KOSAKI Motohiro <kosaki.motohiro / gmail.com> wrote:
> After while thinking, I conclude I was wrong. If rb_io_fptr_cleanup()
> raise a exception, We don't have to kill other threads. So, now I'm
> incline to revert r31230. Hmm...

I think we can still fix the issues revolving around r31230.  We've
already found at least two (this and #4555) bugs.  As for #4555
(connect() failing on EINTR) it was a bug anyways, but just very hard to
expose before.

-- 
Eric Wong