inline swp_pager_isondev() function and use it 2 other places

[ota_j.email.ne.jp]

Summary

Simplify code and take advantage of inline function.

Ref: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223640

Test Plan

Use memory to trigger and cause swapins and swapouts.

[alc]

sys/vm/swap_pager.c

766 ↗

(On Diff #46686)

Please eliminate the space between "swp_pager_isondev" and "(".

800 ↗

(On Diff #46686)

Ditto.

[kib]

sys/vm/swap_pager.c

752 ↗

(On Diff #46686)

Does inline change anything in the generated code ?

[ota_j.email.ne.jp]

Style fix.

[ota_j.email.ne.jp]

sys/vm/swap_pager.c

1749 ↗

(On Diff #46758)

This had been the only caller.

752 ↗

(On Diff #46686)

C99 introduced inline function.
Compiler can choose to embed code into each callers instead of making a function call at each location.

If a compiler optimizes, this change won't have a function call overhead and will be the same. We have one existing caller that can benefit inline.

I didn't check resulted code in FreeBSD, though; I saw lots of other inline functions in FreeBSD code and expected we had been taking advantages.

[alc]

sys/vm/swap_pager.c

752 ↗

(On Diff #46686)

I believe that Kostik's point is that clang will automatically inline this function without the need for explicitly defining the function as inline.

[alc]

sys/vm/swap_pager.c

752 ↗

(On Diff #46686)

The generated code is the same with or without the explicit inline.

[kib]

Prefer to remove the inline keyword.

[alc]

In D16712#385624, @kib wrote:

Prefer to remove the inline keyword.

Okay, I'm going to commit this change in a few hours without the inline keyword.

[alc]

In D16712#385630, @alc wrote:

In D16712#385624, @kib wrote:

Prefer to remove the inline keyword.

Okay, I'm going to commit this change in a few hours without the inline keyword.

I'm also going to change the return type for swp_pager_isondev() to bool.