[dpdk-dev,v2] ethdev: Prefetch driver variable structure
Commit Message
Adds ethdev driver prefetch of variable structure to CPU cache 0
while calling into tx or rx device driver operation.
RFC 2544 test of NIC task test measurement points show improvement
of lower latency and/or better packet throughput indicating clock
cycles saved.
Signed-off-by: Mike A. Polehn <mike.a.polehn@intel.com>
---
lib/librte_ether/rte_ethdev.h | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
Comments
2015-11-11 13:59, Polehn, Mike A:
> Adds ethdev driver prefetch of variable structure to CPU cache 0
> while calling into tx or rx device driver operation.
>
> RFC 2544 test of NIC task test measurement points show improvement
> of lower latency and/or better packet throughput indicating clock
> cycles saved.
>
> Signed-off-by: Mike A. Polehn <mike.a.polehn@intel.com>
There are still some spacing issues with this patch.
Bruce, what is your opinion about this prefetch?
On 11/11/2015 1:59 PM, mike.a.polehn at intel.com (Polehn, Mike A) wrote:
> Adds ethdev driver prefetch of variable structure to CPU cache 0
> while calling into tx or rx device driver operation.
>
> RFC 2544 test of NIC task test measurement points show improvement
> of lower latency and/or better packet throughput indicating clock
> cycles saved.
>
> Signed-off-by: Mike A. Polehn <mike.a.polehn at intel.com>
Hi Mike,
This patch is sitting on patchwork since 2015 without review, I am updating it
as rejected, if it is still relevant please let us know.
Sorry for any inconvenience caused.
@@ -2458,12 +2458,17 @@ rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
struct rte_mbuf **rx_pkts, const uint16_t nb_pkts)
{
struct rte_eth_dev *dev;
+ int16_t nb_rx;
dev = &rte_eth_devices[port_id];
- int16_t nb_rx = (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id],
- rx_pkts, nb_pkts);
+ { /* limit scope of rxq variable */
+ /* rxq is going to be immediately used, prefetch it */
+ void *rxq = dev->data->rx_queues[queue_id];
+ rte_prefetch0(rxq);
+ nb_rx = (*dev->rx_pkt_burst)(rxq, rx_pkts, nb_pkts);
+ }
#ifdef RTE_ETHDEV_RXTX_CALLBACKS
struct rte_eth_rxtx_callback *cb = dev->post_rx_burst_cbs[queue_id];
@@ -2600,6 +2605,7 @@ rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
{
struct rte_eth_dev *dev;
+ void *txq;
dev = &rte_eth_devices[port_id];
@@ -2615,7 +2621,11 @@ rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
}
#endif
- return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
+ /* txq is going to be immediately used, prefetch it */
+ txq = dev->data->tx_queues[queue_id];
+ rte_prefetch0(txq);
+
+ return (*dev->tx_pkt_burst)(txq, tx_pkts, nb_pkts);
}
#endif