On 2/1/2015 9:27 AM, Or Gerlitz wrote:
On 1/30/2015 11:42 PM, Sagi Grimberg wrote:
Btw Moussa, what ever happened to the mlx4 driver change..?
Mossa's mlx4 patch wasn't the correct way to handle this BUG (mlx4
reserved almost all core EQs for mlx4_en and only 3 for RoCE). The
correct fix is heading upstream and hopefully be included in 3.20
Does the benefit brought by this patch comes into play only when either
Moussa's hack or the proposed 3.20 mlx4 patch are there, or this is
standing alone?
My recollection was that this patch on it's own gave a nice benefit
simply because the backend device might be utilizing some cores for
its own processing and iser can choose another core to poll the cq
at that time (luckily unbound workqueues are NUMA aware so they will
first try to pick in the same numa socket).
Nic, I forgot to give Moussa a proper credit in this patch. Can you
add his Signed-off-by please?
Thanks,
Sagi.
--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
