Skip to content
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.

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

Configure compiler to generate 'wasm' executables using wasm-llvm-aot. #5079

Merged
merged 2 commits into from
Mar 12, 2021

Conversation

ScottTodd
Copy link
Member

Progress on #4019

There's still more work to do with build/test configurations and filtering, but this at least gets us to the point where IREE's LLVM AOT compiler path can generate flatbuffers suitable for a wasm runtime HAL to handle.

I'm using these flags:

  • compiler: -iree-hal-target-backends=wasm-llvm-aot -iree-llvm-target-triple=wasm32-unknown-unknown
  • runtime: --driver=wasm

@ScottTodd ScottTodd added infrastructure/cmake codegen/llvm LLVM code generation compiler backend labels Mar 12, 2021
@ScottTodd ScottTodd requested a review from benvanik March 12, 2021 01:01
@google-cla google-cla bot added the cla: yes label Mar 12, 2021
@ScottTodd ScottTodd merged commit 0668b55 into iree-org:main Mar 12, 2021
@ScottTodd ScottTodd deleted the wasm-compiler branch March 12, 2021 19:17
copybara-service bot pushed a commit that referenced this pull request Mar 12, 2021
* 161f932 Enable CUDA tests on Turing platforms (#5091)
* e16759c [CodeGen] Avoid eliminating output store unconditionally (#5073)
* e2fe13e Add wasm-micro-runtime submodule and get building with CMake. (#5077)
* 33c4e39 Bump Tracy submodule to v0.7.6. (#5094)
* 3f6335f Use correct tied operand index when creating DispatchWorkgroupsOp (#5090)
* e5cfa53 Allow all matmul variants to be a root op (#5065)
* a5a0f1d Merge google -> main (#5086)
* 1ef76ba Merge pull request #5083 from google/benvanik-flow-passes
* 0668b55 Configure compiler to generate 'wasm' executables using wasm-llvm-aot. (#5079)
* 6d41b9d Create InitTensorOp based on input operand shapes. (#5070)
* bb819df Shuffling the SIP reflection attrs pass around a bit. This is in preparation f..
* 6b63dac Fixing deprecation warning.
* 5c5d2aa Removing old identify dispatch regions. Not yet the new old identify dispatch ..
* 80cdcce Splitting out input passes from flow passes. Eventually the TOSA/HLO stuff wil..
* a2a464d Add missing CMake dep (#5081)
* ca29d25 [CodeGen] NFC: Rename LinalgTileAndFusePass (#5075)
* f3d7654 Add MobileNetV2 to benchmarking targets. (#5064)
* 8aace66 Rename CMake target DYLIB-LLVM-AOT to just LLVM-AOT.
* 98f2135 Add WASM executable format and produce it with LLVMAOTTarget.
* 2795092 Add CMake configuration for WAMR's runtime "iwasm VM core".
* 2b5c7b9 Add third_party/wasm-micro-runtime submodule.

COPYBARA_INTEGRATE_REVIEW=#5097 from ThomasRaoux:main-to-google 161f932
PiperOrigin-RevId: 362617674
iree-github-actions-bot pushed a commit that referenced this pull request Mar 12, 2021
* 161f932 Enable CUDA tests on Turing platforms (#5091)
* e16759c [CodeGen] Avoid eliminating output store unconditionally (#5073)
* e2fe13e Add wasm-micro-runtime submodule and get building with CMake. (#5077)
* 33c4e39 Bump Tracy submodule to v0.7.6. (#5094)
* 3f6335f Use correct tied operand index when creating DispatchWorkgroupsOp (#5090)
* e5cfa53 Allow all matmul variants to be a root op (#5065)
* a5a0f1d Merge google -> main (#5086)
* 1ef76ba Merge pull request #5083 from google/benvanik-flow-passes
* 0668b55 Configure compiler to generate 'wasm' executables using wasm-llvm-aot. (#5079)
* 6d41b9d Create InitTensorOp based on input operand shapes. (#5070)
* bb819df Shuffling the SIP reflection attrs pass around a bit. This is in preparation f..
* 6b63dac Fixing deprecation warning.
* 5c5d2aa Removing old identify dispatch regions. Not yet the new old identify dispatch ..
* 80cdcce Splitting out input passes from flow passes. Eventually the TOSA/HLO stuff wil..
* a2a464d Add missing CMake dep (#5081)
* ca29d25 [CodeGen] NFC: Rename LinalgTileAndFusePass (#5075)
* f3d7654 Add MobileNetV2 to benchmarking targets. (#5064)
* 8aace66 Rename CMake target DYLIB-LLVM-AOT to just LLVM-AOT.
* 98f2135 Add WASM executable format and produce it with LLVMAOTTarget.
* 2795092 Add CMake configuration for WAMR's runtime "iwasm VM core".
* 2b5c7b9 Add third_party/wasm-micro-runtime submodule.

PiperOrigin-RevId: 362617674
ThomasRaoux added a commit that referenced this pull request Mar 13, 2021
* 161f932 Enable CUDA tests on Turing platforms (#5091)
* e16759c [CodeGen] Avoid eliminating output store unconditionally (#5073)
* e2fe13e Add wasm-micro-runtime submodule and get building with CMake. (#5077)
* 33c4e39 Bump Tracy submodule to v0.7.6. (#5094)
* 3f6335f Use correct tied operand index when creating DispatchWorkgroupsOp (#5090)
* e5cfa53 Allow all matmul variants to be a root op (#5065)
* a5a0f1d Merge google -> main (#5086)
* 1ef76ba Merge pull request #5083 from google/benvanik-flow-passes
* 0668b55 Configure compiler to generate 'wasm' executables using wasm-llvm-aot. (#5079)
* 6d41b9d Create InitTensorOp based on input operand shapes. (#5070)
* bb819df Shuffling the SIP reflection attrs pass around a bit. This is in preparation f..
* 6b63dac Fixing deprecation warning.
* 5c5d2aa Removing old identify dispatch regions. Not yet the new old identify dispatch ..
* 80cdcce Splitting out input passes from flow passes. Eventually the TOSA/HLO stuff wil..
* a2a464d Add missing CMake dep (#5081)
* ca29d25 [CodeGen] NFC: Rename LinalgTileAndFusePass (#5075)
* f3d7654 Add MobileNetV2 to benchmarking targets. (#5064)
* 8aace66 Rename CMake target DYLIB-LLVM-AOT to just LLVM-AOT.
* 98f2135 Add WASM executable format and produce it with LLVMAOTTarget.
* 2795092 Add CMake configuration for WAMR's runtime "iwasm VM core".
* 2b5c7b9 Add third_party/wasm-micro-runtime submodule.

PiperOrigin-RevId: 362617674
@GMNGeoffrey GMNGeoffrey added the infrastructure Relating to build systems, CI, or testing label May 11, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
codegen/llvm LLVM code generation compiler backend infrastructure Relating to build systems, CI, or testing
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants