linux-kernel - Re: [PATCH 3/9] memory: mtk-smi: Use clk

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 11 Jul 2021 16:29:59 +0800
From:   Yong Wu <yong.wu@...iatek.com>
To:     Krzysztof Kozlowski <krzysztof.kozlowski@...onical.com>
CC:     Rob Herring <robh+dt@...nel.org>,
        Matthias Brugger <matthias.bgg@...il.com>,
        Krzysztof Kozlowski <krzk@...nel.org>,
        "Joerg Roedel" <joro@...tes.org>, Will Deacon <will@...nel.org>,
        Robin Murphy <robin.murphy@....com>,
        Tomasz Figa <tfiga@...omium.org>,
        <linux-mediatek@...ts.infradead.org>,
        <srv_heupstream@...iatek.com>, <linux-kernel@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <iommu@...ts.linux-foundation.org>, <youlin.pei@...iatek.com>,
        <anan.sun@...iatek.com>, <ming-fan.chen@...iatek.com>,
        <yi.kuo@...iatek.com>, <anthony.huang@...iatek.com>
Subject: Re: [PATCH 3/9] memory: mtk-smi: Use clk_bulk instead of the clk ops

On Thu, 2021-07-08 at 11:32 +0200, Krzysztof Kozlowski wrote:
> On 16/06/2021 13:43, Yong Wu wrote:
> > smi have many clocks: apb/smi/gals.
> > This patch use clk_bulk interface instead of the orginal one to simply
> > the code.
> > 
> > gals is optional clk(some larbs may don't have gals). use clk_bulk_optional
> > instead. and then remove the has_gals flag.
> > 
> > Also remove clk fail logs since bulk interface already output fail log.
> > 
> > Signed-off-by: Yong Wu <yong.wu@...iatek.com>
> > ---
> >  drivers/memory/mtk-smi.c | 124 +++++++++++----------------------------
> >  1 file changed, 34 insertions(+), 90 deletions(-)
> > 
> > diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c
> > index c5fb51f73b34..bcd2bf130655 100644
> > --- a/drivers/memory/mtk-smi.c
> > +++ b/drivers/memory/mtk-smi.c
> > @@ -60,9 +60,18 @@ enum mtk_smi_gen {
> >  	MTK_SMI_GEN2
> >  };
> >  
> > +#define MTK_SMI_CLK_NR_MAX			4
> > +
> > +static const char * const mtk_smi_common_clocks[] = {
> > +	"apb", "smi", "gals0", "gals1", /* glas is optional */
> 
> Typo here - glas.

Will Fix. Thanks.

> 
> > +};
> > +

[snip]

> > @@ -493,7 +449,7 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
> >  	struct device *dev = &pdev->dev;
> >  	struct mtk_smi *common;
> >  	struct resource *res;
> > -	int ret;
> > +	int i, ret;
> >  
> >  	common = devm_kzalloc(dev, sizeof(*common), GFP_KERNEL);
> >  	if (!common)
> > @@ -501,23 +457,13 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
> >  	common->dev = dev;
> >  	common->plat = of_device_get_match_data(dev);
> >  
> > -	common->clk_apb = devm_clk_get(dev, "apb");
> > -	if (IS_ERR(common->clk_apb))
> > -		return PTR_ERR(common->clk_apb);
> > -
> > -	common->clk_smi = devm_clk_get(dev, "smi");
> > -	if (IS_ERR(common->clk_smi))
> > -		return PTR_ERR(common->clk_smi);
> > +	common->clk_num = ARRAY_SIZE(mtk_smi_common_clocks);
> > +	for (i = 0; i < common->clk_num; i++)
> > +		common->clks[i].id = mtk_smi_common_clocks[i];
> >  
> > -	if (common->plat->has_gals) {
> > -		common->clk_gals0 = devm_clk_get(dev, "gals0");
> > -		if (IS_ERR(common->clk_gals0))
> > -			return PTR_ERR(common->clk_gals0);
> > -
> > -		common->clk_gals1 = devm_clk_get(dev, "gals1");
> > -		if (IS_ERR(common->clk_gals1))
> > -			return PTR_ERR(common->clk_gals1);
> > -	}
> > +	ret = devm_clk_bulk_get_optional(dev, common->clk_num, common->clks);
> > +	if (ret)
> > +		return ret;
> 
> How do you handle now missing required clocks?

It looks this is a common issue for this function which supports all the
clocks could be optional. Is there common suggestion for this?

For our case, the apb/smi clocks are required while "gals" are optional.

thus, we should use devm_clk_bulk_get for the necessary clocks and
devm_clk_bulk_get_optional for the optional ones. right?

> 
> >  
> >  	/*
> >  	 * for mtk smi gen 1, we need to get the ao(always on) base to config
> > @@ -561,11 +507,9 @@ static int __maybe_unused mtk_smi_common_resume(struct device *dev)
> >  	u32 bus_sel = common->plat->bus_sel;
> >  	int ret;
> >  
> > -	ret = mtk_smi_clk_enable(common);
> > -	if (ret) {
> > -		dev_err(common->dev, "Failed to enable clock(%d).\n", ret);
> > +	ret = clk_bulk_prepare_enable(common->clk_num, common->clks);
> > +	if (ret)
> >  		return ret;
> > -	}
> >  
> >  	if (common->plat->gen == MTK_SMI_GEN2 && bus_sel)
> >  		writel(bus_sel, common->base + SMI_BUS_SEL);
> > @@ -576,7 +520,7 @@ static int __maybe_unused mtk_smi_common_suspend(struct device *dev)
> >  {
> >  	struct mtk_smi *common = dev_get_drvdata(dev);
> >  
> > -	mtk_smi_clk_disable(common);
> > +	clk_bulk_disable_unprepare(common->clk_num, common->clks);
> >  	return 0;
> >  }
> >  
> > 
> 
> 
> Best regards,
> Krzysztof