From d553c0a257ada6d5f5835eb58202f6bf367ac631 Mon Sep 17 00:00:00 2001 From: Chien-Chun Hung <2679986+chienchunhung@users.noreply.github.com> Date: Mon, 27 Jul 2026 16:48:54 -0700 Subject: [PATCH 1/3] [https://nvbugs/6487038][fix] extend GB300 Kimi KV transfer timeout Signed-off-by: Chien-Chun Hung <2679986+chienchunhung@users.noreply.github.com> --- ...4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml b/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml index 7bbe295f8371..12b245ec091c 100644 --- a/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml +++ b/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml @@ -64,6 +64,7 @@ worker_config: max_tokens_in_buffer: 16384 backend: NIXL transceiver_runtime: PYTHON + kv_transfer_timeout_ms: 600000 disable_overlap_scheduler: true trust_remote_code: true num_postprocess_workers: 4 @@ -90,5 +91,6 @@ worker_config: max_tokens_in_buffer: 16384 backend: NIXL transceiver_runtime: PYTHON + kv_transfer_timeout_ms: 600000 disable_overlap_scheduler: true trust_remote_code: true From 47e01e12d0c69c20596a6e93bd551942d0c347b8 Mon Sep 17 00:00:00 2001 From: Chien-Chun Hung <2679986+chienchunhung@users.noreply.github.com> Date: Mon, 27 Jul 2026 16:56:31 -0700 Subject: [PATCH 2/3] [https://nvbugs/6487038][test] unwaive GB300 Kimi KV timeout cases Signed-off-by: Chien-Chun Hung <2679986+chienchunhung@users.noreply.github.com> --- tests/integration/test_lists/waives.txt | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/integration/test_lists/waives.txt b/tests/integration/test_lists/waives.txt index 71b5595714c3..d4e3787f49aa 100644 --- a/tests/integration/test_lists/waives.txt +++ b/tests/integration/test_lists/waives.txt @@ -360,7 +360,6 @@ perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_deepseek-r1-fp4_128k8 perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_deepseek-v32-fp4_32k4k_con2048_ctx1_dep4_gen1_dep32_eplb288_mtp1_ccb-NIXL] SKIP (https://nvbugs/6550133) perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_deepseek-v32-fp4_8k1k_con4096_ctx1_dep4_gen1_dep32_eplb256_mtp0_ccb-NIXL] SKIP (https://nvbugs/6550133) perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb300_glm-5-fp4_8k1k_con1024_ctx1_dep2_gen1_dep8_eplb256_mtp1_ccb-NIXL] SKIP (https://nvbugs/6566777) -perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6490049) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-b200_deepseek-r1-fp4_8k1k_con1_ctx1_dep4_gen1_tep8_eplb0_mtp3_ccb-NIXL] SKIP (https://nvbugs/6561566) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-b200_deepseek-r1-fp4_8k1k_con256_ctx1_dep4_gen1_dep8_eplb0_mtp1_ccb-NIXL] SKIP (https://nvbugs/6561566) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_8k1k_con1024_ctx1_dep4_gen1_dep32_eplb0_mtp3_ccb-NIXL] SKIP (https://nvbugs/6561566) From ab6ef4b61b076ae4d6090dad701fbcee01e46b65 Mon Sep 17 00:00:00 2001 From: Chien-Chun Hung <2679986+chienchunhung@users.noreply.github.com> Date: Thu, 30 Jul 2026 15:16:48 -0700 Subject: [PATCH 3/3] [https://nvbugs/6487038][test] validate GB300 Kimi with default KV timeout Signed-off-by: Chien-Chun Hung <2679986+chienchunhung@users.noreply.github.com> --- ...4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml | 2 -- 1 file changed, 2 deletions(-) diff --git a/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml b/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml index 12b245ec091c..7bbe295f8371 100644 --- a/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml +++ b/tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL.yaml @@ -64,7 +64,6 @@ worker_config: max_tokens_in_buffer: 16384 backend: NIXL transceiver_runtime: PYTHON - kv_transfer_timeout_ms: 600000 disable_overlap_scheduler: true trust_remote_code: true num_postprocess_workers: 4 @@ -91,6 +90,5 @@ worker_config: max_tokens_in_buffer: 16384 backend: NIXL transceiver_runtime: PYTHON - kv_transfer_timeout_ms: 600000 disable_overlap_scheduler: true trust_remote_code: true