Fix flexattention not realizing inputs before lowering (also refactored runtime estimation)

[Chillee]

Stack from ghstack (oldest at bottom):

• Fix silu test for flexattention #126641
• Prevent partitioner from ever saving views #126446
• -> Fix flexattention not realizing inputs before lowering (also refactored runtime estimation) #126615

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/126615

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

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

❌ 3 New Failures, 9 Unrelated Failures

As of commit 097ff9c with merge base 7dae7d3 ():

NEW FAILURES - The following jobs have failed:

• inductor-periodic / cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks / test (aot_eager_timm, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
  No space left on device : '/home/ec2-user/actions-runner/_work/_actions/pytorch/pytorch/main/torch/csrc/autograd/autograd_not_implemented_fallback.h'
• inductor-periodic / cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks / test (aot_eager_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
  No space left on device
• inductor-periodic / cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks / test (dynamic_aot_eager_timm, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
  No space left on device : '/home/ec2-user/actions-runner/_work/_actions/pytorch/pytorch/main/torch/csrc/autograd/autograd_not_implemented_fallback.h'

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• inductor / linux-jammy-cpu-py3.8-gcc11-inductor / test (inductor_torchbench_cpu_smoketest_perf, 1, 1, linux.24xl.spr-metal) (gh) (similar failure)
  Process completed with exit code 1.

BROKEN TRUNK - The following jobs failed but was present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• inductor-periodic / cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks / test (dynamic_aot_eager_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (trunk failure)
  hf_BigBird
• inductor-periodic / cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks / test (dynamo_eager_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (trunk failure)
  hf_BigBird

UNSTABLE - The following jobs failed but were likely due to flakiness present on trunk and has been marked as unstable:

• inductor / cuda12.1-py3.10-gcc9-sm86 / test (aot_inductor_huggingface, 1, 1, linux.g5.4xlarge.nvidia.gpu, unstable) (gh) ()
  No space left on device
• inductor / cuda12.1-py3.10-gcc9-sm86 / test (aot_inductor_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu, unstable) (gh) ()
  hf_BigBird
• inductor / cuda12.1-py3.10-gcc9-sm86 / test (dynamic_inductor_huggingface, 1, 1, linux.g5.4xlarge.nvidia.gpu, unstable) (gh) ()
  No space left on device : '/home/ec2-user/actions-runner/_work/_actions/pytorch/pytorch/main/torch/csrc/api/include/torch/nn/options/transformerlayer.h'
• inductor / cuda12.1-py3.10-gcc9-sm86 / test (dynamic_inductor_timm, 1, 2, linux.g5.4xlarge.nvidia.gpu, unstable) (gh) ()
  No space left on device
• inductor / cuda12.1-py3.10-gcc9-sm86 / test (dynamic_inductor_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu, unstable) (gh) ()
  hf_BigBird
• inductor / cuda12.1-py3.10-gcc9-sm86 / test (inductor_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu, unstable) (gh) ()
  hf_BigBird

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

[drisspg]

The flex changes look good, the other looks like a lot of code motion

[xmfan]

could we cleanup the now unused globals from torch/_inductor/metrics.py and in reset()

[DanilBaibak]

@pytorchbot revert -m "Break internal build" -c ghfirst

[pytorchmergebot]

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

[pytorchmergebot]

@Chillee your PR has been successfully reverted.

[DanilBaibak]

Hi @Chillee! Sorry, I need to revert your PR because it breaks the internal a lot of inductor tests (triton_kernels and pattern_matcher). Here you can find more information about the issues - D57619189.