diff options
-rw-r--r-- | CHANGELOG | 2 | ||||
-rw-r--r-- | ROADMAP.md | 21 |
2 files changed, 13 insertions, 10 deletions
@@ -3,7 +3,7 @@ Development (next version) - Re-designed and integrated the auto-tuner, no more dependency on CLTune - Made it possible to override the tuning parameters in the clients straight from JSON tuning files - Added OpenCL pre-processor to unroll loops and perform array-to-register promotions for compilers - which don't this themselves (ARM, Qualcomm) - greatly improves performance on these platforms + which don't do this themselves (ARM, Qualcomm) - greatly improves performance on these platforms - Added tuned parameters for various devices (see README) Version 1.2.0 @@ -3,12 +3,15 @@ CLBlast feature road-map This file gives an overview of the main features planned for addition to CLBlast. A first-order indication time-frame for development time is provided: -| Issue# | When | Who | Status | What | -| -----------|-------------|-----------|--------|---------------| -| - | Oct '17 | CNugteren | ✔ | CUDA API for CLBlast | -| [#169](https://github.com/CNugteren/CLBlast/issues/169), [#195](https://github.com/CNugteren/CLBlast/issues/195) | Oct-Nov '17 | CNugteren | ✔ | Auto-tuning the kernel selection parameter | -| [#181](https://github.com/CNugteren/CLBlast/issues/181), [#201](https://github.com/CNugteren/CLBlast/issues/201) | Nov '17 | CNugteren | ✔ | Compilation for Android and testing on a device | -| - | Nov '17 | CNugteren | ✔ | Integration of CLTune for easy testing on Android / fewer dependencies | -| [#128](https://github.com/CNugteren/CLBlast/issues/128), [#205](https://github.com/CNugteren/CLBlast/issues/205) | Nov-Dec '17 | CNugteren | | Pre-processor for loop unrolling and array-to-register-promotion for e.g. ARM Mali | -| [#207](https://github.com/CNugteren/CLBlast/issues/207) | Dec '17 | CNugteren | | Tuning of the TRSM/TRSV routines | -| [#169](https://github.com/CNugteren/CLBlast/issues/169) | '17 | dividiti | | Problem-specific tuning parameter selection | +| Issue# | When | Who | Status | What | +| ---------------------------------------------------------------|-------------|-----------|--------|---------------| +| - | Oct '17 | CNugteren | ✔ | CUDA API for CLBlast | +| [#169](https://github.com/CNugteren/CLBlast/issues/169) & #195 | Oct-Nov '17 | CNugteren | ✔ | Auto-tuning the kernel selection parameter | +| [#181](https://github.com/CNugteren/CLBlast/issues/181) & #201 | Nov '17 | CNugteren | ✔ | Compilation for Android and testing on a device | +| - | Nov '17 | CNugteren | ✔ | Integration of CLTune for easy testing on Android / fewer dependencies | +| [#128](https://github.com/CNugteren/CLBlast/issues/128) & #205 | Nov-Dec '17 | CNugteren | ✔ | Pre-processor for loop unrolling and array-to-register-promotion for e.g. ARM Mali | +| [#207](https://github.com/CNugteren/CLBlast/issues/207) | Dec '17 | CNugteren | | Tuning of the TRSM/TRSV routines | +| [#195](https://github.com/CNugteren/CLBlast/issues/195) | Jan '18 | CNugteren | | Extra GEMM API with pre-allocated temporary buffer | +| [#224](https://github.com/CNugteren/CLBlast/issues/224) | Jan-Feb '18 | CNugteren | | Implement Hadamard product (element-wise vector-vector product) | +| [#223](https://github.com/CNugteren/CLBlast/issues/223) | Feb '18 | CNugteren | | Python OpenCL interface | +| [#169](https://github.com/CNugteren/CLBlast/issues/169) | ?? | dividiti | | Problem-specific tuning parameter selection | |