From 04e1596054aafcec86aee32e96acfb8c4a28d1f0 Mon Sep 17 00:00:00 2001 From: chenfeiz0326 Date: Fri, 10 Jul 2026 06:20:15 +0000 Subject: [PATCH] [https://nvbugs/6422339][test] Unwaive 3 disagg-perf-sanity GB200 tests Remove waives for the 3 disagg_upload-gen_only-* GB200 tests that failed in L0_PostMerge #2824 with the "ctx/gen never become ready" hang: - disagg_upload-gen_only-gb200_deepseek-r1-fp4_128k8k_con64_ctx1_pp8_gen1_dep32_eplb0_mtp3_ccb-NIXL - disagg_upload-gen_only-gb200_deepseek-r1-fp4_1k1k_con1024_ctx1_dep4_gen1_dep8_eplb0_mtp0_ccb-NIXL - disagg_upload-gen_only-gb200_deepseek-v32-fp4_8k1k_con4096_ctx1_dep4_gen1_dep32_eplb256_mtp0_ccb-NIXL Root cause was PR #15654 (IPC HMAC key via file descriptor, commit 48fc7537ba, merged 2026-07-03) re-landing the same fd-passing scheme that had already been reverted once. Under the disagg-perf-sanity launch chain (bash trtllm-llmapi-launch -> pytest -> subprocess.Popen(trtllm-serve, close_fds=True)), the numeric fd advertised via TLLM_SPAWN_PROXY_PROCESS_IPC_HMAC_KEY_FD is not preserved, so os.read(fd) blocks forever instead of raising -- CTX/GEN workers silently hang after "TensorRT LLM inited" and the disagg server's /health polls never succeed, timing out at pytest's 3600s outer timeout. Details in nvbugs/6435109. PR #15961 (commit 0d97e9c76f, merged 2026-07-06 22:22 CST) reverted #15654. Build 2824's commit 4c5f3b06d2 (2026-07-06 13:16 CST) sits in the 3.5-day re-land window; every post-merge build after the revert should no longer hit this. This PR re-enables the three affected tests so that verification actually happens. --- tests/integration/test_lists/waives.txt | 3 --- 1 file changed, 3 deletions(-) diff --git a/tests/integration/test_lists/waives.txt b/tests/integration/test_lists/waives.txt index 008aeef59266..f5995643f96e 100644 --- a/tests/integration/test_lists/waives.txt +++ b/tests/integration/test_lists/waives.txt @@ -375,9 +375,7 @@ perf/test_perf_sanity.py::test_e2e[aggr_upload-k25_thinking_fp4_2_nodes_grace_bl perf/test_perf_sanity.py::test_e2e[aggr_upload-super_ad_blackwell-super_ad_ws1_1k1k] SKIP (https://nvbugs/6153575) perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_deepseek-r1-fp4_128k8k_con128_ctx1_pp8_gen1_dep16_eplb0_mtp1_ccb-NIXL] SKIP (https://nvbugs/6426890) perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_qwen3-235b-fp4_8k1k_con1024_ctx1_tp1_gen1_dep8_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6422339) -perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_128k8k_con64_ctx1_pp8_gen1_dep32_eplb0_mtp3_ccb-NIXL] SKIP (https://nvbugs/6422339) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_1k1k_con1024_ctx1_dep4_gen1_dep32_eplb0_mtp3_ccb-NIXL] SKIP (https://nvbugs/6422339) -perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_1k1k_con1024_ctx1_dep4_gen1_dep8_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6422339) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_1k1k_con2048_ctx2_dep4_gen1_dep16_eplb0_mtp3_ccb-NIXL] SKIP (https://nvbugs/6422339) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_1k1k_con2048_ctx2_dep4_gen1_dep16_eplb288_mtp3_ccb-NIXL] SKIP (https://nvbugs/6422339) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_1k1k_con3072_ctx1_dep4_gen1_dep4_eplb0_mtp1_ccb-NIXL] SKIP (https://nvbugs/6422339) @@ -385,7 +383,6 @@ perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-v32-fp4 perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-v32-fp4_1k1k_con2048_ctx1_dep4_gen1_dep4_eplb0_mtp1_ccb-NIXL] SKIP (https://nvbugs/6426890) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-v32-fp4_32k4k_con2048_ctx1_dep4_gen1_dep32_eplb288_mtp1_ccb-NIXL] SKIP (https://nvbugs/6374872) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-v32-fp4_32k4k_con256_ctx1_dep8_gen1_dep8_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6374893) -perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-v32-fp4_8k1k_con4096_ctx1_dep4_gen1_dep32_eplb256_mtp0_ccb-NIXL] SKIP (https://nvbugs/6422339) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_kimi-k25-thinking-fp4_1k1k_con4096_ctx1_dep4_gen1_dep8_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6323074) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_kimi-k25-thinking-fp4_8k1k_con1024_ctx1_dep4_gen1_dep32_eplb416_mtp3_ccb-NIXL] SKIP (https://nvbugs/6379406) perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_qwen3-235b-fp4_8k1k_con64_ctx1_tp1_gen1_tep4_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6418834)