Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
171 commits
Select commit Hold shift + click to select a range
4dac7bc
Begin working on IndexTree transformations
AK2000 Nov 1, 2023
20f916a
Fixing some of the problems introduced on merge
AK2000 Nov 1, 2023
864c9f8
Resolved including of device mapping attribute
AK2000 Nov 1, 2023
0447561
Fixed type inclusion, parsing and printing
AK2000 Nov 1, 2023
cf505bd
V1 - Lower TA to new IndexTree ops, but removed everything else
AK2000 Nov 6, 2023
3f4e849
Fixes to TA to change how file is included
AK2000 Nov 6, 2023
d6bc7fd
Creating new block for index tree
AK2000 Nov 8, 2023
48fe5fd
Implement domain inference pass, fix to index ordering
AK2000 Nov 10, 2023
2d6ba90
[WIP] Fragile version of index tree to SCF lowering
AK2000 Nov 30, 2023
972ee7e
Fix carrying tensors inside loop, refactor domain concretization
AK2000 Dec 12, 2023
c29ee1f
Adding TA to index tree patterns for elementwise operations
AK2000 Dec 13, 2023
8d44653
[WIP] Trying to implement intersection op lowering
AK2000 Dec 19, 2023
f25fb52
[WIP] Got domain intersection working, but only with dense output
AK2000 Jan 3, 2024
35cf1c5
[WIP] Minor fix to ordering of reduce args
AK2000 Jan 3, 2024
04b31ba
[WIP] Beginning support for sparse output tensors with new index tree
AK2000 Jan 10, 2024
1c06a26
[WIP] Inlined itree op, got hacky version of removing set op working
AK2000 Jan 11, 2024
00801b1
[WIP] Included lowering to LLVM, lowering print op does not work
AK2000 Jan 12, 2024
352da6a
[WIP] Almost got print op lowering working
AK2000 Jan 15, 2024
9436939
[WIP] Fixed bufferization
AK2000 Jan 18, 2024
ac98adf
[WIP] Generate symbolic pass for sparse tensor declarations
AK2000 Jan 24, 2024
935079d
[WIP] Lots of changes for first try at symbolic domain pass and works…
AK2000 Feb 13, 2024
0c2d98d
[WIP] Broke everythink trying to redo tensor conversion infrastructure
AK2000 Feb 21, 2024
bb32d70
Changing alot to create new sparse tensor types, and appropriate lowe…
AK2000 May 15, 2024
ac27cfb
Fixing some problems with tests, ad pure ops
AK2000 Jun 5, 2024
52495d4
Fixed inconsistencies in test suite
AK2000 Jun 5, 2024
27dd373
Fixing more of the test cases
AK2000 Jun 13, 2024
8bc6187
Fixed dense transpose and print elapsed time
AK2000 Jun 17, 2024
3f777dd
Fixing errors in typing
AK2000 Jun 19, 2024
8cb0d2a
Fixing errors in typing and set op
AK2000 Jun 19, 2024
dc0e445
Adding back ttgt pass
AK2000 Jun 20, 2024
ffe69f2
Fixed delete before use errors
AK2000 Oct 14, 2024
1a7eb01
Another bug found with asan
AK2000 Oct 14, 2024
c32a1cd
Rebased pull request with master
pthomadakis Oct 15, 2024
2153358
1) Fixed minor bugs in IndexTreeToSCF conversion
pthomadakis Oct 16, 2024
e0c95d7
Fixed another use-after-erase bug
pthomadakis Oct 17, 2024
ebc2efc
Fixed some more bugs coming from merging
pthomadakis Oct 17, 2024
5ac3316
Enforce tensor semantics on scalar op
AK2000 Oct 17, 2024
cdfa968
Added semiring operations to work correct assumption of multiplicativ…
AK2000 Oct 29, 2024
83660c7
Merge pull request #67 from AK2000/fix-scalar-op
AK2000 Oct 29, 2024
2705351
Merge pull request #68 from AK2000/fix-semiring-tests
AK2000 Oct 29, 2024
aad9afc
[WIP] Enabling function support
pthomadakis Oct 25, 2024
870b0e6
[WIP] Enabling function support. Dense transpose compound expressions…
pthomadakis Oct 29, 2024
8dd185f
[WIP] Enabling function support. Made transpose operation a littel mo…
pthomadakis Oct 29, 2024
27eacee
Refactored SpConstructOp
pthomadakis Oct 30, 2024
cc73ecd
Fixed func.return when returning SparseTensorType
pthomadakis Oct 30, 2024
aaca5f9
Changed ta.sum to return a scalar instead of a memref.
pthomadakis Nov 1, 2024
e565571
Decouple semiring attribute from domain inference
AK2000 Nov 5, 2024
5b66b30
"fixed" triangle counting test :)
AK2000 Nov 5, 2024
6efa6e3
Merge pull request #72 from AK2000/fix-semiring-tests
gkestor Nov 6, 2024
e4a987a
Merge pull request #73 from AK2000/fix-triangle-counting
gkestor Nov 6, 2024
880c9cc
[COMETPY] Fixed issue with COO formats
pthomadakis Nov 7, 2024
de6469d
[COMETPY] Update tests to make use of pytest and removed custom test …
pthomadakis Nov 7, 2024
5cfd397
[COMETPY] Added data directory to tests
pthomadakis Nov 7, 2024
bfd5993
[COMETPY] Added missing file
pthomadakis Nov 7, 2024
f4fe43a
Added support for f32 data #69
pthomadakis Nov 1, 2024
836f8ce
Updated numpy-scipy side to support f32 #69
pthomadakis Nov 2, 2024
664a7df
Updated SparseTensor type to use the respective enum instead of raw n…
pthomadakis Nov 7, 2024
878c711
Sparse tensor indices crd(_tile),pos(_tile) now have an explicit inte…
pthomadakis Nov 9, 2024
89ee8c4
[COMETPY][WIP] Enabling i32 indices for sparse matrices
pthomadakis Nov 10, 2024
001cf29
Fixed bugs in Index to Integer casts and vice-versa
pthomadakis Nov 10, 2024
6aacaa5
[COMETPY] Finished support for i32,i64 sparse matrix indices.
pthomadakis Nov 10, 2024
978b77c
[WIP] reviewing new index tree dialect and related test cases (added …
gkestor Oct 25, 2024
ac744a9
blis_interface is updated - debug prints removed
gkestor Oct 29, 2024
304f20e
Bug is fixed in the dense transpose optimization. Testcases are organ…
gkestor Nov 13, 2024
4433b75
Fixed path in CMakeLists.txt
pthomadakis Nov 15, 2024
33621c7
Changing how index tree to SCF conversion pass is structured
AK2000 Aug 19, 2024
5557cb3
[WIP] adding semiring zero
AK2000 Nov 7, 2024
4176e61
[WIP] Symbolic pass and workspace transform not working
AK2000 Nov 11, 2024
814240f
Fixed issues with workspace transformation and symbolic pass
AK2000 Nov 11, 2024
5112041
Resolved conflicts with restructure SCF conversion
pthomadakis Nov 15, 2024
a25befd
[COMETPY] Updated sparse tensor representation
pthomadakis Nov 17, 2024
4253d33
[WIP] Got masking partially working but broke other things
AK2000 Nov 18, 2024
b93b044
[WIP] Fixed other problems, but still issues with Sandia_LL with masking
AK2000 Nov 18, 2024
d3551f4
Helped compilation for SandiaLL w Masking
AK2000 Nov 18, 2024
28f5cf7
Merge pull request #76 from AK2000/rewrite_masking
AK2000 Nov 20, 2024
b0fa0ae
Rearranging comet.cpp to emit it at different stage
AK2000 Dec 3, 2024
3ed0cf5
Initial implementation of parallel loops (#77)
AK2000 Dec 10, 2024
3081c87
Minors updates: added dump, trimmed whitespaces, reindented
johnpzh Nov 6, 2024
a4fe9d5
minors change for formats and dumping
johnpzh Dec 3, 2024
729b74a
[WIP] add KernelFusion pass but have errors
johnpzh Dec 3, 2024
6ce9bb7
Fixing build errors caused by dependent dialects
AK2000 Dec 3, 2024
c0674b0
[WIP] Handled a constant RHS operand in IndexTreeToSCF
johnpzh Dec 13, 2024
adeea5b
[WIP] Handle a constant LSH operand.
johnpzh Dec 14, 2024
45b3d5b
Initial version of index tree with proper use of block arguments for …
AK2000 Jan 8, 2025
d3eeef2
1st version of Fusion with new Index Tree dialect.
johnpzh Feb 5, 2025
d65b183
First fix of inferring domain of intermeidate tensors
AK2000 Feb 10, 2025
9301d10
Fixing simplifying union domain with same operands
AK2000 Feb 10, 2025
328c053
Change how reduction varaibles are identified, fixed SPDMM
AK2000 Feb 12, 2025
c3936d8
Fixed small change in creating union op
AK2000 Feb 12, 2025
58a03b4
[WIP] divide fusion into two passes
johnpzh Feb 18, 2025
f3699d0
Adding sort to workspace
AK2000 Feb 25, 2025
2afc1e8
Adding loop invariant code motion
AK2000 Feb 25, 2025
686636b
Adding optimizations to workspace insert and read
AK2000 Feb 25, 2025
c348dc3
1st version Dimension Recudtion pass.
johnpzh Feb 27, 2025
4b1e50f
Adding masked domain to support different means of intersection
AK2000 Mar 3, 2025
c04eaf4
Moving bit tensor decleration to outside of index tree
AK2000 Mar 10, 2025
c57f5e7
Modified affine_map for tensor ops in TensorAlgebra
johnpzh Mar 14, 2025
a33f8c5
Fixing crash during gnn dimension reduction
AK2000 Mar 17, 2025
1bd440a
Modified positions of DomainConcretizationPass and WorkspaceTransform…
johnpzh Mar 18, 2025
132cd5e
Updated DimensionReduction pass's resetting
johnpzh Mar 19, 2025
c58232d
Fixed inconsistencies from cherry-picking
pthomadakis Mar 22, 2025
3bfabf4
[WIP] Issue with parallel for
pthomadakis Mar 22, 2025
c41f83b
[WIP] FIxing the parallel for issues
pthomadakis Mar 22, 2025
22011f3
Fixed issue with scf.forall
pthomadakis Mar 22, 2025
cf4497a
Fixed blis_interface.cpp for BlueSky
johnpzh Mar 24, 2025
f1c1044
Update Ubuntu-latest.yml
pthomadakis Mar 24, 2025
2a31872
Added debug option debug-ta-labels-alphabet-order
johnpzh Mar 25, 2025
2fcb1d6
Upgraded llvm to version 18.1.8 in order to prevent llvm execution en…
pthomadakis Mar 26, 2025
6c6daf9
Reverted upgrade of LLVM as it mismatched Triton's version
pthomadakis Mar 26, 2025
7d22b8c
[WIP] Reenabling support for GPU targets.
pthomadakis Jan 24, 2025
b0e727a
Reenabled GPU target support for matrix multiplication (dense)
pthomadakis Jan 24, 2025
3a8ea51
[WIP] Reenabling FPGA target
pthomadakis Feb 15, 2025
2e8ce5a
[WIP] Added script for SPIRV binary to .xclbin conversion
pthomadakis Feb 24, 2025
5a59dc8
Added submodules needed for FPGA lowering in tools directory
pthomadakis Feb 27, 2025
8833902
Added README for tools directory
pthomadakis Feb 27, 2025
a165355
[WIP] Working version of FPGA backend both for dense and sparse compu…
pthomadakis Mar 4, 2025
162b7f2
[WIP] Made FPGA target support optional during compilation.
pthomadakis Mar 4, 2025
ef17f7d
[WIP] Removed MinUIOp from FPGA lowering path (was there from Triton …
pthomadakis Mar 5, 2025
bb2cdf0
[WIP] Added if conditions (guards) for fpga block sizes to fit within…
pthomadakis Mar 5, 2025
929cfdd
Fixed failure on chained spmm+dense_mm operation
pthomadakis Mar 6, 2025
98b3098
Fixed dense transpose optimization to work with tensors and in cometp…
pthomadakis Mar 6, 2025
3fcef45
[CPU] Reenabled support for parallel CPU execution using OpenMP
pthomadakis Mar 21, 2025
b3ddcfd
[CPU] Added missing files
pthomadakis Mar 21, 2025
34a3da4
[CPU] Added OpenMP linkage to test files
pthomadakis Mar 21, 2025
2e6d3ef
Handle --convert-to-triton correctly
pthomadakis Mar 22, 2025
b93caa2
1. Link with OpenMP library in cometpy.
pthomadakis Mar 24, 2025
7cef31f
Removed debugging prints.
pthomadakis Mar 24, 2025
838b995
Updated test file
pthomadakis Mar 24, 2025
368bd3b
Update Ubuntu-latest.yml
pthomadakis Mar 25, 2025
f296395
Update Ubuntu-latest.yml
pthomadakis Mar 25, 2025
ed23971
Parallel execution does not work properly when also using fusion, so …
pthomadakis Mar 25, 2025
f72e1bd
Changed default values for block size when GPU, FPGA co-exist
pthomadakis Mar 26, 2025
aff03bc
Fix for ccsd optimizations
pthomadakis Sep 20, 2024
a3dbf01
Corrected: Fix for ccsd optimizations in cometpy
pthomadakis Sep 20, 2024
a8d0de0
Fixed loop issue in cometpy
pthomadakis Sep 21, 2024
a20879d
Changed print_flops and get_time function declaration to only be incl…
pthomadakis Mar 26, 2025
bd6e9c0
[COMETPY] Distinguish kernels of same name but different optimization…
pthomadakis Sep 22, 2024
331cab4
[GPU] Fixed issue where not all gpu kernels would be converted to triton
pthomadakis Mar 26, 2025
b83d329
Fixed dimension Z in GPU
pthomadakis Sep 23, 2024
756933b
[GPU] Fixed performance issue
pthomadakis Mar 26, 2025
4c85ca3
Enable lowering code to GPU even if not CUDA device is found
pthomadakis Oct 10, 2024
6df6b12
Improved Triton backend to code generation to create a separate modul…
pthomadakis Oct 10, 2024
104d554
1. Added the ability to lower GPU code to either binary (cubin) or as…
pthomadakis Oct 11, 2024
47d2804
Fixed typos and some error messages
pthomadakis Oct 11, 2024
e212c9c
[GPU] Fixed test case "Dense_chain_mult_matrix.ta" to correctly targ…
pthomadakis Mar 26, 2025
3b73299
[GPU] Make sure CUDA context is destroyed before exiting
pthomadakis Mar 26, 2025
7339990
[GPU] Ignore unsupported test cases for now
pthomadakis Mar 26, 2025
91f742b
Adding a Fill and Zero Mask node to the index tree dialect (#89)
AK2000 Mar 29, 2025
3912a2c
Upgraded COMET and Triton to LLVM v20.0.0 hash: 61f8a7f618901797ee866…
pthomadakis Mar 31, 2025
5c6eda1
Fixed or silenced all warnings
pthomadakis Mar 27, 2025
525bc0b
[GPU] Enabled AMDGPU execution
pthomadakis Mar 31, 2025
f581fe4
Make sure FPGA and GPU can co-exist
pthomadakis Apr 3, 2025
e79443e
TC-to-TTGT optimization now works with dynamic inputs by implementing…
pthomadakis Apr 3, 2025
28d6dd7
Improve GPU code generation approach
pthomadakis Apr 4, 2025
cf771a2
Updated CuGpuUtils to match HIP new design
pthomadakis Apr 7, 2025
ccf03b9
Fixed by in TTGT optimization and changed code generation to directly…
pthomadakis Apr 7, 2025
d5fcce9
[3D-CSF] Fixed bug in lowering CSF format for 3D tensor (issue #80)
johnpzh Apr 7, 2025
f922015
Parallel TC-to-TTGT and dense transpose
pthomadakis Apr 10, 2025
b589592
Introduce interoperability with cupy, pytorch, numba
pthomadakis Apr 10, 2025
c440786
Enabled parallel execution with fusion.
pthomadakis Apr 10, 2025
1d8c997
Disable parallel + fusion again as there is another bug in it -> scf …
pthomadakis Apr 10, 2025
c9fd9d9
Fixes for TCtoTTGT pass
pthomadakis Apr 16, 2025
3feccf9
GPU lowering can now support nested scf.parallel operations with othe…
pthomadakis Apr 16, 2025
4e0c29d
linalg.matvec will now lower to parallel loops
pthomadakis Apr 17, 2025
e66d5de
[COMETPY] Revamp cometpy
pthomadakis Apr 20, 2025
d0a0ba8
Removed redudant tensor format attributes (e.g. Dense,CSR) (#106)
pthomadakis Apr 21, 2025
47963b1
Further optimized TC-to-TTGT (#108)
pthomadakis Apr 30, 2025
c6d7faa
[GPU] Support explicit block sizes, SPMM (#109)
pthomadakis Jun 27, 2025
89aaa1d
[COMETPY][GPU] Fixed bug in interoperability with other GPU frontends…
pthomadakis Jun 27, 2025
c8fa059
Gpu remove redundant copies (#113)
pthomadakis Jun 27, 2025
7c5ba6c
Automate build (#114)
pthomadakis Aug 22, 2025
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
104 changes: 56 additions & 48 deletions .github/workflows/Ubuntu-latest.yml
Original file line number Diff line number Diff line change
@@ -1,20 +1,21 @@
# This starter workflow is for a CMake project running on a single platform. There is a different starter workflow if you need cross-platform coverage.
# See: https://github.com/actions/starter-workflows/blob/main/ci/cmake-multi-platform.yml
name: Test on Ubuntu Linux

on:
push:
branches: [ "master" ]
branches: [ "master", "dev-new"]
pull_request:
branches: [ "master" ]
branches: [ "master", "dev-new"]
workflow_dispatch:

env:
# Customize the CMake build type here (Release, Debug, RelWithDebInfo, etc.)
BUILD_TYPE: Release
COMET_SRC: ${{github.workspace}}

COMETPY_COMET_PATH: ${{github.workspace}}/build/comet/
COMETPY_LLVM_PATH: ${{github.workspace}}/build/llvm/src/llvm-build
jobs:
build-and-test-comet:
build-comet-and-dependencies:
# The CMake configure and build commands are platform agnostic and should work equally well on Windows or Mac.
# You can convert this to a matrix build if you need cross-platform coverage.
# See: https://docs.github.com/en/free-pro-team@latest/actions/learn-github-actions/managing-complex-workflows#using-a-build-matrix
Expand All @@ -33,51 +34,46 @@ jobs:
with:
path: |
${{github.workspace}}/llvm
${{github.workspace}}/blis/
${{github.workspace}}/install/
${{github.workspace}}/build/
${{github.workspace}}/build/llvm
${{github.workspace}}/build/blis
key: ${{ runner.os }}-submodules

- name: Update git submodules
- name: Build clean
if: steps.cache-submodule.outputs.cache-hit != 'true'
uses: actions/checkout@v3
with:
submodules: recursive
run: mkdir ${{github.workspace}}/build && cd ${{github.workspace}}/build && cmake ../ -DCMAKE_BUILD_TYPE=Release && make

- name: Build llvm
if: steps.cache-submodule.outputs.cache-hit != 'true'
run: mkdir ${{github.workspace}}/llvm/build && cd llvm/build/ && cmake -G Ninja ../llvm -DLLVM_ENABLE_PROJECTS="mlir;openmp;clang" -DLLVM_TARGETS_TO_BUILD="X86" -DLLVM_ENABLE_ASSERTIONS=ON -DCMAKE_BUILD_TYPE=Release && ninja
- name: Build from cached
if: steps.cache-submodule.outputs.cache-hit == 'true'
run: cd ${{github.workspace}}/build && cmake ../ -DCMAKE_BUILD_TYPE=Release -DLLVM_CUSTOM_BUILD_PATH=${{github.workspace}}/build/llvm/src/llvm-build -DCMAKE_BUILD_TYPE=Release && make

- name: Build blis
if: steps.cache-submodule.outputs.cache-hit != 'true'
run: cd ${{github.workspace}} && patch -s -p0 < comet-blis.patch && cd blis && ./configure --prefix=$COMET_SRC/install --disable-shared auto && make && make install

- name: Build COMET
# Build your program with the given configuration
run: rm -rf ${{github.workspace}}/build && mkdir ${{github.workspace}}/build && cd ${{github.workspace}}/build && cmake -G Ninja .. -DMLIR_DIR=${{github.workspace}}/llvm/build/lib/cmake/mlir -DLLVM_DIR=${{github.workspace}}/llvm/build/lib/cmake/llvm -DLLVM_ENABLE_ASSERTIONS=ON -DCMAKE_BUILD_TYPE=Release && ninja
test-comet-backend:

# test-comet-backend:

# needs: build-comet
# runs-on: ubuntu-latest
needs: build-comet-and-dependencies
runs-on: ubuntu-latest

# steps:
# - uses: actions/checkout@v4
# # - uses: lukka/get-cmake@latest
# - name: Install CMake
# run: sudo apt-get install cmake && sudo apt-get install ninja-build
steps:
- uses: actions/checkout@v4
# - uses: lukka/get-cmake@latest
- name: Install CMake
run: sudo apt-get install cmake && sudo apt-get install ninja-build

# - name: Cache Submodules
# id: cache-submodule
# uses: actions/cache@v4
# # if: always()
# with:
# path: |
# ${{github.workspace}}/llvm
# ${{github.workspace}}/blis/
# ${{github.workspace}}/install/
# ${{github.workspace}}/build/
# key: ${{ runner.os }}-submodules

- name: Cache Submodules
id: cache-submodule
uses: actions/cache@v4
# if: always()
with:
path: |
${{github.workspace}}/llvm
${{github.workspace}}/build/llvm
${{github.workspace}}/build/blis
key: ${{ runner.os }}-submodules

- name: Build COMET
working-directory: ${{github.workspace}}/build/
# Build your program with the given configuration
run: cmake .. -DLLVM_CUSTOM_BUILD_PATH=${{github.workspace}}/build/llvm/src/llvm-build -DCMAKE_BUILD_TYPE=Release && make

- name: Initialize Python 3.11
uses: actions/setup-python@v4
Expand All @@ -91,7 +87,7 @@ jobs:
sudo apt-get install -y python3-pip

- name: Test
working-directory: ${{github.workspace}}/build
working-directory: ${{github.workspace}}/build/comet/
shell: bash

run: ninja check-comet-integration
Expand Down Expand Up @@ -129,7 +125,7 @@ jobs:
# cargo test

test-cometpy:
needs: build-and-test-comet
needs: build-comet-and-dependencies
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
Expand All @@ -141,9 +137,8 @@ jobs:
with:
path: |
${{github.workspace}}/llvm
${{github.workspace}}/blis/
${{github.workspace}}/install/
${{github.workspace}}/build/
${{github.workspace}}/build/llvm
${{github.workspace}}/build/blis
key: ${{ runner.os }}-submodules

- name: Initialize Python 3.11
Expand All @@ -152,10 +147,23 @@ jobs:
with:
python-version: 3.11

- name: Install python dependencies
run: |
sudo apt-get install -y python3-pip

- name: Build COMET
working-directory: ${{github.workspace}}/build/
# Build your program with the given configuration
run: cmake .. -DLLVM_CUSTOM_BUILD_PATH=${{github.workspace}}/build/llvm/src/llvm-build -DCMAKE_BUILD_TYPE=Release && make


- name: Setup cometPy
run: cd ${{github.workspace}}/frontends/numpy-scipy/ && pip3 install -e .


- name: Install PyTest
run: pip3 install -U pytest

- name: Test CometPy
working-directory: ${{github.workspace}}/frontends/numpy-scipy/integration_tests/

run: python3 numpy_integration.py -v
run: pytest
4 changes: 3 additions & 1 deletion .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -19,4 +19,6 @@ lit.site.cfg.py
.vscode
/.idea

# External software
# Clangd
/.cache

15 changes: 15 additions & 0 deletions .gitmodules
Original file line number Diff line number Diff line change
Expand Up @@ -7,3 +7,18 @@
[submodule "triton"]
path = triton
url = https://github.com/triton-lang/triton.git
[submodule "tools/llvm-spirv"]
path = tools/llvm-spirv
url = https://github.com/llvm/llvm-project.git
branch = release/7.x
[submodule "tools/spirv-llvm-tranlate"]
path = tools/spirv-llvm-tranlate
url = https://github.com/KhronosGroup/SPIRV-LLVM-Translator.git
branch = llvm_release_70
[submodule "tools/spirv-llvm-translate"]
path = tools/spirv-llvm-translate
url = https://github.com/KhronosGroup/SPIRV-LLVM-Translator.git
branch = llvm_release_70
[submodule "runtimes/mcl"]
path = runtimes/mcl
url = https://github.com/pnnl/mcl.git
Loading
Loading