On 12/19/09 2:41 AM, "Simon Horman" <horms@verge.net.au> wrote:

quoted text
> On Fri, Dec 18, 2009 at 06:09:46PM -0800, Scott Feldman wrote:
>> for (i = 0; i < enic->rq_count; i++) {
>> -  err = vnic_rq_fill(&enic->rq[i], enic->rq_alloc_buf);
>> -  if (err) {
>> +  vnic_rq_fill(&enic->rq[i], enic->rq_alloc_buf);
>> +  /* Need at least one buffer on ring to get going */
>> +  if (vnic_rq_desc_used(&enic->rq[i]) == 0) {
>> printk(KERN_ERR PFX
>> "%s: Unable to alloc receive buffers.\n",
>> netdev->name);
>> +   err = -ENOMEM;
>> goto err_out_notify_unset;
>> }
>> }
> 
> Also, in the case where a failure occurs for i > 0,
> it it necessary to unwind the previous rq allocations?

Sorry Simon, I replied wrongly on this one.  The code only cares if _all_
allocations failed (i.e. The ring is empty), and takes the err path.  Since
nothing was allocated, the err path has nothing to cleanup.

If at lease one skb was allocated, then we don't go down the err path, even
if not all of the allocations succeeded.  Bottom line is we only need one
skb on the ring to get the ball rolling.  If we start out with a full
rings-worth of skbs (the expected case), then that's even better.

-scott

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html