mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kevin Klues <>
Subject Re: Review Request 48376: Changed semantics for granting access to /dev/nvidiactl, etc.
Date Sat, 11 Jun 2016 03:05:05 GMT

This is an automatically generated e-mail. To reply, visit:

(Updated June 11, 2016, 3:05 a.m.)

Review request for mesos and Benjamin Mahler.


Rebased for

Bugs: MESOS-5555

Repository: mesos


Previously, access to `/dev/nvidiactl` and `/dev/nvidia-uvm` was
only granted to / revoked from a container as GPUs were added and
removed from them. On some level, this makes sense because most jobs
don't need access to these devices unless they are also using a GPU.
However, there are cases when access to these files is appropriate,
even when not making use of a GPU. Running `nvidia-smi` to control the
global state of the underlying nvidia driver, for example.

This commit adds `/dev/nvidiactl` and `/dev/nvidia-uvm` to the default
whitelist of devices to include in every container when the
`gpu/nvidia` isolator is enabled. This allows a container to run
standard nvidia driver tools (such as `nvidia-smi`) without failing
with abnormal errors when no GPUs have been granted to it. As such,
these tools will now report that no GPUs are installed instead of
failing abnormally.

NOTE: Once we allow GPUs to be granted to containers with filesystem
isolation turned on, other criteria will be used to determine when /
if to grant access to these control devices.

Diffs (updated)

  src/slave/containerizer/mesos/isolators/gpu/isolator.cpp d7557a0c338e8c0e51461b2326600c03f89c2e8b



GTEST_FILTER="" make -j check && sudo GTEST_FILTER="*NVIDIA*" src/mesos-tests


Kevin Klues

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message