< :前の番号
^ :番号順リスト
> :次の番号
P :前の記事
N :次の記事
|<:スレッドの先頭
>|:次のスレッド
^ :返事先
_:自分への返事
>:同じ返事先を持つ記事(前)
<:同じ返事先を持つ記事(後)
---:分割してスレッド表示、再表示
| :分割して(縦)スレッド表示、再表示
~ :スレッドのフレーム消去
.:インデックス
..:インデックスのインデックス
> On second thought, I feel exit! is more essential solution.
>
> Because there may be other objects which behaves as "flush at finalization",
> flushing IO objects is not enough.
>
> The problem is finalization is occur twice (at parent and child).
> If ruby uses exit! for forked process, we can avoid dual finalization.
>
> However it means users of fork must finalize objects explicitly in
> forked process.
> For example, fork { print "a" } may lost "a" because buffered "a" is
> not flushed.
Also, Process.exit! escape to run at_exit handler. I think exit! is too drastic.
it help many situation than flushing, but also makes much side effect. I suspect
"fork { print "a" }" type output lost makes lots test-all break.
Instead of, providing fork callback likes pthread_atfork() and uses it?