[Backend Tester] Increase linux runner size

[GregoryComer]

QNN test jobs are failing with what looks like resource utilization issues since the update. 8x size hosts seem to resolve it.

[GregoryComer]

Stack from ghstack (oldest at bottom):

• -> [Backend Tester] Increase linux runner size #14200
• [Backend Tester] Add ARM TOSA flow #14190
• [Backend Tester] Skip tests with undelegated conv3d ops #14185
• [Backend Tester] Refactor workflows into one workflow per backend #14137
• [Backend Tester] Add flow-specific skipped tests, skip argmin/max on Core ML due to native crashes, float16/64 on Vulkan #13992
• [Backend Tester] Add error message field to CSV report #13991
• [Backend Tester] Skip inf and nan operator tests #13990
• [Backend Tester] Skip in-place activation tests due to lack of support in ET #13989

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14200

• 📄 Preview Python docs built from this PR

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

❌ 6 New Failures, 3 Cancelled Jobs, 2 Unrelated Failures

As of commit 0fedd15 with merge base ed0ab94 ():

NEW FAILURES - The following jobs have failed:

• pull / test-llama-runner-linux (fp32, xnnpack+custom+quantize_kv, linux.2xlarge, executorch-ubuntu-22.04... / linux-job (gh)
  RuntimeError: Command docker exec -t 8040b5af9d004f5666ede2d86f25fac13a2f2c1bb5779a6f329909ee9204f7e3 /exec failed with exit code 1
• pull / test-llama-runner-linux (fp32, xnnpack+quantize_kv, linux.2xlarge, executorch-ubuntu-22.04-clang12) / linux-job (gh)
  RuntimeError: Command docker exec -t 8f8d5e2b8fb0b658749c616e82e4b325003b67d63626fc544e0512dc32219565 /exec failed with exit code 1
• pull / test-llava-runner-linux / linux-job (gh)
  RuntimeError: Command docker exec -t 4ab9248e0ec54c366a619ebea703d24b33a1f7bc42a28ca43044e8e4301920b5 /exec failed with exit code 139
• pull / test-models-linux (mv2, xnnpack-quantization-delegation, linux.2xlarge) / linux-job (gh)
  RuntimeError: Command docker exec -t 0af1612b39201660351497a5a030fea31bd1d33d5e8a05ebae506a480e5d725c /exec failed with exit code 1
• pull / unittest / macos / macos-job (gh)
  backends/xnnpack/test/ops/test_conv2d.py::TestConv2d::test_fp32_conv2d
• pull / unittest-editable / macos / macos-job (gh)
  backends/xnnpack/test/ops/test_sub.py::TestSub::test_add_with_alpha

CANCELLED JOBS - The following jobs were cancelled. Please retry:

• pull / test-openvino-linux / linux-job (gh)
  ##[error]The operation was canceled.
• Test ARM Backend / test-arm / test-backend-linux (arm_tosa, models) / linux-job (gh)
  ##[error]The operation was canceled.
• Test CoreML Backend / test-coreml / test-backend-macos (coreml_static_int8, operators) / macos-job (gh)
  ##[error]The operation was canceled.

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

• pull / test-models-linux (resnet50, portable, linux.2xlarge) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
• pull / test-models-linux-basic (mv3, xnnpack-quantization-delegation, buck2, linux.2xlarge, executorch-u... / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

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

[digantdesai]

should we do it only for QNN?