Open
Description
I run the docker command in the README, but failed to pass the check:
sudo docker run --rm -it --privileged --pull always --network=host --name=torch_musa_dev --env MTHREADS_VIS
IBLE_DEVICES=all --shm-size=80g registry.mthreads.com/mcconline/musa-pytorch-dev-public:rc3.1.0-v1.3.0-S80-py310 /bin/bas
h
rc3.1.0-v1.3.0-S80-py310: Pulling from mcconline/musa-pytorch-dev-public
Digest: sha256:70bc3e7494ac4538867706bdf7e0b996fa8ab35fcc08d257da6346c22e55118c
Status: Image is up to date for registry.mthreads.com/mcconline/musa-pytorch-dev-public:rc3.1.0-v1.3.0-S80-py310
Error! mthreads-gmi did not output anything! Please check the underlying drivers and container tools
/usr/bin/which: this version of `which' is deprecated; use `command -v' in scripts instead.
If the size of mthreads-gmi is 0, musa container tool is abnormal.
Error! In the output of clinfo, the value of 'Platform Vendor' is not 'MooreThreads'!
/usr/bin/ld: warning: libdrm.so.2, needed by /usr/lib/x86_64-linux-gnu/musa/libsrv_um_MUSA.so, not found (try using -rpath or -rpath-link)
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVReleaseDeviceMappingDirectX'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVMapToDeviceDirect'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVDevmemXAllocUnifiedVirtualRange'
/usr/bin/ld: /usr/lib/x86_64-linux-gnu/musa/libsrv_um_MUSA.so: undefined reference to `drmFreeVersion'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `musa_KickCETQ'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVDevMemXFreeUnifiedVirtualRange'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVGetFabricType'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVGetVersion'
/usr/bin/ld: /usr/lib/x86_64-linux-gnu/musa/libsrv_um_MUSA.so: undefined reference to `drmGetDevices2'
/usr/bin/ld: /usr/lib/x86_64-linux-gnu/musa/libsrv_um_MUSA.so: undefined reference to `drmGetVersion'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `RGXDecodeHWRInfo'
/usr/bin/ld: /usr/local/musa/lib/libmusa.so.1.0: undefined reference to `PVRSRVGetHeapTotalSize'
clang-14: error: linker command failed with exit code 1 (use -v to see invocation)
MCC failed to compile a simple demo musa program. The compilation command is:mcc test_musa.mu -o test_musa -mtgpu -O2 -lmusart -L/lib
__ __ ___ ___ ____ _____ _____ _ _ ____ _____ _ ____ ____
| \/ |/ _ \ / _ \| _ \| ____| |_ _| | | | _ \| ____| / \ | _ \/ ___|
| |\/| | | | | | | | |_) | _| | | | |_| | |_) | _| / _ \ | | | \___ \
| | | | |_| | |_| | _ <| |___ | | | _ | _ <| |___ / ___ \| |_| |___) |
|_| |_|\___/ \___/|_| \_\_____| |_| |_| |_|_| \_\_____/_/ \_\____/|____/
_____ _ _ _ _
| ___|_ _(_) | ___ __| | |
| |_ / _` | | |/ _ \/ _` | |
| _| (_| | | | __/ (_| |_|
|_| \__,_|_|_|\___|\__,_(_)
(py310) root@localhost:/home#
I'm sure everything is ok. I run docker run --rm --env MTHREADS_VISIBLE_DEVICES=all ubuntu:20.04 mthreads-gmi
to check if container toolkit works, and shows this:
sudo docker run --rm --env MTHREADS_VISIBLE_DEVICES=all ubuntu:20.04 mthreads-gmi
Wed Apr 9 11:32:44 2025
---------------------------------------------------------------
mthreads-gmi:1.8.2 Driver Version:N/A
---------------------------------------------------------------
ID Name |PCIe |%GPU Mem
Device Type |Pcie Lane Width |Temp MPC Capable
+-------------------------------------------------------------+
0 MTT S80 |00000000:01:00.0 |0% 0MiB(16384MiB)
Physical |16x(16x) |48C YES
---------------------------------------------------------------
---------------------------------------------------------------
Processes:
ID PID Process name GPU Memory
Usage
+-------------------------------------------------------------+
No running processes found
---------------------------------------------------------------
So it is probably a problem of torch-musa
container.
Metadata
Metadata
Assignees
Labels
No labels