home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!paladin.american.edu!howland.reston.ans.net!wupost!micro-heart-of-gold.mit.edu!uw-beaver!news.u.washington.edu!serval!hlu
- From: hlu@eecs.wsu.edu (H.J. Lu)
- Subject: Re: bug in signal.h
- Message-ID: <1993Jan9.002140.28879@serval.net.wsu.edu>
- Sender: news@serval.net.wsu.edu (USENET News System)
- Organization: School of EECS, Washington State University
- References: <1993Jan8.212309.10182@athena.mit.edu>
- Date: Sat, 9 Jan 93 00:21:40 GMT
- Lines: 24
-
- In article <1993Jan8.212309.10182@athena.mit.edu>, shap@halifax.syncomas.com writes:
- |> The macro for sigismember() needs to test that the signal number
- |> passed is <= _NSIG. Some programs depend on sigismember responding
- ^^
- It should be `< _NSIG' (0 - 31).
-
- |> with '0' if the value is out of range.
- |>
- |> I found this when trying to build CVS.
- |>
- |>
- |> Thanks
- |>
- |>
- |> Jonathan S. Shapiro
-
- Fixed. Use this one:
-
- --------
- #define __sigismember(set, sig) \
- ((((sig) < _NSIG) && (*(set) & __sigmask (sig))) ? 1 : 0)
- --------
-
- H.J.
-