mratsim laser issues - Githubissues

mratsim / laser

The HPC toolbox: fused matrix multiplication, convolution, data-parallel strided tensor primitives, OpenMP facilities, SIMD, JIT Assembler, CPU detection, state-of-the-art vectorized BLAS for floats and integers

Apache License 2.0

281 stars 15 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

I can't use openMP in nim_2.0, and it needs to put dll files ,like libgomp-1, the same folder to built exe file to execute it.

#43 kiyoken1594 opened 2 months ago
0
performance of avx512 bit ops and popcounts

#41 brentp opened 5 years ago
4
Mysterious 2x perf regression on GEMM

#40 mratsim opened 5 years ago
2
Add float32 implementation of min/max/sum

#39 mratsim closed 5 years ago
0
[Benchmarks] Cleanup fp_reduction_latency benchmarks

#38 mratsim opened 5 years ago
0
[Showstopper regression] emit does not generate proper symbol

#37 mratsim closed 5 years ago
1
parallel reduction

#36 brentp closed 5 years ago
6
Added Win32 executable memory support

#35 awr1 closed 5 years ago
0
Lux refactor v3 - Frontend

#34 mratsim closed 5 years ago
0
[Gemm] Nim devel compiler gets stuck when compiling older commits

#33 mratsim closed 5 years ago
1
[GEMM] Significant performance regression (divided by 5)

#32 mratsim closed 5 years ago
2
[Lux] Multithreading for JIT code

#31 mratsim opened 5 years ago
0
NUMA-aware memory allocation and computation

#30 mratsim opened 5 years ago
0
Lux AST refactor - frontend done

#29 mratsim closed 5 years ago
0
WIP - Fix 27 and 26

#28 mratsim closed 5 years ago
0
Regression on GEMM allocation

#27 mratsim closed 5 years ago
3
Prepacked gemm

#26 mratsim closed 5 years ago
0
System Profile Dual Xeon Gold 6154

#25 Laurae2 closed 5 years ago
0
Optimize serial gemm + Fix parallel result

#24 mratsim closed 5 years ago
2
performance of gemm_strided vs numpy

#23 timotheecour opened 5 years ago
1
gemm_strided: error: always_inline function '_mm256_setzero_pd' requires target feature 'xsave'

#22 timotheecour opened 5 years ago
1
[GEMM] Enhance serial implementation

#21 mratsim opened 5 years ago
1
Improve gemm threading

#20 mratsim closed 5 years ago
1
Fast vectorized exponential float32 implementation (SSE2, AVX2, AVX512)

#19 mratsim closed 5 years ago
0
Fused assignation shortcut

#18 mratsim opened 5 years ago
0
Fast image loading primitives

#17 mratsim opened 5 years ago
0
Try to workaround static generic regression with static param

#16 mratsim closed 5 years ago
1
Devel regression "object constructor needs an object type"

#15 mratsim closed 5 years ago
1
AVX512 GEMM kernel

#14 mratsim closed 5 years ago
1
Transpose does not scale well with multithread

#13 Laurae2 opened 5 years ago
0
Create a benchmark script

#12 mratsim opened 5 years ago
0
Exponential: Dual Xeon Gold 6154 result

#11 mratsim opened 5 years ago
3
Benchmark example using Intel MKL (for history)

#10 Laurae2 opened 5 years ago
1
Matrix multiplication: Nested parallelism

#9 mratsim opened 5 years ago
1
Optimised random sampling methods

#8 mratsim opened 5 years ago
1
Jit assembler

#7 mratsim closed 5 years ago
0
Generalize gemm

#6 mratsim closed 6 years ago
0
Parallel strided iteration does not scale linearly

#5 mratsim opened 6 years ago
0
Introduce forEach multi-stage domain specific language

#4 mratsim closed 6 years ago
0
Update for devel OpenMP

#3 mratsim closed 6 years ago
0
[Design] Error model

#2 mratsim opened 6 years ago
0
Iteration code size comparison

#1 mratsim closed 6 years ago
0