Re: [PATCH net] net: usb: pegasus: validate USB endpoints

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Mon, Feb 23, 2026 at 09:39:52AM -0500, Alan Stern wrote:
> On Mon, Feb 23, 2026 at 01:58:48PM +0100, Greg Kroah-Hartman wrote:
> > The pegasus driver should validate that the device it is probing has the
> > proper number and types of USB endpoints it is expecting before it binds
> > to it.  If a malicious device were to not have the same urbs the driver
> > will crash later on when it blindly accesses these endpoints.
> > 
> > Cc: Petko Manolov <petkan@nucleusys.com>
> > Cc: stable <stable@kernel.org>
> > Assisted-by: gkh_clanker_2000
> > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > ---
> 
> This does much the same thing as
> 
> https://lore.kernel.org/linux-usb/20260222050633.410165-1-n7l8m4@u.northwestern.edu/T/#u
> 
> and that patch also removes some magic numbers.

Yes it does, that's a much nicer patch than mine.

> BTW, what is gkh_clanker_2000?

A hacked up system of tools/scripts I'm running here to find stuff like
"take this previously applied commit that fixed a problem, does the same
pattern need to be also done anywhere else in the tree"?  It finds a lot
of stuff and then I sift through it and see if anything is actually real
or not and if so, make up a patch for it.  It was my "merge window is
giving me a respite from reviewing patches" hobby project this past
week.

Now if I was really good, I could turn the output into a coccinelle
script, as this is just simple patterns.

Also it seems that we aren't running the coccinelle scripts anymore, as
many things it has found are already covered by that, I wonder why that
is...

thanks,

greg k-h