Re: [PATCH] ipheth.c: Enable IP header alignment

["L. Alberto Giménez" <agimenez@sysvalve.es>]

On Mon, May 02, 2011 at 12:46:22PM -0700, David Miller wrote:
> 
> Why did this break things?

Hi, I don't know. As upstream is unresponsive and is applying patches to his
private repo without submitting them to the list (which I can understand), I
decided to submit the particular fix so mainline users can get tethering working
again.

I received a forwarded email with the patch (I think that's because I submitted
the driver to mainline) asking for the mainline driver status and if it was
being maintained.

> 
> I'm not applying a fix when nobody can explain the reason why:
> 
> 1) Things broke in the first place
> 2) Forcing reservation of 2 bytes fixes things

Honestly, I can't answer either of those ones. I just submitted a patch that
*seemed* to fix the problem (I don't own an iPhone device since long time ago),
after explictly requesting upstream to submit by himself, and getting a
negative.


> Where is the built in assumption about "2" and why does it exist?  Why
> can't we fix this code not to have such assumptions in the first
> place?

Ditto.

At this point, I think that David, Diego or Daniel should step in if they want
to keep on with this discussion. I won't have problems if you want to take this
off-list.

Best regards,

-- 
L. Alberto Giménez
JabberID agimenez@jabber.sysvalve.es
GnuPG key ID 0x3BAABDE1