Search results

Error is too big with ssyrk, ssyr2k on apple m1

correctness

Status: Open.

#668 In CNugteren/CLBlast;

· octaveoclx opened on Jul 7, 2026

FP16 with NVIDIA and Apple SILICON

feature request

Status: Open.

#667 In CNugteren/CLBlast;

· octaveoclx opened on Jun 10, 2026

Tuning issue on directgemm on access memory beyond the largest legal address

correctness

Status: Open.

#665 In CNugteren/CLBlast;

· octaveoclx opened on May 24, 2026

Changing of Tuning strategy after skipping or driver problem

feature request

Status: Open.

#664 In CNugteren/CLBlast;

· octaveoclx opened on May 17, 2026

Batched TRSM

feature request

Status: Open.

#663 In CNugteren/CLBlast;

· octaveoclx opened on May 6, 2026

Double-buffer GEMM

performance

Status: Open.

#652 In CNugteren/CLBlast;

· CNugteren opened on Dec 3, 2025

Using Builtin FindOpenCL Instead of Manual CMake

maintenance

Status: Open.

#645 In CNugteren/CLBlast;

· EJainDev opened on Nov 26, 2025

Performance Overhead for Irregular Matrices

performance

Status: Open.

#642 In CNugteren/CLBlast;

· EJainDev opened on Oct 26, 2025

[BUG] Discrepancy Between OpenCL and Host

correctness

Status: Open.

#623 In CNugteren/CLBlast;

· EJainDev opened on Aug 31, 2025

[QUESTION] CUDA API?

feature request

Status: Open.

#619 In CNugteren/CLBlast;

· EJainDev opened on Aug 29, 2025

Intel UHD GPU: Only reaching half the peak GFlops for GEMM

performance

Status: Open.

#618 In CNugteren/CLBlast;

· EJainDev opened on Aug 29, 2025

Using Required Sub Group Size for Intel

feature request

Status: Open.

#611 In CNugteren/CLBlast;

· EJainDev opened on Aug 25, 2025