Return to BSD News archive
Newsgroups: comp.unix.solaris,comp.unix.bsd.misc,comp.unix.internals Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.mira.net.au!news.netspace.net.au!news.mel.connect.com.au!munnari.OZ.AU!spool.mu.edu!news.sgi.com!news.mathworks.com!fu-berlin.de!unihil!baghira.han.de!jum From: jum@anubis.han.de (Jens-Uwe Mager) Subject: Re: SunOS4.1.3 to Solaris2.5.1 Porting issues. Content-Type: text/plain; charset=ISO-8859-1 Distribution: inet X-Newsreader: Yet Another NewsWatcher 2.3.6 Sender: news@baghira.han.de (news at baghira.han.de) Content-Transfer-Encoding: 8bit Organization: At Home Message-ID: <jum-ya023680001612962351490001@baghira.han.de> References: <32986299.AC7@mail.esrin.esa.it> <57ik5l$12i$1@shade.twinsun.com> <x7917mx5gx.fsf@dumbcat.codewright.com> <57shh0$o3u@web.nmti.com> <58bdl4$g68@spitfire.ecsel.psu.edu> <58ccqm$q13@web.nmti.com> <58hj11$bb6@panix.com> <32ADB280.55AD@jpmorgan.com> <58m3ne$hju@innocence.interface-business.de> <casper.32afd36d@mail.fwi.uva.nl> <593j2l$efa@innocence.interface-business.de> Mime-Version: 1.0 Date: Mon, 16 Dec 1996 22:51:49 GMT Lines: 36 Xref: euryale.cc.adfa.oz.au comp.unix.solaris:92714 comp.unix.bsd.misc:1834 comp.unix.internals:11646 In article <593j2l$efa@innocence.interface-business.de>, joerg_wunsch@interface-business.de (Joerg Wunsch) wrote: > casper@fwi.uva.nl (Casper H.S. Dik) wrote: > > > >Well, it looks that BSD does provide better transitional help here > > >(despite of the Solaris people calling us ignorant against those > > >problems :). > > > > sigsetmask is not provided in standard libraries becaus eit doesn't > > do you any good; the interface was badly designed (if designed at all) > > That's still no reason to not help your customers in a transition: > provide a man page, and put it into libcompat.a. State this in the > man page. > > Or do you really believe that customers who have, heck!, _paid_ for > your stuff should get their knowledge from Usenet, as opposed to the > system's documentation? > > I didn't say i would encourage users in using obsolete interfaces > either; just help them to find their new way. If you (i.e., SMI) used > to ship various interfaces in previous systems (SunOS 4), your > customers can expect to at least find a migration guide for the > interfaces that have changed. That's been my point. I think it would be a mistake to provide sigsetmask if you are not able to manipulate all signals using it. As Solaris 2 uses more than 32 signals, a few would simply be discarded by the C integer arithmetic generating the mask and thus would not be masked. I think this produces much more difficult to find errors than a simple link error if the routine is not there. _______________________________________________________________________ Jens-Uwe Mager jum@anubis.han.de 30177 Hannover jum@helios.de Brahmsstr. 3 Tel.: +49 511 660238