You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now the sequencer translation always lowers to bytecode. We should be able to emit C code or LLVM IR just as easily from the low-level sequencer dialect. Whether these targets are represented as new dialects (such as LLVM IR) or directly as a serialization mechanism (bytecode, C, etc) varies, so we should probably have both. This may mainly consist of splitting serialization from the translation (as the SPIR-V dialect is split).
The text was updated successfully, but these errors were encountered:
Prior to the changes all `_` in the path got replaced with `/`:
Before:
```
Test #30: iree/test/e2e/linalg/path:pw_add.mlir.test
Test #31: iree/test/e2e/linalg/path:pw_add_mul.mlir.test
Test #32: iree/test/e2e/linalg/path:pw_add_multiwg.mlir.test
Test #33: iree/test/e2e/linalg/path:reduce.mlir.test
Test #268: iree/samples/custom/modules:custom_modules_test
Test #269: iree/samples/custom/modules/dialect/test:conversion.mlir.test
Test #270: iree/samples/custom/modules/dialect/test:custom_ops.mlir.test
Test #271: iree/samples/simple/embedding:simple_embedding_test
```
After:
```
Test #30: iree/test/e2e/linalg_path:pw_add.mlir.test
Test #31: iree/test/e2e/linalg_path:pw_add_mul.mlir.test
Test #32: iree/test/e2e/linalg_path:pw_add_multiwg.mlir.test
Test #33: iree/test/e2e/linalg_path:reduce.mlir.test
Test #268: iree/samples/custom_modules:custom_modules_test
Test #269: iree/samples/custom_modules/dialect/test:conversion.mlir.test
Test #270: iree/samples/custom_modules/dialect/test:custom_ops.mlir.test
Test #271: iree/samples/simple_embedding:simple_embedding_test
```
Closes#1360
COPYBARA_INTEGRATE_REVIEW=#1360 from iml130:cmake-test-labels 34dc25d
PiperOrigin-RevId: 304524688
Right now the sequencer translation always lowers to bytecode. We should be able to emit C code or LLVM IR just as easily from the low-level sequencer dialect. Whether these targets are represented as new dialects (such as LLVM IR) or directly as a serialization mechanism (bytecode, C, etc) varies, so we should probably have both. This may mainly consist of splitting serialization from the translation (as the SPIR-V dialect is split).
The text was updated successfully, but these errors were encountered: