Skip to content
This repository was archived by the owner on Mar 28, 2023. It is now read-only.

[SYCL][HIP] Fix shuffles and reductions #759

Merged
merged 2 commits into from
Jan 26, 2022
Merged

[SYCL][HIP] Fix shuffles and reductions #759

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
079884f
[SYCL][HIP] Fix shuffles and reductions
npmiller Jan 21, 2022
ba423d2
[SYCL][HIP] Re-enable broadcast tests
npmiller Jan 25, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions SYCL/Reduction/reduction_big_data.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,9 @@
// RUN: %ACC_RUN_PLACEHOLDER %t.out
// RUN: %CPU_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupFMax on AMD, error message `Group algorithms are not
// supported on host device` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device` on Nvidia.
// XFAIL: hip_nvidia
//
// UNSUPPORTED: ze_debug-1,ze_debug4

// RUNx: %HOST_RUN_PLACEHOLDER %t.out
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_nd_s0_dw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,8 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd, __spirv_GroupSMin, __spirv_GroupSMax on AMD, error
// message `Group algorithms are not supported on host device.` on HIP Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on HIP Nvidia.
// XFAIL: hip_nvidia

// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_nd_s0_rw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,8 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd, __spirv_GroupSMin, __spirv_GroupSMax on AMD, error
// message `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_nvidia

// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_nd_s1_dw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,9 +4,8 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd, __spirv_GroupSMin, __spirv_GroupSMax on AMD, error
// message `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_nvidia

// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_nd_s1_rw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,8 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd, __spirv_GroupSMin, __spirv_GroupSMax on AMD, error
// message `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_nvidia

// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
Expand Down
6 changes: 3 additions & 3 deletions SYCL/Reduction/reduction_placeholder.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,9 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd, __spirv_GroupFMin on AMD, error message `Group
// algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_nvidia
//
// UNSUPPORTED: ze_debug-1,ze_debug4

// RUNx: %HOST_RUN_PLACEHOLDER %t.out
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_queue_parallel_for.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,8 @@
// RUN: %ACC_RUN_PLACEHOLDER %t.out
// RUN: %CPU_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd on AMD, error message `Group algorithms are not
// supported on host device.` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_nvidia

// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_usm.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,9 +3,8 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_GroupIAdd, __spirv_GroupFMin, __spirv_GroupFMax on AMD, error
// message `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device.` on Nvidia.
// XFAIL: hip_nvidia

// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
Expand Down
5 changes: 2 additions & 3 deletions SYCL/Reduction/reduction_usm_dw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,9 +6,8 @@
// TODO: test disabled due to sporadic fails in level_zero:gpu RT.
// UNSUPPORTED: linux && level_zero
//
// Missing __spirv_GroupIAdd, __spirv_GroupFMin, __spirv_GroupFMax on AMD, error
// message `Group algorithms are not supported on host device` on Nvidia.
// XFAIL: hip_amd || hip_nvidia
// `Group algorithms are not supported on host device` on Nvidia.
// XFAIL: hip_nvidia

// This test performs basic checks of parallel_for(nd_range, reduction, func)
// with reductions initialized with USM var. It tests only SYCL-2020 reduction
Expand Down
2 changes: 0 additions & 2 deletions SYCL/SubGroup/broadcast.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,6 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out

// XFAIL: hip_amd

//==--------- broadcast.cpp - SYCL sub_group broadcast test ----*- C++ -*---==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
Expand Down
2 changes: 0 additions & 2 deletions SYCL/SubGroup/broadcast_fp16.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,8 +1,6 @@
// RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple %s -o %t.out
// RUN: %GPU_RUN_PLACEHOLDER %t.out

// XFAIL: hip_amd

//==---- broadcast_fp16.cpp - SYCL sub_group broadcast test ----*- C++ -*---==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
Expand Down
2 changes: 0 additions & 2 deletions SYCL/SubGroup/broadcast_fp64.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,6 @@
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out

// XFAIL: hip_amd

//==----- broadcast_fp64.cpp - SYCL sub_group broadcast test ----*- C++ -*--==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
Expand Down
4 changes: 0 additions & 4 deletions SYCL/SubGroup/generic-shuffle.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,13 +1,9 @@
// UNSUPPORTED: hip_amd
// RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple %s -o %t.out
// RUN: %HOST_RUN_PLACEHOLDER %t.out
// RUN: %CPU_RUN_PLACEHOLDER %t.out
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_SubgroupShuffleINTEL, __spirv_SubgroupShuffleUpINTEL,
// __spirv_SubgroupShuffleDownINTEL, __spirv_SubgroupShuffleXorINTEL on AMD
//
//==-- generic_shuffle.cpp - SYCL sub_group generic shuffle test *- C++ -*--==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
Expand Down
4 changes: 0 additions & 4 deletions SYCL/SubGroup/shuffle_fp64.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,13 +1,9 @@
// UNSUPPORTED: hip_amd
// RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple %s -o %t.out
// RUN: %HOST_RUN_PLACEHOLDER %t.out
// RUN: %CPU_RUN_PLACEHOLDER %t.out
// RUN: %GPU_RUN_PLACEHOLDER %t.out
// RUN: %ACC_RUN_PLACEHOLDER %t.out
//
// Missing __spirv_SubgroupShuffleINTEL, __spirv_SubgroupShuffleUpINTEL,
// __spirv_SubgroupShuffleDownINTEL, __spirv_SubgroupShuffleXorINTEL on AMD
//
//==------- shuffle_fp64.cpp - SYCL sub_group shuffle test -----*- C++ -*---==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
Expand Down