v3.0.3

ExaTron v3.0.3

Diff since v3.0.2

Merged pull requests:

• Using Adapt (#55) (@michel2323)

v3.0.2

ExaTron v3.0.2

Diff since v3.0.1

Merged pull requests:

• Fixing potential race conditions (#54) (@kibaekkim)

v3.0.1

ExaTron v3.0.1

Diff since v3.0.0

Closed issues:

• Sporadic hang or freeze in CUDA kernels (#50)
• dgpnorm for CPU vs. GPU (#51)
• Solution difference in 1e-5 from dtron on CPU vs. GPU (#52)

Merged pull requests:

• CI Julia 1.9 (#53) (@michel2323)

v3.0.0

ExaTron v3.0.0

Diff since v2.1.3

Merged pull requests:

• Update CI (#48) (@michel2323)
• Migrate to KA 0.9 (#49) (@michel2323)

v2.1.3

ExaTron v2.1.3

Diff since v2.1.2

Merged pull requests:

• Remove ROCDevice <: KA.GPU dependency (#47) (@michel2323)

v2.1.2

ExaTron v2.1.2

Diff since v2.1.1

Merged pull requests:

• Remove KA.GPU (#46) (@michel2323)

v2.1.1

ExaTron v2.1.1

Diff since v2.1.0

Merged pull requests:

• Adding tron_qp_kernel (#45) (@michel2323)

v2.1.0

ExaTron v2.1.0

Diff since v2.0.0

Closed issues:

• Improving performance on GPUs (1) - replace dsel2() with a GPU friendly top-k algorithm (#2)
• Improving performance on GPUs (2) - reduce getindex() and setindex!() time (#3)
• Improving performance on GPUs (3) - running hundreds of thousands of TRONs in parallel on GPUs (#4)
• Profiling GPU runs (#5)
• GPU compatibility of data types (#13)

v2.0.0

ExaTron v2.0.0

Diff since v1.1.0

Merged pull requests:

• Add KernelAbstractions.jl support (#44) (@michel2323)

v1.1.0

ExaTron v1.1.0

Diff since v1.0.0

Closed issues:

• Release for moving the ADMM part to ExaAdmm.jl (#32)

Merged pull requests:

• Adding CUDA QP API (#43) (@michel2323)