Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2024-07-16 | Don't install private softfloat headers | Andrew Waterman | 1 | -5/+0 | |
2024-06-21 | Merge pull request #1704 from riscv-software-src/thread-local-again | Andrew Waterman | 1 | -2/+1 | |
Fix C/C++ thread-local linkage differently | |||||
2024-06-21 | Fix C/C++ thread-local linkage differently | Andrew Waterman | 1 | -2/+1 | |
Just admit Mac OS is broken, so explicitly special-case it. See #1689 and #1703 | |||||
2024-06-18 | Add several BF16 ops to SoftFloat | Andrew Waterman | 10 | -0/+435 | |
2024-06-17 | Consistently order BF16 routines in Makefile and softfloat.h | Andrew Waterman | 2 | -4/+8 | |
2024-06-13 | Make softfloat's rounding mode thread-local | Andrew Waterman | 1 | -2/+5 | |
This has no effect on Spike itself, but it might matter for anyone who's using Spike as a library in a multithreaded program. | |||||
2023-05-29 | Add convertion function between binary float16 and float32 in softfloat | Weiwei Li | 8 | -0/+316 | |
2022-12-22 | Remove non-installed header lists from Makefile fragments | Andrew Waterman | 1 | -9/+0 | |
It's just dead code. (Dependences on headers are auto-generated as `.d` files.) | |||||
2022-10-04 | Fix unused-variable warnings in softfloat | Andrew Waterman | 1 | -4/+4 | |
2022-10-04 | Suppress unused-paramter warnings in softfloat | Andrew Waterman | 1 | -0/+6 | |
2022-09-23 | Build and install lib and header dependencies for Hammer | Jerin Joy | 1 | -1/+8 | |
https://github.com/rivosinc/hammer Signed-off-by: Jerin Joy <joy@rivosinc.com> | |||||
2021-07-28 | Significantly speed up compilation with GCC | Andrew Waterman | 1 | -2/+0 | |
Precompiled headers were broken because they weren't compiled with the same -fPIC setting as the rest of the code. Fix by just making everything use -fPIC. | |||||
2020-08-31 | softfloat: add reciprocal api | Chih-Min Chao | 3 | -0/+399 | |
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> | |||||
2020-05-28 | sf: add f16_to_[u]i8 APIs | Chih-Min Chao | 9 | -6/+136 | |
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> | |||||
2020-05-04 | sf: fp16: add missing APIs | Han-Kuan Chen | 8 | -0/+266 | |
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> | |||||
2020-03-23 | rvv: sf: handle signaling NaN for fmax/fmin | Chih-Min Chao | 1 | -10/+10 | |
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> | |||||
2020-02-28 | Enable SOFTFLOAT_ROUND_ODD for vfncvt.rod.f.f.w | Andrew Waterman | 1 | -0/+1 | |
2019-10-28 | Implement support for big-endian hosts | Marcus Comstedt | 1 | -0/+3 | |
2019-06-18 | rvv: extend softfloat to support 16-64 min/max function | Chih-Min Chao | 3 | -0/+88 | |
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> | |||||
2019-06-07 | build: build softfloat with pic and provide shared library | Chih-Min Chao | 1 | -0/+4 | |
vector unit tests need identical floating library to generate golden pattern Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> | |||||
2018-03-06 | Fix a missed header file in the softfloat include install. | Prashanth Mundkur | 1 | -0/+1 | |
2017-09-24 | Update SoftFloat | Andrew Waterman | 218 | -1068/+11195 | |
2016-04-02 | Allow configuration of default ISA with --with-isa | Andrew Waterman | 1 | -0/+0 | |
2016-03-02 | Mark SoftFloat routines static inline | Andrew Waterman | 1 | -1/+1 | |
This avoids duplicate definitions on some platforms. | |||||
2016-03-01 | Upgrade to latest SoftFloat | Andrew Waterman | 154 | -5124/+9789 | |
2016-01-29 | Fix NaN propagation for fcvt | Andrew Waterman | 3 | -4/+3 | |
2016-01-12 | Use new NaN discipline | Andrew Waterman | 5 | -3/+7 | |
2015-02-23 | Softfloat fcvt.{w/l}.d now returns -MaxInt for -NaN | Andrew Waterman | 2 | -3/+0 | |
2015-02-22 | Softfloat fcvt.{w/h}.s now returns -MaxInt for -NaN | Andrew Waterman | 2 | -2/+1 | |
Behavior now consistent with RISC-V user spec. | |||||
2014-03-06 | Add fclass.{s|d} instructions | Andrew Waterman | 4 | -0/+70 | |
2014-01-25 | Merge softfloat_riscv into softfloat | Andrew Waterman | 13 | -1/+397 | |
They really aren't independent libraries. | |||||
2012-03-19 | make NaN behavior consistent with hardfloat | Andrew Waterman | 1 | -1/+1 | |
2011-06-19 | temporary undoing of renaming | Andrew Waterman | 119 | -0/+6271 | |
2011-06-12 | [sim] renamed to riscv-isa-run | Andrew Waterman | 119 | -6271/+0 | |
2010-10-15 | [sim] made softfloat files C instead of C++ | Andrew Waterman | 102 | -101/+109 | |
2010-08-18 | [sim] integrated SoftFloat-3 with ISA sim; removed SoftFloat-2b | Andrew Waterman | 138 | -10300/+335 | |
2010-08-17 | [sim] specialized softfloat for riscv | Andrew Waterman | 5 | -8/+7 | |
2010-08-17 | [sim] added riscv folder to softfloat | Andrew Waterman | 12 | -0/+442 | |
2010-08-17 | [sim] added SoftFloat-3 source | Andrew Waterman | 118 | -0/+9111 | |
2010-08-09 | [xcc,sim] implement FP using softfloat | Andrew Waterman | 5 | -253/+268 | |
The intersection of the Hauser FP and MIPS FP is implemented. | |||||
2010-08-09 | [sim] added softfloat | Andrew Waterman | 6 | -0/+6661 | |