This will allow userspace to more easily figure out if ALTQ is built
into the kernel and what disciplines are supported.
Details
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
Sysctls are considered as an expensive activity, so they need to be justified.
We have currently 48 kern.feature entries in the GENERIC stable/12 kernel.
You propose to increase this number by 16%
Can you point me to a relevant use case?
We can also view it as 8 more out of ~8400 sysctls already present. And only on ALTQ enabled kernels, which is not default.
Can you point me to a relevant use case?
My use case for this is to make the new altq test conditional on the presence of ALTQ in the kernel. In much the same way that the pf tests already check for kern.features.vimage
We can also view it as 8 more out of ~8400 sysctls already present. And only on ALTQ enabled kernels, which is not default.
Ok.
Can you point me to a relevant use case?
My use case for this is to make the new altq test conditional on the presence of ALTQ in the kernel. In much the same way that the pf tests already check for kern.features.vimage
How about introducing a hierarchy? Something like kern.feature.altq,.xxx ?
I know, that is not the typical usage, but it might be clearer than structuring with underscores and rely on the alphabetical sort.