* [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment
@ 2018-04-30 16:28 Didier Pallard
2018-04-30 16:28 ` [dpdk-stable] [18.02 2/2] net/vmxnet3: fix Rx offload information in multiseg packets Didier Pallard
2018-04-30 16:32 ` [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment Luca Boccassi
0 siblings, 2 replies; 3+ messages in thread
From: Didier Pallard @ 2018-04-30 16:28 UTC (permalink / raw)
To: stable
[ backported from upstream commit 5e5ac26f958f10ef3f1b443ed74c13898cc82e87 ]
Offloads are split between first and last segment of a packet.
Call a single vmxnet3_rx_offload function that will contain all
offload operations. This patch does not introduce any code modification.
Pass a vmxnet3_hw as parameter to the function, it is not presently
used in this patch, but will be later used for TSO offloads.
Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
Acked-by: Yong Wang <yongwang@vmware.com>
Conflicts:
drivers/net/vmxnet3/vmxnet3_rxtx.c : do not change packet type
behaviour on stable branch; continue to parse packet
to return precise RTE_PTYPE_L3_IPV4, rather than returning
RTE_PTYPE_L3_IPV4_UNKNOWN like in master branch.
---
drivers/net/vmxnet3/vmxnet3_rxtx.c | 66 +++++++++++++++++++++-----------------
1 file changed, 36 insertions(+), 30 deletions(-)
diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index 3a8c62fc16fc..23b69cce1029 100644
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -648,34 +648,46 @@ vmxnet3_post_rx_bufs(vmxnet3_rx_queue_t *rxq, uint8_t ring_id)
/* Receive side checksum and other offloads */
-static void
-vmxnet3_rx_offload(const Vmxnet3_RxCompDesc *rcd, struct rte_mbuf *rxm)
+static inline void
+vmxnet3_rx_offload(struct vmxnet3_hw *hw, const Vmxnet3_RxCompDesc *rcd,
+ struct rte_mbuf *rxm, const uint8_t sop)
{
- /* Check for RSS */
- if (rcd->rssType != VMXNET3_RCD_RSS_TYPE_NONE) {
- rxm->ol_flags |= PKT_RX_RSS_HASH;
- rxm->hash.rss = rcd->rssHash;
- }
+ (void)hw;
+
+ /* Offloads set in sop */
+ if (sop) {
+ /* Check for RSS */
+ if (rcd->rssType != VMXNET3_RCD_RSS_TYPE_NONE) {
+ rxm->ol_flags |= PKT_RX_RSS_HASH;
+ rxm->hash.rss = rcd->rssHash;
+ }
- /* Check packet type, checksum errors, etc. Only support IPv4 for now. */
- if (rcd->v4) {
- struct ether_hdr *eth = rte_pktmbuf_mtod(rxm, struct ether_hdr *);
- struct ipv4_hdr *ip = (struct ipv4_hdr *)(eth + 1);
+ /* Check packet type, checksum errors. Only IPv4 for now. */
+ if (rcd->v4) {
+ struct ether_hdr *eth = rte_pktmbuf_mtod(rxm, struct ether_hdr *);
+ struct ipv4_hdr *ip = (struct ipv4_hdr *)(eth + 1);
- if (((ip->version_ihl & 0xf) << 2) > (int)sizeof(struct ipv4_hdr))
- rxm->packet_type = RTE_PTYPE_L3_IPV4_EXT;
- else
- rxm->packet_type = RTE_PTYPE_L3_IPV4;
+ if (((ip->version_ihl & 0xf) << 2) > (int)sizeof(struct ipv4_hdr))
+ rxm->packet_type = RTE_PTYPE_L3_IPV4_EXT;
+ else
+ rxm->packet_type = RTE_PTYPE_L3_IPV4;
- if (!rcd->cnc) {
- if (!rcd->ipc)
- rxm->ol_flags |= PKT_RX_IP_CKSUM_BAD;
+ if (!rcd->cnc) {
+ if (!rcd->ipc)
+ rxm->ol_flags |= PKT_RX_IP_CKSUM_BAD;
- if ((rcd->tcp || rcd->udp) && !rcd->tuc)
- rxm->ol_flags |= PKT_RX_L4_CKSUM_BAD;
+ if ((rcd->tcp || rcd->udp) && !rcd->tuc)
+ rxm->ol_flags |= PKT_RX_L4_CKSUM_BAD;
+ }
+ } else {
+ rxm->packet_type = RTE_PTYPE_UNKNOWN;
+ }
+ } else { /* Offloads set in eop */
+ /* Check for hardware stripped VLAN tag */
+ if (rcd->ts) {
+ rxm->ol_flags |= (PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED);
+ rxm->vlan_tci = rte_le_to_cpu_16((uint16_t)rcd->tci);
}
- } else {
- rxm->packet_type = RTE_PTYPE_UNKNOWN;
}
}
@@ -807,7 +819,7 @@ vmxnet3_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
}
rxq->start_seg = rxm;
- vmxnet3_rx_offload(rcd, rxm);
+ vmxnet3_rx_offload(hw, rcd, rxm, 1);
} else {
struct rte_mbuf *start = rxq->start_seg;
@@ -823,13 +835,7 @@ vmxnet3_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
if (rcd->eop) {
struct rte_mbuf *start = rxq->start_seg;
- /* Check for hardware stripped VLAN tag */
- if (rcd->ts) {
- start->ol_flags |= (PKT_RX_VLAN |
- PKT_RX_VLAN_STRIPPED);
- start->vlan_tci = rte_le_to_cpu_16((uint16_t)rcd->tci);
- }
-
+ vmxnet3_rx_offload(hw, rcd, start, 0);
rx_pkts[nb_rx++] = start;
rxq->start_seg = NULL;
}
--
2.11.0
^ permalink raw reply [flat|nested] 3+ messages in thread
* [dpdk-stable] [18.02 2/2] net/vmxnet3: fix Rx offload information in multiseg packets
2018-04-30 16:28 [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment Didier Pallard
@ 2018-04-30 16:28 ` Didier Pallard
2018-04-30 16:32 ` [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment Luca Boccassi
1 sibling, 0 replies; 3+ messages in thread
From: Didier Pallard @ 2018-04-30 16:28 UTC (permalink / raw)
To: stable
[ backported from upstream commit d08e3c90d4e0a37618a5174c712dee1bf3cb73a3 ]
In case we are working on a multisegment buffer, most bit are set
in last segment of the buffer. Correctly look at those bits in eop part
of the rx_offload function.
Fixes: 2fdd835f992c ("vmxnet3: support jumbo frames")
Cc: stable@dpdk.org
Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
Acked-by: Yong Wang <yongwang@vmware.com>
---
drivers/net/vmxnet3/vmxnet3_rxtx.c | 13 +++++++------
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index 23b69cce1029..c08d957fb093 100644
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -656,12 +656,19 @@ vmxnet3_rx_offload(struct vmxnet3_hw *hw, const Vmxnet3_RxCompDesc *rcd,
/* Offloads set in sop */
if (sop) {
+ } else { /* Offloads set in eop */
/* Check for RSS */
if (rcd->rssType != VMXNET3_RCD_RSS_TYPE_NONE) {
rxm->ol_flags |= PKT_RX_RSS_HASH;
rxm->hash.rss = rcd->rssHash;
}
+ /* Check for hardware stripped VLAN tag */
+ if (rcd->ts) {
+ rxm->ol_flags |= (PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED);
+ rxm->vlan_tci = rte_le_to_cpu_16((uint16_t)rcd->tci);
+ }
+
/* Check packet type, checksum errors. Only IPv4 for now. */
if (rcd->v4) {
struct ether_hdr *eth = rte_pktmbuf_mtod(rxm, struct ether_hdr *);
@@ -682,12 +689,6 @@ vmxnet3_rx_offload(struct vmxnet3_hw *hw, const Vmxnet3_RxCompDesc *rcd,
} else {
rxm->packet_type = RTE_PTYPE_UNKNOWN;
}
- } else { /* Offloads set in eop */
- /* Check for hardware stripped VLAN tag */
- if (rcd->ts) {
- rxm->ol_flags |= (PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED);
- rxm->vlan_tci = rte_le_to_cpu_16((uint16_t)rcd->tci);
- }
}
}
--
2.11.0
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment
2018-04-30 16:28 [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment Didier Pallard
2018-04-30 16:28 ` [dpdk-stable] [18.02 2/2] net/vmxnet3: fix Rx offload information in multiseg packets Didier Pallard
@ 2018-04-30 16:32 ` Luca Boccassi
1 sibling, 0 replies; 3+ messages in thread
From: Luca Boccassi @ 2018-04-30 16:32 UTC (permalink / raw)
To: Didier Pallard, stable
On Mon, 2018-04-30 at 18:28 +0200, Didier Pallard wrote:
> [ backported from upstream commit
> 5e5ac26f958f10ef3f1b443ed74c13898cc82e87 ]
>
> Offloads are split between first and last segment of a packet.
> Call a single vmxnet3_rx_offload function that will contain all
> offload operations. This patch does not introduce any code
> modification.
>
> Pass a vmxnet3_hw as parameter to the function, it is not presently
> used in this patch, but will be later used for TSO offloads.
>
> Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
> Acked-by: Yong Wang <yongwang@vmware.com>
>
> Conflicts:
> drivers/net/vmxnet3/vmxnet3_rxtx.c : do not change packet type
> behaviour on stable branch; continue to parse packet
> to return precise RTE_PTYPE_L3_IPV4, rather than returning
> RTE_PTYPE_L3_IPV4_UNKNOWN like in master branch.
> ---
> drivers/net/vmxnet3/vmxnet3_rxtx.c | 66 +++++++++++++++++++++-------
> ----------
Thanks, series applied
--
Kind regards,
Luca Boccassi
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-04-30 16:32 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-04-30 16:28 [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment Didier Pallard
2018-04-30 16:28 ` [dpdk-stable] [18.02 2/2] net/vmxnet3: fix Rx offload information in multiseg packets Didier Pallard
2018-04-30 16:32 ` [dpdk-stable] [18.02 1/2] net/vmxnet3: gather offload data on first and last segment Luca Boccassi
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).