| 2 | |
| 3 | CLRadeonExtender 0.1.5: |
| 4 | |
| 5 | * ignore case in an access qualifier name's (Amd and AmdCL2) |
| 6 | * improve handling a '\()' and '\@' |
| 7 | * add SDWA and DPP words to set instruction encoding |
| 8 | * fixing few CLRXDocs typos |
| 9 | * fixes for AMD RX VEGA (GFX900) |
| 10 | * disassembler prints an instruction's position in comments |
| 11 | * update GcnTimings |
| 12 | * update VectorAdd and ReverseBits for LLVM 4.0 and Mesa3D 17.0.0 |
| 13 | * updates in ImageMix (correct workSize calculating for kernel) |
| 14 | * small fixes in disassembler |
| 15 | * disassembler can correctly disassemble GalliumCompute for LLVM 4.0 |
| 16 | * add '--llvmVersion' to clrxdisasm |
| 17 | * dump AMD HSA configuration for GalliumCompute and AmdCL2 (like in ROCm format) |
| 18 | * disassembler add '@' to hwreg and sendmsg to make dump compatible with clrxasm |
| 19 | * add '--HSAConfig' to dump AmdCL2 kernel configuration as AMD HSA config |
| 20 | * add AMD HSA configuration pseudo-ops to GalliumCompute and AmdCL2 binary formats |
| 21 | * update device list for Gallium and ROCm binary formats for recognizing device |
| 22 | * add pseudo-op '.default_hsa_features' to AmdCL2, Gallium and ROCm formats |
| 23 | * update headers in code |
| 24 | * make error handling more compact in assembler's code |
| 25 | * fixed '.machine', '.codeversion' handling (do not print obsolete warnings) |
| 26 | * add pkg-config files to installation |
| 27 | * remove obsolete warnings in CMakeLists.txt |
| 28 | * added GFX901 support (RX VEGA with HBCC ?) |
| 29 | * add Config.h and amdbin/Elf.h headers to Doxygen documentation |
| 30 | * change lowest device for GCN 1.2 to Iceland in GPUId. |
| 31 | * add support for Windows developments environments: CygWin and MinGW |
| 32 | * make detecting of 64-bits more portable in CMakeLists.txt (use compiler to do) |
| 33 | * checking whether std::call_once is available for non full supported std threads |
| 34 | * use only C++ compiler to check features (Int128Detect.cpp) |