Skip to content

[DCU] fix compile error on K100 #63535

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.

Sign up for GitHub

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

Merged
qili93 merged 1 commit into from
Apr 17, 2024
Merged

[DCU] fix compile error on K100 #63535

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 4 additions & 4 deletions cmake/hip.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -142,11 +142,11 @@ set(HIP_CLANG_FLAGS ${HIP_CXX_FLAGS})
# Ask hcc to generate device code during compilation so we can use
# host linker to link.
list(APPEND HIP_HCC_FLAGS -fno-gpu-rdc)
list(APPEND HIP_HCC_FLAGS --offload-arch=gfx906)
list(APPEND HIP_HCC_FLAGS --offload-arch=gfx908)
list(APPEND HIP_HCC_FLAGS --offload-arch=gfx906) # Z100 (ZIFANG)
list(APPEND HIP_HCC_FLAGS --offload-arch=gfx926) # K100 (KONGING)
list(APPEND HIP_CLANG_FLAGS -fno-gpu-rdc)
list(APPEND HIP_CLANG_FLAGS --offload-arch=gfx906)
list(APPEND HIP_CLANG_FLAGS --offload-arch=gfx908)
list(APPEND HIP_CLANG_FLAGS --offload-arch=gfx906) # Z100 (ZIFANG)
list(APPEND HIP_CLANG_FLAGS --offload-arch=gfx926) # K100 (KONGING)

if(HIP_COMPILER STREQUAL clang)
set(hip_library_name amdhip64)
Expand Down
5 changes: 5 additions & 0 deletions python/paddle/distributed/launch/controllers/watcher.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,8 @@
import time
from threading import Thread

import paddle

from ..utils.nvsmi import get_gpu_info, get_gpu_process, get_gpu_util


Expand All @@ -30,6 +32,9 @@ def __init__(self, ctx):
if not self.ctx.args.enable_gpu_log:
return

if paddle.is_compiled_with_rocm():
return

# gpu log file
self.gpus = self.ctx.args.devices or self.ctx.node.device.labels
if len(self.gpus) > 0:
Expand Down