linux-kernel - Re: [PATCH] PCI: imx6: Fix wrong check in imx6_pcie_attach

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:   Tue, 13 Sep 2022 09:39:03 +0200
From:   Lucas Stach <l.stach@...gutronix.de>
To:     Tang Bin <tangbin@...s.chinamobile.com>, hongxing.zhu@....com,
        lorenzo.pieralisi@....com, robh@...nel.org, kw@...ux.com,
        shawnguo@...nel.org, bhelgaas@...gle.com, s.hauer@...gutronix.de,
        kernel@...gutronix.de, festevam@...il.com, linux-imx@....com
Cc:     linux-pci@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] PCI: imx6: Fix wrong check in imx6_pcie_attach_pd()

Am Dienstag, dem 13.09.2022 um 14:59 +0800 schrieb Tang Bin:
> In the function imx6_pcie_attach_pd(),
> dev_pm_domain_attach_by_name() may return NULL in some cases,
> so IS_ERR() doesn't meet the requirements. Thus fix it.
> 
I don't like this added complexity in the driver. IHMO if there is a
real issue, dev_pm_domain_attach_by_name() should just return a error
code, instead of NULL. The fact that you need to pull a error code out
of thin air in the driver is a big hint that this should be fixed in
the called function, not in the return handling in the driver.

A bit down the callstack genpd_dev_pm_attach_by_id() is called, which
is documented like this "Returns the created virtual device if
successfully attached PM domain, NULL when the device don't need a PM
domain [...]". NULL is a valid return code, where the driver should
_not_ stop probing, as the device should work without the power domain
attached.

Regards,
Lucas

> Fixes: 3f7cceeab895 ("PCI: imx: Add multi-pd support")
> Signed-off-by: Tang Bin <tangbin@...s.chinamobile.com>
> ---
>  drivers/pci/controller/dwc/pci-imx6.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/pci/controller/dwc/pci-imx6.c b/drivers/pci/controller/dwc/pci-imx6.c
> index 6619e3caf..65d6ebbba 100644
> --- a/drivers/pci/controller/dwc/pci-imx6.c
> +++ b/drivers/pci/controller/dwc/pci-imx6.c
> @@ -337,8 +337,8 @@ static int imx6_pcie_attach_pd(struct device *dev)
>  		return 0;
>  
>  	imx6_pcie->pd_pcie = dev_pm_domain_attach_by_name(dev, "pcie");
> -	if (IS_ERR(imx6_pcie->pd_pcie))
> -		return PTR_ERR(imx6_pcie->pd_pcie);
> +	if (IS_ERR_OR_NULL(imx6_pcie->pd_pcie))
> +		return PTR_ERR(imx6_pcie->pd_pcie) ? : -ENODATA;
>  	/* Do nothing when power domain missing */
>  	if (!imx6_pcie->pd_pcie)
>  		return 0;
> @@ -352,8 +352,8 @@ static int imx6_pcie_attach_pd(struct device *dev)
>  	}
>  
>  	imx6_pcie->pd_pcie_phy = dev_pm_domain_attach_by_name(dev, "pcie_phy");
> -	if (IS_ERR(imx6_pcie->pd_pcie_phy))
> -		return PTR_ERR(imx6_pcie->pd_pcie_phy);
> +	if (IS_ERR_OR_NULL(imx6_pcie->pd_pcie_phy))
> +		return PTR_ERR(imx6_pcie->pd_pcie_phy) ? : -ENODATA;
>  
>  	link = device_link_add(dev, imx6_pcie->pd_pcie_phy,
>  			DL_FLAG_STATELESS |