Commits · 19ebf40ddb2ded63aa8f2b76d0617ed554189df3 · gaoqiong / MIGraphX

26 May, 2023 25 commits
- Format · 19ebf40d
  Paul authored May 26, 2023
  
  19ebf40d
- Mark the problems · f20af942
  Paul authored May 26, 2023
  
  f20af942
- Format · 7099811c
  Paul authored May 26, 2023
  
  7099811c
- Format · 3d8c71fa
  Paul authored May 26, 2023
  
  3d8c71fa
- Add problem_cache to avoid retuning same configs · bfaec6b4
  Paul authored May 26, 2023
  
  bfaec6b4
- Format · a4e2ee8f
  Paul authored May 26, 2023
  
  a4e2ee8f
- Move data type to is_ck_gemm matcher · 8d02a27a
  Paul authored May 26, 2023
  
  8d02a27a
- Pass tuning parameter · d30df5bb
  Paul authored May 26, 2023
  
  d30df5bb
- Format · 744f236c
  Paul authored May 26, 2023
  
  744f236c
- Generate configs · 5c2a9f2e
  Paul authored May 26, 2023
  
  5c2a9f2e
- Format · 059fa4cc
  Paul authored May 26, 2023
  
  059fa4cc
- Use device name · c027803e
  Paul authored May 26, 2023
  
  c027803e
- Add lift class macro · ee391f22
  Paul authored May 26, 2023
  
  ee391f22
- Format · 9b3e4576
  Paul authored May 26, 2023
  
  9b3e4576
- Refactor · 74f21ca6
  Paul authored May 26, 2023
  
  74f21ca6
- Add banchmarking · 674b3bac
  Paul authored May 26, 2023
  
  674b3bac
- Move time_op to migraphx_gpu · 52223d46
  Paul authored May 26, 2023
  
  52223d46
- Add some comments · da29ed61
  Paul authored May 26, 2023
  
  da29ed61
- Format · acbb2e88
  Paul authored May 26, 2023
  
  acbb2e88
- Update compile_ops to handle tuning config · 3a58dbf9
  Paul authored May 26, 2023
  
  3a58dbf9
- Format · 247ce1d2
  Paul authored May 26, 2023
  
  247ce1d2
- Add interface to enable tuning · 17ee5428
  Paul authored May 26, 2023
  
  17ee5428
- Add unpack array · 2956bb3f
  Paul authored May 25, 2023
  
  2956bb3f
- Format · 3262208c
  Paul authored May 25, 2023
  
  3262208c
- Disable warnings for ck headers · 6b0dfe0f
  Paul authored May 25, 2023
  
  6b0dfe0f
25 May, 2023 13 commits
- Add group stride · 891be203
  Paul authored May 25, 2023
  
  891be203
- Fix include · b8b2e8ab
  Paul authored May 25, 2023
  
  b8b2e8ab
- Update ck version · e734215e
  Paul authored May 25, 2023
  
  e734215e
- Enable pass · e0724664
  Paul authored May 25, 2023
  
  e0724664
- Rename var · 0ee0d6e7
  Paul authored May 25, 2023
  
  0ee0d6e7
- Format · 1f352d61
  Paul authored May 25, 2023
  
  1f352d61
- Simplify fuse_ck · b621b28f
  Paul authored May 25, 2023
  
  b621b28f
- Add include · fee874e3
  Paul authored May 25, 2023
  
  fee874e3
- Format · b256a3b0
  Paul authored May 25, 2023
  
  b256a3b0
- Add additional src files · 96e062d1
  Paul authored May 25, 2023
  
  96e062d1
- Format · 36154263
  Paul authored May 25, 2023
  
  36154263
- Add initial ck_gemm code · af0d45a5
  Paul authored May 25, 2023
  
  af0d45a5
- Update cpp generator to handle inf from float (#1758) · 763dd1da
  Ted Themistokleous authored May 25, 2023
```
Use std::numeric_limits::min/max() functions plus the appropriate value to encode -inf/inf 
```
  763dd1da
24 May, 2023 2 commits
- Change compiler_replace to a class that stores the code objects directly (#1739) · 37f5df20
  Paul Fultz II authored May 24, 2023
```
Enable retrieving the code object to do tuning in the future.
```
  37f5df20
- Update xdlops/rocblas fp32 arch (#1752) · 77042e30
  kahmed10 authored May 24, 2023
```
Refactor supported gfx archs
```
  77042e30