intel graph-compiler issues

intel / graph-compiler

Apache License 2.0

27 stars 14 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

[GpuOclRuntime] Retain input and release created cl_events

#367 AndreyPavlenko opened 7 hours ago
0
[GPU] Automatically dump lowered mlir module on 'enableObjectDump=true'

#366 dchigarev closed 11 hours ago
1
Perf dashbaord, get relevant performance graph using DB

#365 lmontigny opened 1 day ago
0
Update pinned imex version

#364 dchigarev closed 1 day ago
0
Update llvm commit

#363 BRUCE11111 closed 5 days ago
2
Implemented GPU runner

#362 AndreyPavlenko closed 9 hours ago
0
[LinalgToXeGPU] Support `linalg.matmul_transpose_a`

#361 dchigarev opened 1 week ago
0
InsertGPUAllocs does not properly insert dealloc for temporary buffers in kernel code

#360 zhczhong opened 1 week ago
2
Merge cost model

#359 lmontigny opened 1 week ago
0
[BenchGC] add tuner tools for benchgc

#358 xurui1995 opened 1 week ago
2
Update llvm commit

#357 yifeizh2 closed 6 days ago
7
Avoid use DPS interface

#356 WangJialei-A opened 1 week ago
6
GC check fail with clang & release build

#355 WangJialei-A opened 1 week ago
0
Add an option for insert gpu allocs to treat allocations as gpu-native

#354 zhczhong closed 1 week ago
0
optimize thread local cache for brgemm

#353 crazydemo opened 1 week ago
7
No verbose on correctness check script

#352 WangJialei-A opened 1 week ago
0
[Transform][vector] lowering dynamic shape of tensor.pack to vector

#351 BRUCE11111 opened 1 week ago
0
add thread local cache for brgemm

#350 crazydemo closed 1 week ago
0
Minor build script enhancements

#349 AndreyPavlenko opened 1 week ago
1
[Transform][Fusion] align file name with upstream

#348 Yun-Fly closed 2 weeks ago
0
[LinalgToXeGPU] Lower `linalg.matmul_transpose_b` into `xegpu.dpas`

#347 dchigarev closed 2 days ago
4
Get a baseline performance for the IMEX-based path

#346 kurapov-peter opened 2 weeks ago
0
Allow for passing an allocator to the GPU pipeline

#345 kurapov-peter opened 2 weeks ago
0
Add an option for insert gpu allocs to treat allocations as gpu-native

#344 kurapov-peter opened 2 weeks ago
0
Implemented GPU OpenCL runtime

#343 AndreyPavlenko closed 1 day ago
4
[Runtime] Constant cache manager and runtime pipeline

#342 niuxiaog opened 2 weeks ago
0
Constant cache manager and runtime pipeline

#341 niuxiaog opened 2 weeks ago
0
[`LinalgToXeGPU`] Support conversion for `linalg.matmul` with `transpose_b`

#340 dchigarev closed 2 days ago
0
benchgc: support transpose op

#339 WangJialei-A closed 2 weeks ago
0
BenchGC should support linalg.transpose in correctness check

#338 WangJialei-A closed 2 weeks ago
0
fix BenchGC execution issue when driver is mlir

#337 xurui1995 closed 3 weeks ago
0
benchgc bench driver=mlir problem

#336 BRUCE11111 closed 3 weeks ago
0
Deprecate all linalgx matmul ops

#335 LongshengDu closed 3 weeks ago
0
Reuse code, fix Coverity scans

#334 kwasd closed 3 weeks ago
1
Convert a subset of GPU dialect ops to the OpenCL GPU runtime calls

#333 AndreyPavlenko closed 1 week ago
0
[`IterativeTilingAndFusionPass`] Wrap linalg.ops in a loop even if the shape is smaller than min tiling size

#332 dchigarev opened 3 weeks ago
1
Convert large vector to physical register vector

#331 BRUCE11111 opened 3 weeks ago
0
add tuner mode for BenchGC to support auto-tuning

#330 xurui1995 opened 3 weeks ago
0
[GPU] Register initial GPU pipeline that uses IMEX

#329 dchigarev closed 2 weeks ago
0
[Build] fix issue when GC_DEV_LINK_LLVM_DYLIB is ON

#328 xurui1995 closed 3 weeks ago
0
Build error when `GC_DEV_LINK_LLVM_DYLIB` is ON

#327 xurui1995 closed 3 weeks ago
0
`CPUPhysicalRegister` pass failed when processing 2D4D fp32 matmul

#326 yifeizh2 closed 1 week ago
4
Refinements on `microkernel` dialect lowering

#325 huanghaixin008 closed 3 weeks ago
0
[Transform] Refinements on `microkernel` dialect lowering

#324 huanghaixin008 closed 3 weeks ago
1
Performance regression caused by read lock in brgemm

#323 zhczhong closed 1 week ago
2
Nightly Performance Result

#322 github-actions[bot] opened 3 weeks ago
0
Nightly Test Result

#321 WangJialei-A closed 3 weeks ago
1
bf16 matmul's corresponding `tensor.pack` not properly optimized

#320 yifeizh2 opened 3 weeks ago
4
Install clang in our runner image

#319 WangJialei-A closed 3 weeks ago
6
`IterativeTilingAndFusion` cause performance regression

#318 yifeizh2 closed 4 weeks ago
1