linux-kernel - Re: Re: [PATCH] hwrng: ks-sa - fix runtime pm imbalance on error

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: <945e695.ba3c9.17235e473d9.Coremail.dinghao.liu@zju.edu.cn>
Date:   Thu, 21 May 2020 14:21:03 +0800 (GMT+08:00)
From:   dinghao.liu@....edu.cn
To:     "Sverdlin, Alexander (Nokia - DE/Ulm)" <alexander.sverdlin@...ia.com>
Cc:     "stern@...land.harvard.edu" <stern@...land.harvard.edu>,
        "kjlu@....edu" <kjlu@....edu>, "mpm@...enic.com" <mpm@...enic.com>,
        "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
        "herbert@...dor.apana.org.au" <herbert@...dor.apana.org.au>,
        "ben.dooks@...ethink.co.uk" <ben.dooks@...ethink.co.uk>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "arnd@...db.de" <arnd@...db.de>,
        "allison@...utok.net" <allison@...utok.net>,
        "yuehaibing@...wei.com" <yuehaibing@...wei.com>,
        "rfontana@...hat.com" <rfontana@...hat.com>,
        "linux-crypto@...r.kernel.org" <linux-crypto@...r.kernel.org>,
        "tglx@...utronix.de" <tglx@...utronix.de>
Subject: Re: Re: [PATCH] hwrng: ks-sa - fix runtime pm imbalance on error

Hi Alexander,

There are large amounts of cases that assume pm_runtime_get_sync()
will modify runtime PM usage counter on error. Fixing this in PM 
subsystem will influence all callers of pm_runtime_get_sync() and
introduce new bugs. Therefore I think the better solution is to fix
misused cases individually.

Dinghao

> Hello Dinghao,
> 
> On Wed, 2020-05-20 at 21:29 +0800, Dinghao Liu wrote:
> > pm_runtime_get_sync() increments the runtime PM usage counter even
> > the call returns an error code. Thus a pairing decrement is needed
> > on the error handling path to keep the counter balanced.
> 
> I believe, this is the wrong place for such kind of fix.
> pm_runtime_get_sync() has obviously a broken semantics with regards to
> your observation but no other driver does what you propose.
> I think the proper fix belong into PM subsystem, please take a look
> onto commit 15bcb91d7e60 "PM / Runtime: Implement autosuspend support".
> 
> > Signed-off-by: Dinghao Liu <dinghao.liu@....edu.cn>
> > ---
> >  drivers/char/hw_random/ks-sa-rng.c | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/drivers/char/hw_random/ks-sa-rng.c b/drivers/char/hw_random/ks-sa-rng.c
> > index e2330e757f1f..85c81da4a8af 100644
> > --- a/drivers/char/hw_random/ks-sa-rng.c
> > +++ b/drivers/char/hw_random/ks-sa-rng.c
> > @@ -244,6 +244,7 @@ static int ks_sa_rng_probe(struct platform_device *pdev)
> >  	ret = pm_runtime_get_sync(dev);
> >  	if (ret < 0) {
> >  		dev_err(dev, "Failed to enable SA power-domain\n");
> > +		pm_runtime_put_sync(dev);
> >  		pm_runtime_disable(dev);
> >  		return ret;
> >  	}
> -- 
> Alexander Sverdlin.
>