[AMD] Update aiter to v0.1.10.post2 (#18423)

Co-authored-by: kkHuang-amd <wunhuang@amd.com>
Co-authored-by: YC Tseng <yctseng@amd.com>

test/registered/attention/test_triton_attention_backend.py

@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
 
 # Triton attention backend integration test with latency benchmark and MMLU eval
 register_cuda_ci(est_time=200, suite="stage-b-test-large-1-gpu")
-register_amd_ci(est_time=1110, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=1400, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestTritonAttnBackend(CustomTestCase):

test/registered/dllm/test_llada2_mini_amd.py

@@ -20,7 +20,7 @@ from sglang.test.test_utils import (
     write_github_step_summary,
 )
 
-register_amd_ci(est_time=520, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=1000, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestLLaDA2MiniAMD(CustomTestCase):

test/registered/eval/test_eval_accuracy_large.py

@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
 )
 
 register_cuda_ci(est_time=300, suite="stage-b-test-small-1-gpu")
-register_amd_ci(est_time=300, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=420, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestEvalAccuracyLarge(CustomTestCase):

test/registered/mla/test_mla_fp8.py

@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
 
 # MLA FP8 KV cache test with MGSM evaluation
 register_cuda_ci(est_time=77, suite="stage-b-test-large-1-gpu")
-register_amd_ci(est_time=360, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=800, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestMLA(CustomTestCase):

test/registered/models/test_vlm_models.py

@@ -17,7 +17,7 @@ from sglang.test.test_utils import is_in_ci
 
 
 register_cuda_ci(est_time=228, suite="stage-b-test-large-1-gpu")
-register_amd_ci(est_time=420, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=850, suite="stage-b-test-small-1-gpu-amd")
 
 _is_hip = is_hip()
 # VLM models for testing

test/registered/perf/test_bench_serving_2gpu.py

@@ -15,7 +15,7 @@ from sglang.test.test_utils import (
 )
 
 register_cuda_ci(est_time=600, suite="stage-b-test-large-2-gpu")
-register_amd_ci(est_time=600, suite="stage-b-test-large-2-gpu-amd")
+register_amd_ci(est_time=1100, suite="stage-b-test-large-2-gpu-amd")
 
 
 class TestBenchServing2GPU(CustomTestCase):

test/registered/quant/test_eval_fp8_accuracy.py

@@ -15,7 +15,7 @@ from sglang.test.test_utils import (
 )
 
 register_cuda_ci(est_time=250, suite="stage-b-test-large-1-gpu")
-register_amd_ci(est_time=303, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=600, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestEvalFP8Accuracy(CustomTestCase):

test/registered/quant/test_torchao.py

@@ -7,7 +7,7 @@ from sglang import Engine
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
 register_cuda_ci(est_time=103, suite="stage-b-test-small-1-gpu")
-register_amd_ci(est_time=106, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=230, suite="stage-b-test-small-1-gpu-amd")
 from sglang.lang.chat_template import get_chat_template_by_model_path
 from sglang.srt.utils import kill_process_tree
 from sglang.test.run_eval import run_eval

test/registered/scheduler/test_abort.py

@@ -18,7 +18,7 @@ from sglang.test.test_utils import (
 )
 
 register_cuda_ci(est_time=131, suite="stage-b-test-small-1-gpu")
-register_amd_ci(est_time=51, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=300, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestAbort(CustomTestCase):

test/registered/scheduler/test_retract_decode.py

@@ -18,7 +18,7 @@ from sglang.test.test_utils import (
 from sglang.utils import is_in_ci
 
 register_cuda_ci(est_time=311, suite="stage-b-test-small-1-gpu")
-register_amd_ci(est_time=450, suite="stage-b-test-small-1-gpu-amd")
+register_amd_ci(est_time=600, suite="stage-b-test-small-1-gpu-amd")
 
 
 class TestRetractDecode(CustomTestCase):