Another round of subr_blist.c fixes

[alc]

Summary

Address the remaining integer overflow issues with the "skip" parameters and "next_skip" variables. The "skip" value in struct blist has long been a 64-bit quantity but various functions have implicitly truncated this value to 32 bits. Now, all arithmetic involving the "skip" value is 64 bits wide. (This should allow us to relax the size limit on a swap device in the swap pager.)

Maintain the ability to test this allocator as a user-space application by including <stdbool.h>.

Remove an unused variable from blst_radix_print().

[kib]

kern/subr_blist.c

629 ↗

(On Diff #30064)

Is child size enough for this ? If yes, might be it makes sense to use daddr_t still but assert that it is under one node.

[alc]

kern/subr_blist.c

629 ↗

(On Diff #30064)

Yes, it is. Throughout this code the variable name "radix" is a misnomer. The variables named "radix" actually represent the maximum number of possible leaves in the subtree, and not what you and I expect "radix" to mean. So, when we divide "(freeBlk - blk)" by "radix" we are operating on numbers that differ by at most a factor of BLIST_META_RADIX, or 16.

I think that I can #define MPASS() to be assert() for user-space compilation and sprinkle MPASS(child >= 0 && child < BLIST_META_RADIX) throughout the code.