Linux 6.0 compat fix for register_shrinker() now accepting var-arg name via printf-like interface

[ckane]

Motivation and Context

The 6.0 kernel added a printf-style var-arg for args > 0 to the register_shrinker function, in order to add names to shrinkers, in commit e33c267ab70de4249d22d7eab1cc7d68a889bac2. This enables the shrinkers to have friendly names exposed in /sys/kernel/debug/shrinker/.

Description

Added a new test inside of config/kernel-shrink.m4 that looks for the var-arg version of register_shrinker() first, and if found, it also assumes the split-callback shrinker structure API, as well (which is a pre-requisite for the register_shrinker test). Since spl_register_shrinker is only called in one place (the arc.c code), I embedded the "zfs-arc-shrinker" name in the macro located in include/os/linux/spl/sys/shrinker.h, in order to minimize changes to lines of code. Should we ever want to implement more than one shrinker using this code, we'll have to move the naming out of the macro and into the module code.

How Has This Been Tested?

So far have build tested, will be run-testing this week

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Performance enhancement (non-breaking change which improves efficiency)
• Code cleanup (non-breaking change which makes code smaller or more readable)
• Breaking change (fix or feature that would cause existing functionality to change)
• Library ABI change (libzfs, libzfs_core, libnvpair, libuutil and libzfsbootenv)
• Documentation (a change to man pages or other documentation)

Checklist:

• My code follows the OpenZFS code style requirements.
• I have updated the documentation accordingly.
• I have read the contributing document.
• I have added tests to cover my changes.
• I have run the ZFS Test Suite with this change applied.
• All commit messages are properly formatted and contain Signed-off-by.

[ckane]

@behlendorf I have got zero clue why the CI/CD test builds fail around the "checking for parallel" line in configure, but I've built this on my Arch linux system multiple times and it works there. the kernel boots and seems to work, although latest pulls in some other regressions related to AMDGPU that were reverted in 5.19, which causes my kernel to crash shortly after starting, so I don't get much run-testing.

[ckane]

Ok, looks like an issue with the older versions of automake/autoconf that come with Ubuntu and how it handles newlines within [] constructs. I tried the configure again in a Ubuntu docker, was able to reproduce the issue seen in the GitHub workflows, and cleaned up Line 138 in config/kernel-shrink.m4 and it resolved the syntax error messages.