nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 4 Packages Projects Releases Wiki Activity

[SYCL] Support OP EXPM1, support all UT cases of FLOOR, TRUNC, ROUND (#24363)

Browse Source 
* support OP EXPM1, support all UT cases of FLOOR, TRUNC, ROUND

* fix conflict

* rebase, support new UT case of repeat, concat
This commit is contained in:
Neo Zhang
2026-06-16 13:34:29 +08:00
committed by GitHub
parent 4196b477da
commit fdd109883d
7 changed files with 101 additions and 96 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/ops.md
View File

@@ -44,10 +44,10 @@ Legend:
| DUP | ❌ | ✅ | ✅ | 🟡 | 🟡 | 🟡 | ✅ | ✅ | ❌ | ❌ | ❌ |
| ELU | ❌ | ✅ | ✅ | 🟡 | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| EXP | ❌ | ✅ | ✅ | 🟡 | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| EXPM1 | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ |
| EXPM1 | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| FILL | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| FLASH_ATTN_EXT | ❌ | 🟡 | ✅ | 🟡 | 🟡 | 🟡 | 🟡 | 🟡 | 🟡 | ❌ | ❌ |
| FLOOR | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | 🟡 | ✅ | ✅ | ❌ | ❌ |
| FLOOR | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| GATED_DELTA_NET | ❌ | ❌ | ✅ | ❌ | 🟡 | ❌ | ✅ | 🟡 | ✅ | ❌ | ❌ |
| GATED_LINEAR_ATTN | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ❌ |
| GEGLU | ❌ | ✅ | ✅ | ✅ | 🟡 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ |
@@ -89,7 +89,7 @@ Legend:
| ROLL | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ |
| ROPE | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ |
| ROPE_BACK | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ |
| ROUND | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | 🟡 | ✅ | ✅ | ❌ | ❌ |
| ROUND | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| RWKV_WKV6 | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ |
| RWKV_WKV7 | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ |
| SCALE | ❌ | 🟡 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ |
@@ -118,6 +118,6 @@ Legend:
| TIMESTEP_EMBEDDING | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ |
| TOP_K | ❌ | ❌ | ✅ | ❌ | ✅ | ❌ | 🟡 | 🟡 | ✅ | ❌ | ❌ |
| TRI | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| TRUNC | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | 🟡 | ✅ | ✅ | ❌ | ❌ |
| TRUNC | ❌ | ❌ | ✅ | 🟡 | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
| UPSCALE | ❌ | 🟡 | ✅ | ✅ | ✅ | 🟡 | ✅ | ✅ | ✅ | ❌ | ❌ |
| XIELU | ❌ | ❌ | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ |

72
docs/ops/SYCL.csv
View File

@@ -27,20 +27,20 @@
"SYCL0","HARDSIGMOID","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","EXP","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","EXP","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[128,2,2,2],v=0","support","0","no","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[5,7,11,13],v=0","support","0","no","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[128,2,2,2],v=0","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[5,7,11,13],v=0","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[128,2,2,2],v=0","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[5,7,11,13],v=0","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[128,2,2,2],v=0","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[5,7,11,13],v=0","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","ABS","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","ABS","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","SGN","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
@@ -69,20 +69,20 @@
"SYCL0","HARDSIGMOID","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","EXP","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","EXP","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","ABS","type=f32,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","ABS","type=f32,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","SGN","type=f32,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
@@ -111,8 +111,8 @@
"SYCL0","HARDSIGMOID","type=f32,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","EXP","type=f32,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","EXP","type=f32,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[128,2,2,2],v=0","support","0","no","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[5,7,11,13],v=0","support","0","no","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f32,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f32,ne_a=[5,7,11,13],v=0","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f32,ne_a=[128,2,2,2],v=0","support","1","yes","SYCL"
@@ -153,20 +153,20 @@
"SYCL0","HARDSIGMOID","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","EXP","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","EXP","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","EXPM1","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","SOFTPLUS","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","GELU_ERF","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f32,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","FLOOR","type=f32,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","FLOOR","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","CEIL","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","CEIL","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","ROUND","type=f32,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","ROUND","type=f32,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","TRUNC","type=f32,ne_a=[128,2,2,2],v=1","support","0","no","SYCL"
"SYCL0","TRUNC","type=f32,ne_a=[5,7,11,13],v=1","support","0","no","SYCL"
"SYCL0","ROUND","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","ROUND","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f32,ne_a=[128,2,2,2],v=1","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f32,ne_a=[5,7,11,13],v=1","support","1","yes","SYCL"
"SYCL0","REGLU","type=f16,ne_a=[128,2,2,2],v=0,swapped=0","support","1","yes","SYCL"
"SYCL0","REGLU","type=f16,ne_a=[5,7,11,13],v=0,swapped=0","support","1","yes","SYCL"
"SYCL0","REGLU","type=f16,ne_a=[128,2,2,2],v=0,swapped=1","support","1","yes","SYCL"
@@ -5105,6 +5105,7 @@
"SYCL0","REPEAT","type=f32,ne=[10,5,4,1],nr=[1,1,1,2]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=i32,ne=[10,5,4,1],nr=[2,1,1,1]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=i16,ne=[10,5,4,1],nr=[1,1,1,2]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=bf16,ne=[10,5,4,1],nr=[2,1,1,1]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=f32,ne=[10,5,4,3],nr=[1,1,1,1]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=f32,ne=[10,5,4,3],nr=[2,1,1,1]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=f32,ne=[10,5,4,3],nr=[1,2,1,1]","support","1","yes","SYCL"
@@ -5112,6 +5113,7 @@
"SYCL0","REPEAT","type=f32,ne=[10,5,4,3],nr=[1,1,1,2]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=i32,ne=[10,5,4,3],nr=[2,1,1,1]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=i16,ne=[10,5,4,3],nr=[1,1,1,2]","support","1","yes","SYCL"
"SYCL0","REPEAT","type=bf16,ne=[10,5,4,3],nr=[2,1,1,1]","support","1","yes","SYCL"
"SYCL0","REPEAT_BACK","type=f32,ne=[8,6,4,2],nr=[1,1,1,1],v=0","support","1","yes","SYCL"
"SYCL0","REPEAT_BACK","type=f32,ne=[8,6,4,2],nr=[2,1,1,1],v=0","support","1","yes","SYCL"
"SYCL0","REPEAT_BACK","type=f32,ne=[8,6,4,2],nr=[1,2,1,1],v=0","support","1","yes","SYCL"
@@ -9748,10 +9750,10 @@
"SYCL0","COS","type=f16,ne=[10,2,2,2]","support","0","no","SYCL"
"SYCL0","CLAMP","type=f16,ne=[10,5,4,3],min=-0.500000,max=0.500000","support","0","no","SYCL"
"SYCL0","LEAKY_RELU","type=f16,ne_a=[10,5,4,3],negative_slope=0.100000","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne=[10,2,2,2]","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne=[10,2,2,2]","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne=[10,2,2,2]","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne=[10,2,2,2]","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne=[10,2,2,2]","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne=[10,2,2,2]","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne=[10,2,2,2]","support","1","yes","SYCL"
"SYCL0","SQR","type=f16,ne=[7,1,5,3]","support","0","no","SYCL"
"SYCL0","SQR","type=f16,ne=[1024,1024,1,1]","support","0","no","SYCL"
"SYCL0","SQRT","type=f16,ne=[7,1,5,3]","support","0","no","SYCL"
@@ -9766,14 +9768,14 @@
"SYCL0","CLAMP","type=f16,ne=[1024,1024,1,1],min=-0.500000,max=0.500000","support","0","no","SYCL"
"SYCL0","LEAKY_RELU","type=f16,ne_a=[7,1,5,3],negative_slope=0.100000","support","1","yes","SYCL"
"SYCL0","LEAKY_RELU","type=f16,ne_a=[1024,1024,1,1],negative_slope=0.100000","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne=[7,1,5,3]","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne=[1024,1024,1,1]","support","0","no","SYCL"
"SYCL0","FLOOR","type=f16,ne=[7,1,5,3]","support","1","yes","SYCL"
"SYCL0","FLOOR","type=f16,ne=[1024,1024,1,1]","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne=[7,1,5,3]","support","1","yes","SYCL"
"SYCL0","CEIL","type=f16,ne=[1024,1024,1,1]","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne=[7,1,5,3]","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne=[1024,1024,1,1]","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne=[7,1,5,3]","support","0","no","SYCL"
"SYCL0","TRUNC","type=f16,ne=[1024,1024,1,1]","support","0","no","SYCL"
"SYCL0","ROUND","type=f16,ne=[7,1,5,3]","support","1","yes","SYCL"
"SYCL0","ROUND","type=f16,ne=[1024,1024,1,1]","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne=[7,1,5,3]","support","1","yes","SYCL"
"SYCL0","TRUNC","type=f16,ne=[1024,1024,1,1]","support","1","yes","SYCL"
"SYCL0","SQR","type=f32,ne=[10,5,4,3]","support","1","yes","SYCL"
"SYCL0","SQRT","type=f32,ne=[10,3,3,2]","support","1","yes","SYCL"
"SYCL0","LOG","type=f32,ne=[10,5,4,3]","support","1","yes","SYCL"
Can't render this file because it is too large.