Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
1–10 of 13 posts
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#2Renaming fopen64 to intercept library calls feels like a brittle hack masquerading as "sandboxing." Why not just upstream this hardware support to nvtop instead of fragmenting the ecosystem?
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#3Renaming fopen64 to intercept library calls feels like a brittle hack masquerading as "sandboxing." Why not just upstream this hardware support to nvtop instead of fragmenting the ecosystem?
sadly, sandboxing is something that can't be upstreamed. this way, sandboxing is kept in zml instead of patching mesa.
as for nvtop, great program, but we missed a few features (such as sandboxing)
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#4Looks cool!
nvtop can actually support TPUs too via https://github.com/rdyro/libtpuinfo/ https://github.com/Syllo/nvtop/blob/76890233d759199f50ad3bdb...
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#5Renaming fopen64 to intercept library calls feels like a brittle hack masquerading as "sandboxing." Why not just upstream this hardware support to nvtop instead of fragmenting the ecosystem?
sadly, sandboxing is something that can't be upstreamed. this way, sandboxing is kept in zml instead of patching mesa. as for nvtop, great program, but we missed a few features (such as sandboxing)
It looks cool and I was excited to get monitoring for the NPU on my Ryzen AI 395+, unfortunately it does not show. NPU support in linux really seems to be an afterthought.
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#6Earlier quoted context omitted.
sadly, sandboxing is something that can't be upstreamed. this way, sandboxing is kept in zml instead of patching mesa. as for nvtop, great program, but we missed a few features (such as sandboxing)
It looks cool and I was excited to get monitoring for the NPU on my Ryzen AI 395+, unfortunately it does not show. NPU support in linux really seems to be an afterthought.
Weird, because we tried it. It doesn’t show anything?
We use the amdsmi to get metrics. I’ll investigate.
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#7Renaming fopen64 to intercept library calls feels like a brittle hack masquerading as "sandboxing." Why not just upstream this hardware support to nvtop instead of fragmenting the ecosystem?
If this logic were pushed into nvtop, wouldn't the codebase become unmaintainable? Each vendor's interception method is going to be different.
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#8Renaming fopen64 to intercept library calls feels like a brittle hack masquerading as "sandboxing." Why not just upstream this hardware support to nvtop instead of fragmenting the ecosystem?
If this logic were pushed into nvtop, wouldn't the codebase become unmaintainable? Each vendor's interception method is going to be different.
[dead]
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#9"NPU" seems to refer to trainium only?
Re: Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs
#10Look into all-smi https://github.com/lablup/all-smi
It supports all GPUs thinkable including Apple Silicon and many AI accelerator cards.