tech-net archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: iwi(4) patch



In article <49B30FBB.2090009%free.fr@localhost>,
Jean-Yves Migeon  <jeanyves.migeon%free.fr@localhost> wrote:
>Hi,
>
>With a Xen debug kernel (PAE off), iwi(4) is not able to attach during 
>autoconf ("XENMEM_decrease_reservation failed!" issue) on my laptop. 
>This should not be fatal to the boot process, but in fact, it is; errors 
>are (from my PoV) incorrectly handled in if_iwi.c.
>
>I have local fixes to it [1]. However, as it is not a part of the kernel 
>I am familiar with, I would like more experienced people to review my patch:
>
>- some bus_dma fix, with checks against invalid dmap.
>
>- move the (RX and TX) rings' allocation code later in iwi_attach() to 
>avoid NULL pointer dereference if allocation fails (iwi_detach() 
>manipulates structure variables that are not set in case of a failed 
>allocation)
>
>- avoid double free in case of failure during attach. If an allocation 
>fails, do not free it directly, this is handled by iwi_detach() routine.
>
>- only set ring->count for RX/TX rings when ring malloc() is successful, 
>or else the for loop during detach will fail with a NULL dereference.
>
>Tested with current GENERIC, XEN3_DOM0 and XEN3PAE_DOM0. Did not notice 
>any regression so far.
>
>[1] http://www.netbsd.org/~jym/if_iwi.c.diff

Looks ok to me.

christos



Home | Main Index | Thread Index | Old Index