3/3 vfs: refactor vputx and add more comments

[mjg]

Summary

the vnode interlock relocking for vchr in vrele may seem problematic, but it almost never happens in practice. I take it for readability.

[mjg]

tested by pho

[jeff]

sys/kern/vfs_subr.c

3218

Can we rename this vput_final or vput_last or something. At first glance it was confusing to me for historical reasons.

3332

It's not clear to me why all of the release routines don't have this VCHR code.

[jeff]

sys/kern/vfs_subr.c

2802–2804

The caller is required to hold the interlock when transitioning a VCHR usecount to zero. This prevents a race with devfs_reclaim_vchr() that would leak a si_usecount reference. The vnode lock will also prevent this race if it is held while dropping the last ref.

2810

vrele()
decr v_usecount

2812

v_usecount == 0, does not decrement si_usecount
v_rdev = NULL

2817

v_decr_devcount()
v_dev == NULL, does not decrement si_usecount