Re: [PATCH 1/4] ASoC: Intel: sof_cs42l42: support JSL DAI link sequence - Pierre-Louis Bossart

From: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
To: "Lu, Brent" <brent.lu@intel.com>,
	"alsa-devel@alsa-project.org" <alsa-devel@alsa-project.org>
Cc: "Rojewski, Cezary" <cezary.rojewski@intel.com>,
	Liam Girdwood <liam.r.girdwood@linux.intel.com>,
	Jie Yang <yang.jie@linux.intel.com>,
	Mark Brown <broonie@kernel.org>, Jaroslav Kysela <perex@perex.cz>,
	Takashi Iwai <tiwai@suse.com>,
	"Wang, Rander" <rander.wang@intel.com>,
	Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Ranjani Sridharan <ranjani.sridharan@linux.intel.com>,
	Kai Vehmanen <kai.vehmanen@linux.intel.com>,
	"R, Dharageswari" <dharageswari.r@intel.com>,
	"Nujella, Sathyanarayana" <sathyanarayana.nujella@intel.com>,
	Bard Liao <yung-chuan.liao@linux.intel.com>,
	Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>,
	"M, Naveen" <naveen.m@intel.com>,
	"Zhi, Yong" <yong.zhi@intel.com>,
	"Gopal, Vamshi Krishna" <vamshi.krishna.gopal@intel.com>,
	Fred Oh <fred.oh@linux.intel.com>,
	Tzung-Bi Shih <tzungbi@google.com>
Subject: Re: [PATCH 1/4] ASoC: Intel: sof_cs42l42: support JSL DAI link sequence
Date: Mon, 7 Jun 2021 15:49:09 -0500	[thread overview]
Message-ID: <a3b5d6be-17eb-088f-197c-8a0b30c2f32d@linux.intel.com> (raw)
In-Reply-To: <DM6PR11MB36429F4E8A6F077F8D2374F997389@DM6PR11MB3642.namprd11.prod.outlook.com>

> Cs42l42 is sharing topology with DA7219's topology source file sof-glk-da7219.m4
> on GLK platform. The configuration is:
> 
> dai link id 0 is for spk
> #SSP 1 (ID: 0) with 19.2 MHz mclk with MCLK_ID 1 (unused), 1.536 MHz blck
> DAI_CONFIG(SSP, 1, 0, SSP1-Codec,
> 
> dai link id 1 is for headphone
> #SSP 2 (ID: 1) with 19.2 MHz mclk with MCLK_ID 1, 1.92 MHz bclk
> DAI_CONFIG(SSP, 2, 1, SSP2-Codec,
> 
> dai link id 2 is for dmic
> DAI_CONFIG(DMIC, 0, 2, dmic01,
> 
> dai link id 3/4/5 is for hdmi
> DAI_CONFIG(HDA, 3, 3, iDisp1,
> DAI_CONFIG(HDA, 4, 4, iDisp2,
> DAI_CONFIG(HDA, 5, 5, iDisp3,
> 
> When on JSL, we plan to share topology with rt5682 which has different dai link
> sequence:
> sof-jsl-rt5682.m4:
> 
> dai link id 0 is for headphone
> DAI_CONFIG(SSP, 0, 0, SSP0-Codec,
> 
> dai link id 6 is for spk
> # SSP 1 (ID: 6)
> DAI_CONFIG(SSP, SPK_INDEX, 6, SPK_NAME,
> 	SET_SSP_CONFIG)
> 
> dai link id 3/4/5 is for hdmi
> # 4 HDMI/DP outputs (ID: 3,4,5)
> DAI_CONFIG(HDA, 0, 3, iDisp1,
> DAI_CONFIG(HDA, 1, 4, iDisp2,
> DAI_CONFIG(HDA, 2, 5, iDisp3,
> 
> I'm not sure if there is convention about the sequence to follow?

ok, now I get what you are trying to do.

Unfortunately there are no conventions so far, and since we have to be 
backwards-compatible with topology files already released we will need 
to deal with the different configurations in this machine driver, you're 
right about this.

The code you suggested is fine, but we can future-proof it a bit.

Instead of assuming any order depending on GLK or !GLK, we can add a BE 
'base' for headphone, amp, DMIC and DMIC each (represented as a constant 
structure) and point to different configurations depending on a quirk. 
That way we can deal with other permutations such as HP - SPK - HDMI - DMIC