Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
From:   "Sriranjani P" <sriranjani.p@samsung.com>
To:     "'Krzysztof Kozlowski'" <krzysztof.kozlowski@linaro.org>,
        <davem@davemloft.net>, <edumazet@google.com>, <kuba@kernel.org>,
        <pabeni@redhat.com>, <robh+dt@kernel.org>,
        <krzysztof.kozlowski+dt@linaro.org>, <conor+dt@kernel.org>,
        <richardcochran@gmail.com>, <alexandre.torgue@foss.st.com>,
        <joabreu@synopsys.com>, <mcoquelin.stm32@gmail.com>,
        <alim.akhtar@samsung.com>, <linux-fsd@tesla.com>,
        <pankaj.dubey@samsung.com>, <swathi.ks@samsung.com>,
        <ravi.patel@samsung.com>
Cc:     <netdev@vger.kernel.org>, <devicetree@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>,
        <linux-samsung-soc@vger.kernel.org>,
        <linux-arm-kernel@lists.infradead.org>,
        "'Chandrasekar R'" <rcsekar@samsung.com>,
        "'Suresh Siddha'" <ssiddha@tesla.com>
In-Reply-To: <b224ccaf-d70f-8f65-4b2f-6f4798841558@linaro.org>
Subject: RE: [PATCH v3 2/4] net: stmmac: dwc-qos: Add FSD EQoS support
Date:   Wed, 16 Aug 2023 12:08:50 +0530
Message-ID: <001201d9d00c$5413a9a0$fc3afce0$@samsung.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Thread-Index: AQG0kE2cByMDcfrFjxkR49X5VWx9JAI1Xm+tAnXeS/UBtEgXlbAD+MpQ
Content-Language: en-in
Content-Type: text/plain; charset="utf-8"
CMS-TYPE: 105P
DLP-Filter: Pass
References: <20230814112539.70453-1-sriranjani.p@samsung.com>
        <CGME20230814112612epcas5p275cffb4d3dae86c6090ca246083631c4@epcas5p2.samsung.com>
        <20230814112539.70453-3-sriranjani.p@samsung.com>
        <b224ccaf-d70f-8f65-4b2f-6f4798841558@linaro.org>
Precedence: bulk


> -----Original Message-----
> From: Krzysztof Kozlowski =5Bmailto:krzysztof.kozlowski=40linaro.org=5D
> Sent: 15 August 2023 01:21
> To: Sriranjani P <sriranjani.p=40samsung.com>; davem=40davemloft.net;
> edumazet=40google.com; kuba=40kernel.org; pabeni=40redhat.com;
> robh+dt=40kernel.org; krzysztof.kozlowski+dt=40linaro.org;
> conor+dt=40kernel.org; richardcochran=40gmail.com;
> alexandre.torgue=40foss.st.com; joabreu=40synopsys.com;
> mcoquelin.stm32=40gmail.com; alim.akhtar=40samsung.com; linux-
> fsd=40tesla.com; pankaj.dubey=40samsung.com; swathi.ks=40samsung.com;
> ravi.patel=40samsung.com
> Cc: netdev=40vger.kernel.org; devicetree=40vger.kernel.org; linux-
> kernel=40vger.kernel.org; linux-samsung-soc=40vger.kernel.org; linux-arm-
> kernel=40lists.infradead.org; Chandrasekar R <rcsekar=40samsung.com>;
> Suresh Siddha <ssiddha=40tesla.com>
> Subject: Re: =5BPATCH v3 2/4=5D net: stmmac: dwc-qos: Add FSD EQoS suppor=
t
>=20
> On 14/08/2023 13:25, Sriranjani P wrote:
> > The FSD SoC contains two instance of the Synopsys DWC ethernet QOS IP
> core.
> > The binding that it uses is slightly different from existing ones
> > because of the integration (clocks, resets).
> >
> > For FSD SoC, a mux switch is needed between internal and external clock=
s.
> > By default after reset internal clock is used but for receiving
> > packets properly, external clock is needed. Mux switch to external
> > clock happens only when the external clock is present.
> >
> > Signed-off-by: Chandrasekar R <rcsekar=40samsung.com>
> > Signed-off-by: Suresh Siddha <ssiddha=40tesla.com>
> > Signed-off-by: Swathi K S <swathi.ks=40samsung.com>
> > Signed-off-by: Sriranjani P <sriranjani.p=40samsung.com>
> > ---
>=20
>=20
> > +static int dwc_eqos_setup_rxclock(struct platform_device *pdev, int
> > +ins_num) =7B
> > +	struct device_node *np =3D pdev->dev.of_node;
> > +	struct regmap *syscon;
> > +	unsigned int reg;
> > +
> > +	if (np && of_property_read_bool(np, =22fsd-rx-clock-skew=22)) =7B
> > +		syscon =3D syscon_regmap_lookup_by_phandle_args(np,
> > +							      =22fsd-rx-clock-
> skew=22,
> > +							      1, &reg);
> > +		if (IS_ERR(syscon)) =7B
> > +			dev_err(&pdev->dev,
> > +				=22couldn't get the rx-clock-skew syscon=21=5Cn=22);
> > +			return PTR_ERR(syscon);
> > +		=7D
> > +
> > +		regmap_write(syscon, reg, rx_clock_skew_val=5Bins_num=5D);
> > +	=7D
> > +
> > +	return 0;
> > +=7D
> > +
> > +static int fsd_eqos_clk_init(struct fsd_eqos_plat_data *plat,
> > +			     struct plat_stmmacenet_data *data) =7B
> > +	int ret =3D 0, i;
> > +
> > +	const struct fsd_eqos_variant *fsd_eqos_v_data =3D
> > +						plat->fsd_eqos_inst_var;
> > +
> > +	plat->clks =3D devm_kcalloc(plat->dev, fsd_eqos_v_data->num_clks,
> > +				  sizeof(*plat->clks), GFP_KERNEL);
> > +	if (=21plat->clks)
> > +		return -ENOMEM;
> > +
> > +	for (i =3D 0; i < fsd_eqos_v_data->num_clks; i++)
> > +		plat->clks=5Bi=5D.id =3D fsd_eqos_v_data->clk_list=5Bi=5D;
> > +
> > +	ret =3D devm_clk_bulk_get(plat->dev, fsd_eqos_v_data->num_clks,
> > +				plat->clks);
>=20
> Instead of duplicating entire clock management with existing code, you
> should extend/rework existing one.
>=20
> This code is unfortunately great example how not to stuff vendor code int=
o
> upstream project. :(

I will check again if I can extend existing one to support FSD platform spe=
cific requirement.

>=20
> > +
> > +	return ret;
> > +=7D
> > +
> > +static int fsd_clks_endisable(void *priv, bool enabled) =7B
> > +	int ret, num_clks;
> > +	struct fsd_eqos_plat_data *plat =3D priv;
> > +
> > +	num_clks =3D plat->fsd_eqos_inst_var->num_clks;
> > +
> > +	if (enabled) =7B
> > +		ret =3D clk_bulk_prepare_enable(num_clks, plat->clks);
> > +		if (ret) =7B
> > +			dev_err(plat->dev, =22Clock enable failed, err =3D %d=5Cn=22,
> ret);
> > +			return ret;
> > +		=7D
> > +	=7D else =7B
> > +		clk_bulk_disable_unprepare(num_clks, plat->clks);
> > +	=7D
> > +
> > +	return 0;
> > +=7D
> > +
> > +static int fsd_eqos_probe(struct platform_device *pdev,
> > +			  struct plat_stmmacenet_data *data,
> > +			  struct stmmac_resources *res)
> > +=7B
> > +	struct fsd_eqos_plat_data *priv_plat;
> > +	struct device_node *np =3D pdev->dev.of_node;
> > +	int ret =3D 0;
> > +
> > +	priv_plat =3D devm_kzalloc(&pdev->dev, sizeof(*priv_plat),
> GFP_KERNEL);
> > +	if (=21priv_plat) =7B
> > +		ret =3D -ENOMEM;
>=20
> return -ENOMEM

Will fix this in v4.

>=20
> > +		goto error;
> > +	=7D
> > +
> > +	priv_plat->dev =3D &pdev->dev;
> > +	data->bus_id =3D of_alias_get_id(np, =22eth=22);
>=20
> No, you cannot do like this. Aliases are board specific and are based on
> labeling on the board.

So if I understood this correctly, I need to move alias in the board specif=
ic DTS file and I can use this, because we have to handle rx-clock-skew dif=
ferently for the two instances in the FSD platform. Another approach we too=
k in v1, by specifying the value to be programmed in rx-clock-skew property=
 itself, but it seems it is not a preferred approach.=20
I can see that in drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +43=
6 common code is already using this API and getting alias id, so I can prob=
ably use the same rather getting same again here, but I have to specify ali=
as in DTS file.

>=20
> > +
> > +	priv_plat->fsd_eqos_inst_var =3D &fsd_eqos_clk_info=5Bdata->bus_id=5D=
;
> > +
> > +	ret =3D fsd_eqos_clk_init(priv_plat, data);
> > +
> > +	data->bsp_priv =3D priv_plat;
> > +	data->clks_config =3D fsd_clks_endisable;
> > +	data->rxmux_setup =3D dwc_eqos_rxmux_setup;
> > +
> > +	ret =3D fsd_clks_endisable(priv_plat, true);
> > +	if (ret)
> > +		goto error;
> > +
> > +	ret =3D dwc_eqos_setup_rxclock(pdev, data->bus_id);
> > +	if (ret) =7B
> > +		fsd_clks_endisable(priv_plat, false);
> > +		dev_err_probe(&pdev->dev, ret, =22Unable to setup
> rxclock=5Cn=22);
>=20
> The syntax is: return dev_err_probe().

Will fix it in v4.

>=20
> > +	=7D
> > +
> > +error:
> > +	return ret;
> > +=7D
>=20
> ....
>=20
>=20
> Best regards,
> Krzysztof