Questions tagged [sse]
Streaming SIMD Extensions (SSE) is the first generation of SIMD Intel's instruction sets available on modern x86-compatible CPUs. SSE offers single-precision floating point arithmetic and integer arithmetic (excluding division) and logical operations on packed or single operands of sizes from 8 to 64 bits.
3
votes
1answer
50 views
Generic pixel class to seamlessly alpha-blend and convert between different pixel structure layouts
Does what it says in the title. I just finished this and wanted to share with someone.
Looking for possible optimizations, bugs (most of it is tested to work) or any constructive criticism.
...
4
votes
0answers
134 views
Fast Hardy-Weinberg equilibrium simulation
I was very bored over one of my breaks this year, so I built a Hardy-Weinberg equilibrium simulator for two unrelated alleles of the same gene. Hardy-Weinberg equilibrium is when there is no evolution,...
6
votes
1answer
267 views
SIMD memcpy assembler implementation
I am fairly rusty with assembler, let alone the AT&T syntax. I would appreciate it if someone with more experience could please review the following memcpy implementation. Note that this will only ...
1
vote
1answer
276 views
Fast affine transformations of many 3D points by one 3×4 matrix
I wrote a function to batch-transform 3D vectors by a single 3x4 matrix using SSE2:
...
6
votes
3answers
610 views
Converting Array of Floats to UINT8 (`char`) or UINT16 (`unsigned short`) Using SSE4
The problem is given image in 32 Bit Floating Point Format (float) how to convert it to UINT8 (char) or UNIT16 (...
0
votes
1answer
233 views
Finding the Minimum and Maximum Value in an Image
Given an image which is padded to support aligning (SSE) I need to find its minimum and maximum value as fast as possible.
Mind you the padded values are not defined and can't be assumed to have ...
5
votes
2answers
2k views
AVX SIMD in matrix multiplication
I have coded the following C function for multiplying two NxN matrices and using AVX vectors to speed up the calculation. It works but the speedup is not what is to be expected(some scalar code is ...
8
votes
2answers
559 views
Vectorized and Multi Threaded Image Convolution
I created code for Image Convolution. The code is in my Image Convolution GitHub Repository.
It includes the case for arbitrary Image Convolution and for Separable Kernel Convolution.
The code is a ...
10
votes
3answers
309 views
SSE loop to walk likely primes
This is a continuation of a discussion that was started here. While there are some interesting points there about instruction timing and latency, it is not necessary to read that Question to ...
1
vote
0answers
252 views
MD5 implementation with SSE2
My implementation of MD5 (hash function) works correctly, though it has terrible performance. As a laboratory work, I must write the algorithm myself and implement the brute force.
full code with ...
5
votes
0answers
151 views
Cache conscious SIMD matrix multiply of unsigned integers
The goal of the code review by order of importance (i.e. What I hope to hear from you):
I've verified correctness using a straightforward matrix multiply function though I am open to those who want ...
1
vote
1answer
237 views
HPC kernel for DGEMM: compiler v.s. assembly
This is a correct version, for computing a small matrix multiplication: C += A * B, where C is ...
4
votes
0answers
103 views
Parallel popcount of packed bytes
I want to do a popcount of bytes in an array.
Not a total popcount of everything, but lots of little popcounts for every individual byte to determine the neighborhood count in GoL.
The following ...
6
votes
1answer
4k views
Optimizing bilinear interpolation
I have spent countless hours trying to speed up my bilinear interpolation up. I even tried implementing an SSE version (a double version and a float version), but that was even slower than this ...
3
votes
0answers
529 views
Grayscale bilinear patch extraction - SSE optimization
My program makes an intensive use of small sub-images extracted using bilinear interpolation from larger grayscale images (some call this problem subpixel translation).
I am using the following C++ ...
6
votes
1answer
406 views
Optimizing multiplication of square matrices for full CPU utilization
Problem
I am learning about HPC and code optimization. I attempt to replicate the results in Goto's seminal matrix multiplication paper. Despite my best efforts, I cannot get over ~50% maximum ...
4
votes
0answers
197 views
SSE optimisation for audio resampling
I'm learning SSE for the first time and trying to optimise some code. Using oprofile shows that the CPU usage in this function went down from 2.5% to 0.9% using the ...
8
votes
1answer
546 views
Optimizing multiplication of two polynomials in GF(2^32)
I have the following method for multiplying two polynomials in GF(232):
...
4
votes
2answers
876 views
Is this C++ program correctly implemented in SSE?
I have to implement a simple program in SSE, and I don't know if I have done it in the right way (it's my first program in SSE).
This is the C++ program:
...
1
vote
1answer
2k views
SSE Intrinsics Gaussian Blur Approximation
The following is the current iteration of a gaussian blur approximation code I am using.
A typical naive convolution operation is O(N*M), where N is the number of ...
3
votes
1answer
672 views
Trying to improve and better understand Intel SSE intrinsics
I've been writing a collection of signal processing function optimized with SSE intrinsics (mostly for audio processing). Here is a linear interpolation function I wrote. It works well and is quite ...
7
votes
1answer
435 views
Writing SIMD libraries for C++ on FASM in x86-64 Linux
I have recently started a project of SIMD libraries development for C++ on FASM for x86-64 Linux.
I would be glad to hear any opinion or feedback about the project, cleanness of the code and ...
4
votes
1answer
2k views
SSE2 assembly optimization - multiply unsigned shorts and add the result
I am attempting to optimize a piece of C code which aims to multiply a series of pairs of unsigned shorts and add the result. I am only concerned about the high 16 bits of the result, and I can ...
36
votes
9answers
11k views
Copying 80 bytes as fast as possible
I am running a math-oriented computation that spends a significant amount of its time doing memcpy, always copying 80 bytes from one location to the next, an array ...
