linux-kernel - Re: [PATCH v3 2/2] memory: mtk-smi: mt8188: Add SMI reset and clamp for MT8188

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] [day] [month] [year] [list]
Message-ID: <c4ce6e5c6350e6cfb5ff9bebf3118be0ac76edbd.camel@mediatek.com>
Date: Wed, 22 Jan 2025 07:39:43 +0000
From: Friday Yang (杨阳) <Friday.Yang@...iatek.com>
To: "p.zabel@...gutronix.de" <p.zabel@...gutronix.de>, "robh@...nel.org"
	<robh@...nel.org>, Yong Wu (吴勇)
	<Yong.Wu@...iatek.com>, "conor+dt@...nel.org" <conor+dt@...nel.org>,
	"krzk@...nel.org" <krzk@...nel.org>, AngeloGioacchino Del Regno
	<angelogioacchino.delregno@...labora.com>, "matthias.bgg@...il.com"
	<matthias.bgg@...il.com>
CC: "linux-arm-kernel@...ts.infradead.org"
	<linux-arm-kernel@...ts.infradead.org>, "linux-kernel@...r.kernel.org"
	<linux-kernel@...r.kernel.org>, "linux-mediatek@...ts.infradead.org"
	<linux-mediatek@...ts.infradead.org>, "devicetree@...r.kernel.org"
	<devicetree@...r.kernel.org>, Project_Global_Chrome_Upstream_Group
	<Project_Global_Chrome_Upstream_Group@...iatek.com>
Subject: Re: [PATCH v3 2/2] memory: mtk-smi: mt8188: Add SMI reset and clamp
 for MT8188

On Tue, 2025-01-21 at 09:53 +0100, Philipp Zabel wrote:
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> 
> 
> On Di, 2025-01-21 at 14:49 +0800, Friday Yang wrote:
> > From: "friday.yang" <friday.yang@...iatek.com>
> > 
> > To prevent handling glitch signals during MTCMOS on/off
> > transitions,
> > SMI requires clamp and reset operations. Parse the reset settings
> > for
> > SMI LARBs and the clamp settings for the SMI Sub-Common. Register
> > genpd callback for the SMI LARBs located in image, camera and IPE
> > subsystems, and apply reset and clamp operations within the
> > callback.
> > 
> > Signed-off-by: Friday Yang <friday.yang@...iatek.com>
> > ---
> >  drivers/memory/mtk-smi.c | 141
> > +++++++++++++++++++++++++++++++++++++--
> >  1 file changed, 137 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c
> > index 5710348f72f6..aaeb80379ec1 100644
> > --- a/drivers/memory/mtk-smi.c
> > +++ b/drivers/memory/mtk-smi.c
> 
> [...]
> > @@ -528,6 +598,53 @@ static int mtk_smi_dts_clk_init(struct device
> > *dev, struct mtk_smi *smi,
> >       return ret;
> >  }
> > 
> > +static int mtk_smi_larb_parse_clamp_optional(struct mtk_smi_larb
> > *larb)
> > +{
> > +     struct device *dev = larb->dev;
> > +     const struct mtk_smi_larb_gen *larb_gen = larb->larb_gen;
> > +     u32 larb_id;
> > +     int ret;
> > +
> > +     /**
> > +      * Only SMI LARBs in camera, image and IPE subsys need to
> > +      * apply clamp and reset operations, others can be skipped.
> > +      */
> > +     ret = of_property_read_u32(dev->of_node, "mediatek,larb-id",
> > &larb_id);
> > +     if (ret)
> > +             return -EINVAL;
> > +     if (!larb_gen->clamp_port || !larb_gen->clamp_port[larb_id])
> > +             return 0;
> > +
> > +     larb->sub_comm_inport = larb_gen->clamp_port[larb_id];
> > +     larb->sub_comm_syscon = syscon_regmap_lookup_by_phandle(dev-
> > >of_node, "mediatek,smi");
> > +     if (IS_ERR(larb->sub_comm_syscon)) {
> > +             larb->sub_comm_syscon = NULL;
> > +             return dev_err_probe(dev, -EINVAL,
> > +                                  "Unknown clamp port for larb
> > %d\n", larb_id);
> > +     }
> > +
> > +     return 0;
> > +}
> > +
> > +static int mtk_smi_larb_parse_reset_optional(struct mtk_smi_larb
> > *larb)
> > +{
> > +     struct device *dev = larb->dev;
> > +     int ret;
> > +
> > +     larb->rst_con = devm_reset_control_get(dev, "larb");
> 
> Please use devm_reset_control_get_exclusive() directly.
> Or use devm_reset_control_get_optional_exclusive(), which returns
> NULL
> instead of -ENOENT. That way you can ...

Thanks for your comment, I will fix it in this way.

> 
> > +     if (IS_ERR(larb->rst_con))
> > +             return dev_err_probe(dev, PTR_ERR(larb->rst_con),
> > +                                  "Failed to get larb reset
> > controller\n");
> 
> ... suppress this error message in case of -ENOENT and return with:
> 
>         if (!larb->rst_con)
>                 return 0;
> 
> here.
> 
> > +
> > +     larb->nb.notifier_call = mtk_smi_genpd_callback;
> > +     ret = dev_pm_genpd_add_notifier(dev, &larb->nb);
> > +     if (ret)
> > +             return dev_err_probe(dev, -EINVAL,
> > +                                  "Failed to add genpd callback
> > %d\n", ret);
> > +
> > +     return 0;
> > +}
> > +
> >  static int mtk_smi_larb_probe(struct platform_device *pdev)
> >  {
> >       struct mtk_smi_larb *larb;
> > @@ -538,6 +655,7 @@ static int mtk_smi_larb_probe(struct
> > platform_device *pdev)
> >       if (!larb)
> >               return -ENOMEM;
> > 
> > +     larb->dev = dev;
> >       larb->larb_gen = of_device_get_match_data(dev);
> >       larb->base = devm_platform_ioremap_resource(pdev, 0);
> >       if (IS_ERR(larb->base))
> > @@ -554,15 +672,24 @@ static int mtk_smi_larb_probe(struct
> > platform_device *pdev)
> >       if (ret < 0)
> >               return ret;
> > 
> > -     pm_runtime_enable(dev);
> > +     ret = mtk_smi_larb_parse_clamp_optional(larb);
> > +     if (ret)
> > +             goto err_link_remove;
> > +
> > +     ret = mtk_smi_larb_parse_reset_optional(larb);
> > +     if (ret && ret != -ENOENT)
> 
> The ret != -ENOENT check could be dropped if you use
> devm_reset_control_get_optional_exclusive() above.
> 

OK, I will fix it.

> 
> regards
> Philipp