From: Moti Haimovsky <motih@mellanox.com>
To: pascal.mazon@6wind.com
Cc: ferruh.yigit@intel.com, dev@dpdk.org,
Moti Haimovsky <motih@mellanox.com>
Subject: [dpdk-dev] [PATCH V5 2/2] net/tap: use new Rx offloads API
Date: Wed, 17 Jan 2018 16:04:34 +0200 [thread overview]
Message-ID: <1516197874-133169-2-git-send-email-motih@mellanox.com> (raw)
In-Reply-To: <1516197874-133169-1-git-send-email-motih@mellanox.com>
Ethdev Rx offloads API has changed since:
commit ce17eddefc20 ("ethdev: introduce Rx queue offloads API")
This commit adds support for the new Rx offloads API.
Signed-off-by: Moti Haimovsky <motih@mellanox.com>
---
V5:
* Fixed compilation errors caused by not using PRIx64 in log messages
when displaying uint64_t values.
* Modified patch name
V4:
Modifications according to inputs from Pascal Mazon
* Removed extra braces.
V3:
* Fixed coding style warnings
V2:
* Fixed coding style warnings
---
drivers/net/tap/rte_eth_tap.c | 67 +++++++++++++++++++++++++++++++++++++------
1 file changed, 59 insertions(+), 8 deletions(-)
diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c
index 7357dce..2eb8734 100644
--- a/drivers/net/tap/rte_eth_tap.c
+++ b/drivers/net/tap/rte_eth_tap.c
@@ -253,6 +253,43 @@ enum ioctl_mode {
}
}
+static uint64_t
+tap_rx_offload_get_port_capa(void)
+{
+ /*
+ * In order to support legacy apps,
+ * report capabilities also as port capabilities.
+ */
+ return DEV_RX_OFFLOAD_SCATTER |
+ DEV_RX_OFFLOAD_IPV4_CKSUM |
+ DEV_RX_OFFLOAD_UDP_CKSUM |
+ DEV_RX_OFFLOAD_TCP_CKSUM;
+}
+
+static uint64_t
+tap_rx_offload_get_queue_capa(void)
+{
+ return DEV_RX_OFFLOAD_SCATTER |
+ DEV_RX_OFFLOAD_IPV4_CKSUM |
+ DEV_RX_OFFLOAD_UDP_CKSUM |
+ DEV_RX_OFFLOAD_TCP_CKSUM;
+}
+
+static bool
+tap_rxq_are_offloads_valid(struct rte_eth_dev *dev, uint64_t offloads)
+{
+ uint64_t port_offloads = dev->data->dev_conf.rxmode.offloads;
+ uint64_t queue_supp_offloads = tap_rx_offload_get_queue_capa();
+ uint64_t port_supp_offloads = tap_rx_offload_get_port_capa();
+
+ if ((offloads & (queue_supp_offloads | port_supp_offloads)) !=
+ offloads)
+ return false;
+ if ((port_offloads ^ offloads) & port_supp_offloads)
+ return false;
+ return true;
+}
+
/* Callback to handle the rx burst of packets to the correct interface and
* file descriptor(s) in a multi-queue setup.
*/
@@ -277,8 +314,9 @@ enum ioctl_mode {
int len;
len = readv(rxq->fd, *rxq->iovecs,
- 1 + (rxq->rxmode->enable_scatter ?
- rxq->nb_rx_desc : 1));
+ 1 +
+ (rxq->rxmode->offloads & DEV_RX_OFFLOAD_SCATTER ?
+ rxq->nb_rx_desc : 1));
if (len < (int)sizeof(struct tun_pi))
break;
@@ -333,7 +371,7 @@ enum ioctl_mode {
seg->next = NULL;
mbuf->packet_type = rte_net_get_ptype(mbuf, NULL,
RTE_PTYPE_ALL_MASK);
- if (rxq->rxmode->hw_ip_checksum)
+ if (rxq->rxmode->offloads & DEV_RX_OFFLOAD_CHECKSUM)
tap_verify_csum(mbuf);
/* account for the receive frame */
@@ -694,12 +732,12 @@ enum ioctl_mode {
dev_info->min_rx_bufsize = 0;
dev_info->pci_dev = NULL;
dev_info->speed_capa = tap_dev_speed_capa();
- dev_info->rx_offload_capa = (DEV_RX_OFFLOAD_IPV4_CKSUM |
- DEV_RX_OFFLOAD_UDP_CKSUM |
- DEV_RX_OFFLOAD_TCP_CKSUM);
+ dev_info->rx_queue_offload_capa = tap_rx_offload_get_queue_capa();
+ dev_info->rx_offload_capa = tap_rx_offload_get_port_capa() |
+ dev_info->rx_queue_offload_capa;
dev_info->tx_queue_offload_capa = tap_tx_offload_get_queue_capa();
- dev_info->tx_offload_capa = dev_info->tx_queue_offload_capa |
- tap_tx_offload_get_port_capa();
+ dev_info->tx_offload_capa = tap_tx_offload_get_port_capa() |
+ dev_info->tx_queue_offload_capa;
}
static int
@@ -1015,6 +1053,19 @@ enum ioctl_mode {
return -1;
}
+ /* Verify application offloads are valid for our port and queue. */
+ if (!tap_rxq_are_offloads_valid(dev, rx_conf->offloads)) {
+ rte_errno = ENOTSUP;
+ RTE_LOG(ERR, PMD,
+ "%p: Rx queue offloads 0x%" PRIx64
+ " don't match port offloads 0x%" PRIx64
+ " or supported offloads 0x%" PRIx64 "\n",
+ (void *)dev, rx_conf->offloads,
+ dev->data->dev_conf.rxmode.offloads,
+ (tap_rx_offload_get_port_capa() |
+ tap_rx_offload_get_queue_capa()));
+ return -rte_errno;
+ }
rxq->mp = mp;
rxq->trigger_seen = 1; /* force initial burst */
rxq->in_port = dev->data->port_id;
--
1.8.3.1
next prev parent reply other threads:[~2018-01-17 14:04 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-04 19:18 [dpdk-dev] [PATCH V3 0/2] net/tap: convert to new ethdev " Moti Haimovsky
2018-01-04 19:18 ` [dpdk-dev] [PATCH V3 1/2] net/tap: convert to new Tx " Moti Haimovsky
2018-01-05 8:18 ` Pascal Mazon
2018-01-04 19:18 ` [dpdk-dev] [PATCH V3 2/2] net/tap: convert to new Rx " Moti Haimovsky
2018-01-05 8:26 ` Pascal Mazon
2018-01-10 16:20 ` [dpdk-dev] [PATCH V4 1/2] net/tap: convert to new Tx " Moti Haimovsky
2018-01-10 16:20 ` [dpdk-dev] [PATCH V4 2/2] net/tap: convert to new Rx " Moti Haimovsky
2018-01-10 16:42 ` Pascal Mazon
2018-01-17 14:04 ` [dpdk-dev] [PATCH V5 1/2] net/tap: use new Tx " Moti Haimovsky
2018-01-17 14:04 ` Moti Haimovsky [this message]
2018-03-02 21:44 ` [dpdk-dev] [PATCH V5 2/2] net/tap: use new Rx " Ferruh Yigit
2018-03-12 14:20 ` Shahaf Shuler
2018-03-12 16:59 ` Ferruh Yigit
2018-03-12 17:58 ` Shahaf Shuler
2018-03-12 19:05 ` Ferruh Yigit
2018-03-13 7:08 ` Shahaf Shuler
2018-03-13 11:56 ` Ferruh Yigit
2018-03-14 5:49 ` Shahaf Shuler
2018-03-14 22:40 ` Ferruh Yigit
2018-03-15 6:16 ` Shahaf Shuler
2018-03-15 14:34 ` Ferruh Yigit
2018-01-18 14:02 ` [dpdk-dev] [PATCH V5 1/2] net/tap: use new Tx " Pascal Mazon
2018-01-18 15:19 ` Ferruh Yigit
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1516197874-133169-2-git-send-email-motih@mellanox.com \
--to=motih@mellanox.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=pascal.mazon@6wind.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).