From mboxrd@z Thu Jan 1 00:00:00 1970 From: Phil Reid Subject: [RFC PATCH 0/1] i2c: i2c-mux-pca954x: Add interrupt controller support Date: Wed, 27 Jul 2016 11:05:55 +0800 Message-ID: <1469588756-70579-1-git-send-email-preid@electromag.com.au> Return-path: Received: from 203-59-230-133.perm.iinet.net.au ([203.59.230.133]:51396 "EHLO preid-centos7.electromag.com.au" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751561AbcG0DO1 (ORCPT ); Tue, 26 Jul 2016 23:14:27 -0400 Sender: linux-i2c-owner@vger.kernel.org List-Id: linux-i2c@vger.kernel.org To: wsa@the-dreams.de, peda@axentia.se, linux-i2c@vger.kernel.org Cc: Phil Reid G'day All, I'm seeking some feedback on this. Not really happy with it but can't think of a good solution. This patch adds support to read the irq status register from the a PCA9543 i2c mux. This chip has two active low interrupt inputs with are combined into a single active low interrupt output. It does not provide the ability to mask interrupts. Only to read which interrupt line is active. On the hardware I'm interfacing too I have an ltc1760 sbs manager on each multiplexed bus. On booting these may assert there SMBALERT irq outputs immediately, again these devices have no ability to mask there interrupt. So the irq is asserted during boot until the ltc1760 driver is started. And there's no way to mask it. My work around is to enable the irq until both irq pins are unmasked. Not sure what else to do. The masking could be a dts property to make it configurable. ltc1760 driver is currently being review for submission in Karl-Heinz Schneider patch series. Appreciate any feedback on what could be done here. Phil Reid (1): i2c: i2c-mux-pca954x: Add interrupt controller support. drivers/i2c/muxes/i2c-mux-pca954x.c | 116 ++++++++++++++++++++++++++++++++++++ 1 file changed, 116 insertions(+) -- 1.8.3.1