You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@mxnet.apache.org by GitBox <gi...@apache.org> on 2021/07/24 13:38:01 UTC
[GitHub] [incubator-mxnet] matteosal opened a new issue #20467: Build failure with CUDA 11.4
matteosal opened a new issue #20467:
URL: https://github.com/apache/incubator-mxnet/issues/20467
Building with CUDA 11.4 on Linux produces these errors:
```
/home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/storage_profiler.cc: In member function ‘void mxnet::profiler::GpuDeviceStorageProfiler::DumpProfile() const’:
/home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/storage_profiler.cc:113:78: error: cannot convert ‘nvmlProcessInfo_st*’ to ‘nvmlProcessInfo_v1_t*’ {aka ‘nvmlProcessInfo_v1_st*’}
113 | nvmlDeviceGetComputeRunningProcesses(nvml_device, &info_count, infos.data());
| ~~~~~~~~~~^~
| |
| nvmlProcessInfo_st*
In file included from /home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/storage_profiler.cc:22:
/usr/local/cuda/include/nvml.h:7973:127: note: initializing argument 3 of ‘nvmlReturn_t nvmlDeviceGetComputeRunningProcesses(nvmlDevice_t, unsigned int*, nvmlProcessInfo_v1_t*)’
7973 | nvmlReturn_t DECLDIR nvmlDeviceGetComputeRunningProcesses(nvmlDevice_t device, unsigned int *infoCount, nvmlProcessInfo_v1_t *infos);
| ~~~~~~~~~~~~~~~~~~~~~~^~~~~
In file included from /home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/storage_profiler.cc:31:
/home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/storage_profiler.cc:115:88: error: cannot convert ‘nvmlProcessInfo_st*’ to ‘nvmlProcessInfo_v1_t*’ {aka ‘nvmlProcessInfo_v1_st*’}
115 | NVML_CALL(nvmlDeviceGetComputeRunningProcesses(nvml_device, &info_count, infos.data()));
| ~~~~~~~~~~^~
| |
| nvmlProcessInfo_st*
/home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/../common/cuda/utils.h:190:28: note: in definition of macro ‘NVML_CALL’
190 | nvmlReturn_t result = (func); \
| ^~~~
In file included from /home/matteo/Git/mxnet-build/Build/Linux-x86-64/CUDA/mxnet/src/profiler/storage_profiler.cc:22:
/usr/local/cuda/include/nvml.h:7973:127: note: initializing argument 3 of ‘nvmlReturn_t nvmlDeviceGetComputeRunningProcesses(nvmlDevice_t, unsigned int*, nvmlProcessInfo_v1_t*)’
7973 | nvmlReturn_t DECLDIR nvmlDeviceGetComputeRunningProcesses(nvmlDevice_t device, unsigned int *infoCount, nvmlProcessInfo_v1_t *infos);
```
Is there place one can go check for a list of supported/recommended CUDA versions?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org
[GitHub] [incubator-mxnet] leezu commented on issue #20467: Build failure with CUDA 11.4
Posted by GitBox <gi...@apache.org>.
leezu commented on issue #20467:
URL: https://github.com/apache/incubator-mxnet/issues/20467#issuecomment-887870099
Do you need nvml? Otherwise you may workaround this issue by setting `USE_NVML=0`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org
[GitHub] [incubator-mxnet] matteosal closed issue #20467: Build failure with CUDA 11.4
Posted by GitBox <gi...@apache.org>.
matteosal closed issue #20467:
URL: https://github.com/apache/incubator-mxnet/issues/20467
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org
[GitHub] [incubator-mxnet] matteosal commented on issue #20467: Build failure with CUDA 11.4
Posted by GitBox <gi...@apache.org>.
matteosal commented on issue #20467:
URL: https://github.com/apache/incubator-mxnet/issues/20467#issuecomment-888313240
Yes, `USE_NVML=0` fixes the problem. Thanks!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org
[GitHub] [incubator-mxnet] leezu commented on issue #20467: Build failure with CUDA 11.4 and NVML
Posted by GitBox <gi...@apache.org>.
leezu commented on issue #20467:
URL: https://github.com/apache/incubator-mxnet/issues/20467#issuecomment-888403497
FYI @TristonC @ptrendx
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org