

score 2.4k · Answer 4 · 2015-06-26 09:13:46

2.4k

Views

Is it really efficient to use Karatsuba algorithm in 64-bit x 64-bit multiplication?

Published on 26 June 2015 at 09:13

score 5.1k · Answer 5 · 2016-12-19 17:42:54

5.1k

Views

How can I convert a vector of float to short int using avx instructions?

Published on 19 December 2016 at 17:42

score 2.7k · Answer 6 · 2016-12-30 22:54:56

2.7k

Views

Can I use the AVX FMA units to do bit-exact 52 bit integer multiplications?

Published on 30 December 2016 at 22:54

score 1.5k · Answer 7 · 2017-01-01 19:41:34

1.5k

Views

How to force pointer returned by new operator to be 32-byte aligned

Published on 01 January 2017 at 19:41

score 712 · Answer 8 · 2017-01-06 05:42:11

712

Views

Speeding up gather

Published on 06 January 2017 at 05:42

score 1.8k · Answer 9 · 2014-11-11 12:09:40

1.8k

Views

Why doesn't MSVC's auto-vectorization use AVX2?

Published on 11 November 2014 at 12:09

score 2.7k · Answer 10 · 2014-11-19 09:32:52

2.7k

Views

Optimal SIMD algorithm to rotate or transpose an array

Published on 19 November 2014 at 09:32

score 551 · Answer 11 · 2014-11-21 22:23:34

551

Views

AVX2 rotate vector

Published on 21 November 2014 at 22:23

score 2.2k · Answer 12 · 2013-12-15 06:26:03

2.2k

Views

_mm256_loadu2_m128i intrinsic not available under g++?

Published on 15 December 2013 at 06:26

score 914 · Answer 13 · 2017-08-24 16:49:32

914

Views

Efficient (on Ryzen) way to extract the odd elements of a m256 into a m128?

Published on 24 August 2017 at 16:49

score 1.8k · Answer 14 · 2017-08-30 18:32:03

1.8k

Views

Having 4 bits, how to produce a mask for AVX register?

Published on 30 August 2017 at 18:32

score 410 · Answer 15 · 2017-09-01 14:23:40

410

Views

SIMD -> uint16_t array to float array work on float then back to uint16_t

Published on 01 September 2017 at 14:23

score 1.5k · Answer 16 · 2017-09-04 13:44:06

1.5k

Views

Unpack m128i/m256i to m64 (MMX, SSE2, AVX2)

Published on 04 September 2017 at 13:44

score 1.7k · Answer 17 · 2017-09-03 15:16:36

1.7k

Views

Efficient way to set first N or last N bits of __m256i to 1, the rest to 0

Published on 03 September 2017 at 15:16

score 3.6k · Answer 18 · 2017-09-18 05:44:56

3.6k

Views

Bitwise NOT/complement in AVX2

Published on 18 September 2017 at 05:44

score 424 · Answer 19 · 2020-09-24 10:12:30

424

Views

AVX512 Vector Multiplication Speed

Published on 24 September 2020 at 10:12

score 317 · Answer 20 · 2020-10-12 00:04:16

317

Views

Using SIMD, how do I conditionally move only the pixels with an alpha channel value of 255?

Published on 12 October 2020 at 00:04

List Question

CMake: Enable '/arch:AVX2' in Visual Studio 2013 projects

For some reason serial code runs faster than SIMD code

SSE - AVX conversion from double to char

Is it really efficient to use Karatsuba algorithm in 64-bit x 64-bit multiplication?

How can I convert a vector of float to short int using avx instructions?

Can I use the AVX FMA units to do bit-exact 52 bit integer multiplications?

How to force pointer returned by new operator to be 32-byte aligned

Speeding up gather

Why doesn't MSVC's auto-vectorization use AVX2?

Optimal SIMD algorithm to rotate or transpose an array

AVX2 rotate vector

_mm256_loadu2_m128i intrinsic not available under g++?

Efficient (on Ryzen) way to extract the odd elements of a m256 into a m128?

Having 4 bits, how to produce a mask for AVX register?

SIMD -> uint16_t array to float array work on float then back to uint16_t

Unpack m128i/m256i to m64 (MMX, SSE2, AVX2)

Efficient way to set first N or last N bits of __m256i to 1, the rest to 0

Bitwise NOT/complement in AVX2

AVX512 Vector Multiplication Speed

Using SIMD, how do I conditionally move only the pixels with an alpha channel value of 255?

Trending Questions

Popular # Hahtags

Popular Questions