Re: [PATCH] net: via-velocity.c fix sleep-with-spinlock bug during MTU change

Francois Romieu a écrit :

Séguier Régis [off-list ref] :
[...]
  

quoted

But if I generate a paquet with a size more than around 3825, the driver 
don't send it and no paquet is send after, work again after an interface 
down/up and until no paquet > 3825.
    

Does it qualify as a regression ?

  

I couldn't say yes now.

To be honest, I never test mtu > 1500 with this driver before.

This can be detected before with the difficulty of mtu change?

I'll test the driver before patch with a bigger MTU. (i'll try to do 
this next week)

-- 
Régis