.gitignore
.gitmodules
.pre-commit-config.yaml
MANIFEST.in
README.md
pyproject.toml
setup.py
uv.lock
.github/workflows/build_wheels.yml
.github/workflows/docs.yml
.github/workflows/publish_pypi.yml
docs/README.md
docs/conf.py
docs/index.md
docs/installation.md
docs/optimization-runs.md
docs/quick-start.md
docs/requirements.in
docs/requirements.txt
docs/_hosting/.nojekyll
docs/_hosting/CNAME
docs/_hosting/README.md
docs/_hosting/index.html
docs/_static/agent-loop.svg
docs/_static/custom.css
docs/components/kcoral.md
docs/components/tirx-lite.md
docs/components/tools.md
docs/development/add-workload.md
docs/development/build-the-docs.md
docs/development/contribute-kernel.md
docs/development/report-and-fix-bugs.md
evolution/README.md
evolution/__init__.py
evolution/setup.py
evolution/benchmark/__init__.py
evolution/benchmark/adapter.py
evolution/benchmark/flashinfer_bench_evolve/__init__.py
evolution/benchmark/flashinfer_bench_evolve/benchmark_common.py
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_02d6ae9c64ab42ff93f05c23c53bcb7d.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_05f6de657db543ae9e4c46796522843a.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_0a63b87bb2e54e9db1ca3c4c53a1d521.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_0c23b10c7b7645719517828c12eaa1d2.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_2207f0fdc96347c59d106e4976cfad57.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_232ed014bafc4835b9881bb308c659b0.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_3838996164a94d728710f913477feba8.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_385742b2717e4f02b918c7349dde23d8.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_4c46a94ba2364dc7ab476286dee8dce3.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_5096e459ce3f4cdf82773ce1a0c73c8a.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_564007ac354e4662a62cc4d6352dc494.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_68d6817dcfd1433aa8d2ddeefd54b6ea.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_78b2e11c30004cceb84355722a7c6b0a.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_7a389715dcc3479e9b0512309f9d1d56.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_9d4a5f21268e484ea05a2f2af91d9fa7.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_9f3f891bfbe24776adfcd5a579d093dd.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_ae4219a95f044f45bd10e17ab63c6e8f.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_b7668cfd194c4b95ab600feb205ebac6.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_d57eb9e19f0642e8af9bd76ad0823303.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_ddfa9e340b264f76abe7418692faa876.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_e6b849f2900446148c01d81152efae23.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_f77df5ce21634f7ca0586af7e3c13434.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/dsa_attention/dsa_sparse_attention_h16_ckv512_kpe64_topk2048_ps64_fc85411e250c41879b6d6a1edb80f0a7.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_02c1e5f0-0d21-4c3c-a22d-40ad45a93549_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_043e74e4-0297-4988-808a-0090be48a90e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_056224b8-0a27-4634-b93b-36beb658e135_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_06f21bb1-6cbd-4d55-b620-fb4d62181a71_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_078a41ea-3a59-4b4f-b3c2-95cb6e538a6e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_07aa7922-1848-48a9-830a-54216b5553b3_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_08d4f2c4-dd07-4cc0-9d93-975c91b17577_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_109addb1-15e0-4ff2-9b39-df3e79746af0_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_15856e8c-d105-478d-b0de-4ef8141b5a77_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_19c6ab20-df55-4b2c-b9a7-d76870b50e3e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_1aa8cf18-0fca-4e32-a7ff-7ed29d24a3d2_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_1b441950-6f77-4565-bada-12294714fb26_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_1cf8e175-9f87-4add-984a-6c87a380fe9d_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_1d0cc342-9e3d-41e3-9ade-532551f86c17_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_1efaf2a9-05db-4737-8f41-6880ba1bb487_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_25d9c14d-90ad-442d-8b0f-9452ad198832_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_26244fb4-d295-4d4b-bf61-235d17fd0149_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_2683c087-e726-4a48-af25-674a95a889f2_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_27f44fd6-0b0d-44bc-a771-617fd3e9e61c_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_28b70283-08fd-4eb9-8831-147abbf8d0f2_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_2ba465c0-7405-4bf1-ba4e-7d632d95f141_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_2c9693b4-6375-459a-a29f-650c24abe472_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_3215fe5f-4a3b-4eb6-af20-4e17368d87a9_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_339a7ff4-0a55-40d0-99d1-f33e3b1ffe4a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_33a38713-868b-43f7-8d75-fbe472e30387_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_352c9ace-2650-481f-ab51-69394fd0f366_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_35ea9bbe-0189-4089-878c-4fc21461297e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_3a77dfec-f1c1-40aa-82e4-fafdfc84e540_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_410794d4-6f70-4bb4-aed1-4669be9a610f_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_43bf9699-adb3-4161-b741-8ff07b553336_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_49ef89d2-da00-4cd5-b912-708cbfc7106d_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_4b6143dd-0e5f-499f-93cb-076d9635bcd0_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_4b94d568-35ce-45a5-91eb-f8dc4b7077a7_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_54856fec-3b50-4866-815d-19bfa16cf58b_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_5835a2bc-8d60-43fc-b1ed-d4729ea62693_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_5a91aa02-8cee-49f9-aa2b-145d6e254ee4_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_5b8a0e4b-5ed7-45d0-aec3-c2c154e515b1_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_5d26ac5b-3c04-409d-a80d-5f1ac455916c_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_5d3fc66a-6bba-4602-8319-a1da07045354_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_618df04a-540c-42fd-b2ce-285dfdf78422_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_62447caf-853a-4bf8-a62e-d12885248881_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_685d26ff-9ad1-4378-b15a-b279d2db1bcb_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_6f1ad833-3d8a-4613-b5bd-a5d69263db1d_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_6fbc155c-fcd0-4423-b1bf-970eef91d99b_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_73b8cc85-11ec-4ac4-8469-fce7b52ee44a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_77daf91d-0660-4c4b-8c32-336a69281cd9_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_7a7deca8-255f-4400-9fc6-367d4671ddcc_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_7ba9d519-7c0b-4eac-b2ad-ae403dce2f95_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_85d7becb-2969-4cc8-9120-1ec26a78d258_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_87bff084-1b7a-478e-99fd-5952c989d80e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_8e7ef744-5f4e-40e7-aae3-471e013657bd_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_9343fd82-a06d-493a-9918-1044b0c1cbd1_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_977d19f8-5d0f-4c19-91e0-b05b500b34fc_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_9a5d694b-7d4c-4ee6-8315-a13053ab6f92_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_9c1ef562-578a-4df4-986e-e0c1aaa4727a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_a01a3f93-19b0-4b12-95e3-6132a41d628e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_a0eb2dc2-9de3-47fc-8ab4-0beee6150015_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_a2629e02-611f-4264-8ee7-5c431f5fee4f_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_a39aa135-dd03-4abb-862d-020cf52fc7dd_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_a87ded8a-8202-405c-8ced-17521a8c8379_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_a9540651-862b-44af-ad75-57e2d9caf37a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_aaa378be-7365-4381-b2d3-35951ef88c7a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_b8c8dc3c-3dfe-4334-a784-c477213e2f1a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_ba08a83e-e151-4e16-bc70-abee6851604c_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_ba95d412-06d5-47d9-99b8-185146c2f869_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_bfd8f7b6-3953-46c3-af04-409b2fbe353c_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_c2931c92-bbe9-43d9-bc55-9aed77901266_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_c2bb4f66-6955-4e0f-b223-b802c1cd2348_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_c358edcd-3fda-43bb-98a2-ac32da9038b6_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_c5257f65-c411-4dbb-9dc1-ad4abcf00254_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_c7846f96-c9ff-44d3-94ab-5eab99a1431b_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_cc241d2e-f8b7-448f-95ec-1d5d2809c068_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_cc310f94-60ea-4dcd-9e63-135dea91b83a_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_cd979341-7112-4887-a2f3-dfb269a49bbc_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_ce832e76-c4c2-421f-ad75-ebda2032c401_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d0ce7b5d-49e2-4a0b-b2ce-a087139b7d6b_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d2b5a221-f890-4cdd-83bc-bdbff1f0d0b0_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d3dc3577-8a65-4937-a63a-4aa7204181a7_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d49df0b2-7838-4865-9a51-7ec7877fe27f_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d5aa60dc-f159-4d02-a1d4-f68bd2a62f33_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d5f5c00c-c159-45c8-aa2a-2bb8a00e2fe8_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_d8f4a9ae-d391-4f70-8069-f5deb510a2d1_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_e286a4f4-4a93-43bd-9465-b2915f0f8f96_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_e44ba4d3-b2d7-4a2a-b860-0fcbdda6c351_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_e92dafeb-3d50-4bce-90c3-6d0172e189d3_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_e9e1e445-b41a-404b-a20f-bc869acc6cf0_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_ea27be17-3dfa-423e-aa85-e9001c24daed_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_eaa0fd47-ac57-47d0-a896-e3b1a914a5da_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_ed66c791-0393-45d5-946b-4cd5d3e8b395_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_ef9515b6-ad88-4a3e-bd89-31384ddd53ad_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f105eda8-d047-4624-90a2-f5d5039e2f4e_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f203fdcd-6b72-4687-9fcf-a010066eab5c_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f2f01c2c-3327-4cae-a0d6-a12ec0e94fc3_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f3d30cb9-65d2-452a-92ef-10fb8b78473d_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f4926229-e4ba-49a3-8b2f-df2ec79c333b_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f5619793-6267-4a96-8a23-6c6a937183e7_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_f622a11d-8b79-40b6-9949-5509afe827c8_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_fc7a2bcb-25d4-4840-a095-959499223267_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_fd072ba6-2190-4ce6-b96c-5212d4caf6a0_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/gdn_prefill/gdn_prefill_qk4_v8_d128_k_last_fdf5f1f4-1135-4d3f-99e4-8965958317fc_cu_seqlens.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_13f420b30bbe43548bf54556a0185219.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_25ff432053b5474d86dda63b7daf1734.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_547d24f37f554e2fab107fb57a41e73e.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_5b074f60cf5d449db9375e6ff0fa89b6.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_5d8ecbc401024d0697fec81a0dab7163.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_62adbb89946c4512881ab23d3819f740.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_6c6354ff369c41f3b577e01b70a015a7.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_886eb15cbc2f49ad8bd3d4244c38ed91.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_8d2c57225e994c41aca633b282f92c78.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_90235a13a92d4cb0b552a5d13b119b2e.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_96384077a7da46bcb65b7c694586960d.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_963f2f970f2e4f1688595246c239372e.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_a071bbc5a57648f6ac1346515429d92d.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_ad0adeae69b443b192dc6c7394302345.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_b7a56d54b53740e1b95e08d5b31977f0.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_bcb1fd5c45f84a46b6e094fe7a988890.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_c0d3fb257396413c85038b179a27d3a1.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_d517fa44dc8245ef9053f097b6b0cb76.safetensors
evolution/benchmark/flashinfer_bench_evolve/blob/moe/moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048_e82e1e6107d24eb9ba26b15acd580832.safetensors
evolution/benchmark/flashinfer_bench_evolve/tasks/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/alphamoe/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/alphamoe/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/alphamoe/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/alphamoe/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/alphamoe/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/dsa_attention/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/dsa_attention/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/dsa_attention/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/dsa_attention/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/dsa_attention/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_backward/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_backward/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_backward/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_backward/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_backward/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_forward/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_forward/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_forward/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_forward/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/fa4_forward/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/fp16_gemm_floor/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fp16_gemm_floor/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fp16_gemm_floor/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/fp16_gemm_floor/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/fp16_gemm_floor/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_decode/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_decode/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_decode/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_decode/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_decode/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_prefill/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_prefill/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_prefill/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_prefill/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/gdn_prefill/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_backward/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_backward/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_backward/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_backward/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_backward/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_decode/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_decode/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_decode/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_decode/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_decode/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_forward/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_forward/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_forward/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_forward/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_forward/holdout.py
evolution/benchmark/flashinfer_bench_evolve/tasks/kda_forward/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/mla_dsv4/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/mla_dsv4/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/mla_dsv4/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/mla_dsv4/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/mla_dsv4/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/moe/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/moe/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/moe/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/moe/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/moe/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_decode/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_decode/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_decode/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_decode/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_decode/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_prefill/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_prefill/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_prefill/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_prefill/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/msa_prefill/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/nvfp4_attention/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/nvfp4_attention/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/nvfp4_attention/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/nvfp4_attention/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/nvfp4_attention/workload.jsonl
evolution/benchmark/flashinfer_bench_evolve/tasks/vsa/__init__.py
evolution/benchmark/flashinfer_bench_evolve/tasks/vsa/baseline.py
evolution/benchmark/flashinfer_bench_evolve/tasks/vsa/benchmark.py
evolution/benchmark/flashinfer_bench_evolve/tasks/vsa/definition.json
evolution/benchmark/flashinfer_bench_evolve/tasks/vsa/workload.jsonl
evolution/preparation/__init__.py
evolution/preparation/claude.py
evolution/preparation/codex.py
evolution/preparation/declare.py
evolution/preparation/environment.py
evolution/preparation/guards.py
evolution/preparation/live_references.py
evolution/preparation/requirements.txt
evolution/preparation/sandbox.py
evolution/preparation/worktree.py
evolution/prompts/PROMPT.md
evolution/prompts/PROMPT_remote.md
evolution/prompts/__init__.py
evolution/prompts/render.py
evolution/prompts/rules/rule_banned_reads.md
evolution/prompts/rules/rule_gpu.md
evolution/prompts/rules/rule_tirx_lite_authoring.md
evolution/remote/__init__.py
evolution/remote/kcoral_compute_sanitizer.py
evolution/remote/kcoral_driver.py
evolution/remote/kcoral_iket.py
evolution/remote/kcoral_ncu.py
evolution/remote/kcoral_python.py
evolution/remote/kcoral_remote.py
evolution/remote/kcoral_tool.py
evolution/remote/kcoral_tool_worker.py
evolution/tasks/README.md
evolution/tasks/alphamoe.yaml
evolution/tasks/alphamoe_m128_e512_topk10_k2048_i128_fp8.yaml
evolution/tasks/dsa.yaml
evolution/tasks/flash_attn_bwd.yaml
evolution/tasks/fp16_gemm_floor.yaml
evolution/tasks/gdn_decode.yaml
evolution/tasks/gdn_prefill.yaml
evolution/tasks/gdn_prefill_native.yaml
evolution/tasks/kda_backward.yaml
evolution/tasks/kda_backward_b1_t8192_h96.yaml
evolution/tasks/kda_decode.yaml
evolution/tasks/kda_decode_b128_t1_h16_hv32_d128_bf16.yaml
evolution/tasks/kda_forward.yaml
evolution/tasks/kda_forward_b1_t8192_h96.yaml
evolution/tasks/mla_dsv4.yaml
evolution/tasks/mla_dsv4_prefill_b2_qsum386_qmax257_h128_d512_swa16384_c16384_topk1152_bf16_hnd_varlen.yaml
evolution/tasks/moe.yaml
evolution/tasks/msa_decode.yaml
evolution/tasks/msa_decode_b128_q16_kv4096_hq64_hkv4_d128_topk16_bf16_flat.yaml
evolution/tasks/msa_prefill.yaml
evolution/tasks/msa_prefill_b1_q4096_kv4096_hq64_hkv4_d128_topk16_bf16_flat.yaml
evolution/tasks/nvfp4_attention.yaml
evolution/tasks/vsa.yaml
evolution/tasks/vsa_s80000_h8_d128_blk128_topk156_bf16.yaml
evolution/tests/__init__.py
evolution/tests/test_entrypoints.py
evolution/tests/remote/__init__.py
evolution/tests/remote/test_kcoral_tools.py
evolution/tests/remote/test_profiler_environment.py
evolution/toolsets/kda_flow.yaml
scripts/smoke_wheel.py
skills/README.md
skills/external.json
skills/tirx-debug-kernel/SKILL.md
skills/tirx-debug-kernel/agents/openai.yaml
skills/tirx-debug-kernel/references/compute-sanitizer.md
skills/tirx-debug-kernel/references/numsim.md
skills/tirx-debug-kernel/references/racecheck.md
skills/tirx-debug-kernel/references/synccheck.md
skills/tirx-profile-kernel/SKILL.md
skills/tirx-profile-kernel/agents/openai.yaml
skills/tirx-profile-kernel/references/dump-source.md
skills/tirx-profile-kernel/references/iket.md
skills/tirx-profile-kernel/references/ncu.md
skills/tirx-wiki/SKILL.md
skills/tirx-wiki/requirements.txt
skills/tirx-wiki/agents/openai.yaml
skills/tirx-wiki/optimization/INDEX.md
skills/tirx-wiki/optimization/kernel-catalog.md
skills/tirx-wiki/optimization/playbook.md
skills/tirx-wiki/optimization/swp-edge-cutting.md
skills/tirx-wiki/optimization/perf_checklist/section_1_persistent_kernel_structure.md
skills/tirx-wiki/optimization/perf_checklist/section_2_static_source_patterns.md
skills/tirx-wiki/optimization/perf_checklist/section_3_scalar_dataflow.md
skills/tirx-wiki/optimization/perf_checklist/section_4_dead_work.md
skills/tirx-wiki/optimization/perf_checklist/section_5_smem.md
skills/tirx-wiki/optimization/perf_checklist/section_6_register_pressure.md
skills/tirx-wiki/references/INDEX.md
skills/tirx-wiki/references/sources.json
skills/tirx-wiki/references/blogs/INDEX.md
skills/tirx-wiki/references/blogs/colfax/a-note-on-the-algebra-of-cute-layouts.md
skills/tirx-wiki/references/blogs/colfax/a-users-guide-to-flexattention-in-flash-attention-cute-dsl.md
skills/tirx-wiki/references/blogs/colfax/adding-fp8-to-flashattention.md
skills/tirx-wiki/references/blogs/colfax/categorical-foundations-for-cute-layouts.md
skills/tirx-wiki/references/blogs/colfax/custom-cuda-and-python.md
skills/tirx-wiki/references/blogs/colfax/cutlass-3-x-apis-orthogonal-reusable-and-composable-abstractions-for-gemm-kernel-design-external.md
skills/tirx-wiki/references/blogs/colfax/cutlass-tutorial-design-of-a-gemm-kernel.md
skills/tirx-wiki/references/blogs/colfax/cutlass-tutorial-gemm-with-thread-block-clusters-on-nvidia-blackwell-gpus.md
skills/tirx-wiki/references/blogs/colfax/cutlass-tutorial-persistent-kernels-and-stream-k.md
skills/tirx-wiki/references/blogs/colfax/cutlass-tutorial-sub-byte-gemm-on-nvidia-blackwell-gpus.md
skills/tirx-wiki/references/blogs/colfax/cutlass-tutorial-wgmma-hopper.md
skills/tirx-wiki/references/blogs/colfax/cutlass-tutorial-writing-gemm-kernels-using-tensor-memory-for-nvidia-blackwell-gpus.md
skills/tirx-wiki/references/blogs/colfax/deepseek-r1-and-fp8-mixed-precision-training.md
skills/tirx-wiki/references/blogs/colfax/epilogue_visitor_tree.md
skills/tirx-wiki/references/blogs/colfax/flashattention-3-fast-and-accurate-attention-with-asynchrony-and-low-precision.md
skills/tirx-wiki/references/blogs/colfax/flashattention-3-for-inference-int8-quantization-and-query-head-packing-for-mqa-gqa-external.md
skills/tirx-wiki/references/blogs/colfax/gpu-mode-cutlass-and-flashattention-3.md
skills/tirx-wiki/references/blogs/colfax/introduction-to-transformers.md
skills/tirx-wiki/references/blogs/colfax/narrow-bit-width-formats-for-deep-learning.md
skills/tirx-wiki/references/blogs/colfax/nvidia-data-center-gpu-manager.md
skills/tirx-wiki/references/blogs/colfax/nvidia-hopper-flashattention-2.md
skills/tirx-wiki/references/blogs/colfax/nvidia-hopper-gemm-cutlass.md
skills/tirx-wiki/references/blogs/colfax/sharing-nvidia-gpus-at-the-system-level-time-sliced-and-mig-backed-vgpus.md
skills/tirx-wiki/references/blogs/colfax/software-pipelining-in-the-nvidia-h100-architecture.md
skills/tirx-wiki/references/blogs/colfax/tutorial-hopper-tma.md
skills/tirx-wiki/references/blogs/colfax/tutorial-matrix-transpose-in-cutlass.md
skills/tirx-wiki/references/blogs/colfax/tutorial-python-binding-for-cuda-libraries-in-pytorch.md
skills/tirx-wiki/references/blogs/thunderkittens/2024-05-12-quick-tk.md
skills/tirx-wiki/references/blogs/thunderkittens/2024-05-12-tk.md
skills/tirx-wiki/references/blogs/thunderkittens/2024-10-29-tk2.md
skills/tirx-wiki/references/blogs/thunderkittens/2024-11-27-tk-fp8.md
skills/tirx-wiki/references/blogs/thunderkittens/2024-11-28-tk-mlx.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-03-04-thundermla.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-03-15-tk-blackwell.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-05-27-no-bubbles.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-09-22-pgl.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-09-28-tp-llama-intro.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-09-28-tp-llama-main.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-11-09-amd-brr.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-11-09-hk.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-11-17-fluffy-kittens.md
skills/tirx-wiki/references/blogs/thunderkittens/2025-11-17-pk.md
skills/tirx-wiki/references/blogs/thunderkittens/2026-02-19-tk-2.md
skills/tirx-wiki/references/manuals/INDEX.md
skills/tirx-wiki/references/repos/INDEX.md
skills/tirx-wiki/scripts/fetch_references.py
thirdparty/README.md
tirx_harness/AGENTS.md
tirx_harness/CLAUDE.md
tirx_harness/README.md
tirx_harness/pyproject.toml
tirx_harness/docs/numsim/HOW_TO_EXTEND_NUMSIM.md
tirx_harness/frontend-rs/Cargo.lock
tirx_harness/frontend-rs/Cargo.toml
tirx_harness/frontend-rs/src/decode.rs
tirx_harness/frontend-rs/src/dtypes.rs
tirx_harness/frontend-rs/src/host_abi.rs
tirx_harness/frontend-rs/src/ir_tests.rs
tirx_harness/frontend-rs/src/lib.rs
tirx_harness/frontend-rs/src/registry.rs
tirx_harness/frontend-rs/src/schema.rs
tirx_harness/frontend-rs/src/serialization.rs
tirx_harness/frontend-rs/src/tables.rs
tirx_harness/frontend-rs/src/tvm_compat.rs
tirx_harness/frontend-rs/src/analyze/buffers.rs
tirx_harness/frontend-rs/src/analyze/cuda_arch.rs
tirx_harness/frontend-rs/src/analyze/expression_bindings.rs
tirx_harness/frontend-rs/src/analyze/frontend.rs
tirx_harness/frontend-rs/src/analyze/layout.rs
tirx_harness/frontend-rs/src/analyze/memory.rs
tirx_harness/frontend-rs/src/analyze/mod.rs
tirx_harness/frontend-rs/src/analyze/pointer_targets.rs
tirx_harness/frontend-rs/src/analyze/shapes.rs
tirx_harness/frontend-rs/src/analyze/topology.rs
tirx_harness/frontend-rs/src/analyze/util.rs
tirx_harness/frontend-rs/src/analyze/vector.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/coordinates.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/copy.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/gemm.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/mod.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/numeric.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/parse.rs
tirx_harness/frontend-rs/src/analyze/tile_forms/tcgen_layout.rs
tirx_harness/frontend-rs/src/decode/pointers.rs
tirx_harness/frontend-rs/src/decode/ptx.rs
tirx_harness/frontend-rs/src/emit/abi.rs
tirx_harness/frontend-rs/src/emit/async_copy.rs
tirx_harness/frontend-rs/src/emit/atomic_bulk.rs
tirx_harness/frontend-rs/src/emit/calls.rs
tirx_harness/frontend-rs/src/emit/clc.rs
tirx_harness/frontend-rs/src/emit/control.rs
tirx_harness/frontend-rs/src/emit/cuda_helper.rs
tirx_harness/frontend-rs/src/emit/diagnostics.rs
tirx_harness/frontend-rs/src/emit/expr.rs
tirx_harness/frontend-rs/src/emit/layout.rs
tirx_harness/frontend-rs/src/emit/matrix.rs
tirx_harness/frontend-rs/src/emit/matrix_variants.rs
tirx_harness/frontend-rs/src/emit/memory_ops.rs
tirx_harness/frontend-rs/src/emit/memory_support.rs
tirx_harness/frontend-rs/src/emit/mod.rs
tirx_harness/frontend-rs/src/emit/module.rs
tirx_harness/frontend-rs/src/emit/module_template.rs
tirx_harness/frontend-rs/src/emit/ptx_addr.rs
tirx_harness/frontend-rs/src/emit/ptx_address.rs
tirx_harness/frontend-rs/src/emit/ptx_arithmetic.rs
tirx_harness/frontend-rs/src/emit/ptx_bit_carrier.rs
tirx_harness/frontend-rs/src/emit/ptx_bitops.rs
tirx_harness/frontend-rs/src/emit/ptx_cache_policy.rs
tirx_harness/frontend-rs/src/emit/ptx_clmad.rs
tirx_harness/frontend-rs/src/emit/ptx_compare.rs
tirx_harness/frontend-rs/src/emit/ptx_cvt.rs
tirx_harness/frontend-rs/src/emit/ptx_cvt_pack.rs
tirx_harness/frontend-rs/src/emit/ptx_cvt_variants.rs
tirx_harness/frontend-rs/src/emit/ptx_integer_arithmetic.rs
tirx_harness/frontend-rs/src/emit/ptx_lop3.rs
tirx_harness/frontend-rs/src/emit/ptx_minmax.rs
tirx_harness/frontend-rs/src/emit/ptx_move.rs
tirx_harness/frontend-rs/src/emit/ptx_set_packed.rs
tirx_harness/frontend-rs/src/emit/ptx_spdecompress.rs
tirx_harness/frontend-rs/src/emit/ptx_unary.rs
tirx_harness/frontend-rs/src/emit/ptx_warp.rs
tirx_harness/frontend-rs/src/emit/pure.rs
tirx_harness/frontend-rs/src/emit/raw_memory.rs
tirx_harness/frontend-rs/src/emit/raw_tcgen.rs
tirx_harness/frontend-rs/src/emit/raw_tma.rs
tirx_harness/frontend-rs/src/emit/register_call.rs
tirx_harness/frontend-rs/src/emit/scaffold.rs
tirx_harness/frontend-rs/src/emit/shared_blocks.rs
tirx_harness/frontend-rs/src/emit/stmt.rs
tirx_harness/frontend-rs/src/emit/sync.rs
tirx_harness/frontend-rs/src/emit/tcgen05.rs
tirx_harness/frontend-rs/src/emit/tcgen_descriptor.rs
tirx_harness/frontend-rs/src/emit/tile.rs
tirx_harness/frontend-rs/src/emit/tile_async_copy.rs
tirx_harness/frontend-rs/src/emit/tile_common.rs
tirx_harness/frontend-rs/src/emit/tile_gemm_async.rs
tirx_harness/frontend-rs/src/emit/tile_tcgen05.rs
tirx_harness/frontend-rs/src/emit/tmem.rs
tirx_harness/frontend-rs/src/emit/wait_until.rs
tirx_harness/frontend-rs/src/emit/raw_tma/host_prelude.rs
tirx_harness/frontend-rs/src/emit/shared_blocks/tests.rs
tirx_harness/frontend-rs/tests/architecture.rs
tirx_harness/src/tirx_harness/__init__.py
tirx_harness/src/tirx_harness/_report.py
tirx_harness/src/tirx_harness/dump_kernel.py
tirx_harness/src/tirx_harness.egg-info/PKG-INFO
tirx_harness/src/tirx_harness.egg-info/SOURCES.txt
tirx_harness/src/tirx_harness.egg-info/dependency_links.txt
tirx_harness/src/tirx_harness.egg-info/requires.txt
tirx_harness/src/tirx_harness.egg-info/scm_file_list.json
tirx_harness/src/tirx_harness.egg-info/scm_version.json
tirx_harness/src/tirx_harness.egg-info/top_level.txt
tirx_harness/src/tirx_harness/numsim/AGENTS.md
tirx_harness/src/tirx_harness/numsim/CLAUDE.md
tirx_harness/src/tirx_harness/numsim/__init__.py
tirx_harness/src/tirx_harness/numsim/abi.py
tirx_harness/src/tirx_harness/numsim/api.py
tirx_harness/src/tirx_harness/numsim/bindings.py
tirx_harness/src/tirx_harness/numsim/cases.py
tirx_harness/src/tirx_harness/numsim/checker_render.py
tirx_harness/src/tirx_harness/numsim/checker_report.py
tirx_harness/src/tirx_harness/numsim/checker_runner.py
tirx_harness/src/tirx_harness/numsim/checkers.py
tirx_harness/src/tirx_harness/numsim/dtype_abi.py
tirx_harness/src/tirx_harness/numsim/dtype_registry.json
tirx_harness/src/tirx_harness/numsim/errors.py
tirx_harness/src/tirx_harness/numsim/host_abi.py
tirx_harness/src/tirx_harness/numsim/report.py
tirx_harness/src/tirx_harness/numsim/value_analysis.py
tirx_harness/src/tirx_harness/numsim/engine-rs/.gitignore
tirx_harness/src/tirx_harness/numsim/engine-rs/Cargo.lock
tirx_harness/src/tirx_harness/numsim/engine-rs/Cargo.toml
tirx_harness/src/tirx_harness/numsim/engine-rs/SUPPORTED_OPS.md
tirx_harness/src/tirx_harness/numsim/engine-rs/artifact-template/Cargo.lock
tirx_harness/src/tirx_harness/numsim/engine-rs/artifact-template/Cargo.toml
tirx_harness/src/tirx_harness/numsim/engine-rs/artifact-template/src/lib.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/fp-env/Cargo.toml
tirx_harness/src/tirx_harness/numsim/engine-rs/fp-env/src/lib.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/host-buffer/Cargo.toml
tirx_harness/src/tirx_harness/numsim/engine-rs/host-buffer/src/lib.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/analysis_search.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/artifact_support.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/async_groups.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/async_token.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/cluster_barriers.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/collectives.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/completion.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/completion_action_id.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/context.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/deferred_drop.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/deferred_payload.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/deferred_payload_tests.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/effect.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/engine_mode.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/executor.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/hardware_barriers.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/hardware_barriers_tests.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/instruction_codec.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/instruction_profile.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/kernel_engine.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/lib.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/mask.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/memory.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/mode_completions.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/numpy_backend.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/operation.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/ordering.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/physical_access.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/profile.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/resolved_transition.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/scalar.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/scheduling.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/setmaxnreg.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/setmaxnreg_tests.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/spaces.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/strict_cluster_barrier.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/strict_named_barrier.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/tcgen.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/topology.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/typed_async.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/typed_copy.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/warp_value.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/worker_affinity.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/mod.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/addr.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/async_copy.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/collective.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/control.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/error.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/instruction.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/matrix.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/mem.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/mod.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/reg.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/sync.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/tcgen05.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/tile.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/tmem.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/transport.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/abi/v2/warp.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/artifact_support/frontend_expr.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/checker_launch.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/sync_causality.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/racecheck/global_race.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/racecheck/race_check.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/racecheck/race_check_python.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/racecheck/race_shadow.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/racecheck/tcgen_fence.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/racecheck/transactional_interval_map.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/setmaxnreg_verifier.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/setmaxnreg_verifier_tests.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/strict_mbarrier.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/sync_check.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/sync_check_python.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/sync_fixed_unified.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/sync_fixed_verifier.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/native_analysis/synccheck/sync_partial_order.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/abi_transport.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/io.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/launch.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/matrix_ops.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/memory_ops.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/mod.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/operand.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/python.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/sync.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tcgen_lifecycle.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tcgen_ops.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tcgen_work.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tensor_map.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tensor_map_registry.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tmem.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/warp_ops.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/async_copy.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/collective.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/control.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/matrix.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/mem.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/mod.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/mode_axis.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/mov_unpack.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/packed_fma.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/packed_mul.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/prmt.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/reg.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/reg_compare.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/sparse_compress.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/sync.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/tcgen05.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/tile.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/tmem.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/instructions/warp.rs
tirx_harness/src/tirx_harness/numsim/engine-rs/src/runtime/tcgen_ops/integer.rs
tirx_harness/src/tirx_harness/numsim/transpiler/__init__.py
tirx_harness/src/tirx_harness/numsim/transpiler/artifact_template.py
tirx_harness/src/tirx_harness/numsim/transpiler/build.py
tirx_harness/src/tirx_harness/numsim/transpiler/cache.py
tirx_harness/src/tirx_harness/numsim/transpiler/compile_cache.py
tirx_harness/src/tirx_harness/numsim/transpiler/frontend.py
tirx_harness/src/tirx_harness/numsim/transpiler/host_prelude.py
tirx_harness/src/tirx_harness/numsim/transpiler/native_frontend.py
tirx_harness/src/tirx_harness/numsim/transpiler/ptx_dialect.py
tirx_harness/src/tirx_harness/numsim/transpiler/semantic_ir.py
tirx_harness/src/tirx_harness/numsim/transpiler/shared_helpers.py
tirx_harness/src/tirx_harness/numsim/transpiler/source_map.py
tirx_harness/src/tirx_harness/numsim/transpiler/support_matrix.py
tirx_harness/src/tirx_harness/numsim/transpiler/suspend_scaffold.py
tirx_harness/tests/AGENTS.md
tirx_harness/tests/CLAUDE.md
tirx_harness/tests/README.md
tirx_harness/tests/__init__.py
tirx_harness/tests/conftest.py
tirx_harness/tests/test_dump_kernel.py
tirx_harness/tests/test_packaging.py
tirx_harness/tests/analysis_tools/README.md
tirx_harness/tests/analysis_tools/__init__.py
tirx_harness/tests/analysis_tools/test_report.py
tirx_harness/tests/analysis_tools/racecheck/__init__.py
tirx_harness/tests/analysis_tools/racecheck/_native_race_trace.py
tirx_harness/tests/analysis_tools/racecheck/conftest.py
tirx_harness/tests/analysis_tools/racecheck/test_bulk_reduction_widths.py
tirx_harness/tests/analysis_tools/racecheck/test_declared_word_regressions.py
tirx_harness/tests/analysis_tools/racecheck/test_declared_word_shapes.py
tirx_harness/tests/analysis_tools/racecheck/test_native_alias_advisory.py
tirx_harness/tests/analysis_tools/racecheck/test_native_async_lifetime_contracts.py
tirx_harness/tests/analysis_tools/racecheck/test_native_atomic_semantics.py
tirx_harness/tests/analysis_tools/racecheck/test_native_exact_oob.py
tirx_harness/tests/analysis_tools/racecheck/test_native_global_access_count.py
tirx_harness/tests/analysis_tools/racecheck/test_native_global_scoped_hb.py
tirx_harness/tests/analysis_tools/racecheck/test_native_global_scoped_hb_matrix.py
tirx_harness/tests/analysis_tools/racecheck/test_native_global_write_seed.py
tirx_harness/tests/analysis_tools/racecheck/test_native_kernel_contracts.py
tirx_harness/tests/analysis_tools/racecheck/test_native_matrix_collective_sync.py
tirx_harness/tests/analysis_tools/racecheck/test_native_ordered_b128_load.py
tirx_harness/tests/analysis_tools/racecheck/test_native_proxy_async_fence.py
tirx_harness/tests/analysis_tools/racecheck/test_native_public_api.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_arrive_snapshot.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_artifact.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_exact_control.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_lane_order.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_per_warp.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_release_rmw_handoff.py
tirx_harness/tests/analysis_tools/racecheck/test_native_racecheck_tmem_arrive_snapshot.py
tirx_harness/tests/analysis_tools/racecheck/test_native_raw_async_copy_footprints.py
tirx_harness/tests/analysis_tools/racecheck/test_native_register_copy_sync.py
tirx_harness/tests/analysis_tools/racecheck/test_native_same_warp_tmem_review.py
tirx_harness/tests/analysis_tools/racecheck/test_native_shared_publication.py
tirx_harness/tests/analysis_tools/racecheck/test_native_sub_word_accesses.py
tirx_harness/tests/analysis_tools/racecheck/test_native_tcgen_thread_fence.py
tirx_harness/tests/analysis_tools/racecheck/test_native_vector_destination_loads.py
tirx_harness/tests/analysis_tools/racecheck/test_signal_diagnostics.py
tirx_harness/tests/analysis_tools/racecheck/corpus/__init__.py
tirx_harness/tests/analysis_tools/racecheck/corpus/test_canonical_kernels_racecheck.py
tirx_harness/tests/analysis_tools/racecheck/wiki/__init__.py
tirx_harness/tests/analysis_tools/racecheck/wiki/_cases.py
tirx_harness/tests/analysis_tools/racecheck/wiki/test_wiki_racecheck.py
tirx_harness/tests/analysis_tools/shared/__init__.py
tirx_harness/tests/analysis_tools/shared/test_checker_defaults.py
tirx_harness/tests/analysis_tools/shared/test_codegen_dependency_paths.py
tirx_harness/tests/analysis_tools/shared/test_instruction_profile.py
tirx_harness/tests/analysis_tools/shared/test_known_cuda_func_artifact.py
tirx_harness/tests/analysis_tools/shared/test_native_dense_cta2_mma_ordering.py
tirx_harness/tests/analysis_tools/shared/test_native_dynamic_launch.py
tirx_harness/tests/analysis_tools/shared/test_native_predicated_reduction.py
tirx_harness/tests/analysis_tools/shared/test_native_shared_descriptor_table.py
tirx_harness/tests/analysis_tools/shared/test_native_tile_config_scalars.py
tirx_harness/tests/analysis_tools/synccheck/__init__.py
tirx_harness/tests/analysis_tools/synccheck/test_native_break_continue.py
tirx_harness/tests/analysis_tools/synccheck/test_native_drain_tail.py
tirx_harness/tests/analysis_tools/synccheck/test_native_frontend_source_anchor.py
tirx_harness/tests/analysis_tools/synccheck/test_native_gate_intrinsics.py
tirx_harness/tests/analysis_tools/synccheck/test_native_kernel_contracts.py
tirx_harness/tests/analysis_tools/synccheck/test_native_lost_wake_regression_fanout.py
tirx_harness/tests/analysis_tools/synccheck/test_native_lost_wake_regression_pingpong.py
tirx_harness/tests/analysis_tools/synccheck/test_native_polling_control.py
tirx_harness/tests/analysis_tools/synccheck/test_native_shared_control_relay.py
tirx_harness/tests/analysis_tools/synccheck/test_native_shared_mbar_pointer.py
tirx_harness/tests/analysis_tools/synccheck/test_native_split_site_named_barrier.py
tirx_harness/tests/analysis_tools/synccheck/test_native_synccheck_artifact.py
tirx_harness/tests/analysis_tools/synccheck/test_native_synccheck_exact_control.py
tirx_harness/tests/analysis_tools/synccheck/test_native_uninitialized_read_source.py
tirx_harness/tests/analysis_tools/synccheck/test_register_allocation.py
tirx_harness/tests/analysis_tools/synccheck/test_report_readability.py
tirx_harness/tests/analysis_tools/synccheck/test_reported_tool_regressions.py
tirx_harness/tests/analysis_tools/synccheck/test_sync_site_manifest.py
tirx_harness/tests/analysis_tools/synccheck/corpus/__init__.py
tirx_harness/tests/analysis_tools/synccheck/corpus/test_tirx_kernels_synccheck.py
tirx_harness/tests/analysis_tools/synccheck/registry/__init__.py
tirx_harness/tests/analysis_tools/synccheck/runtime/__init__.py
tirx_harness/tests/analysis_tools/synccheck/runtime/test_device_additional_payload_ops.py
tirx_harness/tests/analysis_tools/synccheck/runtime/test_device_payload_ops.py
tirx_harness/tests/analysis_tools/synccheck/runtime/test_device_protocol_ops.py
tirx_harness/tests/numsim/AGENTS.md
tirx_harness/tests/numsim/CLAUDE.md
tirx_harness/tests/numsim/__init__.py
tirx_harness/tests/numsim/conftest.py
tirx_harness/tests/numsim/abi/__init__.py
tirx_harness/tests/numsim/abi/test_abi_version.py
tirx_harness/tests/numsim/abi/test_engine_public_api.py
tirx_harness/tests/numsim/corpus/__init__.py
tirx_harness/tests/numsim/corpus/canonical_cases.py
tirx_harness/tests/numsim/corpus/canonical_verdict_rationale.md
tirx_harness/tests/numsim/corpus/test_canonical_kernels.py
tirx_harness/tests/numsim/corpus/test_flash_attention4_case.py
tirx_harness/tests/numsim/corpus/test_flash_attention_backward_case.py
tirx_harness/tests/numsim/corpus/test_fp16_bf16_gemm_case.py
tirx_harness/tests/numsim/corpus/test_fp8_blockwise_gemm_case.py
tirx_harness/tests/numsim/corpus/test_mqa_cases.py
tirx_harness/tests/numsim/corpus/test_native_alphamoe.py
tirx_harness/tests/numsim/corpus/test_native_kda_backward.py
tirx_harness/tests/numsim/corpus/test_nvfp4_gemm_case.py
tirx_harness/tests/numsim/corpus/test_recurrent_cases.py
tirx_harness/tests/numsim/corpus/test_tf32_hc_prenorm_case.py
tirx_harness/tests/numsim/corpus/kernels/__init__.py
tirx_harness/tests/numsim/corpus/kernels/activation.py
tirx_harness/tests/numsim/corpus/kernels/attention.py
tirx_harness/tests/numsim/corpus/kernels/blockscaled_gemm.py
tirx_harness/tests/numsim/corpus/kernels/cascade.py
tirx_harness/tests/numsim/corpus/kernels/cudnn.py
tirx_harness/tests/numsim/corpus/kernels/deepgemm.py
tirx_harness/tests/numsim/corpus/kernels/flashmla.py
tirx_harness/tests/numsim/corpus/kernels/flex_attention.py
tirx_harness/tests/numsim/corpus/kernels/gemm.py
tirx_harness/tests/numsim/corpus/kernels/msa.py
tirx_harness/tests/numsim/corpus/kernels/native_kda_backward.py
tirx_harness/tests/numsim/corpus/kernels/native_moe.py
tirx_harness/tests/numsim/corpus/kernels/native_multishape.py
tirx_harness/tests/numsim/corpus/kernels/normalization.py
tirx_harness/tests/numsim/corpus/kernels/quantization.py
tirx_harness/tests/numsim/corpus/kernels/recurrent.py
tirx_harness/tests/numsim/corpus/kernels/state_update.py
tirx_harness/tests/numsim/corpus/kernels/topk.py
tirx_harness/tests/numsim/integration/__init__.py
tirx_harness/tests/numsim/integration/test_api.py
tirx_harness/tests/numsim/integration/test_artifact_build.py
tirx_harness/tests/numsim/integration/test_bindings.py
tirx_harness/tests/numsim/integration/test_block_scaled_gemm_artifact.py
tirx_harness/tests/numsim/integration/test_build_workspace.py
tirx_harness/tests/numsim/integration/test_compile_cache.py
tirx_harness/tests/numsim/integration/test_dependency_isolation.py
tirx_harness/tests/numsim/integration/test_deterministic_reexecution.py
tirx_harness/tests/numsim/integration/test_direct_memory_artifact.py
tirx_harness/tests/numsim/integration/test_emission_errors.py
tirx_harness/tests/numsim/integration/test_float8_artifact.py
tirx_harness/tests/numsim/integration/test_fp8_cta1_descriptor_layout.py
tirx_harness/tests/numsim/integration/test_fp8_tmem_a_effects.py
tirx_harness/tests/numsim/integration/test_frontend.py
tirx_harness/tests/numsim/integration/test_gemm_async_artifact.py
tirx_harness/tests/numsim/integration/test_global_alias_artifact.py
tirx_harness/tests/numsim/integration/test_host_prelude.py
tirx_harness/tests/numsim/integration/test_instruction_profile.py
tirx_harness/tests/numsim/integration/test_integer_addresses.py
tirx_harness/tests/numsim/integration/test_large_buffer_artifact.py
tirx_harness/tests/numsim/integration/test_memory_artifact.py
tirx_harness/tests/numsim/integration/test_memory_plan.py
tirx_harness/tests/numsim/integration/test_mismatch_diagnostics.py
tirx_harness/tests/numsim/integration/test_multi_kernel_artifact.py
tirx_harness/tests/numsim/integration/test_native_frontend.py
tirx_harness/tests/numsim/integration/test_native_ir.py
tirx_harness/tests/numsim/integration/test_native_statement_tree.py
tirx_harness/tests/numsim/integration/test_numpy_backend.py
tirx_harness/tests/numsim/integration/test_opaque_helper_artifact.py
tirx_harness/tests/numsim/integration/test_packaging.py
tirx_harness/tests/numsim/integration/test_packed_ptx_cvt_and_return.py
tirx_harness/tests/numsim/integration/test_permute_layout_artifact.py
tirx_harness/tests/numsim/integration/test_profile_build.py
tirx_harness/tests/numsim/integration/test_ptx_schema_compatibility.py
tirx_harness/tests/numsim/integration/test_raw_memory_artifact.py
tirx_harness/tests/numsim/integration/test_raw_versus_typed_cta2_mma.py
tirx_harness/tests/numsim/integration/test_register_layout_artifact.py
tirx_harness/tests/numsim/integration/test_reported_layout_regressions.py
tirx_harness/tests/numsim/integration/test_runtime_sized_global_alias.py
tirx_harness/tests/numsim/integration/test_scalar_parameter_artifact.py
tirx_harness/tests/numsim/integration/test_scheduler_polling_artifact.py
tirx_harness/tests/numsim/integration/test_source_map.py
tirx_harness/tests/numsim/integration/test_stmatrix_artifact.py
tirx_harness/tests/numsim/integration/test_tcgen_transfer_artifact.py
tirx_harness/tests/numsim/integration/test_tirx_kernels.py
tirx_harness/tests/numsim/integration/test_tmem_artifact.py
tirx_harness/tests/numsim/integration/test_topology_artifact.py
tirx_harness/tests/numsim/integration/test_vector_ir_artifact.py
tirx_harness/tests/numsim/integration/test_warp_gemm_artifact.py
tirx_harness/tests/numsim/integration/test_warp_ops_artifact.py
tirx_harness/tests/numsim/integration/test_written_buffer_inference.py
tirx_harness/tests/numsim/microtests/__init__.py
tirx_harness/tests/numsim/microtests/harness.py
tirx_harness/tests/numsim/microtests/test_atomic_f32_noftz.py
tirx_harness/tests/numsim/microtests/test_bar_reduce.py
tirx_harness/tests/numsim/microtests/test_bitwise_nodes.py
tirx_harness/tests/numsim/microtests/test_cvt_f32_half.py
tirx_harness/tests/numsim/microtests/test_float_atomics.py
tirx_harness/tests/numsim/microtests/test_gpu_harness.py
tirx_harness/tests/numsim/microtests/test_ldmatrix_b8.py
tirx_harness/tests/numsim/microtests/test_mapped_shared_offsets.py
tirx_harness/tests/numsim/microtests/test_matrix_memory.py
tirx_harness/tests/numsim/microtests/test_mbarrier_drop.py
tirx_harness/tests/numsim/microtests/test_memory_coverage_next.py
tirx_harness/tests/numsim/microtests/test_memory_sync_coverage.py
tirx_harness/tests/numsim/microtests/test_mma_sync.py
tirx_harness/tests/numsim/microtests/test_multicast32.py
tirx_harness/tests/numsim/microtests/test_mxf8_cta2.py
tirx_harness/tests/numsim/microtests/test_pointer_null.py
tirx_harness/tests/numsim/microtests/test_ptx_cvt_fp8.py
tirx_harness/tests/numsim/microtests/test_ptx_cvt_narrow.py
tirx_harness/tests/numsim/microtests/test_ptx_cvt_scalar.py
tirx_harness/tests/numsim/microtests/test_ptx_numeric.py
tirx_harness/tests/numsim/microtests/test_raw_memory_forms.py
tirx_harness/tests/numsim/microtests/test_remaining_memory_ops.py
tirx_harness/tests/numsim/microtests/test_shared_descriptor_choices.py
tirx_harness/tests/numsim/microtests/test_sm107_register_predicates.py
tirx_harness/tests/numsim/microtests/test_tcgen05_advanced.py
tirx_harness/tests/numsim/microtests/test_tcgen05_core.py
tirx_harness/tests/numsim/microtests/test_tcgen05_fp8_ts.py
tirx_harness/tests/numsim/microtests/test_tcgen05_mma_forms.py
tirx_harness/tests/numsim/microtests/test_tcgen05_scale_abi.py
tirx_harness/tests/numsim/microtests/test_tcgen05_transfer.py
tirx_harness/tests/numsim/microtests/test_tcgen_commit_predicates.py
tirx_harness/tests/numsim/microtests/test_tcgen_descriptor_dispatch.py
tirx_harness/tests/numsim/microtests/test_tcgen_inactive_boundaries.py
tirx_harness/tests/numsim/microtests/test_tma.py
tirx_harness/tests/numsim/microtests/test_tma_interleave_hints.py
tirx_harness/tests/numsim/microtests/test_tma_overrides.py
tirx_harness/tests/numsim/microtests/test_wait_until.py
tirx_harness/tests/numsim/microtests/test_warp_collectives.py
tirx_harness/tests/numsim/microtests/cases/float_atomics.py
tirx_harness/tests/numsim/microtests/cases/matrix_memory.py
tirx_harness/tests/numsim/microtests/cases/mma_sync.py
tirx_harness/tests/numsim/microtests/cases/mxf8_cta2.py
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_fp8.py
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_fp8_goldens.cu
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_fp8_goldens.py
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_narrow.py
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_narrow_goldens.cu
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_narrow_goldens.py
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_scalar.py
tirx_harness/tests/numsim/microtests/cases/ptx_cvt_scalar_goldens.py
tirx_harness/tests/numsim/microtests/cases/ptx_numeric.py
tirx_harness/tests/numsim/microtests/cases/raw_memory_forms.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_advanced_mma.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_fp8_ts.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_ld_red.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_lifecycle_ldst.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_mma_forms.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_scale_abi.py
tirx_harness/tests/numsim/microtests/cases/tcgen05_transfer_semantics.py
tirx_harness/tests/numsim/microtests/cases/tma.py
tirx_harness/tests/numsim/registry/__init__.py
tirx_harness/tests/numsim/registry/test_bitwise_nodes.py
tirx_harness/tests/numsim/registry/test_call_registry.py
tirx_harness/tests/numsim/registry/test_call_resolution.py
tirx_harness/tests/numsim/registry/test_copy_dispatch_contract.py
tirx_harness/tests/numsim/registry/test_engine_support_matrix.py
tirx_harness/tests/numsim/registry/test_op_registry_contract.py
tirx_harness/tests/numsim/registry/test_operation_runtime_cases.py
tirx_harness/tests/numsim/registry/test_ptx_cache_hint_registry.py
tirx_harness/tests/numsim/registry/test_ptx_dialect_decoder.py
tirx_harness/tests/numsim/registry/test_ptx_lop3_registry.py
tirx_harness/tests/numsim/registry/test_ptx_spdecompress_registry.py
tirx_harness/tests/numsim/registry/test_ptx_warp_registry.py
tirx_harness/tests/numsim/registry/test_tile_dispatch_invariance.py
tirx_harness/tests/numsim/registry/test_tile_registry_contract.py
tirx_harness/tests/numsim/runtime/__init__.py
tirx_harness/tests/numsim/runtime/test_approximate_f32_contract.py
tirx_harness/tests/numsim/runtime/test_async_group_domain_oracle.py
tirx_harness/tests/numsim/runtime/test_async_release.py
tirx_harness/tests/numsim/runtime/test_atomic_domain_oracle.py
tirx_harness/tests/numsim/runtime/test_atomic_f32_noftz.py
tirx_harness/tests/numsim/runtime/test_bar_reduce.py
tirx_harness/tests/numsim/runtime/test_bitwise_nodes.py
tirx_harness/tests/numsim/runtime/test_bulk_copy_scopes.py
tirx_harness/tests/numsim/runtime/test_bulk_g2s_scopes.py
tirx_harness/tests/numsim/runtime/test_bulk_reduce_s2g_f32.py
tirx_harness/tests/numsim/runtime/test_byte_copy_vector_store.py
tirx_harness/tests/numsim/runtime/test_byte_pointer_view_alias.py
tirx_harness/tests/numsim/runtime/test_cache_hint_ops.py
tirx_harness/tests/numsim/runtime/test_cache_policy.py
tirx_harness/tests/numsim/runtime/test_compare_predicates.py
tirx_harness/tests/numsim/runtime/test_copy_multicast32.py
tirx_harness/tests/numsim/runtime/test_cp_async_mbarrier_visibility.py
tirx_harness/tests/numsim/runtime/test_cuda_generic_scalar_forms.py
tirx_harness/tests/numsim/runtime/test_cuda_ldg_overloads.py
tirx_harness/tests/numsim/runtime/test_cuda_low_precision_reductions.py
tirx_harness/tests/numsim/runtime/test_cuda_packed_vector_forms.py
tirx_harness/tests/numsim/runtime/test_cvt_carriers.py
tirx_harness/tests/numsim/runtime/test_cvt_integer_sat.py
tirx_harness/tests/numsim/runtime/test_cvt_predicates.py
tirx_harness/tests/numsim/runtime/test_dense_mma_forms.py
tirx_harness/tests/numsim/runtime/test_discard.py
tirx_harness/tests/numsim/runtime/test_dtype_abi.py
tirx_harness/tests/numsim/runtime/test_dynamic_pure_call_runtime_domains.py
tirx_harness/tests/numsim/runtime/test_expr_codegen.py
tirx_harness/tests/numsim/runtime/test_fetch_register_domain_oracle.py
tirx_harness/tests/numsim/runtime/test_flashkda_cuda_helpers.py
tirx_harness/tests/numsim/runtime/test_float8_address_shuffle.py
tirx_harness/tests/numsim/runtime/test_ftz_rounding.py
tirx_harness/tests/numsim/runtime/test_gate_intrinsics.py
tirx_harness/tests/numsim/runtime/test_im2col_cache_hints.py
tirx_harness/tests/numsim/runtime/test_launch_resource_facts.py
tirx_harness/tests/numsim/runtime/test_layout_lowering_contract.py
tirx_harness/tests/numsim/runtime/test_ldmatrix_b8.py
tirx_harness/tests/numsim/runtime/test_log2.py
tirx_harness/tests/numsim/runtime/test_logic_carriers.py
tirx_harness/tests/numsim/runtime/test_loop_bounds.py
tirx_harness/tests/numsim/runtime/test_matrix_instruction_codegen.py
tirx_harness/tests/numsim/runtime/test_matrix_memory_domain_oracle.py
tirx_harness/tests/numsim/runtime/test_mbarrier_drop.py
tirx_harness/tests/numsim/runtime/test_mbarrier_lane_semantics.py
tirx_harness/tests/numsim/runtime/test_mbarrier_maintenance.py
tirx_harness/tests/numsim/runtime/test_mbarrier_multicast.py
tirx_harness/tests/numsim/runtime/test_mbarrier_pending_count.py
tirx_harness/tests/numsim/runtime/test_mbarrier_report.py
tirx_harness/tests/numsim/runtime/test_memory_coverage_next.py
tirx_harness/tests/numsim/runtime/test_memory_ops.py
tirx_harness/tests/numsim/runtime/test_memory_sync_coverage.py
tirx_harness/tests/numsim/runtime/test_mixed_version_reads.py
tirx_harness/tests/numsim/runtime/test_mov_forms.py
tirx_harness/tests/numsim/runtime/test_multicast16.py
tirx_harness/tests/numsim/runtime/test_mxf8_cta2.py
tirx_harness/tests/numsim/runtime/test_non_tensor_bulk_forms.py
tirx_harness/tests/numsim/runtime/test_ordering_calls.py
tirx_harness/tests/numsim/runtime/test_packed_f16x2_arithmetic.py
tirx_harness/tests/numsim/runtime/test_packed_f32x2_float32_destination.py
tirx_harness/tests/numsim/runtime/test_packed_float4_global_views.py
tirx_harness/tests/numsim/runtime/test_physical_pointer_ownership.py
tirx_harness/tests/numsim/runtime/test_pointer_null.py
tirx_harness/tests/numsim/runtime/test_pointer_slot_arrays.py
tirx_harness/tests/numsim/runtime/test_ptx_addr_runtime.py
tirx_harness/tests/numsim/runtime/test_ptx_atomic_cas.py
tirx_harness/tests/numsim/runtime/test_ptx_bitops.py
tirx_harness/tests/numsim/runtime/test_ptx_clmad.py
tirx_harness/tests/numsim/runtime/test_ptx_compare_semantics.py
tirx_harness/tests/numsim/runtime/test_ptx_cvt_94.py
tirx_harness/tests/numsim/runtime/test_ptx_cvt_packed_forms.py
tirx_harness/tests/numsim/runtime/test_ptx_cvt_pzo.py
tirx_harness/tests/numsim/runtime/test_ptx_float_register_ops.py
tirx_harness/tests/numsim/runtime/test_ptx_integer_arithmetic.py
tirx_harness/tests/numsim/runtime/test_ptx_lop3.py
tirx_harness/tests/numsim/runtime/test_ptx_mixed_vector_arithmetic.py
tirx_harness/tests/numsim/runtime/test_ptx_move_pack_extended.py
tirx_harness/tests/numsim/runtime/test_ptx_new_register_semantics.py
tirx_harness/tests/numsim/runtime/test_ptx_register_bits.py
tirx_harness/tests/numsim/runtime/test_ptx_rm_zero_sign.py
tirx_harness/tests/numsim/runtime/test_ptx_set_packed.py
tirx_harness/tests/numsim/runtime/test_ptx_spdecompress.py
tirx_harness/tests/numsim/runtime/test_ptx_warp_collectives.py
tirx_harness/tests/numsim/runtime/test_raw_memory_state_spaces.py
tirx_harness/tests/numsim/runtime/test_raw_pointer_runtime_domain.py
tirx_harness/tests/numsim/runtime/test_raw_tcgen_codegen.py
tirx_harness/tests/numsim/runtime/test_raw_tensor_map_descriptors.py
tirx_harness/tests/numsim/runtime/test_raw_tma_codegen.py
tirx_harness/tests/numsim/runtime/test_readonly_proxy.py
tirx_harness/tests/numsim/runtime/test_red_async.py
tirx_harness/tests/numsim/runtime/test_register_extensions.py
tirx_harness/tests/numsim/runtime/test_reinterpret_runtime_domain.py
tirx_harness/tests/numsim/runtime/test_runtime_form_domain_oracles.py
tirx_harness/tests/numsim/runtime/test_scalar_control.py
tirx_harness/tests/numsim/runtime/test_scalar_f64_rounding.py
tirx_harness/tests/numsim/runtime/test_scalar_helpers.py
tirx_harness/tests/numsim/runtime/test_shared_descriptor_choices.py
tirx_harness/tests/numsim/runtime/test_sm107_register_predicates.py
tirx_harness/tests/numsim/runtime/test_sparse_mma_forms.py
tirx_harness/tests/numsim/runtime/test_store_sinks.py
tirx_harness/tests/numsim/runtime/test_sync_instruction_predicates.py
tirx_harness/tests/numsim/runtime/test_sync_runtime_domain_oracle.py
tirx_harness/tests/numsim/runtime/test_tcgen05_block_tmem.py
tirx_harness/tests/numsim/runtime/test_tcgen05_i8.py
tirx_harness/tests/numsim/runtime/test_tcgen05_ld_red.py
tirx_harness/tests/numsim/runtime/test_tcgen05_ld_spcompress.py
tirx_harness/tests/numsim/runtime/test_tcgen05_restricted_commit.py
tirx_harness/tests/numsim/runtime/test_tcgen05_sparse_b16.py
tirx_harness/tests/numsim/runtime/test_tcgen05_sparse_block.py
tirx_harness/tests/numsim/runtime/test_tcgen05_sparse_k128.py
tirx_harness/tests/numsim/runtime/test_tcgen05_sparse_narrow.py
tirx_harness/tests/numsim/runtime/test_tcgen05_tf32.py
tirx_harness/tests/numsim/runtime/test_tcgen05_ti16.py
tirx_harness/tests/numsim/runtime/test_tcgen_ashift.py
tirx_harness/tests/numsim/runtime/test_tcgen_codegen.py
tirx_harness/tests/numsim/runtime/test_tcgen_collectors.py
tirx_harness/tests/numsim/runtime/test_tcgen_commit_predicates.py
tirx_harness/tests/numsim/runtime/test_tcgen_inactive_boundaries.py
tirx_harness/tests/numsim/runtime/test_tcgen_lut_b.py
tirx_harness/tests/numsim/runtime/test_tcgen_misc_runtime_domain.py
tirx_harness/tests/numsim/runtime/test_tensormap_predicates.py
tirx_harness/tests/numsim/runtime/test_tensormap_publication.py
tirx_harness/tests/numsim/runtime/test_tf32_layout_f_poison.py
tirx_harness/tests/numsim/runtime/test_tile_codegen.py
tirx_harness/tests/numsim/runtime/test_tile_general_semantics.py
tirx_harness/tests/numsim/runtime/test_tile_owner_transport.py
tirx_harness/tests/numsim/runtime/test_tile_reduction_variants.py
tirx_harness/tests/numsim/runtime/test_tile_unary_codegen.py
tirx_harness/tests/numsim/runtime/test_tma_atomicity.py
tirx_harness/tests/numsim/runtime/test_tma_dtype_runtime_domain.py
tirx_harness/tests/numsim/runtime/test_tma_im2col.py
tirx_harness/tests/numsim/runtime/test_tma_im2col_multiissuer.py
tirx_harness/tests/numsim/runtime/test_tma_interleave_hints.py
tirx_harness/tests/numsim/runtime/test_tma_multiissuer.py
tirx_harness/tests/numsim/runtime/test_tma_overrides.py
tirx_harness/tests/numsim/runtime/test_tma_swizzle96.py
tirx_harness/tests/numsim/runtime/test_tma_u6.py
tirx_harness/tests/numsim/runtime/test_variable_identity.py
tirx_harness/tests/numsim/runtime/test_wait_until.py
tirx_harness/tests/numsim/support/__init__.py
tirx_harness/tests/numsim/support/_tirx_kernels.py
tirx_harness/tests/numsim/support/cvt_cases.py
tirx_harness/tests/numsim/support/execution.py
tirx_harness/tests/numsim/support/kernels.py
tirx_harness/tests/numsim/support/manifest.py
tirx_harness/tests/numsim/support/multi_kernels.py
tirx_harness/tests/numsim/support/paths.py
tirx_harness/tests/numsim/support/register_cases.py
tirx_harness/tests/numsim/support/remote_mbarrier.py
tirx_harness/tests/numsim/support/runtime_cases.py
tirx_harness/tests/numsim/support/runtime_domains.py
tirx_harness/tests/numsim/support/tcgen_descriptor.py
tirx_harness/tests/numsim/support/three_way.py
tirx_harness/tests/numsim/support/wait_until.py
tirx_harness/tests/support/__init__.py
tirx_harness/tests/support/tirx_device_surface.py