TornadoVM 1.0.6

[jjfumero]

Improvements

• [fix][rt] support for query multiple SPIR-V supported versions #442: Support for multiple SPIR-V device versions (>= 1.2).
• Enabling release memory (device memory deallocations) mode after each run from the Execution Plan #444: Enabling automatic device memory clean-up after each run from the execution plan.
• [feat] Query Memory Consumption and expanded profiler with memory consumption #448: API extension to query device memory consumption at the TaskGraph granularity.
• Option to select the SPIR-V Runtime Dispatcher #451: Option to select the default SPIR-V runtime.
• API Refactoring and Documentation #455: Refactoring the API and documentation updated.
• Refactor tests and update to use the API try-with-resources #460: Refactoring all examples to use try-with-resources execution plans by default.
• Support conditional copy-by-reference in local memory  #462: Support for copy array references from private to private memory on the hardware accelerator.

Compatibility

• Remove writeFile for Julia in unit-test #438: No writes for intermediate files to avoid permissions issues with Jenkins.
• [feat] Removed conflicts that prohibited the build of remote branches #440: Update Jenkinsfile for CI/CD testing.
• Update Jenkins to run SPIRV unit-tests for OpenCL and LevelZero runtimes #443: Level Zero and OpenCL runtimes for SPIR-V included in the Jenkins CI/CD.
• Update tornado-benchmarks.py, add --properties option #450: TornadoVM benchmark script improved to report dimensions and sizes.
• Update Jenkinsfile regarding the execution of SPIR-V with the OpenCL and Intel LevelZero runtimes #453: Update Jenkinsfile with regards to the runtime for SPIR-V.

Bug Fixes

• Hotfix to bypass clCreateWithBinaryIL for non-spirv devices on OSx #434: Fix for building TornadoVM on OSx after integration with SPIR-V binaries for OpenCL.
• Fix failing unit-tests for PTX in Jenkins #441: Fix PTX unit-tests.
• Guard Nvidia GPU scheduler for minor version to prevent incorrect thread scheduling #446: Fix NVIDIA thread-block scheduler for new GPU drivers.
• Fix recompilation when batch processing is not triggered #447: Fix recompilation when batch processing is not triggered.
• CPU Virtual tests removed, since the scheduler between CPU/GPU is the same #463: Fix unit-tests for CPU virtual devices.

---

This discussion was created from the release TornadoVM 1.0.6.