From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mika Westerberg Subject: Re: [PATCH] spi/spidev: Convert to use unified device property API Date: Tue, 30 Dec 2014 09:37:14 +0200 Message-ID: <20141230073714.GD1386@lahna.fi.intel.com> References: <1419846074-240217-1-git-send-email-mika.westerberg@linux.intel.com> <20141229160316.GO17800@sirena.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-spi@vger.kernel.org, Grant Likely , Rob Herring , "Rafael J. Wysocki" , linux-kernel@vger.kernel.org To: Mark Brown Return-path: Content-Disposition: inline In-Reply-To: <20141229160316.GO17800@sirena.org.uk> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-spi.vger.kernel.org On Mon, Dec 29, 2014 at 04:03:16PM +0000, Mark Brown wrote: > On Mon, Dec 29, 2014 at 11:41:14AM +0200, Mika Westerberg wrote: > > This will allow the driver to match using DT compatible property if the > > device has ACPI _HID of "PRP0001" and accompanying "compatible" property > > listed in _DSD. > > Which nobody should be doing since it's already totally broken for > something to be using this on the DT side except in the one case of the > Rohm device which is listed there. I don't want to merge anything which > allows the breakage we're seeing with people putting spidev in their DTs > to be propagated into ACPI, at most we should have something that > specifically identifies individual devices only. It is pretty convenient for testing SPI bus and that's why I thought it would be good to have possibility to enumerate this in similar way than DT does but I understand your point.