Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932411Ab1BCSlr (ORCPT ); Thu, 3 Feb 2011 13:41:47 -0500 Received: from mail-wy0-f174.google.com ([74.125.82.174]:45689 "EHLO mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932312Ab1BCSlq convert rfc822-to-8bit (ORCPT ); Thu, 3 Feb 2011 13:41:46 -0500 MIME-Version: 1.0 In-Reply-To: <1296685682.3336.302.camel@work-vm> References: <1296577820-27456-1-git-send-email-mroberto@cpti.cetuc.puc-rio.br> <1296615605.3336.239.camel@work-vm> <1296685479.3336.298.camel@work-vm> <1296685682.3336.302.camel@work-vm> From: Marcelo Roberto Jimenez Date: Thu, 3 Feb 2011 16:41:25 -0200 Message-ID: Subject: Re: [PATCH] RTC: Fix for issues in the kernel RTC API. To: john stultz Cc: a.zummo@towertech.it, rtc-linux@googlegroups.com, linux-kernel@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1973 Lines: 44 Hi John, On Wed, Feb 2, 2011 at 20:28, john stultz wrote: > On Wed, 2011-02-02 at 14:24 -0800, John Stultz wrote: >> On Wed, 2011-02-02 at 15:58 -0200, Marcelo Roberto Jimenez wrote: >> > Again, what the rtc-test kernel RTC and the strongarm RTC user space >> > behavior have changed. Alarm interrupts and update interrupts were >> > generated by a different interrupts in the strongarm driver, and the >> > rtc-test driver also behaved similarly, i.e., an update interrupt did >> > not trigger an alarm interrupt. Currently, rtc_handle_legacy_irq() >> > centralizes the irq processing, and by not checking the generated >> > interrupt, it allows the new behavior, which seemed broken to me. >> >> So > > Sorry. I didn't finish my thought here. (I *did not* mean "So?" :) > > So... yes, we should make sure its not broken. Could you explain some > more details about the different interrupts fro the driver? ?If the two > interrupt types come from different sources, is there a problem actually > using the alarm irq to emulate the update irq? In other words, can we > just skip the update irq management code? Or is there a draw back to > that? In a real timer device, I don't see a problem, as time passes the same for everything, including update and alarm interrupts. This is an issue only in the rtc-test driver, which is just a debug device. Since everything is now emulated in timers, the word IRQ does not make sense anymore, the right thing to do would be a global rename to remove references to that string in the RTC framework. The rtc-test driver is based on the concept of an interrupt, which is no longer the case. I think your implementation is fine. > thanks > -john Regards, Marcelo. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/