net/ionic: support vdev for embedded applications

Add support for running DPDK applications directly on AMD Pensando
embedded HW. The platform exposes the device BARs through UIO. The
UIO code in th

net/ionic: support vdev for embedded applications

Add support for running DPDK applications directly on AMD Pensando
embedded HW. The platform exposes the device BARs through UIO. The
UIO code in the common/ionic library walks the sysfs filesystem
to identify the relevant BARs and map them into process memory.

The SoCs are named 'Capri' and 'Elba'.

The vdev device interface code is located in ionic_dev_vdev.c.

Some datapath operations are #ifdef-ed out to save on resources when
running in embedded mode.

Some control path operations are skipped by the ionic_is_embedded()
helper function.

Before ringing the doorbell, use an ARM 'dsb st' barrier. The normal
barrier inside rte_write64() is insufficient on these devices due to
a chip errata.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Neel Patel <neel.patel@amd.com>
Signed-off-by: R Mohamed Shah <mohamedshah.r@amd.com>
Signed-off-by: Alfredo Cardigliano <cardigliano@ntop.org>
Acked-by: Ferruh Yigit <ferruh.yigit@amd.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>

show more ...

net/ionic: optimize device close

Use a single device reset command to speed up dev_close(). The LIF stop
and port reset commands are not needed.
This reduces the outage window when restarting the pr

net/ionic: optimize device close

Use a single device reset command to speed up dev_close(). The LIF stop
and port reset commands are not needed.
This reduces the outage window when restarting the process by about 2ms
plus another 1ms per queue.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

show more ...

net/ionic: fix device close

The close routine should release all resources, but not
call rte_eth_dev_destroy(). As written this code will call
rte_eth_dev_release_port() twice and segfault.

Instead

net/ionic: fix device close

The close routine should release all resources, but not
call rte_eth_dev_destroy(). As written this code will call
rte_eth_dev_release_port() twice and segfault.

Instead, move rte_eth_dev_destroy() to the remove routine.
eth_ionic_dev_uninit() will call close if necessary.

Fixes: 175e4e7ed760 ("net/ionic: complete release on close")
Cc: stable@dpdk.org

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

show more ...

net/ionic: add flexible firmware xstats

Assign 32 counters for flexible firmware events. These can be used as
per-port or per-queue counters in certain firmware configurations.
They are displayed as

net/ionic: add flexible firmware xstats

Assign 32 counters for flexible firmware events. These can be used as
per-port or per-queue counters in certain firmware configurations.
They are displayed as fw_flex_eventX in xstats.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Brad Larson <bradley.larson@amd.com>

show more ...

net/ionic: support 1G and 200G link speeds

The hardware supports these speeds,
so we should report them correctly.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Vamsi Krishna At

net/ionic: support 1G and 200G link speeds

The hardware supports these speeds,
so we should report them correctly.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Vamsi Krishna Atluri <vamsi.atluri@amd.com>

show more ...

net/ionic: fix RSS query

The routine that copies out the RSS config can't use memcpy() because
'reta_conf->reta' is an array of uint16_t while 'lif->rss_ind_tbl' is
an array of uint8_t. Instead, cop

net/ionic: fix RSS query

The routine that copies out the RSS config can't use memcpy() because
'reta_conf->reta' is an array of uint16_t while 'lif->rss_ind_tbl' is
an array of uint8_t. Instead, copy the values individually.

Fixes: 22e7171bc63b ("net/ionic: support RSS")
Cc: stable@dpdk.org

Signed-off-by: Akshay Dorwat <akshay.dorwat@amd.com>
Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

show more ...

net/ionic: add optimized non-scattered Rx/Tx

The code is very similar, but the simple case can skip a few branches
in the hot path. This improves PPS when 10KB mbufs are used.

S/G is enabled on the

net/ionic: add optimized non-scattered Rx/Tx

The code is very similar, but the simple case can skip a few branches
in the hot path. This improves PPS when 10KB mbufs are used.

S/G is enabled on the Rx side by offload DEV_RX_OFFLOAD_SCATTER.
S/G is enabled on the Tx side by offload DEV_TX_OFFLOAD_MULTI_SEGS.

S/G is automatically enabled on the Rx side if the provided mbufs are
too small to hold the maximum possible frame.

To enable S/G in testpmd, add these args:
--rx-offloads=0x2000 --tx-offloads=0x8000

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: R Mohamed Shah <mohamedshah.r@amd.com>

show more ...

net/ionic: allow to specify Tx free threshold

Some clients have opinions about how often to flush the
transmit ring.

The default value is the number of Tx descriptors minus the
default Tx burst siz

net/ionic: allow to specify Tx free threshold

Some clients have opinions about how often to flush the
transmit ring.

The default value is the number of Tx descriptors minus the
default Tx burst size.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

show more ...

net/ionic: add alignment and socket info in allocations

This will avoid memory access penalties on NUMA systems.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

net/ionic: use calloc for array allocations

Use rte_calloc() where appropriate.
This makes the code clearer.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

net/ionic: add Q-in-CMB option

When 'ionic_cmb' is set to '1', queue memory will be allocated from
the device's onboard memory (Controller Memory Buffer). In some
configurations, this will dramatica

net/ionic: add Q-in-CMB option

When 'ionic_cmb' is set to '1', queue memory will be allocated from
the device's onboard memory (Controller Memory Buffer). In some
configurations, this will dramatically reduce packet latency and
increase PPS.

Add the WC_ACTIVATE flag to the PCI driver flags.
Write combining must be enabled to achieve the maximum PPS.

When the queue is in the CMB, descriptors cannot be prefetched.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Neel Patel <neel.patel@amd.com>

show more ...

net/ionic: support Tx descriptor status

This may be useful for clients.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>

net/ionic: support Rx descriptor status

These may be useful for clients.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>

net/ionic: advertise supported packet types

This improves performance, since clients may be able to skip SW
packet classification.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

net/ionic: support mbuf fast free

Use a put() rather than a free() in the optimized case.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

net/ionic: move PCI-specific code to separate file

For future support of virtual devices, move the PCI code to its own
file. Create a new device interface, struct ionic_dev_intf, to plug
in to commo

net/ionic: move PCI-specific code to separate file

For future support of virtual devices, move the PCI code to its own
file. Create a new device interface, struct ionic_dev_intf, to plug
in to common code.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: Neel Patel <neel.patel@amd.com>
Signed-off-by: R Mohamed Shah <mohamedshah.r@amd.com>

show more ...

net/ionic: allocate interrupts only if required

There is no need to allocate the interrupt vector list if
datapath packet interrupts are not enabled.
This conserves resources.

Signed-off-by: Andrew

net/ionic: allocate interrupts only if required

There is no need to allocate the interrupt vector list if
datapath packet interrupts are not enabled.
This conserves resources.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

show more ...

net/ionic: update MTU calculations

Test min and max MTU against values read from firmware, for correctness.
Update the firmware field name, for clarity.
The device must be stopped before changing MT

net/ionic: update MTU calculations

Test min and max MTU against values read from firmware, for correctness.
Update the firmware field name, for clarity.
The device must be stopped before changing MTU, for correctness.
Store the calculated frame size in the queue, for performance.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>
Signed-off-by: R Mohamed Shah <mohamedshah.r@amd.com>

show more ...

net/ionic: update license terms to remove GPL

Remove GPL2 and leave only BSD-3-Clause. This is more in line with
the norms of the DPDK community.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

net/ionic: update documentation and copyrights

Pensando Systems has been acquired by AMD.
Update all copyright strings and email addresses.

Signed-off-by: Andrew Boyer <andrew.boyer@amd.com>

bus/pci: make driver-only headers private

The pci bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.

While at it, cleanup the code:
- fix indentatio

bus/pci: make driver-only headers private

The pci bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.

While at it, cleanup the code:
- fix indentation,
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions and macro manipulating the bus singleton object,
- remove inclusion of rte_bus.h and fix the code that relied on implicit
inclusion,

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>

show more ...

drivers: remove direct access to interrupt handle

Removing direct access to interrupt handle structure fields,
rather use respective get set APIs for the same.
Making changes to all the drivers acce

drivers: remove direct access to interrupt handle

Removing direct access to interrupt handle structure fields,
rather use respective get set APIs for the same.
Making changes to all the drivers access the interrupt handle fields.

Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Tested-by: Raslan Darawsheh <rasland@nvidia.com>

show more ...

ethdev: add namespace

Add 'RTE_ETH' namespace to all enums & macros in a backward compatible
way. The macros for backward compatibility can be removed in next LTS.
Also updated some struct names to

ethdev: add namespace

Add 'RTE_ETH' namespace to all enums & macros in a backward compatible
way. The macros for backward compatibility can be removed in next LTS.
Also updated some struct names to have 'rte_eth' prefix.

All internal components switched to using new names.

Syntax fixed on lines that this patch touches.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Tyler Retzlaff <roretzla@linux.microsoft.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Wisam Jaddo <wisamm@nvidia.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Chenbo Xia <chenbo.xia@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>

show more ...

ethdev: remove jumbo offload flag

Removing 'DEV_RX_OFFLOAD_JUMBO_FRAME' offload flag.

Instead of drivers announce this capability, application can deduct the
capability by checking reported 'dev_in

ethdev: remove jumbo offload flag

Removing 'DEV_RX_OFFLOAD_JUMBO_FRAME' offload flag.

Instead of drivers announce this capability, application can deduct the
capability by checking reported 'dev_info.max_mtu' or
'dev_info.max_rx_pktlen'.

And instead of application setting this flag explicitly to enable jumbo
frames, this can be deduced by driver by comparing requested 'mtu' to
'RTE_ETHER_MTU'.

Removing this additional configuration for simplification.

Suggested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Huisong Li <lihuisong@huawei.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

show more ...

ethdev: fix max Rx packet length

There is a confusion on setting max Rx packet length, this patch aims to
clarify it.

'rte_eth_dev_configure()' API accepts max Rx packet size via
'uint32_t max_rx_p

ethdev: fix max Rx packet length

There is a confusion on setting max Rx packet length, this patch aims to
clarify it.

'rte_eth_dev_configure()' API accepts max Rx packet size via
'uint32_t max_rx_pkt_len' field of the config struct 'struct
rte_eth_conf'.

Also 'rte_eth_dev_set_mtu()' API can be used to set the MTU, and result
stored into '(struct rte_eth_dev)->data->mtu'.

These two APIs are related but they work in a disconnected way, they
store the set values in different variables which makes hard to figure
out which one to use, also having two different method for a related
functionality is confusing for the users.

Other issues causing confusion is:
* maximum transmission unit (MTU) is payload of the Ethernet frame. And
'max_rx_pkt_len' is the size of the Ethernet frame. Difference is
Ethernet frame overhead, and this overhead may be different from
device to device based on what device supports, like VLAN and QinQ.
* 'max_rx_pkt_len' is only valid when application requested jumbo frame,
which adds additional confusion and some APIs and PMDs already
discards this documented behavior.
* For the jumbo frame enabled case, 'max_rx_pkt_len' is an mandatory
field, this adds configuration complexity for application.

As solution, both APIs gets MTU as parameter, and both saves the result
in same variable '(struct rte_eth_dev)->data->mtu'. For this
'max_rx_pkt_len' updated as 'mtu', and it is always valid independent
from jumbo frame.

For 'rte_eth_dev_configure()', 'dev->data->dev_conf.rxmode.mtu' is user
request and it should be used only within configure function and result
should be stored to '(struct rte_eth_dev)->data->mtu'. After that point
both application and PMD uses MTU from this variable.

When application doesn't provide an MTU during 'rte_eth_dev_configure()'
default 'RTE_ETHER_MTU' value is used.

Additional clarification done on scattered Rx configuration, in
relation to MTU and Rx buffer size.
MTU is used to configure the device for physical Rx/Tx size limitation,
Rx buffer is where to store Rx packets, many PMDs use mbuf data buffer
size as Rx buffer size.
PMDs compare MTU against Rx buffer size to decide enabling scattered Rx
or not. If scattered Rx is not supported by device, MTU bigger than Rx
buffer size should fail.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Huisong Li <lihuisong@huawei.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>

show more ...