From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarkko Nikula Subject: Re: [PATCH] ASoC: rx51: Correct DAPM route for capture Date: Mon, 02 Feb 2015 21:29:10 +0200 Message-ID: <54CFD006.6080505@bitmer.com> References: <1422603163-13346-1-git-send-email-peter.ujfalusi@ti.com> <54CBCE44.6040907@bitmer.com> <54CFB01D.7080908@metafoo.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from bitmer.com (bitmer.com [213.157.87.50]) by alsa0.perex.cz (Postfix) with ESMTP id 581732606FC for ; Mon, 2 Feb 2015 20:29:35 +0100 (CET) In-Reply-To: <54CFB01D.7080908@metafoo.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: alsa-devel-bounces@alsa-project.org Sender: alsa-devel-bounces@alsa-project.org To: Lars-Peter Clausen , Peter Ujfalusi , Mark Brown Cc: alsa-devel@alsa-project.org, Liam Girdwood , pavel@ucw.cz List-Id: alsa-devel@alsa-project.org On 02/02/2015 07:13 PM, Lars-Peter Clausen wrote: > On 01/30/2015 07:32 PM, Jarkko Nikula wrote: > [...] >> Here's additional data. Without this patch there are these errors below >> when registering the card: >> >> [ 8.804626] rx51-audio n900-audio: Connecting non-supply widget to >> supply widget is not supported (DMic -> Mic Bias) >> [ 8.815795] rx51-audio n900-audio: ASoC: no dapm match for DMic --> >> (null) --> Mic Bias >> [ 8.824249] rx51-audio n900-audio: ASoC: Failed to add route DMic -> >> direct -> Mic Bias >> [ 8.832763] rx51-audio n900-audio: Connecting non-supply widget to >> supply widget is not supported (HS Mic -> b Mic Bias) >> [ 8.844238] rx51-audio n900-audio: ASoC: no dapm match for HS Mic --> >> (null) --> b Mic Bias >> [ 8.853057] rx51-audio n900-audio: ASoC: Failed to add route HS Mic >> -> direct -> b Mic Bias >> >> This don't happen on v3.18 so in that sense Peter's patch is kind of >> half-regression fix for v3.19. > > What's new in v3.19 is that the kernel will shout if it encounters such > a invalid route, previously it was just silently ignored. But either way > it never actually worked, so it's not really a v3.19 regression. > It used to work in the past, not sure about 3.18 since I didn't actually test recording due missing mic bias from DT data (i.e. recording has been actually broken quite many releases) but looks like your e409dfbfccf9 ("ASoC: dapm: Add a few supply widget sanity checks") made it visible. Fortunately Peter saw it from sources even without above report. Good to have active users that report issues :-) -- Jarkko