[openib-general] Re: CM misuse of in_atomic/irqs_disabled

[Sean Hefty]

Hal Rosenstock wrote:
> Hi Sean,
> 
> Should the following in the CM be changed:
> 
> int ib_cm_establish(struct ib_cm_id *cm_id)
> {
> ...
>         work = kmalloc(sizeof *work, (in_atomic() || irqs_disabled()) ?
>                                       GFP_ATOMIC : GFP_KERNEL);
> to just
>         work = kmalloc(sizeof *work, GFP_ATOMIC);
> 
> similar to the other core changes for this issue ?

I think so.  This was structured similar to the MAD code.  I'll commit 
a patch to change this.

- Sean