linux-kernel - Re: [PATCH v2 22/28] mfd: core: Ensure disabled devices are skiped without aborting

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]

Message-ID: <20230727121834.475541d0@bootlin.com>
Date:   Thu, 27 Jul 2023 12:18:34 +0200
From:   Herve Codina <herve.codina@...tlin.com>
To:     Lee Jones <lee@...nel.org>
Cc:     "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>, Andrew Lunn <andrew@...n.ch>,
        Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Conor Dooley <conor+dt@...nel.org>,
        Linus Walleij <linus.walleij@...aro.org>,
        Qiang Zhao <qiang.zhao@....com>, Li Yang <leoyang.li@....com>,
        Liam Girdwood <lgirdwood@...il.com>,
        Mark Brown <broonie@...nel.org>,
        Jaroslav Kysela <perex@...ex.cz>,
        Takashi Iwai <tiwai@...e.com>,
        Shengjiu Wang <shengjiu.wang@...il.com>,
        Xiubo Li <Xiubo.Lee@...il.com>,
        Fabio Estevam <festevam@...il.com>,
        Nicolin Chen <nicoleotsuka@...il.com>,
        Christophe Leroy <christophe.leroy@...roup.eu>,
        Randy Dunlap <rdunlap@...radead.org>, netdev@...r.kernel.org,
        linuxppc-dev@...ts.ozlabs.org, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-gpio@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org, alsa-devel@...a-project.org,
        Thomas Petazzoni <thomas.petazzoni@...tlin.com>
Subject: Re: [PATCH v2 22/28] mfd: core: Ensure disabled devices are skiped
 without aborting

Hi Lee,

On Thu, 27 Jul 2023 10:22:09 +0100
Lee Jones <lee@...nel.org> wrote:

> On Wed, 26 Jul 2023, Herve Codina wrote:
> 
> > The loop searching for a matching device based on its compatible
> > string is aborted when a matching disabled device is found.
> > This abort avoid to add devices as soon as one disabled device
> > is found.
> > 
> > Continue searching for an other device instead of aborting on the
> > first disabled one fixes the issue.
> > 
> > Fixes: 22380b65dc70 ("mfd: mfd-core: Ensure disabled devices are ignored without error")
> > Signed-off-by: Herve Codina <herve.codina@...tlin.com>
> > ---
> >  drivers/mfd/mfd-core.c | 18 +++++++++++++-----
> >  1 file changed, 13 insertions(+), 5 deletions(-)
> > 
> > diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c
> > index 0ed7c0d7784e..bcc26e64639a 100644
> > --- a/drivers/mfd/mfd-core.c
> > +++ b/drivers/mfd/mfd-core.c
> > @@ -146,6 +146,7 @@ static int mfd_add_device(struct device *parent, int id,
> >  	struct platform_device *pdev;
> >  	struct device_node *np = NULL;
> >  	struct mfd_of_node_entry *of_entry, *tmp;
> > +	bool disabled;
> >  	int ret = -ENOMEM;
> >  	int platform_id;
> >  	int r;
> > @@ -181,13 +182,13 @@ static int mfd_add_device(struct device *parent, int id,
> >  		goto fail_res;
> >  
> >  	if (IS_ENABLED(CONFIG_OF) && parent->of_node && cell->of_compatible) {
> > +		disabled = false;  
> 
> This does not appear to reside in a loop.
> 
> Why not set it to false on declaration?

Indeed, I will change in the next iteration and set the value to false at
the declaration.

> 
> >  		for_each_child_of_node(parent->of_node, np) {
> >  			if (of_device_is_compatible(np, cell->of_compatible)) {
> > -				/* Ignore 'disabled' devices error free */
> > +				/* Skip 'disabled' devices */
> >  				if (!of_device_is_available(np)) {
> > -					of_node_put(np);  
> 
> Doesn't this result in a resource leak?

No because we change from 'goto fail_alias' to 'continue' and so we don't
exit from the for_each_child_of_node().
The for_each_child_of_node() calls of_get_next_child() and, in turn, calls
of_node_put().

Regards,
Hervé

> 
> > -					ret = 0;
> > -					goto fail_alias;
> > +					disabled = true;
> > +					continue;
> >  				}
> >  
> >  				ret = mfd_match_of_node_to_dev(pdev, np, cell);
> > @@ -197,10 +198,17 @@ static int mfd_add_device(struct device *parent, int id,
> >  				if (ret)
> >  					goto fail_alias;
> >  
> > -				break;
> > +				goto match;
> >  			}
> >  		}
> >  
> > +		if (disabled) {
> > +			/* Ignore 'disabled' devices error free */
> > +			ret = 0;
> > +			goto fail_alias;
> > +		}
> > +
> > +match:
> >  		if (!pdev->dev.of_node)
> >  			pr_warn("%s: Failed to locate of_node [id: %d]\n",
> >  				cell->name, platform_id);
> > -- 
> > 2.41.0
> >   
>