Skip to content

[BugFix] Add pybind11 check and Windows extension pattern fix #3310

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
vmoens merged 4 commits into from
Jan 8, 2026
Merged

[BugFix] Add pybind11 check and Windows extension pattern fix #3310

vmoens merged 4 commits into from
Jan 8, 2026

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Jan 8, 2026 •
edited
Loading

Stack from ghstack (oldest at bottom):

  • Add early check for pybind11 dependency with clear error message
  • Fix extension file pattern for Windows (.pyd) vs Unix (.so) in stale build cleanup

@vmoens
Update
b951d2b 
[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Jan 8, 2026 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3310

Note: Links to docs will display an error until the docs builds have been completed.

❌ 2 New Failures, 14 Pending

As of commit d91e2c2 with merge base cc917ba (image):

NEW FAILURES - The following jobs have failed:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@meta-cla meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jan 8, 2026
This was referenced Jan 8, 2026
Merged
Merged
@vmoens
Update
0483926 
[ghstack-poisoned]
This was referenced Jan 8, 2026
Merged
Merged
@vmoens
Update
daba2e0 
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Jan 8, 2026
@vmoens
[BugFix] Add pybind11 check and Windows extension pattern fix
9a7a3b1 
- Add early check for pybind11 dependency with clear error message
- Fix extension file pattern for Windows (.pyd) vs Unix (.so) in stale build cleanup

ghstack-source-id: df81b40
Pull-Request: #3310
@github-actions
Copy link

github-actions bot commented Jan 8, 2026 •
edited
Loading

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 164. Improved: $\large\color{#35bf28}24$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 84.8229μs 83.4437μs 11.9841 KOps/s 12.3058 KOps/s $\color{#d91a1a}-2.61\%$
test_tensor_to_bytestream_speed[torch.save] 0.1460ms 0.1447ms 6.9085 KOps/s 7.1210 KOps/s $\color{#d91a1a}-2.98\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1261s 0.1259s 7.9435 Ops/s 8.2835 Ops/s $\color{#d91a1a}-4.10\%$
test_tensor_to_bytestream_speed[numpy] 2.7506μs 2.7240μs 367.1091 KOps/s 380.7854 KOps/s $\color{#d91a1a}-3.59\%$
test_tensor_to_bytestream_speed[safetensors] 37.4659μs 37.2315μs 26.8590 KOps/s 25.5177 KOps/s $\textbf{\color{#35bf28}+5.26\%}$
test_simple 0.5469s 0.5437s 1.8391 Ops/s 1.7507 Ops/s $\textbf{\color{#35bf28}+5.05\%}$
test_transformed 1.1193s 1.1170s 0.8953 Ops/s 0.8767 Ops/s $\color{#35bf28}+2.12\%$
test_serial 1.6711s 1.6647s 0.6007 Ops/s 0.5966 Ops/s $\color{#35bf28}+0.70\%$
test_parallel 1.1168s 1.0990s 0.9099 Ops/s 0.8607 Ops/s $\textbf{\color{#35bf28}+5.72\%}$
test_step_mdp_speed[True-True-True-True-True] 0.3061ms 44.9357μs 22.2540 KOps/s 22.0490 KOps/s $\color{#35bf28}+0.93\%$
test_step_mdp_speed[True-True-True-True-False] 59.5310μs 25.0264μs 39.9579 KOps/s 39.6276 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[True-True-True-False-True] 56.2610μs 25.2266μs 39.6407 KOps/s 39.7782 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-True-True-False-False] 44.4910μs 13.9854μs 71.5030 KOps/s 71.8572 KOps/s $\color{#d91a1a}-0.49\%$
test_step_mdp_speed[True-True-False-True-True] 89.5110μs 47.6324μs 20.9941 KOps/s 20.5506 KOps/s $\color{#35bf28}+2.16\%$
test_step_mdp_speed[True-True-False-True-False] 62.3410μs 27.6749μs 36.1338 KOps/s 35.8188 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[True-True-False-False-True] 75.2320μs 27.6712μs 36.1386 KOps/s 35.7564 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[True-True-False-False-False] 50.0310μs 16.4965μs 60.6188 KOps/s 59.3346 KOps/s $\color{#35bf28}+2.16\%$
test_step_mdp_speed[True-False-True-True-True] 96.0320μs 50.7805μs 19.6926 KOps/s 19.2195 KOps/s $\color{#35bf28}+2.46\%$
test_step_mdp_speed[True-False-True-True-False] 59.9110μs 30.5757μs 32.7057 KOps/s 32.3343 KOps/s $\color{#35bf28}+1.15\%$
test_step_mdp_speed[True-False-True-False-True] 83.5920μs 27.9992μs 35.7154 KOps/s 35.7114 KOps/s $\color{#35bf28}+0.01\%$
test_step_mdp_speed[True-False-True-False-False] 47.1810μs 16.6511μs 60.0562 KOps/s 59.3917 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[True-False-False-True-True] 97.7020μs 53.7172μs 18.6160 KOps/s 18.5560 KOps/s $\color{#35bf28}+0.32\%$
test_step_mdp_speed[True-False-False-True-False] 64.9010μs 32.9253μs 30.3718 KOps/s 29.9524 KOps/s $\color{#35bf28}+1.40\%$
test_step_mdp_speed[True-False-False-False-True] 64.1510μs 30.3521μs 32.9467 KOps/s 32.4895 KOps/s $\color{#35bf28}+1.41\%$
test_step_mdp_speed[True-False-False-False-False] 84.6720μs 19.1691μs 52.1672 KOps/s 50.9922 KOps/s $\color{#35bf28}+2.30\%$
test_step_mdp_speed[False-True-True-True-True] 94.7010μs 50.0620μs 19.9752 KOps/s 19.8646 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-True-True-True-False] 64.0920μs 30.6888μs 32.5852 KOps/s 32.5794 KOps/s $\color{#35bf28}+0.02\%$
test_step_mdp_speed[False-True-True-False-True] 2.3671ms 31.7578μs 31.4883 KOps/s 31.4647 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[False-True-True-False-False] 53.0810μs 18.2301μs 54.8543 KOps/s 54.1350 KOps/s $\color{#35bf28}+1.33\%$
test_step_mdp_speed[False-True-False-True-True] 85.9920μs 53.1267μs 18.8229 KOps/s 18.6287 KOps/s $\color{#35bf28}+1.04\%$
test_step_mdp_speed[False-True-False-True-False] 94.3410μs 33.2258μs 30.0971 KOps/s 29.5960 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[False-True-False-False-True] 99.7020μs 33.3850μs 29.9536 KOps/s 29.1390 KOps/s $\color{#35bf28}+2.80\%$
test_step_mdp_speed[False-True-False-False-False] 49.3400μs 20.6794μs 48.3573 KOps/s 46.6545 KOps/s $\color{#35bf28}+3.65\%$
test_step_mdp_speed[False-False-True-True-True] 86.1920μs 56.1678μs 17.8038 KOps/s 17.7851 KOps/s $\color{#35bf28}+0.11\%$
test_step_mdp_speed[False-False-True-True-False] 72.4210μs 36.1299μs 27.6779 KOps/s 26.7560 KOps/s $\color{#35bf28}+3.45\%$
test_step_mdp_speed[False-False-True-False-True] 71.0820μs 34.0854μs 29.3381 KOps/s 28.9375 KOps/s $\color{#35bf28}+1.38\%$
test_step_mdp_speed[False-False-True-False-False] 54.1710μs 20.8966μs 47.8548 KOps/s 46.8294 KOps/s $\color{#35bf28}+2.19\%$
test_step_mdp_speed[False-False-False-True-True] 93.8520μs 58.0122μs 17.2378 KOps/s 17.1273 KOps/s $\color{#35bf28}+0.64\%$
test_step_mdp_speed[False-False-False-True-False] 73.8120μs 38.6372μs 25.8818 KOps/s 25.8198 KOps/s $\color{#35bf28}+0.24\%$
test_step_mdp_speed[False-False-False-False-True] 68.5220μs 35.9140μs 27.8443 KOps/s 27.5512 KOps/s $\color{#35bf28}+1.06\%$
test_step_mdp_speed[False-False-False-False-False] 52.1110μs 23.6388μs 42.3033 KOps/s 42.1192 KOps/s $\color{#35bf28}+0.44\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8741s 0.7761s 1.2885 Ops/s 1.2900 Ops/s $\color{#d91a1a}-0.12\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7411s 0.6401s 1.5624 Ops/s 1.5764 Ops/s $\color{#d91a1a}-0.89\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7812s 1.7025s 0.5874 Ops/s 0.5910 Ops/s $\color{#d91a1a}-0.62\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5614s 1.4774s 0.6769 Ops/s 0.6813 Ops/s $\color{#d91a1a}-0.65\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0371s 1.9424s 0.5148 Ops/s 0.5177 Ops/s $\color{#d91a1a}-0.56\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7960s 1.7122s 0.5840 Ops/s 0.5832 Ops/s $\color{#35bf28}+0.15\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8120s 4.7054s 0.2125 Ops/s 0.2146 Ops/s $\color{#d91a1a}-0.95\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.5287s 4.4761s 0.2234 Ops/s 0.2228 Ops/s $\color{#35bf28}+0.27\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.1389s 2.0119s 0.4970 Ops/s 0.5002 Ops/s $\color{#d91a1a}-0.64\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7655s 1.6855s 0.5933 Ops/s 0.5975 Ops/s $\color{#d91a1a}-0.70\%$
test_values[generalized_advantage_estimate-True-True] 10.2249ms 10.0226ms 99.7748 Ops/s 98.5750 Ops/s $\color{#35bf28}+1.22\%$
test_values[vec_generalized_advantage_estimate-True-True] 15.1700ms 11.0847ms 90.2142 Ops/s 89.8816 Ops/s $\color{#35bf28}+0.37\%$
test_values[td0_return_estimate-False-False] 0.2233ms 0.1308ms 7.6478 KOps/s 7.6815 KOps/s $\color{#d91a1a}-0.44\%$
test_values[td1_return_estimate-False-False] 27.7485ms 27.4469ms 36.4340 Ops/s 36.2751 Ops/s $\color{#35bf28}+0.44\%$
test_values[vec_td1_return_estimate-False-False] 11.5549ms 11.1344ms 89.8119 Ops/s 90.4547 Ops/s $\color{#d91a1a}-0.71\%$
test_values[td_lambda_return_estimate-True-False] 40.8642ms 40.4071ms 24.7481 Ops/s 24.5502 Ops/s $\color{#35bf28}+0.81\%$
test_values[vec_td_lambda_return_estimate-True-False] 11.3072ms 11.1167ms 89.9547 Ops/s 90.4097 Ops/s $\color{#d91a1a}-0.50\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.0634ms 8.9080ms 112.2586 Ops/s 112.0275 Ops/s $\color{#35bf28}+0.21\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7058ms 1.5235ms 656.3975 Ops/s 673.8151 Ops/s $\color{#d91a1a}-2.58\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4986ms 0.4198ms 2.3822 KOps/s 2.3719 KOps/s $\color{#35bf28}+0.43\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.6214ms 30.4508ms 32.8399 Ops/s 33.5963 Ops/s $\color{#d91a1a}-2.25\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.8803ms 1.7283ms 578.6129 Ops/s 584.3825 Ops/s $\color{#d91a1a}-0.99\%$
test_dqn_speed[False-None] 1.5510ms 1.4087ms 709.8507 Ops/s 697.0869 Ops/s $\color{#35bf28}+1.83\%$
test_dqn_speed[False-backward] 2.0062ms 1.9323ms 517.5071 Ops/s 505.9888 Ops/s $\color{#35bf28}+2.28\%$
test_dqn_speed[True-None] 0.6641ms 0.5270ms 1.8977 KOps/s 1.8690 KOps/s $\color{#35bf28}+1.53\%$
test_dqn_speed[True-backward] 1.0198ms 0.9793ms 1.0211 KOps/s 886.4946 Ops/s $\textbf{\color{#35bf28}+15.19\%}$
test_dqn_speed[reduce-overhead-None] 0.5914ms 0.5215ms 1.9174 KOps/s 1.8259 KOps/s $\textbf{\color{#35bf28}+5.01\%}$
test_dqn_speed[reduce-overhead-backward] 1.0129ms 0.9692ms 1.0318 KOps/s 1.0221 KOps/s $\color{#35bf28}+0.94\%$
test_ddpg_speed[False-None] 3.2502ms 2.8880ms 346.2602 Ops/s 346.8913 Ops/s $\color{#d91a1a}-0.18\%$
test_ddpg_speed[False-backward] 4.2318ms 4.1434ms 241.3479 Ops/s 240.7721 Ops/s $\color{#35bf28}+0.24\%$
test_ddpg_speed[True-None] 1.4638ms 1.3870ms 720.9763 Ops/s 715.6545 Ops/s $\color{#35bf28}+0.74\%$
test_ddpg_speed[True-backward] 2.4431ms 2.3842ms 419.4212 Ops/s 402.3447 Ops/s $\color{#35bf28}+4.24\%$
test_ddpg_speed[reduce-overhead-None] 1.5374ms 1.3837ms 722.6759 Ops/s 715.9904 Ops/s $\color{#35bf28}+0.93\%$
test_ddpg_speed[reduce-overhead-backward] 2.7965ms 2.3821ms 419.7986 Ops/s 422.0703 Ops/s $\color{#d91a1a}-0.54\%$
test_sac_speed[False-None] 8.5752ms 8.0978ms 123.4908 Ops/s 125.1386 Ops/s $\color{#d91a1a}-1.32\%$
test_sac_speed[False-backward] 11.7821ms 11.4178ms 87.5823 Ops/s 89.1174 Ops/s $\color{#d91a1a}-1.72\%$
test_sac_speed[True-None] 2.5561ms 2.1530ms 464.4627 Ops/s 419.9181 Ops/s $\textbf{\color{#35bf28}+10.61\%}$
test_sac_speed[True-backward] 4.3246ms 4.0249ms 248.4531 Ops/s 214.2405 Ops/s $\textbf{\color{#35bf28}+15.97\%}$
test_sac_speed[reduce-overhead-None] 2.6363ms 2.1571ms 463.5781 Ops/s 457.9446 Ops/s $\color{#35bf28}+1.23\%$
test_sac_speed[reduce-overhead-backward] 4.2203ms 4.0734ms 245.4968 Ops/s 233.5381 Ops/s $\textbf{\color{#35bf28}+5.12\%}$
test_redq_speed[False-None] 10.9442ms 10.4933ms 95.2992 Ops/s 95.1434 Ops/s $\color{#35bf28}+0.16\%$
test_redq_speed[False-backward] 18.8483ms 17.9853ms 55.6009 Ops/s 55.6392 Ops/s $\color{#d91a1a}-0.07\%$
test_redq_speed[True-None] 4.7827ms 4.4620ms 224.1159 Ops/s 224.5656 Ops/s $\color{#d91a1a}-0.20\%$
test_redq_speed[True-backward] 10.9206ms 10.0513ms 99.4894 Ops/s 98.8474 Ops/s $\color{#35bf28}+0.65\%$
test_redq_speed[reduce-overhead-None] 4.5976ms 4.4226ms 226.1101 Ops/s 231.9786 Ops/s $\color{#d91a1a}-2.53\%$
test_redq_speed[reduce-overhead-backward] 10.2661ms 10.0707ms 99.2977 Ops/s 98.9536 Ops/s $\color{#35bf28}+0.35\%$
test_redq_deprec_speed[False-None] 11.8590ms 11.2577ms 88.8279 Ops/s 88.8292 Ops/s $-0.00\%$
test_redq_deprec_speed[False-backward] 16.5851ms 16.1771ms 61.8159 Ops/s 62.2405 Ops/s $\color{#d91a1a}-0.68\%$
test_redq_deprec_speed[True-None] 4.3799ms 3.7563ms 266.2171 Ops/s 268.4369 Ops/s $\color{#d91a1a}-0.83\%$
test_redq_deprec_speed[True-backward] 7.8987ms 7.6763ms 130.2705 Ops/s 123.3744 Ops/s $\textbf{\color{#35bf28}+5.59\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.8039ms 3.6472ms 274.1829 Ops/s 269.5907 Ops/s $\color{#35bf28}+1.70\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.6056ms 7.3987ms 135.1588 Ops/s 133.6751 Ops/s $\color{#35bf28}+1.11\%$
test_td3_speed[False-None] 8.2406ms 8.0831ms 123.7152 Ops/s 115.8700 Ops/s $\textbf{\color{#35bf28}+6.77\%}$
test_td3_speed[False-backward] 11.7355ms 11.0715ms 90.3222 Ops/s 90.5026 Ops/s $\color{#d91a1a}-0.20\%$
test_td3_speed[True-None] 1.8641ms 1.8182ms 549.9886 Ops/s 538.3430 Ops/s $\color{#35bf28}+2.16\%$
test_td3_speed[True-backward] 3.8352ms 3.6642ms 272.9145 Ops/s 252.5791 Ops/s $\textbf{\color{#35bf28}+8.05\%}$
test_td3_speed[reduce-overhead-None] 1.9929ms 1.8200ms 549.4411 Ops/s 546.4891 Ops/s $\color{#35bf28}+0.54\%$
test_td3_speed[reduce-overhead-backward] 3.7858ms 3.6959ms 270.5668 Ops/s 260.2408 Ops/s $\color{#35bf28}+3.97\%$
test_cql_speed[False-None] 28.6284ms 26.0193ms 38.4331 Ops/s 38.0794 Ops/s $\color{#35bf28}+0.93\%$
test_cql_speed[False-backward] 38.2362ms 35.3090ms 28.3214 Ops/s 28.1377 Ops/s $\color{#35bf28}+0.65\%$
test_cql_speed[True-None] 12.4564ms 12.0361ms 83.0830 Ops/s 79.1702 Ops/s $\color{#35bf28}+4.94\%$
test_cql_speed[True-backward] 18.1321ms 17.7438ms 56.3576 Ops/s 54.9170 Ops/s $\color{#35bf28}+2.62\%$
test_cql_speed[reduce-overhead-None] 12.6156ms 12.2228ms 81.8142 Ops/s 80.8292 Ops/s $\color{#35bf28}+1.22\%$
test_cql_speed[reduce-overhead-backward] 18.0358ms 17.6738ms 56.5808 Ops/s 53.9597 Ops/s $\color{#35bf28}+4.86\%$
test_a2c_speed[False-None] 5.6916ms 5.4552ms 183.3111 Ops/s 182.3791 Ops/s $\color{#35bf28}+0.51\%$
test_a2c_speed[False-backward] 12.4232ms 12.0224ms 83.1782 Ops/s 82.8883 Ops/s $\color{#35bf28}+0.35\%$
test_a2c_speed[True-None] 4.1649ms 3.7258ms 268.4000 Ops/s 253.0744 Ops/s $\textbf{\color{#35bf28}+6.06\%}$
test_a2c_speed[True-backward] 8.8300ms 8.6507ms 115.5974 Ops/s 113.9113 Ops/s $\color{#35bf28}+1.48\%$
test_a2c_speed[reduce-overhead-None] 4.0874ms 3.7008ms 270.2112 Ops/s 269.4624 Ops/s $\color{#35bf28}+0.28\%$
test_a2c_speed[reduce-overhead-backward] 9.2612ms 8.7540ms 114.2334 Ops/s 113.2331 Ops/s $\color{#35bf28}+0.88\%$
test_ppo_speed[False-None] 6.1918ms 5.8847ms 169.9310 Ops/s 170.5742 Ops/s $\color{#d91a1a}-0.38\%$
test_ppo_speed[False-backward] 13.0448ms 12.7582ms 78.3813 Ops/s 79.9017 Ops/s $\color{#d91a1a}-1.90\%$
test_ppo_speed[True-None] 3.8454ms 3.6070ms 277.2373 Ops/s 268.1925 Ops/s $\color{#35bf28}+3.37\%$
test_ppo_speed[True-backward] 8.6902ms 8.4286ms 118.6433 Ops/s 117.2498 Ops/s $\color{#35bf28}+1.19\%$
test_ppo_speed[reduce-overhead-None] 3.7253ms 3.6006ms 277.7297 Ops/s 276.7242 Ops/s $\color{#35bf28}+0.36\%$
test_ppo_speed[reduce-overhead-backward] 8.9416ms 8.7765ms 113.9409 Ops/s 114.5279 Ops/s $\color{#d91a1a}-0.51\%$
test_reinforce_speed[False-None] 5.6994ms 4.6618ms 214.5115 Ops/s 217.1303 Ops/s $\color{#d91a1a}-1.21\%$
test_reinforce_speed[False-backward] 7.7572ms 7.4723ms 133.8282 Ops/s 132.3859 Ops/s $\color{#35bf28}+1.09\%$
test_reinforce_speed[True-None] 3.0355ms 2.8147ms 355.2780 Ops/s 348.0283 Ops/s $\color{#35bf28}+2.08\%$
test_reinforce_speed[True-backward] 7.8936ms 7.7023ms 129.8316 Ops/s 121.7474 Ops/s $\textbf{\color{#35bf28}+6.64\%}$
test_reinforce_speed[reduce-overhead-None] 3.0746ms 2.8353ms 352.6960 Ops/s 343.7757 Ops/s $\color{#35bf28}+2.59\%$
test_reinforce_speed[reduce-overhead-backward] 8.4325ms 8.0580ms 124.1002 Ops/s 114.7896 Ops/s $\textbf{\color{#35bf28}+8.11\%}$
test_iql_speed[False-None] 24.9255ms 20.5108ms 48.7547 Ops/s 48.9586 Ops/s $\color{#d91a1a}-0.42\%$
test_iql_speed[False-backward] 31.4981ms 30.8179ms 32.4487 Ops/s 32.4098 Ops/s $\color{#35bf28}+0.12\%$
test_iql_speed[True-None] 8.7833ms 8.5427ms 117.0596 Ops/s 109.0226 Ops/s $\textbf{\color{#35bf28}+7.37\%}$
test_iql_speed[True-backward] 17.3085ms 16.8741ms 59.2626 Ops/s 58.3339 Ops/s $\color{#35bf28}+1.59\%$
test_iql_speed[reduce-overhead-None] 9.0784ms 8.6947ms 115.0122 Ops/s 113.4928 Ops/s $\color{#35bf28}+1.34\%$
test_iql_speed[reduce-overhead-backward] 17.6563ms 17.0344ms 58.7048 Ops/s 57.7160 Ops/s $\color{#35bf28}+1.71\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.3050ms 6.0916ms 164.1615 Ops/s 163.3588 Ops/s $\color{#35bf28}+0.49\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5806ms 0.3084ms 3.2430 KOps/s 2.8648 KOps/s $\textbf{\color{#35bf28}+13.20\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5420ms 0.2864ms 3.4910 KOps/s 3.2902 KOps/s $\textbf{\color{#35bf28}+6.10\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1597ms 5.7820ms 172.9513 Ops/s 168.9966 Ops/s $\color{#35bf28}+2.34\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9145ms 0.3195ms 3.1303 KOps/s 2.9301 KOps/s $\textbf{\color{#35bf28}+6.83\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5424ms 0.3137ms 3.1881 KOps/s 3.5251 KOps/s $\textbf{\color{#d91a1a}-9.56\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5967ms 1.3502ms 740.6340 Ops/s 763.2477 Ops/s $\color{#d91a1a}-2.96\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4834ms 1.2621ms 792.3242 Ops/s 810.3815 Ops/s $\color{#d91a1a}-2.23\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0224ms 5.9262ms 168.7414 Ops/s 165.0200 Ops/s $\color{#35bf28}+2.26\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.5865ms 0.4290ms 2.3310 KOps/s 2.1356 KOps/s $\textbf{\color{#35bf28}+9.15\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6381ms 0.4115ms 2.4301 KOps/s 2.1805 KOps/s $\textbf{\color{#35bf28}+11.45\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8722ms 5.7760ms 173.1305 Ops/s 169.8020 Ops/s $\color{#35bf28}+1.96\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8390ms 0.3241ms 3.0851 KOps/s 3.4717 KOps/s $\textbf{\color{#d91a1a}-11.14\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5585ms 0.3248ms 3.0791 KOps/s 3.7596 KOps/s $\textbf{\color{#d91a1a}-18.10\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0242ms 5.6837ms 175.9428 Ops/s 171.1575 Ops/s $\color{#35bf28}+2.80\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7911ms 0.3269ms 3.0591 KOps/s 2.9706 KOps/s $\color{#35bf28}+2.98\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5402ms 0.3058ms 3.2700 KOps/s 3.4528 KOps/s $\textbf{\color{#d91a1a}-5.29\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0340ms 5.9450ms 168.2095 Ops/s 165.1880 Ops/s $\color{#35bf28}+1.83\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9785ms 0.4783ms 2.0908 KOps/s 1.9984 KOps/s $\color{#35bf28}+4.62\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7401ms 0.4405ms 2.2702 KOps/s 2.1069 KOps/s $\textbf{\color{#35bf28}+7.75\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.4178ms 5.0720ms 197.1625 Ops/s 50.8535 Ops/s $\textbf{\color{#35bf28}+287.71\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.2833ms 2.3239ms 430.3146 Ops/s 605.3840 Ops/s $\textbf{\color{#d91a1a}-28.92\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.7314ms 1.1728ms 852.6464 Ops/s 813.5272 Ops/s $\color{#35bf28}+4.81\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.6202s 17.3627ms 57.5949 Ops/s 193.5124 Ops/s $\textbf{\color{#d91a1a}-70.24\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.8682ms 1.7741ms 563.6683 Ops/s 476.6078 Ops/s $\textbf{\color{#35bf28}+18.27\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 10.7738ms 1.2699ms 787.4779 Ops/s 836.3673 Ops/s $\textbf{\color{#d91a1a}-5.85\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.2055ms 5.2190ms 191.6062 Ops/s 188.5746 Ops/s $\color{#35bf28}+1.61\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 7.9463ms 2.2187ms 450.7139 Ops/s 446.7157 Ops/s $\color{#35bf28}+0.90\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.2806ms 1.3840ms 722.5375 Ops/s 816.7115 Ops/s $\textbf{\color{#d91a1a}-11.53\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.9369ms 33.7456ms 29.6335 Ops/s 29.0704 Ops/s $\color{#35bf28}+1.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.3223ms 17.5755ms 56.8972 Ops/s 56.8342 Ops/s $\color{#35bf28}+0.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 37.0772ms 34.7873ms 28.7461 Ops/s 27.8808 Ops/s $\color{#35bf28}+3.10\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.2526ms 17.7806ms 56.2411 Ops/s 32.3166 Ops/s $\textbf{\color{#35bf28}+74.03\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 38.9046ms 36.6085ms 27.3160 Ops/s 26.6341 Ops/s $\color{#35bf28}+2.56\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.8785ms 19.4166ms 51.5024 Ops/s 50.8077 Ops/s $\color{#35bf28}+1.37\%$

@vmoens
Update
d91e2c2 
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Jan 8, 2026
@vmoens
[BugFix] Add pybind11 check and Windows extension pattern fix
62f347a 
- Add early check for pybind11 dependency with clear error message
- Fix extension file pattern for Windows (.pyd) vs Unix (.so) in stale build cleanup

ghstack-source-id: 09e4bfd
Pull-Request: #3310
vmoens added a commit that referenced this pull request Jan 8, 2026
@vmoens
[BugFix] Add pybind11 check and Windows extension pattern fix
5e89e4e 
- Add early check for pybind11 dependency with clear error message
- Fix extension file pattern for Windows (.pyd) vs Unix (.so) in stale build cleanup

ghstack-source-id: 09e4bfd
Pull-Request: #3310
@vmoens vmoens merged commit d91e2c2 into gh/vmoens/188/base Jan 8, 2026
91 of 106 checks passed
@vmoens vmoens deleted the gh/vmoens/188/head branch January 8, 2026 14:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@vmoens