> -----Original Message----- > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > Sent: Thursday, March 19, 2015 4:49 PM > To: Butler, Siobhan A; Iremonger, Bernard > Cc: dev at dpdk.org > Subject: [PATCH v2 3/3] doc: nics guide > > Create nics guide by moving chapters about Intel and Mellanox NICs. > > Signed-off-by: Thomas Monjalon <thomas.monjalon at 6wind.com> > --- > MAINTAINERS | 15 +- > doc/guides/index.rst | 1 + > .../driver_vm_emul_dev.rst => nics/e1000em.rst} | 0 > doc/guides/{prog_guide => nics}/img/console.png | Bin > .../{prog_guide => nics}/img/fast_pkt_proc.png | Bin > .../{prog_guide => nics}/img/forward_stats.png | Bin > .../{prog_guide => nics}/img/host_vm_comms.png | Bin > .../img/host_vm_comms_qemu.png | Bin > .../{prog_guide => nics}/img/inter_vm_comms.png | Bin > .../{prog_guide => nics}/img/perf_benchmark.png | Bin > .../{prog_guide => nics}/img/single_port_nic.png | Bin > .../{prog_guide => nics}/img/vm_vm_comms.png | Bin > .../{prog_guide => nics}/img/vmxnet3_int.png | Bin > doc/guides/{prog_guide => nics}/img/vswitch_vm.png | Bin > doc/guides/{ => nics}/index.rst | 29 ++-- > .../intel_vf.rst} | 8 +- > doc/guides/nics/ixgbe.rst | 184 > +++++++++++++++++++++ > .../mlx4_poll_mode_drv.rst => nics/mlx4.rst} | 0 > .../pcap_ring.rst} | 0 > .../virtio.rst} | 4 +- > .../vmxnet3.rst} | 0 > doc/guides/prog_guide/index.rst | 17 -- > doc/guides/prog_guide/poll_mode_drv.rst | 152 ----------------- > 23 files changed, 217 insertions(+), 193 deletions(-) rename > doc/guides/{prog_guide/driver_vm_emul_dev.rst => nics/e1000em.rst} (100%) > rename > doc/guides/{prog_guide => nics}/img/console.png (100%) rename > doc/guides/{prog_guide => > nics}/img/fast_pkt_proc.png (100%) rename doc/guides/{prog_guide => > nics}/img/forward_stats.png > (100%) rename doc/guides/{prog_guide => nics}/img/host_vm_comms.png (100%) > rename > doc/guides/{prog_guide => nics}/img/host_vm_comms_qemu.png (100%) rename > doc/guides/{prog_guide => nics}/img/inter_vm_comms.png (100%) rename > doc/guides/{prog_guide > => nics}/img/perf_benchmark.png (100%) rename doc/guides/{prog_guide => > nics}/img/single_port_nic.png (100%) rename doc/guides/{prog_guide => > nics}/img/vm_vm_comms.png (100%) rename doc/guides/{prog_guide => > nics}/img/vmxnet3_int.png (100%) rename doc/guides/{prog_guide => > nics}/img/vswitch_vm.png > (100%) copy doc/guides/{ => nics}/index.rst (88%) rename > doc/guides/{prog_guide/i40e_ixgbe_igb_virt_func_drv.rst => nics/intel_vf.rst} > (99%) create mode > 100644 doc/guides/nics/ixgbe.rst rename > doc/guides/{prog_guide/mlx4_poll_mode_drv.rst => > nics/mlx4.rst} (100%) rename > doc/guides/{prog_guide/libpcap_ring_based_poll_mode_drv.rst => > nics/pcap_ring.rst} (100%) rename > doc/guides/{prog_guide/poll_mode_drv_emulated_virtio_nic.rst > => nics/virtio.rst} (99%) rename > doc/guides/{prog_guide/poll_mode_drv_paravirtual_vmxnets_nic.rst => > nics/vmxnet3.rst} (100%) > > diff --git a/MAINTAINERS b/MAINTAINERS > index 24f0a4c..0afb3f4 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -213,15 +213,20 @@ F: lib/librte_pmd_enic/ > > Intel e1000 > F: lib/librte_pmd_e1000/ > +F: doc/guides/nics/e1000em.rst > +F: doc/guides/nics/intel_vf.rst > > Intel ixgbe > M: Helin Zhang <helin.zhang at intel.com> > M: Konstantin Ananyev <konstantin.ananyev at intel.com> > F: lib/librte_pmd_ixgbe/ > +F: doc/guides/nics/ixgbe.rst > +F: doc/guides/nics/intel_vf.rst > > Intel i40e > M: Helin Zhang <helin.zhang at intel.com> > F: lib/librte_pmd_i40e/ > +F: doc/guides/nics/intel_vf.rst > > Intel fm10k > M: Jing Chen <jing.d.chen at intel.com> > @@ -230,12 +235,12 @@ F: lib/librte_pmd_fm10k/ Mellanox mlx4 > M: Adrien Mazarguil <adrien.mazarguil at 6wind.com> > F: lib/librte_pmd_mlx4/ > -F: doc/guides/prog_guide/mlx4_poll_mode_drv.rst > +F: doc/guides/nics/mlx4.rst > > RedHat virtio > M: Changchun Ouyang <changchun.ouyang at intel.com> > F: lib/librte_pmd_virtio/ > -F: doc/guides/prog_guide/poll_mode_drv_emulated_virtio_nic.rst > +F: doc/guides/nics/virtio.rst > F: lib/librte_vhost/ > F: doc/guides/prog_guide/vhost_lib.rst > F: examples/vhost/ > @@ -244,16 +249,16 @@ F: doc/guides/sample_app_ug/vhost.rst > VMware vmxnet3 > M: Yong Wang <yongwang at vmware.com> > F: lib/librte_pmd_vmxnet3/ > -F: doc/guides/prog_guide/poll_mode_drv_paravirtual_vmxnets_nic.rst > +F: doc/guides/nics/vmxnet3.rst > > PCAP PMD > F: lib/librte_pmd_pcap/ > -F: doc/guides/prog_guide/libpcap_ring_based_poll_mode_drv.rst > +F: doc/guides/nics/pcap_ring.rst > > Ring PMD > M: Bruce Richardson <bruce.richardson at intel.com> > F: lib/librte_pmd_ring/ > -F: doc/guides/prog_guide/ring_lib.rst > +F: doc/guides/nics/pcap_ring.rst > F: app/test/test_pmd_ring.c > > Null PMD > diff --git a/doc/guides/index.rst b/doc/guides/index.rst index > 53f1be1..44e8432 100644 > --- a/doc/guides/index.rst > +++ b/doc/guides/index.rst > @@ -41,6 +41,7 @@ Contents: > freebsd_gsg/index > xen/index > prog_guide/index > + nics/index > sample_app_ug/index > testpmd_app_ug/index > rel_notes/index > diff --git a/doc/guides/prog_guide/driver_vm_emul_dev.rst > b/doc/guides/nics/e1000em.rst similarity > index 100% rename from doc/guides/prog_guide/driver_vm_emul_dev.rst > rename to doc/guides/nics/e1000em.rst > diff --git a/doc/guides/prog_guide/img/console.png > b/doc/guides/nics/img/console.png similarity > index 100% rename from doc/guides/prog_guide/img/console.png > rename to doc/guides/nics/img/console.png diff --git > a/doc/guides/prog_guide/img/fast_pkt_proc.png > b/doc/guides/nics/img/fast_pkt_proc.png > similarity index 100% > rename from doc/guides/prog_guide/img/fast_pkt_proc.png > rename to doc/guides/nics/img/fast_pkt_proc.png > diff --git a/doc/guides/prog_guide/img/forward_stats.png > b/doc/guides/nics/img/forward_stats.png > similarity index 100% > rename from doc/guides/prog_guide/img/forward_stats.png > rename to doc/guides/nics/img/forward_stats.png > diff --git a/doc/guides/prog_guide/img/host_vm_comms.png > b/doc/guides/nics/img/host_vm_comms.png > similarity index 100% > rename from doc/guides/prog_guide/img/host_vm_comms.png > rename to doc/guides/nics/img/host_vm_comms.png > diff --git a/doc/guides/prog_guide/img/host_vm_comms_qemu.png > b/doc/guides/nics/img/host_vm_comms_qemu.png > similarity index 100% > rename from doc/guides/prog_guide/img/host_vm_comms_qemu.png > rename to doc/guides/nics/img/host_vm_comms_qemu.png > diff --git a/doc/guides/prog_guide/img/inter_vm_comms.png > b/doc/guides/nics/img/inter_vm_comms.png > similarity index 100% > rename from doc/guides/prog_guide/img/inter_vm_comms.png > rename to doc/guides/nics/img/inter_vm_comms.png > diff --git a/doc/guides/prog_guide/img/perf_benchmark.png > b/doc/guides/nics/img/perf_benchmark.png > similarity index 100% > rename from doc/guides/prog_guide/img/perf_benchmark.png > rename to doc/guides/nics/img/perf_benchmark.png > diff --git a/doc/guides/prog_guide/img/single_port_nic.png > b/doc/guides/nics/img/single_port_nic.png > similarity index 100% > rename from doc/guides/prog_guide/img/single_port_nic.png > rename to doc/guides/nics/img/single_port_nic.png > diff --git a/doc/guides/prog_guide/img/vm_vm_comms.png > b/doc/guides/nics/img/vm_vm_comms.png > similarity index 100% > rename from doc/guides/prog_guide/img/vm_vm_comms.png > rename to doc/guides/nics/img/vm_vm_comms.png > diff --git a/doc/guides/prog_guide/img/vmxnet3_int.png > b/doc/guides/nics/img/vmxnet3_int.png > similarity index 100% > rename from doc/guides/prog_guide/img/vmxnet3_int.png > rename to doc/guides/nics/img/vmxnet3_int.png > diff --git a/doc/guides/prog_guide/img/vswitch_vm.png > b/doc/guides/nics/img/vswitch_vm.png > similarity index 100% > rename from doc/guides/prog_guide/img/vswitch_vm.png > rename to doc/guides/nics/img/vswitch_vm.png > diff --git a/doc/guides/index.rst b/doc/guides/nics/index.rst similarity > index 88% copy from > doc/guides/index.rst copy to doc/guides/nics/index.rst index 53f1be1..6eb2e94 > 100644 > --- a/doc/guides/index.rst > +++ b/doc/guides/nics/index.rst > @@ -28,19 +28,22 @@ > (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE > OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. > > -DPDK documentation > -================== > +Network Interface Controller Drivers > +==================================== > > -Contents: > +|today| > + > + > +**Contents** > > .. toctree:: > - :maxdepth: 1 > - :titlesonly: > - > - linux_gsg/index > - freebsd_gsg/index > - xen/index > - prog_guide/index > - sample_app_ug/index > - testpmd_app_ug/index > - rel_notes/index > + :maxdepth: 3 > + :numbered: > + > + e1000em > + ixgbe > + intel_vf > + mlx4 > + virtio > + vmxnet3 > + pcap_ring > diff --git a/doc/guides/prog_guide/i40e_ixgbe_igb_virt_func_drv.rst > b/doc/guides/nics/intel_vf.rst > similarity index 99% rename from > doc/guides/prog_guide/i40e_ixgbe_igb_virt_func_drv.rst > rename to doc/guides/nics/intel_vf.rst > index 68d7107..f97b68c 100644 > --- a/doc/guides/prog_guide/i40e_ixgbe_igb_virt_func_drv.rst > +++ b/doc/guides/nics/intel_vf.rst > @@ -72,7 +72,7 @@ For more detail on SR-IOV, please refer to the following > documents: > > * `Scalable I/O Virtualized Servers > <http://www.intel.com/content/www/us/en/virtualization/server-virtualization/scalable-i-o- > virtualized-servers-paper.html>`_ > > -.. _pg_figure_10: > +.. _nic_figure_1: > > **Figure 10. Virtualization for a Single Port NIC in SR-IOV Mode** > > @@ -514,7 +514,7 @@ The setup procedure is as follows: > > * The Virtual Machine Monitor (see Figure 11) is equivalent to a Host > OS with KVM installed as > described in the instructions. > > -.. _pg_figure_11: > +.. _nic_figure_2: > > **Figure 11. Performance Benchmark Setup** > > @@ -538,7 +538,7 @@ DPI can be offloaded on the host fast path. > Figure 12 shows the scenario where some VMs directly communicate externally > via a VFs, while > others connect to a virtual switch and share the same uplink bandwidth. > > -.. _pg_figure_12: > +.. _nic_figure_3: > > **Figure 12. Fast Host-based Packet Processing** > > @@ -566,7 +566,7 @@ that is, the packet is forwarded to the correct PF pool. > The SR-IOV NIC switch forwards the packet to a specific VM according to the > MAC destination address > which belongs to the destination VF on the VM. > > -.. _pg_figure_13: > +.. _nic_figure_4: > > **Figure 13. Inter-VM Communication** > > diff --git a/doc/guides/nics/ixgbe.rst b/doc/guides/nics/ixgbe.rst new file > mode 100644 index > 0000000..d2c124e > --- /dev/null > +++ b/doc/guides/nics/ixgbe.rst > @@ -0,0 +1,184 @@ > +.. BSD LICENSE > + Copyright(c) 2010-2014 Intel Corporation. All rights reserved. > + All rights reserved. > + > + Redistribution and use in source and binary forms, with or without > + modification, are permitted provided that the following conditions > + are met: > + > + * Redistributions of source code must retain the above copyright > + notice, this list of conditions and the following disclaimer. > + * Redistributions in binary form must reproduce the above copyright > + notice, this list of conditions and the following disclaimer in > + the documentation and/or other materials provided with the > + distribution. > + * Neither the name of Intel Corporation nor the names of its > + contributors may be used to endorse or promote products derived > + from this software without specific prior written permission. > + > + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS > + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT > + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR > + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT > + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, > + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT > + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, > + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY > + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT > + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE > + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. > + > +IXGBE Driver > +============ > + > +Vector PMD for IXGBE > +-------------------- > + > +Vector PMD uses Intel? SIMD instructions to optimize packet I/O. > +It improves load/store bandwidth efficiency of L1 data cache by using a > wider SSE/AVX register 1 (1). > +The wider register gives space to hold multiple packet buffers so as to save > instruction number when > processing bulk of packets. > + > +There is no change to PMD API. The RX/TX handler are the only two entries > for vPMD packet I/O. > +They are transparently registered at runtime RX/TX execution if all > condition checks pass. > + > +1. To date, only an SSE version of IX GBE vPMD is available. > + To ensure that vPMD is in the binary code, ensure that the option > CONFIG_RTE_IXGBE_INC_VECTOR=y is in the configure file. > + > +Some constraints apply as pre-conditions for specific optimizations on bulk > packet transfers. > +The following sections explain RX and TX constraints in the vPMD. > + > +RX Constraints > +~~~~~~~~~~~~~~ > + > +Prerequisites and Pre-conditions > +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > + > +The following prerequisites apply: > + > +* To enable vPMD to work for RX, bulk allocation for Rx must be allowed. > + > +* The RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC=y configuration MACRO must be > set before > compiling the code. > + > +Ensure that the following pre-conditions are satisfied: > + > +* rxq->rx_free_thresh >= RTE_PMD_IXGBE_RX_MAX_BURST > + > +* rxq->rx_free_thresh < rxq->nb_rx_desc > + > +* (rxq->nb_rx_desc % rxq->rx_free_thresh) == 0 > + > +* rxq->nb_rx_desc < (IXGBE_MAX_RING_DESC - RTE_PMD_IXGBE_RX_MAX_BURST) > + > +These conditions are checked in the code. > + > +Scattered packets are not supported in this mode. > +If an incoming packet is greater than the maximum acceptable length of > +one "mbuf" data size (by default, the size is 2 KB), vPMD for RX would be > disabled. > + > +By default, IXGBE_MAX_RING_DESC is set to 4096 and > RTE_PMD_IXGBE_RX_MAX_BURST is set to > 32. > + > +Feature not Supported by RX Vector PMD > +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > + > +Some features are not supported when trying to increase the throughput in > vPMD. > +They are: > + > +* IEEE1588 > + > +* FDIR > + > +* Header split > + > +* RX checksum off load > + > +Other features are supported using optional MACRO configuration. They > include: > + > +* HW VLAN strip > + > +* HW extend dual VLAN > + > +* Enabled by RX_OLFLAGS (RTE_IXGBE_RX_OLFLAGS_DISABLE=n) > + > + > +To guarantee the constraint, configuration flags in dev_conf.rxmode will be > checked: > + > +* hw_vlan_strip > + > +* hw_vlan_extend > + > +* hw_ip_checksum > + > +* header_split > + > +* dev_conf > + > +fdir_conf->mode will also be checked. > + > +RX Burst Size > +^^^^^^^^^^^^^ > + > +As vPMD is focused on high throughput, it assumes that the RX burst size is > equal to or greater than > 32 per burst. > +It returns zero if using nb_pkt < 32 as the expected packet number in the > receive handler. > + > +TX Constraint > +~~~~~~~~~~~~~ > + > +Prerequisite > +^^^^^^^^^^^^ > + > +The only prerequisite is related to tx_rs_thresh. > +The tx_rs_thresh value must be greater than or equal to > +RTE_PMD_IXGBE_TX_MAX_BURST, but less or equal to > RTE_IXGBE_TX_MAX_FREE_BUF_SZ. > +Consequently, by default the tx_rs_thresh value is in the range 32 to 64. > + > +Feature not Supported by RX Vector PMD > +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > + > +TX vPMD only works when txq_flags is set to IXGBE_SIMPLE_FLAGS. > + > +This means that it does not support TX multi-segment, VLAN offload and TX > csum offload. > +The following MACROs are used for these three features: > + > +* ETH_TXQ_FLAGS_NOMULTSEGS > + > +* ETH_TXQ_FLAGS_NOVLANOFFL > + > +* ETH_TXQ_FLAGS_NOXSUMSCTP > + > +* ETH_TXQ_FLAGS_NOXSUMUDP > + > +* ETH_TXQ_FLAGS_NOXSUMTCP > + > + > +Sample Application Notes > +~~~~~~~~~~~~~~~~~~~~~~~~ > + > +testpmd > +^^^^^^^ > + > +By default, using CONFIG_RTE_IXGBE_RX_OLFLAGS_DISABLE=n: > + > +.. code-block:: console > + > + ./x86_64-native-linuxapp-gcc/app/testpmd -c 300 -n 4 -- -i > + --burst=32 --rxfreet=32 --mbcache=250 --txpt=32 --rxht=8 --rxwt=0 > + --txfreet=32 --txrst=32 --txqflags=0xf01 > + > +When CONFIG_RTE_IXGBE_RX_OLFLAGS_DISABLE=y, better performance can be > achieved: > + > +.. code-block:: console > + > + ./x86_64-native-linuxapp-gcc/app/testpmd -c 300 -n 4 -- -i > + --burst=32 --rxfreet=32 --mbcache=250 --txpt=32 --rxht=8 --rxwt=0 > + --txfreet=32 --txrst=32 --txqflags=0xf01 --disable-hw-vlan > + > +If scatter gather lists are not required, set > CONFIG_RTE_MBUF_SCATTER_GATHER=n for better > throughput. > + > +l3fwd > +^^^^^ > + > +When running l3fwd with vPMD, there is one thing to note. > +In the configuration, ensure that port_conf.rxmode.hw_ip_checksum=0. > +Otherwise, by default, RX vPMD is disabled. > + > +load_balancer > +^^^^^^^^^^^^^ > + > +As in the case of l3fwd, set configure port_conf.rxmode.hw_ip_checksum=0 to > enable vPMD. > +In addition, for improved performance, use -bsz "(32,32),(64,64),(32,32)" in > load_balancer to avoid > using the default burst size of 144. > diff --git a/doc/guides/prog_guide/mlx4_poll_mode_drv.rst > b/doc/guides/nics/mlx4.rst similarity > index 100% rename from doc/guides/prog_guide/mlx4_poll_mode_drv.rst > rename to doc/guides/nics/mlx4.rst > diff --git a/doc/guides/prog_guide/libpcap_ring_based_poll_mode_drv.rst > b/doc/guides/nics/pcap_ring.rst similarity index 100% rename from > doc/guides/prog_guide/libpcap_ring_based_poll_mode_drv.rst > rename to doc/guides/nics/pcap_ring.rst > diff --git a/doc/guides/prog_guide/poll_mode_drv_emulated_virtio_nic.rst > b/doc/guides/nics/virtio.rst similarity index 99% rename from > doc/guides/prog_guide/poll_mode_drv_emulated_virtio_nic.rst > rename to doc/guides/nics/virtio.rst > index 113a0cb..5e2a283 100644 > --- a/doc/guides/prog_guide/poll_mode_drv_emulated_virtio_nic.rst > +++ b/doc/guides/nics/virtio.rst > @@ -106,7 +106,7 @@ Virtio with kni vhost Back End > > This section demonstrates kni vhost back end example setup for Phy-VM > Communication. > > -.. _pg_figure_14: > +.. _nic_figure_5: > > **Figure 14. Host2VM Communication Example Using kni vhost Back End** > > @@ -189,7 +189,7 @@ Host2VM communication example Virtio with qemu virtio > Back End > -------------------------------- > > -.. _pg_figure_15: > +.. _nic_figure_6: > > **Figure 15. Host2VM Communication Example Using qemu vhost Back End** > > diff --git a/doc/guides/prog_guide/poll_mode_drv_paravirtual_vmxnets_nic.rst > b/doc/guides/nics/vmxnet3.rst similarity index 100% rename from > doc/guides/prog_guide/poll_mode_drv_paravirtual_vmxnets_nic.rst > rename to doc/guides/nics/vmxnet3.rst > diff --git a/doc/guides/prog_guide/index.rst > b/doc/guides/prog_guide/index.rst index > b263d31..648757e 100644 > --- a/doc/guides/prog_guide/index.rst > +++ b/doc/guides/prog_guide/index.rst > @@ -48,12 +48,7 @@ Programmer's Guide > mempool_lib > mbuf_lib > poll_mode_drv > - i40e_ixgbe_igb_virt_func_drv > - driver_vm_emul_dev > ivshmem_lib > - poll_mode_drv_emulated_virtio_nic > - poll_mode_drv_paravirtual_vmxnets_nic > - libpcap_ring_based_poll_mode_drv > link_bonding_poll_mode_drv_lib > mlx4_poll_mode_drv > timer_lib > @@ -104,18 +99,6 @@ Programmer's Guide > > :ref:`Figure 9. An mbuf with Three Segments <pg_figure_9>` > > -:ref:`Figure 10. Virtualization for a Single Port NIC in SR-IOV Mode > <pg_figure_10>` > - > -:ref:`Figure 11. Performance Benchmark Setup <pg_figure_11>` > - > -:ref:`Figure 12. Fast Host-based Packet Processing <pg_figure_12>` > - > -:ref:`Figure 13. Inter-VM Communication <pg_figure_13>` > - > -:ref:`Figure 14. Host2VM Communication Example Using kni vhost Back End > <pg_figure_14>` > - > -:ref:`Figure 15. Host2VM Communication Example Using qemu vhost Back End > <pg_figure_15>` > -
Hi Thomas, The references pg_figure_10 to 15 have been renamed to nic_figure_1 to 6. The above :ref: lines which have been deleted should be move to nics/index.rst and reworded. Regards, Bernard. > :ref:`Figure 16. Memory Sharing inthe Intel? DPDK Multi-process Sample > Application <pg_figure_16>` > > :ref:`Figure 17. Components of an Intel? DPDK KNI Application > <pg_figure_17>` diff --git > a/doc/guides/prog_guide/poll_mode_drv.rst > b/doc/guides/prog_guide/poll_mode_drv.rst > index e9e1132..b8b2d1a 100644 > --- a/doc/guides/prog_guide/poll_mode_drv.rst > +++ b/doc/guides/prog_guide/poll_mode_drv.rst > @@ -288,155 +288,3 @@ Ethernet Device API ~~~~~~~~~~~~~~~~~~~ > > The Ethernet device API exported by the Ethernet PMDs is described in the > *DPDK API Reference*. > - > -Vector PMD for IXGBE > --------------------- > - > -Vector PMD uses Intel? SIMD instructions to optimize packet I/O. > -It improves load/store bandwidth efficiency of L1 data cache by using a > wider SSE/AVX register 1 (1). > -The wider register gives space to hold multiple packet buffers so as to save > instruction number when > processing bulk of packets. > - > -There is no change to PMD API. The RX/TX handler are the only two entries > for vPMD packet I/O. > -They are transparently registered at runtime RX/TX execution if all > condition checks pass. > - > -1. To date, only an SSE version of IX GBE vPMD is available. > - To ensure that vPMD is in the binary code, ensure that the option > CONFIG_RTE_IXGBE_INC_VECTOR=y is in the configure file. > - > -Some constraints apply as pre-conditions for specific optimizations on bulk > packet transfers. > -The following sections explain RX and TX constraints in the vPMD. > - > -RX Constraints > -~~~~~~~~~~~~~~ > - > -Prerequisites and Pre-conditions > -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > - > -The following prerequisites apply: > - > -* To enable vPMD to work for RX, bulk allocation for Rx must be allowed. > - > -* The RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC=y configuration MACRO must be > set before > compiling the code. > - > -Ensure that the following pre-conditions are satisfied: > - > -* rxq->rx_free_thresh >= RTE_PMD_IXGBE_RX_MAX_BURST > - > -* rxq->rx_free_thresh < rxq->nb_rx_desc > - > -* (rxq->nb_rx_desc % rxq->rx_free_thresh) == 0 > - > -* rxq->nb_rx_desc < (IXGBE_MAX_RING_DESC - RTE_PMD_IXGBE_RX_MAX_BURST) > - > -These conditions are checked in the code. > - > -Scattered packets are not supported in this mode. > -If an incoming packet is greater than the maximum acceptable length of one > "mbuf" data size (by > default, the size is 2 KB), -vPMD for RX would be disabled. > - > -By default, IXGBE_MAX_RING_DESC is set to 4096 and > RTE_PMD_IXGBE_RX_MAX_BURST is set to > 32. > - > -Feature not Supported by RX Vector PMD > -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > - > -Some features are not supported when trying to increase the throughput in > vPMD. > -They are: > - > -* IEEE1588 > - > -* FDIR > - > -* Header split > - > -* RX checksum off load > - > -Other features are supported using optional MACRO configuration. They > include: > - > -* HW VLAN strip > - > -* HW extend dual VLAN > - > -* Enabled by RX_OLFLAGS (RTE_IXGBE_RX_OLFLAGS_DISABLE=n) > - > - > -To guarantee the constraint, configuration flags in dev_conf.rxmode will be > checked: > - > -* hw_vlan_strip > - > -* hw_vlan_extend > - > -* hw_ip_checksum > - > -* header_split > - > -* dev_conf > - > -fdir_conf->mode will also be checked. > - > -RX Burst Size > -^^^^^^^^^^^^^ > - > -As vPMD is focused on high throughput, it assumes that the RX burst size is > equal to or greater than > 32 per burst. > -It returns zero if using nb_pkt < 32 as the expected packet number in the > receive handler. > - > -TX Constraint > -~~~~~~~~~~~~~ > - > -Prerequisite > -^^^^^^^^^^^^ > - > -The only prerequisite is related to tx_rs_thresh. > -The tx_rs_thresh value must be greater than or equal to > RTE_PMD_IXGBE_TX_MAX_BURST, -but less > or equal to RTE_IXGBE_TX_MAX_FREE_BUF_SZ. > -Consequently, by default the tx_rs_thresh value is in the range 32 to 64. > - > -Feature not Supported by RX Vector PMD > -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > - > -TX vPMD only works when txq_flags is set to IXGBE_SIMPLE_FLAGS. > - > -This means that it does not support TX multi-segment, VLAN offload and TX > csum offload. > -The following MACROs are used for these three features: > - > -* ETH_TXQ_FLAGS_NOMULTSEGS > - > -* ETH_TXQ_FLAGS_NOVLANOFFL > - > -* ETH_TXQ_FLAGS_NOXSUMSCTP > - > -* ETH_TXQ_FLAGS_NOXSUMUDP > - > -* ETH_TXQ_FLAGS_NOXSUMTCP > - > - > -Sample Application Notes > -~~~~~~~~~~~~~~~~~~~~~~~~ > - > -testpmd > -^^^^^^^ > - > -By default, using CONFIG_RTE_IXGBE_RX_OLFLAGS_DISABLE=n: > - > -.. code-block:: console > - > - ./x86_64-native-linuxapp-gcc/app/testpmd -c 300 -n 4 -- -i --burst=32 > --rxfreet=32 --mbcache=250 - > -txpt=32 --rxht=8 --rxwt=0 --txfreet=32 --txrst=32 --txqflags=0xf01 > - > -When CONFIG_RTE_IXGBE_RX_OLFLAGS_DISABLE=y, better performance can be > achieved: > - > -.. code-block:: console > - > - ./x86_64-native-linuxapp-gcc/app/testpmd -c 300 -n 4 -- -i --burst=32 > --rxfreet=32 --mbcache=250 - > -txpt=32 --rxht=8 --rxwt=0 --txfreet=32 --txrst=32 --txqflags=0xf01 > --disable-hw-vlan > - > -If scatter gather lists are not required, set > CONFIG_RTE_MBUF_SCATTER_GATHER=n for better > throughput. > - > -l3fwd > -^^^^^ > - > -When running l3fwd with vPMD, there is one thing to note. > -In the configuration, ensure that port_conf.rxmode.hw_ip_checksum=0. > -Otherwise, by default, RX vPMD is disabled. > - > -load_balancer > -^^^^^^^^^^^^^ > - > -As in the case of l3fwd, set configure port_conf.rxmode.hw_ip_checksum=0 to > enable vPMD. > -In addition, for improved performance, use -bsz "(32,32),(64,64),(32,32)" in > load_balancer to avoid > using the default burst size of 144. > -- > 2.2.2
