From owner-freebsd-current Thu Nov 16 16:29: 6 2000 Delivered-To: freebsd-current@freebsd.org Received: from birch.ripe.net (birch.ripe.net [193.0.1.96]) by hub.freebsd.org (Postfix) with ESMTP id F02B937B479 for ; Thu, 16 Nov 2000 16:29:03 -0800 (PST) Received: from office.ripe.net (office.ripe.net [193.0.1.97]) by birch.ripe.net (8.8.8/8.8.8) with ESMTP id BAA19709 for ; Fri, 17 Nov 2000 01:29:01 +0100 (CET) Received: (from marks@localhost) by office.ripe.net (8.8.8/8.8.5) id BAA11297 for freebsd-current@freebsd.org; Fri, 17 Nov 2000 01:29:01 +0100 (CET) Date: Fri, 17 Nov 2000 01:29:01 +0100 From: Mark Santcroos To: freebsd-current@freebsd.org Subject: strange behaviour of mkioctls in kdump/truss Message-ID: <20001117012901.C3410@ripe.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i X-Handles: MS6-6BONE, MS32260-NIC, MS18417-RIPE Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi, Since I've moved to -CURRENT a few weeks ago [I'm fairly new to FreeBSD] I had build problems with kdump and truss. The problem I report here are as far as I can judge them, sorry for any errors or inconsistencies. I've tried to find something about an earlier bug report but could not find any. The problem consisted of the definition for TELNO_MAX not defined or included in /usr/include/machine/i4b_rbch_ioctl.h. The definition for this is in /usr/include/machine/i4b_ioctl.h. By coincidence the building of kdump/truss succeeds, due to the fact that mkioctl uses find(1) for the retreiving of the include files. And because i4b_ioctl.h is alphabetical in front of i4b_rbch_ioctl.h it will be included before i4b_rbch_ioctl.h and therefor the value for TELNO_MAX is already defined on the moment i4b_rbch_ioctl.h is included. Due to some reason I don't know yet [please enlighten me on this] i4b_rbch_ioctl.h appeared before i4b_ioctl.h in _my_ ioctl.c! [Although like I told find(1) uses alphabetical order ?!?!] This happened still after various builds and cleans, etc. But after I have touched the files a bit [e.g. editing them for testing] they do also appear in the right order for me now. So in my case make buildworld stopped on this error because it did not know the value of TELNO_MAX, which is correct as far as it goes for the compiler part. Anyway, like I said earlier, due to some luck it went well for a long time. [that is why nobody reported it before probably, otherwise shoot me] The solution to this seems to be simple: #include in i4b_rbch_ioctl.h Don't hesitate to ask me anything more (that I forgot). Regards, Mark -- Mark Santcroos RIPE Network Coordination Centre PGP KeyID: 1024/0x3DCBEB8D PGP Fingerprint: BB1E D037 F29D 4B40 0B26 F152 795F FCAB 3DCB EB8D To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message