ruby-dev

In message <200109251459.f8PExBO20666@sharui.nakada.kanuma.tochigi.jp>
	on Wed, 26 Sep 2001 00:05:11 +0900,
	nobu.nakada@nifty.ne.jp wrote:
> > BSD系のOSのselect(2)では、
> > 
> >      requested operation.  select() returns the total number of ready descrip-
> >      tors in all the sets.
> > 
> > と書かれていますので、重複というよりも合計なんでしょう。:-)
> 
> 　てことは3つの数を単純に足してしまってOK?
「元の問題」はあんまり理解してませんが。

もっと細かくソースを追えば、カーネルのselect(2)に対応したsys_select()
の下請け関数selscan()では、

	n = 0;
	for (msk = 0; msk < 3; msk++) {
		for (i = 0; i < nfd; i += NFDBITS) {
			ibits = *ibitp++;
			obits = 0;
			while ((j = ffs(ibits)) && (fd = i + --j) < nfd) {
				ibits &= ~(1 << j);
				if ((fp = fd_getfile(fdp, fd)) == NULL)
					return (EBADF);
				FILE_USE(fp);
				if ((*fp->f_ops->fo_poll)(fp, flag[msk], p)) {
					obits |= (1 << j);
					n++;
				}
				FILE_UNUSE(fp, p);
			}
			*obitp++ = obits;
		}
	}
	*retval = n;

というコードの破片があります。(/usr/src/sys/kern/sys_generic.cより)
n はselect(2)の戻り値になり得る値で、msk は各セットのループ、といった
感じでなんとなく想像できますよね。

--
神戸 隆博(かんべ たかひろ) at home

Thread

Prev Next

In This Thread

Prev Next

[#14614] 大小文字を無視した比較で ' 帰 ' == ' 蟻 ' ? — "K.Umakoshi" <pegacorn@...>

[#14621] suidruby — nobu.nakada@...

[#14626] ARGV — WATANABE Hirofumi <eban@...>

[#14636] Re: [REQ] String#slice(re, n) — Minero Aoki <aamine@...>

[#14642] getnameinfo? — Masatoshi SEKI <m_seki@...>

[#14658] Socekt::SOL_TCP -> Socket::IPPROTO_TCP — m_seki@...

[#14662] How to abondon saved uid privilege — nobu.nakada@...

[#14668] = in conditional — "Akinori MUSHA" <knu@...>

[#14683] ruby 1.7 の eval.c で warning — Kazuhiro NISHIYAMA <zn@...>

[#14697] Forward: [ruby-talk:20735] bug in tk.rb? — matz@... (Yukihiro Matsumoto)

[#14698] Forward: [ruby-talk:20820] Ruby/Tk core dump — matz@... (Yukihiro Matsumoto)

[#14710] SEGV at digest.c — Tietew <tietew@...>

[#14712] extra puts() at digest.c — Tietew <tietew@...>

[#14716] CGI#header bug — Shugo Maeda <shugo@...>

[#14730] ruby-lex.rb: get value from TkSTRING and TkCOMMENT — TAKAHASHI Masayoshi <maki@...>

[#14741] Forward: [ruby-talk:20957] Re: Is Ruby/Tk being maintained? — matz@... (Yukihiro Matsumoto)

[#14742] Forward: [ruby-talk:20964] Re: Is Ruby/Tk being maintained? — matz@... (Yukihiro Matsumoto)

[#14743] [BUG] $_ on thread switching — nobu.nakada@...

[#14749] forwardable.rb : call Kernel.raise — Kenichi Komiya <kom@...1.accsnet.ne.jp>

[#14754] regex.c について — Osamu Shigematsu <shige@...>

[#14765] DISTCLEANFILES in mkmf.rb — nobu.nakada@...

[#14766] Enumerable#inject — OZAWA -Crouton- Sakuro <crouton@...>

[#14768] rational precision 対応 — keiju@... (Keiju ISHITSUKA)

[#14782] coredump in mod_ruby's configure.rb — "Akinori MUSHA" <knu@...>

[#14786] [PATCH] File.utime fails occasionally — nobu.nakada@...

[#14793] Array is not Comparable — "Akinori MUSHA" <knu@...>

[#14798] freeing thread stack etc. (Re: ruby 1.6.5 preview) — matz@... (Yukihiro Matsumoto)

[#14809] WAIT_SELECT value with multithread — nobu.nakada@...

[#14810] ruby1.6.5/marshal.c — "Shin'ya Adzumi" <adzumi@...>

[#14813] option -d differ with --debug — Koji Arai <JCA02266@...>

[#14816] Generator — "Akinori MUSHA" <knu@...>

[#14823] ruby-1.6.5 MacOS X 10.1 patch — SHIROYAMA Takayuki - <psi@...>

[#14824] mailread.rb — Tietew <tietew@...>

[ruby-dev:14820] Re: WAIT_SELECT value with multithread

Thread

In This Thread

[#14823] ruby-1.6.5 MacOS X 10.1 patch — SHIROYAMA Takayuki　- <psi@...>