Skip to content
CI Windows

[WIP] Vectorized convolution with larger transfer reads #1368

Sign in to view logs

[WIP] Vectorized convolution with larger transfer reads

[WIP] Vectorized convolution with larger transfer reads #1368

Triggered via pull request October 30, 2024 22:43
@newlingnewling
opened #867
newling:vectorized_convolution_with_larger_transfer_reads
Status Failure
Total duration 17m 25s
Artifacts 2

ci-windows.yml

on: pull_request
Build and Test (windows, ASSERTIONS)
17m 8s
Build and Test (windows, ASSERTIONS)
E2E Test windows
0s
E2E Test windows
Fit to window
Zoom out
Zoom in

Annotations

10 errors and 11 warnings
Build and Test (windows, ASSERTIONS)
'linalg.fill' op needs logic implemented for handling.
Build and Test (windows, ASSERTIONS)
failed to legalize operation 'aievec.shift' that was explicitly marked illegal
Build and Test (windows, ASSERTIONS)
failed to run translation of source executable to target executable for backend #hal.executable.target<"amd-aie", "amdaie-pdi-fb", {target_device = "npu1_4col", ukernels = "none"}>
Build and Test (windows, ASSERTIONS)
failed to legalize operation 'aievec.shift' that was explicitly marked illegal
Build and Test (windows, ASSERTIONS)
failed to run translation of source executable to target executable for backend #hal.executable.target<"amd-aie", "amdaie-pdi-fb", {target_device = "npu1_4col", ukernels = "none"}>
Build and Test (windows, ASSERTIONS)
'linalg.fill' op needs logic implemented for handling.
Build and Test (windows, ASSERTIONS)
failed to legalize operation 'aievec.shift' that was explicitly marked illegal
Build and Test (windows, ASSERTIONS)
failed to run translation of source executable to target executable for backend #hal.executable.target<"amd-aie", "amdaie-pdi-fb", {target_device = "npu1_4col", ukernels = "none"}>
Build and Test (windows, ASSERTIONS)
failed to legalize operation 'aievec.shift' that was explicitly marked illegal
Build and Test (windows, ASSERTIONS)
failed to run translation of source executable to target executable for backend #hal.executable.target<"amd-aie", "amdaie-pdi-fb", {target_device = "npu1_4col", ukernels = "none"}>
Build and Test (windows, ASSERTIONS)
The following actions use a deprecated Node.js version and will be forced to run on node20: actions/setup-python@v4, actions/cache/restore@v3. For more info: https://github.blog/changelog/2024-03-07-github-actions-all-actions-will-run-on-node20-instead-of-node16-by-default/
Build and Test (windows, ASSERTIONS): compiler/plugins/target/AMD-AIE/aievec/AIEVecTypeConstraints.td#L149
unused template argument: isOperandResultTypePairValidForAIE2MulElem:rhs
Build and Test (windows, ASSERTIONS): compiler/plugins/target/AMD-AIE/aievec/AIEVecTypeConstraints.td#L149
unused template argument: isOperandResultTypePairValidForAIE2MulElem:rhs
Build and Test (windows, ASSERTIONS): compiler/plugins/target/AMD-AIE/aievec/AIEVecTypeConstraints.td#L149
unused template argument: isOperandResultTypePairValidForAIE2MulElem:rhs
Build and Test (windows, ASSERTIONS): compiler/plugins/target/AMD-AIE/aievec/AIEVecTypeConstraints.td#L149
unused template argument: isOperandResultTypePairValidForAIE2MulElem:rhs
Build and Test (windows, ASSERTIONS)
aievec.shift conversion with result vector size 256 is not implemented.
Build and Test (windows, ASSERTIONS)
aievec.shift conversion with result vector size 256 is not implemented.
Build and Test (windows, ASSERTIONS)
aievec.shift conversion with result vector size 256 is not implemented.
Build and Test (windows, ASSERTIONS)
aievec.shift conversion with result vector size 256 is not implemented.
Build and Test (windows, ASSERTIONS)
aievec.shift conversion with result vector size 256 is not implemented.
Build and Test (windows, ASSERTIONS)
aievec.shift conversion with result vector size 256 is not implemented.

Artifacts

Produced during runtime
Name Size
windows_x86_64_iree_packages
359 MB
windows_x86_64_llvm_packages
663 MB