From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from metis.ext.pengutronix.de (unknown [IPv6:2001:6f8:1178:4:290:27ff:fe1d:cc33]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 37E581A0006 for ; Mon, 15 Sep 2014 23:52:24 +1000 (EST) Date: Mon, 15 Sep 2014 15:52:04 +0200 From: Markus Pargmann To: Shengjiu Wang Subject: Re: [PATCH V3] ASoC: fsl_ssi: refine ipg clock usage in this module Message-ID: <20140915135204.GF8844@pengutronix.de> References: <894383e00876763f22988fc5f3f9f232f939f923.1410517971.git.shengjiu.wang@freescale.com> <20140915100541.GD8844@pengutronix.de> <20140915102224.GA23877@audiosh1> <20140915103213.GE8844@pengutronix.de> <20140915103714.GB23877@audiosh1> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="gneEPciiIl/aKvOT" In-Reply-To: <20140915103714.GB23877@audiosh1> Cc: alsa-devel@alsa-project.org, lgirdwood@gmail.com, tiwai@suse.de, Li.Xiubo@freescale.com, timur@tabi.org, perex@perex.cz, nicoleotsuka@gmail.com, broonie@kernel.org, linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , --gneEPciiIl/aKvOT Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Sep 15, 2014 at 06:37:15PM +0800, Shengjiu Wang wrote: > On Mon, Sep 15, 2014 at 12:32:13PM +0200, Markus Pargmann wrote: > > On Mon, Sep 15, 2014 at 06:22:27PM +0800, Shengjiu Wang wrote: > > > On Mon, Sep 15, 2014 at 12:05:41PM +0200, Markus Pargmann wrote: > > > > On Fri, Sep 12, 2014 at 06:35:15PM +0800, Shengjiu Wang wrote: > > > > > Check if ipg clock is in clock-names property, then we can move t= he > > > > > ipg clock enable and disable operation to startup and shutdown, t= hat > > > > > is only enable ipg clock when ssi is working and keep clock is di= sabled > > > > > when ssi is in idle. > > > > > But when the checking is failed, remain the clock control as befo= re. > > > > >=20 > > > > > Signed-off-by: Shengjiu Wang > > > > > --- > > > > > V3 change log: > > > > > update patch according Nicolin and markus's comments > > > > >=20 > > > > >=20 > > > > > sound/soc/fsl/fsl_ssi.c | 53 +++++++++++++++++++++++++++++++++= +++++++------- > > > > > 1 file changed, 45 insertions(+), 8 deletions(-) > > > > >=20 > > > > > diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c > > > > > index 2fc3e66..6d1dfd5 100644 > > > > > --- a/sound/soc/fsl/fsl_ssi.c > > > > > +++ b/sound/soc/fsl/fsl_ssi.c > > > > > @@ -169,6 +169,7 @@ struct fsl_ssi_private { > > > > > u8 i2s_mode; > > > > > bool use_dma; > > > > > bool use_dual_fifo; > > > > > + bool has_ipg_clk_name; > > > > > unsigned int fifo_depth; > > > > > struct fsl_ssi_rxtx_reg_val rxtx_reg_val; > > > > > =20 > > > > > @@ -530,6 +531,11 @@ static int fsl_ssi_startup(struct snd_pcm_su= bstream *substream, > > > > > struct snd_soc_pcm_runtime *rtd =3D substream->private_data; > > > > > struct fsl_ssi_private *ssi_private =3D > > > > > snd_soc_dai_get_drvdata(rtd->cpu_dai); > > > > > + int ret; > > > > > + > > > > > + ret =3D clk_prepare_enable(ssi_private->clk); > > > > > + if (ret) > > > > > + return ret; > > > > > =20 > > > > > /* When using dual fifo mode, it is safer to ensure an even per= iod > > > > > * size. If appearing to an odd number while DMA always starts = its > > > > > @@ -544,6 +550,21 @@ static int fsl_ssi_startup(struct snd_pcm_su= bstream *substream, > > > > > } > > > > > =20 > > > > > /** > > > > > + * fsl_ssi_shutdown: shutdown the SSI > > > > > + * > > > > > + */ > > > > > +static void fsl_ssi_shutdown(struct snd_pcm_substream *substream, > > > > > + struct snd_soc_dai *dai) > > > > > +{ > > > > > + struct snd_soc_pcm_runtime *rtd =3D substream->private_data; > > > > > + struct fsl_ssi_private *ssi_private =3D > > > > > + snd_soc_dai_get_drvdata(rtd->cpu_dai); > > > > > + > > > > > + clk_disable_unprepare(ssi_private->clk); > > > > > + > > > > > +} > > > > > + > > > > > +/** > > > > > * fsl_ssi_set_bclk - configure Digital Audio Interface bit clock > > > > > * > > > > > * Note: This function can be only called when using SSI as DAI = master > > > > > @@ -1043,6 +1064,7 @@ static int fsl_ssi_dai_probe(struct snd_soc= _dai *dai) > > > > > =20 > > > > > static const struct snd_soc_dai_ops fsl_ssi_dai_ops =3D { > > > > > .startup =3D fsl_ssi_startup, > > > > > + .shutdown =3D fsl_ssi_shutdown, > > > > > .hw_params =3D fsl_ssi_hw_params, > > > > > .hw_free =3D fsl_ssi_hw_free, > > > > > .set_fmt =3D fsl_ssi_set_dai_fmt, > > > > > @@ -1168,17 +1190,22 @@ static int fsl_ssi_imx_probe(struct platf= orm_device *pdev, > > > > > u32 dmas[4]; > > > > > int ret; > > > > > =20 > > > > > - ssi_private->clk =3D devm_clk_get(&pdev->dev, NULL); > > > > > + if (ssi_private->has_ipg_clk_name) > > > > > + ssi_private->clk =3D devm_clk_get(&pdev->dev, "ipg"); > > > > > + else > > > > > + ssi_private->clk =3D devm_clk_get(&pdev->dev, NULL); > > > > > if (IS_ERR(ssi_private->clk)) { > > > > > ret =3D PTR_ERR(ssi_private->clk); > > > > > dev_err(&pdev->dev, "could not get clock: %d\n", ret); > > > > > return ret; > > > > > } > > > > > =20 > > > > > - ret =3D clk_prepare_enable(ssi_private->clk); > > > > > - if (ret) { > > > > > - dev_err(&pdev->dev, "clk_prepare_enable failed: %d\n", ret); > > > > > - return ret; > > > > > + if (!ssi_private->has_ipg_clk_name) { > > > > > + ret =3D clk_prepare_enable(ssi_private->clk); > > > > > + if (ret) { > > > > > + dev_err(&pdev->dev, "clk_prepare_enable failed: %d\n", ret); > > > > > + return ret; > > > > > + } > > > > > } > > > > > =20 > > > > > /* For those SLAVE implementations, we ingore non-baudclk cases > > > > > @@ -1236,8 +1263,9 @@ static int fsl_ssi_imx_probe(struct platfor= m_device *pdev, > > > > > return 0; > > > > > =20 > > > > > error_pcm: > > > > > - clk_disable_unprepare(ssi_private->clk); > > > > > =20 > > > > > + if (!ssi_private->has_ipg_clk_name) > > > > > + clk_disable_unprepare(ssi_private->clk); > > > > > return ret; > > > > > } > > > > > =20 > > > > > @@ -1246,7 +1274,8 @@ static void fsl_ssi_imx_clean(struct platfo= rm_device *pdev, > > > > > { > > > > > if (!ssi_private->use_dma) > > > > > imx_pcm_fiq_exit(pdev); > > > > > - clk_disable_unprepare(ssi_private->clk); > > > > > + if (!ssi_private->has_ipg_clk_name) > > > > > + clk_disable_unprepare(ssi_private->clk); > > > > > } > > > > > =20 > > > > > static int fsl_ssi_probe(struct platform_device *pdev) > > > > > @@ -1321,8 +1350,16 @@ static int fsl_ssi_probe(struct platform_d= evice *pdev) > > > > > return -ENOMEM; > > > > > } > > > > > =20 > > > > > - ssi_private->regs =3D devm_regmap_init_mmio(&pdev->dev, iomem, > > > > > + ret =3D of_property_match_string(np, "clock-names", "ipg"); > > > > > + if (ret < 0) { > > > > > + ssi_private->has_ipg_clk_name =3D false; > > > > > + ssi_private->regs =3D devm_regmap_init_mmio(&pdev->dev, iomem, > > > > > &fsl_ssi_regconfig); > > > >=20 > > > > Sorry if I was unclear about that. My suggestion was to enable the = clock > > > > right here: > > > > clk_prepare_enable(ssi_private->clk); > > > >=20 > > > > Then you can remove ssi_private->has_ipg_clk_name and all > > > > clk_prepare_enable() and clk_disable_unprepare() from above. Also y= ou > > > > can move the devm_clk_get() into this block. > > > > > > >=20 > > > ipg clock not only need to be enabled when accessing register, but al= so=20 > > > need to be enabled when ssi is working. So I add clk_enable in startu= p. > > > =20 > > > > It seems you really want to implement this for devicetrees where the > > > > "ipg" clock-name is missing, but I don't understand why? I really c= an't > > > > see any benefit of adding all these clk_prepare_enable() calls for = all > > > > cornercases that may occure. For example the clocks for AC97 are st= ill > > > > missing in this version. > > >=20 > > > When ipg clock-name is missing, I just remain the code logic as befor= e. > > > So AC97 case is same as before too.=20 > > > When have ipg clock-name, register it to regmap for accessing registe= r, > > > and enabling ipg clock in startup also is useful for AC97. > > >=20 > > > I don't understand why you think clock for AC97 is still missing? > >=20 > > Sorry got it all wrong, monday morning tiredness perhaps ;-). > >=20 > > The patch looks good for me at the second look. I will test it later > > today and give you feedback. > >=20 > > Best regards, > >=20 > > Markus >=20 > Thank you very much. Your comments help me a lot. and wait your feedback. Okay, I tested this patch on imx53-qsb with and without the "ipg" clock-name. Both worked, so: Tested-by: Markus Pargmann I noticed that checkpatch shows two minor style issues. It would be good if you could fix them. Best regards, Markus --=20 Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | --gneEPciiIl/aKvOT Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBAgAGBQJUFu8DAAoJEEpcgKtcEGQQIBQP/1NYOZgoJt4matvRdZnDZe7M KRw+PCOdHhst5APhXlrKGBJ3rcXsFdPWaO2NUN1iUAH74hoRD9Q/G41O+hfMHJQk bUGQrdvaXma86ZtoGUbsyQQi0KSFRSyiWcSaLpDGBGvr7pcYIRTcV/YygbZ+WOCW E84+oW7sqlBWRyDxlor1BLSC2Aa0SWhTcM5dMzQyOaGrSmaPxMzzvMtc7/1/rQ8T 1Cb629GtHMciSw5UxW8GXeMYSPgZU/sKVfmzec0/uwysmf7BwzfpKt0zIgWnxHek pIyYvC01g7TR9G4+z+EPRv90XMnWCHMbgPkgeWxJi0NDN8hIEFOiC+z1lHxHK2NG yYnShMBozTNMrSi68wQnMQGBS05SZLl9CQO/6nQUPqs4F55l3HtRYuNFeqHraL8w g3BQII1XM4KDMzEz8ADgfdiv6GVIKoy/dDbzK1eAtpCx9VnH7y9uRzFdqdx2ilD5 orKKTHR7cn5IpZ0AiMHyU553ofw8IZZU3JZYBFUOV1DfGK5Taa0VjFrnH0uvz5dS CTMH3ec7WSoEM3l07+shoyCY2+1sobmmqUY2uBz++JG5SNH30iXsqpWz2g7kYb/k QfFQjZP7UfATZg6eikLJNcYIlSLuvj59e6rXIVSWaFKXbVfUgWwN6C14zyXxjT9J VHoJo9HmoAbcppG0FCGA =hvTM -----END PGP SIGNATURE----- --gneEPciiIl/aKvOT--