linux-kernel - Re: [PATCH 0/2] Style fixes: open code obfuscating macros

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: <20170111065036.GA4111@kroah.com>
Date:   Wed, 11 Jan 2017 07:50:36 +0100
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Logan Gunthorpe <logang@...tatee.com>
Cc:     Joe Perches <joe@...ches.com>, Jon Mason <jdmason@...zu.us>,
        Xiangliang Yu <Xiangliang.Yu@....com>,
        Dave Jiang <dave.jiang@...el.com>,
        Allen Hubbe <Allen.Hubbe@....com>, linux-ntb@...glegroups.com,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 0/2] Style fixes: open code obfuscating macros

On Tue, Jan 10, 2017 at 04:34:41PM -0700, Logan Gunthorpe wrote:
> 
> 
> On 10/01/17 04:06 PM, Joe Perches wrote:
> > It might also be better to add and use YA
> > logging macro like ndev_dbg
> > 
> > #define ndev_dvg(ndev, fmt, ...)	\
> > 	dev_dbg(&(ndev)->ntb.pdev->dev, fmt, ##__VA_ARGS__)
> > 
> > so that the new temporaries that are used just
> > for debugging are not required.
> > 
> 
> I feel like that's going in the exact opposite direction from what Greg
> was looking for... I think the point is to hide less of where the device
> is coming from, not to create a new and different macro that hides
> things in a different way.
> 
> Or, more likely, I _think_ Greg would suggest that this is showing that
> the driver model isn't really being used properly. Perhaps, we should be
> printing against the ntb struct device instead of the pci device so
> there is no dereferencing.

You should be printing against the ntb struct device, as that's the
"correct" structure here, and not the PCI device at all, as that's too
far up the device chain from what the driver is supposed to be doing.

thanks,

greg k-h