Pull requests: iree-org/iree
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add a
--iree-preprocessing-pass-pipeline
to allow user control on preprocessing passes before IREE compilation.
#11986
opened Jan 27, 2023 by
MaheshRavishankar
Loading…
Remove assert for number of results in FuseElementwiseOpsWithMultipleUses
#11983
opened Jan 27, 2023 by
sogartar
Loading…
Turn on aggressive fusion heuristics by default.
#11980
opened Jan 27, 2023 by
MaheshRavishankar
•
Draft
Adding iree_hal_caching_allocator_t.
buildkite:benchmark-android
Invoke Android benchmark pipeline
buildkite:benchmark-cuda
Invoke CUDA benchmark pipeline
buildkite:benchmark-riscv
Invoke RISCV benchmark pipeline (compile stats only for now)
buildkite:benchmark-x86_64
Invoke x86_64 benchmark pipeline
hal/api
IREE's public C hardware abstraction layer API
performance ⚡
Performance/optimization related work across the compiler and runtime
[spirv] Support multiple reductions fused in single dispatch cases
codegen/spirv
SPIR-V code generation compiler backend
#11975
opened Jan 26, 2023 by
antiagainst
Loading…
Try speeding up git submodule init on Windows CI.
infrastructure
Relating to build systems, CI, or testing
yolo
( ͡° ͜ʖ ͡°)
Integrate llvm-project at 05d9cc4799ca and bump dependences
#11959
opened Jan 25, 2023 by
dcaballe
Loading…
Run Relating to build systems, CI, or testing
platform/windows 🚪
Windows-specific build, execution, benchmarking, and deployment
build_test_all_windows
CI job on presubmit.
infrastructure
#11942
opened Jan 23, 2023 by
ScottTodd
Loading…
Support transform dialect matchers in dispatch region formation
#11886
opened Jan 18, 2023 by
ftynse
Loading…
Add memory optimization to transform dialect and enable it for reduction flow
#11854
opened Jan 17, 2023 by
ThomasRaoux
•
Draft
Add a pass to vectorize tensor.pack and tensor.unpack ops.
buildkite:benchmark-x86_64
Invoke x86_64 benchmark pipeline
#11838
opened Jan 13, 2023 by
hanhanW
Loading…
Use task_topology_max_group_count in benchmarks
buildkite:benchmark-android
Invoke Android benchmark pipeline
buildkite:benchmark-cuda
Invoke CUDA benchmark pipeline
buildkite:benchmark-riscv
Invoke RISCV benchmark pipeline (compile stats only for now)
buildkite:benchmark-x86_64
Invoke x86_64 benchmark pipeline
[WIP] Use MaskedVectorOp in the transform script
#11796
opened Jan 11, 2023 by
nicolasvasilache
•
Draft
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.