Re: [RFC PATCH 1/2] net: ethtool: Add capability to retrieve plug-in module EEPROM

[Ben Hutchings <bhutchings@solarflare.com>]

On Thu, 2012-04-12 at 10:18 +0100, Stuart Hodgson wrote:
> On 12/04/12 00:42, Ben Hutchings wrote:
[...]
> > Maybe we should start by refactoring ethtool_get_eeprom() so we can
> > reuse most of its code in ethtool_get_module_eeprom(), rather than
> > having to worry about what the maximum size of a module EEPROM might be
> > and whether we need a loop:
> >
> > Subject: ethtool: Split ethtool_get_eeprom() to allow for additional EEPROM accessors
[...]
> > @@ -771,7 +770,7 @@ static int ethtool_get_eeprom(struct net_device *dev, void __user *useraddr)
> >   		return -EINVAL;
> >
> >   	/* Check for exceeding total eeprom len */
> > -	if (eeprom.offset + eeprom.len>  ops->get_eeprom_len(dev))
> > +	if (eeprom.offset + eeprom.len>  total_len)
> >   		return -EINVAL;
> >
> >   	data = kmalloc(PAGE_SIZE, GFP_USER);
> 
> Should this not be eeprom.len?
[...]

No, because this function loops over PAGE_SIZE chunks.  That's
presumably necessary for large NIC EEPROMs, and if we reuse it we won't
have to wory about whether it's ever necessary for large module EEPROMs.

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.