feat: glibc extension #473
Conversation
This is really cool and looking forward to actually not having to maintain patches or wrappers. What i still don't understand is about kernel modules, would SideroLabs be still shipping them as extensions (I believe that's the case since only machined in talos can load modules)? |
I am sending more PRs, but this is one of the major changes from an architecture POV: the gpu-operator would be allowed to load and unload kernel modules, which means enabling module unloading in the kernel (see siderolabs/pkgs#1031) and not removing SYS_MODULE from containers (see siderolabs/talos#9339). Another important note is that only CDI mode (see https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/cdi.html) works on Talos with this patch. The "legacy" runtime hook require more libraries to be present on the system, whereas the CDI hook is a pure Go program that only requires the glibc dynamic loader and |
We're open to moving to using CDI 👍 |
|
See siderolabs/talos#9339 for main discussion on loading kernel modules. |
jfroy
force-pushed
the
glibc
branch
3 times, most recently
from
d94cf22
to
b4de00e
Compare
|
Patch has been updated to rework the glibc subtree to look like a merged |
The extension is mostly copied from the package that is in the nvidia toolkit extensions, but with a few notable changes. - A symbolic link to `ldconfig` is installed at `/sbin/ldconfig`. This allows the nvidia gpu-operator to work without modification. A patch to the extension validation logic is required to allow the new path. - The `/usr/local/glibc` subtree is now structured as a [merged `/usr` root](https://systemd.io/THE_CASE_FOR_THE_USR_MERGE/). This improves compatbility with ld-linux.so.2 for library discovery. Signed-off-by: Jean-Francois Roy <[email protected]>
This PR adds a glibc extension. The intention is to replace the nvidia extensions entirely and only provide the glibc components required by the nvidia gpu-operator and its components (e.g. the nvidia container toolkit).
The extension is pretty much copied from the package that is in the nvidia extensions, with one major modification: a symbolic link to ldconfig is installed at
/sbin/ldconfig. This change allows the nvidia gpu-operator to work without modification*. This change does require a patch to the extension validation logic, which is provided in separate PRs.