[PATCH 097/117] Staging: hv: storvsc: Add code to handle IDE devices using the storvsc driver

Sasha Levin levinsasha928 at gmail.com
Sat Jul 16 13:01:39 UTC 2011


On Sat, 2011-07-16 at 12:57 +0000, KY Srinivasan wrote:
> 
> > -----Original Message-----
> > From: Christoph Hellwig [mailto:hch at infradead.org]
> > Sent: Friday, July 15, 2011 10:05 PM
> > To: KY Srinivasan
> > Cc: gregkh at suse.de; linux-kernel at vger.kernel.org;
> > devel at linuxdriverproject.org; virtualization at lists.osdl.org; Haiyang Zhang
> > Subject: Re: [PATCH 097/117] Staging: hv: storvsc: Add code to handle IDE devices
> > using the storvsc driver
> > 
> > Thanks, this looks much cleaner than the initial variant.
> > 
> > > +	if (dev_is_ide) {
> > > +		storvsc_get_ide_info(device, &target, &path);
> > > +		host_dev->path = device_info.path_id;
> > > +		host_dev->target = device_info.target_id;
> > > +	} else {
> > > +		host_dev->path = device_info.path_id;
> > > +		host_dev->target = device_info.target_id;
> > > +	}
> > 
> > Is using the device_info values in both branches intentional?  If so
> > there's no need to have these assignments duplicated.
> 
> While we set the values in both the branches, the value set is different;
> The IDE side encodes the bits differently and is appropriately parsed in the
> function storvsc_get_ide_info().

Is think that what Christoph meant was simplifying it to:

	if (dev_is_ide)
		storvsc_get_ide_info(device, &target, &path);
	
	host_dev->path = device_info.path_id;
	host_dev->target = device_info.target_id;

-- 

Sasha.




More information about the devel mailing list