linux-kernel - Re: [PATCH] net/sonic: Fix some resource leaks in error handling paths

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: <alpine.LNX.2.22.394.2005100922240.11@nippy.intranet>
Date:   Sun, 10 May 2020 09:45:04 +1000 (AEST)
From:   Finn Thain <fthain@...egraphics.com.au>
To:     Markus Elfring <Markus.Elfring@....de>
cc:     Christophe Jaillet <christophe.jaillet@...adoo.fr>,
        netdev@...r.kernel.org, kernel-janitors@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>
Subject: Re: [PATCH] net/sonic: Fix some resource leaks in error handling
 paths

On Sat, 9 May 2020, Markus Elfring wrote:

> > While at it, rename a label in order to be slightly more informative and
> > split some too long lines.
> 
> Would you like to add the tag 'Fixes' to the change description?
> 

Sorry but I don't follow your reasoning here. Are you saying that this 
needs to be pushed out to -stable branches? If so, stable-kernel-rules.rst 
would seem to disagree as the bug is theoretical and isn't bothering 
people.

Is there a way to add a Fixes tag that would not invoke the -stable 
process? And was that what you had in mind?

> 
> > +++ b/drivers/net/ethernet/natsemi/macsonic.c
> > @@ -506,10 +506,14 @@ static int mac_sonic_platform_probe(struct platform_device *pdev)
> >
> >  	err = register_netdev(dev);
> >  	if (err)
> > -		goto out;
> > +		goto undo_probe1;
> >
> >  	return 0;
> >
> > +undo_probe1:
> > +	dma_free_coherent(lp->device,
> > +			  SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
> > +			  lp->descriptors, lp->descriptors_laddr);
> >  out:
> How do you think about the possibility to use the label 'free_dma'?

I think 'undo_probe1' is both descriptive and consistent with commit 
10e3cc180e64 ("net/sonic: Fix a resource leak in an error handling path in 
'jazz_sonic_probe()'").

Your suggestion, 'free_dma' is also good. But coming up with good 
alternatives is easy. If every good alternative would be considered there 
would be no obvious way to get a patch merged.