chenyu 4ecd5789ab #include <tgmath.h> in ops_clang (#3927) ...

* different clang sqrt/log2/exp2/sin function based on dtype

fixed softmax_argmax issue in #3552 for clang.

* tgmath.h

* revert those

2024-03-25 17:48:57 -04:00

..

autogen

don't include hip common (#3851)

2024-03-22 08:50:50 -07:00

driver

clean up hsa driver (#3818)

2024-03-20 00:17:41 +03:00

graph

init multidevice cuda graph (#3858)

2024-03-22 13:49:48 -07:00

ops_clang.py

#include <tgmath.h> in ops_clang (#3927)

2024-03-25 17:48:57 -04:00

ops_cuda.py

lru allocator for copyin host buffers (#3918)

2024-03-25 15:57:18 +03:00

ops_disk.py

multioutput ScheduleItem (#3699)

2024-03-13 08:59:38 -07:00

ops_ext.py

ops_ext to replace cpu import (#3409)

2024-02-15 13:03:42 +01:00

ops_gpu.py

delete half_prekernel (#3388)

2024-02-14 15:40:48 +01:00

ops_hsa.py

nv ioctl sniffer (#3892)

2024-03-23 00:29:30 -07:00

ops_llvm.py

faster program start in llvm (#3897)

2024-03-23 15:20:15 +03:00

ops_metal.py

kernel: limit shared memory usage when adding opts (#3705)

2024-03-12 17:06:21 -04:00

ops_python.py

move lines around in ops_python wmma (#3911)

2024-03-24 17:14:26 -04:00

ops_rhip.py

remove hip backend (#3783)

2024-03-17 10:12:16 -07:00