Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / mratsim/laser issues and pull requests

#41 - performance of avx512 bit ops and popcounts

Issue - State: open - Opened by brentp about 5 years ago - 4 comments

#40 - Mysterious 2x perf regression on GEMM

Issue - State: open - Opened by mratsim about 5 years ago - 2 comments

#39 - Add float32 implementation of min/max/sum

Pull Request - State: closed - Opened by mratsim over 5 years ago

#38 - [Benchmarks] Cleanup fp_reduction_latency benchmarks

Issue - State: open - Opened by mratsim over 5 years ago

#37 - [Showstopper regression] emit does not generate proper symbol

Issue - State: closed - Opened by mratsim over 5 years ago - 1 comment
Labels: bug, need upstream fix

#36 - parallel reduction

Issue - State: closed - Opened by brentp over 5 years ago - 6 comments

#35 - Added Win32 executable memory support

Pull Request - State: closed - Opened by awr1 over 5 years ago

#34 - Lux refactor v3 - Frontend

Pull Request - State: closed - Opened by mratsim over 5 years ago

#33 - [Gemm] Nim devel compiler gets stuck when compiling older commits

Issue - State: closed - Opened by mratsim over 5 years ago - 1 comment

#32 - [GEMM] Significant performance regression (divided by 5)

Issue - State: closed - Opened by mratsim over 5 years ago - 2 comments
Labels: bug

#31 - [Lux] Multithreading for JIT code

Issue - State: open - Opened by mratsim over 5 years ago

#30 - NUMA-aware memory allocation and computation

Issue - State: open - Opened by mratsim over 5 years ago

#29 - Lux AST refactor - frontend done

Pull Request - State: closed - Opened by mratsim over 5 years ago

#28 - WIP - Fix 27 and 26

Pull Request - State: closed - Opened by mratsim over 5 years ago

#27 - Regression on GEMM allocation

Issue - State: closed - Opened by mratsim over 5 years ago - 3 comments
Labels: bug

#26 - Prepacked gemm

Pull Request - State: closed - Opened by mratsim almost 6 years ago

#25 - System Profile Dual Xeon Gold 6154

Issue - State: closed - Opened by Laurae2 almost 6 years ago

#24 - Optimize serial gemm + Fix parallel result

Pull Request - State: closed - Opened by mratsim almost 6 years ago - 2 comments

#23 - performance of gemm_strided vs numpy

Issue - State: open - Opened by timotheecour almost 6 years ago - 1 comment

#21 - [GEMM] Enhance serial implementation

Issue - State: open - Opened by mratsim almost 6 years ago - 1 comment

#20 - Improve gemm threading

Pull Request - State: closed - Opened by mratsim almost 6 years ago - 1 comment

#19 - Fast vectorized exponential float32 implementation (SSE2, AVX2, AVX512)

Pull Request - State: closed - Opened by mratsim almost 6 years ago

#18 - Fused assignation shortcut

Issue - State: open - Opened by mratsim almost 6 years ago
Labels: enhancement

#17 - Fast image loading primitives

Issue - State: open - Opened by mratsim almost 6 years ago

#16 - Try to workaround static generic regression with static param

Pull Request - State: closed - Opened by mratsim almost 6 years ago - 1 comment

#15 - Devel regression "object constructor needs an object type"

Issue - State: closed - Opened by mratsim almost 6 years ago - 1 comment
Labels: bug, need upstream fix

#14 - AVX512 GEMM kernel

Pull Request - State: closed - Opened by mratsim almost 6 years ago - 1 comment

#13 - Transpose does not scale well with multithread

Issue - State: open - Opened by Laurae2 almost 6 years ago

#12 - Create a benchmark script

Issue - State: open - Opened by mratsim almost 6 years ago

#11 - Exponential: Dual Xeon Gold 6154 result

Issue - State: open - Opened by mratsim almost 6 years ago - 3 comments

#10 - Benchmark example using Intel MKL (for history)

Issue - State: open - Opened by Laurae2 almost 6 years ago - 1 comment

#9 - Matrix multiplication: Nested parallelism

Issue - State: open - Opened by mratsim almost 6 years ago - 1 comment

#8 - Optimised random sampling methods

Issue - State: open - Opened by mratsim almost 6 years ago - 1 comment

#7 - Jit assembler

Pull Request - State: closed - Opened by mratsim about 6 years ago

#6 - Generalize gemm

Pull Request - State: closed - Opened by mratsim about 6 years ago

#5 - Parallel strided iteration does not scale linearly

Issue - State: open - Opened by mratsim about 6 years ago
Labels: optimisation

#4 - Introduce forEach multi-stage domain specific language

Pull Request - State: closed - Opened by mratsim about 6 years ago

#3 - Update for devel OpenMP

Issue - State: closed - Opened by mratsim about 6 years ago

#2 - [Design] Error model

Issue - State: open - Opened by mratsim about 6 years ago
Labels: RFC

#1 - Iteration code size comparison

Issue - State: closed - Opened by mratsim about 6 years ago
Labels: benchmark, code size