From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thorsten Glaser Subject: Re: [PATCH 1/5] m68k/atari: EtherNAT - change number of Atari interrupts to make room for EtherNAT interrupts Date: Tue, 3 Apr 2012 22:31:08 +0000 (UTC) Message-ID: References: <1327085843-6980-1-git-send-email-geert@linux-m68k.org> <4F4B2BB6.900@gmail.com> <4F77C245.5090002@gmail.com> <4F7B6880.9050605@gmail.com> <4F7B77FF.200@gmail.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from static-87-79-237-121.netcologne.de ([87.79.237.121]:47452 "EHLO herc.mirbsd.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753519Ab2DCWer convert rfc822-to-8bit (ORCPT ); Tue, 3 Apr 2012 18:34:47 -0400 In-Reply-To: <4F7B77FF.200@gmail.com> Sender: linux-m68k-owner@vger.kernel.org List-Id: linux-m68k@vger.kernel.org To: Michael Schmitz Cc: linux-m68k@vger.kernel.org, debian-68k@lists.debian.org Michael Schmitz dixit: >>> For some reason, I have to set NUM_ATARI_SOURCES to 141 to actually= get >>> interrupt source 140 to register. Off-by-one error somewhere in the= generic >>> interrupt code I'd think, but I've not been able to spot it. (Just = increasing >>> =20 >> >> I think 141 interrupt sources covers sources #0 to #140, no? >> =20 > What's interrupt #0? #1 to #7 are level 1 to 7 autovector interrupts.= #8 to I was just guessing. An all-uppercase identifier in a C context that expands to an integer and has NUMber in its name might just be used in the declarartion of a C array, that=E2=80=99s why. > ragnar rings a bell, is that ragnar@chaosdorf.de? Chaosdorf yes, so, probably yes. bye, //mirabilos --=20 =E2=80=9CIt is inappropriate to require that a time represented as seconds since the Epoch precisely represent the number of seconds between the referenced time and the Epoch.=E2=80=9D -- IEEE Std 1003.1b-1993 (POSIX) Section B.2.2.2