From: Ye Xiaolong <xiaolong.ye@intel.com>
To: Xiao Zhang <xiao.zhang@intel.com>
Cc: dev@dpdk.org, qiming.yang@intel.com, qi.z.zhang@intel.com,
stable@dpdk.org
Subject: Re: [dpdk-dev] [v3] net/ice: enable multi-process support
Date: Thu, 25 Jul 2019 18:10:47 +0800 [thread overview]
Message-ID: <20190725101047.GC18332@intel.com> (raw)
In-Reply-To: <1564053515-38009-1-git-send-email-xiao.zhang@intel.com>
On 07/25, Xiao Zhang wrote:
>Add multiple processes support for ice, secondary processes will share
>memory and configuration with primary process, do not need further
>initialization for secondary processes.
>
>Cc: stable@dpdk.org
This is a new feature, it's no need to cc stable@dpdk.org.
Thanks,
Xiaolong
>
>Signed-off-by: Xiao Zhang <xiao.zhang@intel.com>
>---
>v3 Update the title and commit message.
>v2 Remove limitation for secondary processes control path configuration.
>---
> drivers/net/ice/ice_ethdev.c | 12 +++++++
> drivers/net/ice/ice_ethdev.h | 2 ++
> drivers/net/ice/ice_rxtx.c | 74 ++++++++++++++++++++++++++++----------------
> 3 files changed, 62 insertions(+), 26 deletions(-)
>
>diff --git a/drivers/net/ice/ice_ethdev.c b/drivers/net/ice/ice_ethdev.c
>index 9ce730c..532f4db 100644
>--- a/drivers/net/ice/ice_ethdev.c
>+++ b/drivers/net/ice/ice_ethdev.c
>@@ -1408,6 +1408,15 @@ ice_dev_init(struct rte_eth_dev *dev)
> dev->tx_pkt_burst = ice_xmit_pkts;
> dev->tx_pkt_prepare = ice_prep_pkts;
>
>+ /* for secondary processes, we don't initialise any further as primary
>+ * has already done this work.
>+ */
>+ if (rte_eal_process_type() != RTE_PROC_PRIMARY) {
>+ ice_set_rx_function(dev);
>+ ice_set_tx_function(dev);
>+ return 0;
>+ }
>+
> ice_set_default_ptype_table(dev);
> pci_dev = RTE_DEV_TO_PCI(dev->device);
> intr_handle = &pci_dev->intr_handle;
>@@ -1638,6 +1647,9 @@ ice_dev_uninit(struct rte_eth_dev *dev)
> struct ice_pf *pf = ICE_DEV_PRIVATE_TO_PF(dev->data->dev_private);
> struct rte_flow *p_flow;
>
>+ if (rte_eal_process_type() != RTE_PROC_PRIMARY)
>+ return 0;
>+
> ice_dev_close(dev);
>
> dev->dev_ops = NULL;
>diff --git a/drivers/net/ice/ice_ethdev.h b/drivers/net/ice/ice_ethdev.h
>index 8a52239..a083616 100644
>--- a/drivers/net/ice/ice_ethdev.h
>+++ b/drivers/net/ice/ice_ethdev.h
>@@ -282,6 +282,8 @@ struct ice_adapter {
> struct rte_eth_dev *eth_dev;
> struct ice_pf pf;
> bool rx_bulk_alloc_allowed;
>+ bool rx_vec_allowed;
>+ bool tx_vec_allowed;
> bool tx_simple_allowed;
> /* ptype mapping table */
> uint32_t ptype_tbl[ICE_MAX_PKT_TYPE] __rte_cache_min_aligned;
>diff --git a/drivers/net/ice/ice_rxtx.c b/drivers/net/ice/ice_rxtx.c
>index 035ed84..d67de8f 100644
>--- a/drivers/net/ice/ice_rxtx.c
>+++ b/drivers/net/ice/ice_rxtx.c
>@@ -2332,35 +2332,46 @@ ice_set_rx_function(struct rte_eth_dev *dev)
> int i;
> bool use_avx2 = false;
>
>- if (!ice_rx_vec_dev_check(dev)) {
>- for (i = 0; i < dev->data->nb_rx_queues; i++) {
>- rxq = dev->data->rx_queues[i];
>- (void)ice_rxq_vec_setup(rxq);
>- }
>+ if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
>+ if (!ice_rx_vec_dev_check(dev) && ad->rx_bulk_alloc_allowed) {
>+ ad->rx_vec_allowed = true;
>+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
>+ rxq = dev->data->rx_queues[i];
>+ if (rxq && ice_rxq_vec_setup(rxq)) {
>+ ad->rx_vec_allowed = false;
>+ break;
>+ }
>+ }
>+
>+ if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
>+ rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
>+ use_avx2 = true;
>
>- if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
>- rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
>- use_avx2 = true;
>+ } else {
>+ ad->rx_vec_allowed = false;
>+ }
>+ }
>
>+ if (ad->rx_vec_allowed) {
> if (dev->data->scattered_rx) {
> PMD_DRV_LOG(DEBUG,
>- "Using %sVector Scattered Rx (port %d).",
>- use_avx2 ? "avx2 " : "",
>- dev->data->port_id);
>+ "Using %sVector Scattered Rx (port %d).",
>+ use_avx2 ? "avx2 " : "",
>+ dev->data->port_id);
> dev->rx_pkt_burst = use_avx2 ?
>- ice_recv_scattered_pkts_vec_avx2 :
>- ice_recv_scattered_pkts_vec;
>+ ice_recv_scattered_pkts_vec_avx2 :
>+ ice_recv_scattered_pkts_vec;
> } else {
> PMD_DRV_LOG(DEBUG, "Using %sVector Rx (port %d).",
>- use_avx2 ? "avx2 " : "",
>- dev->data->port_id);
>+ use_avx2 ? "avx2 " : "",
>+ dev->data->port_id);
> dev->rx_pkt_burst = use_avx2 ?
>- ice_recv_pkts_vec_avx2 :
>- ice_recv_pkts_vec;
>+ ice_recv_pkts_vec_avx2 :
>+ ice_recv_pkts_vec;
> }
>-
> return;
> }
>+
> #endif
>
> if (dev->data->scattered_rx) {
>@@ -2464,16 +2475,27 @@ ice_set_tx_function(struct rte_eth_dev *dev)
> int i;
> bool use_avx2 = false;
>
>- if (!ice_tx_vec_dev_check(dev)) {
>- for (i = 0; i < dev->data->nb_tx_queues; i++) {
>- txq = dev->data->tx_queues[i];
>- (void)ice_txq_vec_setup(txq);
>- }
>+ if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
>+ if (!ice_tx_vec_dev_check(dev)) {
>+ ad->tx_vec_allowed = true;
>+ for (i = 0; i < dev->data->nb_tx_queues; i++) {
>+ txq = dev->data->tx_queues[i];
>+ if (txq && ice_txq_vec_setup(txq)) {
>+ ad->tx_vec_allowed = false;
>+ break;
>+ }
>+ }
>
>- if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
>- rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
>- use_avx2 = true;
>+ if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
>+ rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
>+ use_avx2 = true;
>+
>+ } else {
>+ ad->tx_vec_allowed = false;
>+ }
>+ }
>
>+ if (ad->tx_vec_allowed) {
> PMD_DRV_LOG(DEBUG, "Using %sVector Tx (port %d).",
> use_avx2 ? "avx2 " : "",
> dev->data->port_id);
>--
>2.7.4
>
prev parent reply other threads:[~2019-07-25 3:29 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-19 16:13 [dpdk-dev] [DPDK] net/ice: CVL " Xiao Zhang
2019-07-19 17:53 ` [dpdk-dev] " Xiao Zhang
2019-07-24 16:56 ` [dpdk-dev] [v2] " Xiao Zhang
2019-07-25 11:18 ` [dpdk-dev] [v3] net/ice: enable " Xiao Zhang
2019-07-25 4:56 ` Zhang, Qi Z
2019-07-25 10:10 ` Ye Xiaolong [this message]
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=20190725101047.GC18332@intel.com \
--to=xiaolong.ye@intel.com \
--cc=dev@dpdk.org \
--cc=qi.z.zhang@intel.com \
--cc=qiming.yang@intel.com \
--cc=stable@dpdk.org \
--cc=xiao.zhang@intel.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).