SSO GGRPs i.e. queue uses DRAM & SRAM buffers to hold in-flight
events. By default the buffers are assigned to the SSO GGRPs to
satisfy minimum HW requirements. SSO is free to assign the remaining
buffers to GGRPs based on a preconfigured threshold.
We can control the QoS of SSO GGRP by modifying the above mentioned
thresholds. GGRPs that have higher importance can be assigned higher
thresholds than the rest.
Example:
--dev "0002:0e:00.0,qos=[1-50-50-50]" // [Qx-XAQ-TAQ-IAQ]
Qx -> Event queue Aka SSO GGRP.
XAQ -> DRAM In-flights.
TAQ & IAQ -> SRAM In-flights.
The values need to be expressed in terms of percentages, 0 represents
default.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Octeontx2 SSO by default is set to use dual workslot mode.
Add devargs option to force legacy mode i.e. single workslot mode.
Example:
--dev "0002:0e:00.0,single_ws=1"
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
OcteonTx2 AP core SSO cache contains two entries each entry caches
state of an single GWS aka event port.
AP core requests events from SSO by using following sequence :
1. Write to SSOW_LF_GWS_OP_GET_WORK
2. Wait for SSO to complete scheduling by polling on SSOW_LF_GWS_TAG[63]
3. SSO notifies core by clearing SSOW_LF_GWS_TAG[63] and if work is
valid SSOW_LF_GWS_WQP is non-zero.
The above sequence uses only one in-core cache entry.
In dual workslot mode we try to use both the in-core cache entries by
triggering GET_WORK on a second workslot as soon as the above sequence
completes. This effectively hides the schedule latency of SSO if there
are enough events with unique flow_tags in-flight.
This mode reserves two SSO GWS lf's for each event port effectively
doubling single core performance.
Dual workslot mode is the default mode of operation in octeontx2.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Add support for retrieving statistics from SSO GWS and GGRP.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Register and implement SSO GWS and GGRP IRQ handlers for error
interrupts.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Links between queues and ports are controlled by setting/clearing GGRP
membership in SSOW_LF_GWS_GRPMSK_CHG.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
The number of events for a *open system* event device is specified
as -1 as per the eventdev specification.
Since, Octeontx2 SSO inflight events are only limited by DRAM size, the
xae_cnt devargs parameter is introduced to provide upper limit for
in-flight events.
Example:
--dev "0002:0e:00.0,xae_cnt=8192"
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add the device configure function that attaches the requested number of
SSO GWS(event ports) and GGRP(event queues) LF's to the PF.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Add the info_get function to return details on the queues, flow,
prioritization capabilities, etc. which this device has.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
SSO object needs to be initialized to communicate with the kernel AF
driver through mbox using the common API's.
Also, initialize the internal eventdev structure to defaults.
Attach NPA lf to the PF if needed.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add the make and meson based build infrastructure along with the
eventdev(SSO) device probe.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
For each driver where we optionally disable it, add in the reason why it's
being disabled, so the user knows how to fix it.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
When configuring with meson we print out a list of enabled components, but
it is also useful to list out the disabled components and the reasons why.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
When running self-tests, the driver needs to know the device on which to
run the tests, so we need to take the device ID as parameter. Only the
skeleton driver is providing this selftest capability right now, so we can
easily update it for this change.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Use a variable value rather than compile-time constant zero as the
device id for the skeleton rawdev tests. This ensures we can make the
tests work even if other rawdevs are present.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Using memory on the stack instead of on the heap by calling malloc
also avoid memory leak in case of test case failures
Fixes: 88d0e47880ec ("raw/skeleton: fix memory leak on test failure")
Cc: stable@dpdk.org
Signed-off-by: Thinh Tran <thinhtr@linux.vnet.ibm.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
The default policy for offload-specific fields is that
they are undefined unless the corresponding offloads are
requested in mbuf ol_flags. This is also the case for outer
L2 and L3 length fields which must not be assumed to contain
zeros for non-tunnel packets. The patch clarifies this behaviour
in the comments and also adds appropriate checks to the PMDs which
do not check any tunnel-related offloads before using the said fields.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Multi-process support for memif PMD.
Primary process handles connection establishment.
Secondary process queries for memory regions.
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
This patch enables need_wakeup flag for Tx and fill rings, when this
flag is set by the driver, it means that the userspace application has
to explicitly wake up the kernel Rx or kernel Tx processing by issuing
a syscall. Poll() can wake up both and sendto() or its alternatives
will wake up Tx processing only.
This feature is to provide efficient support for case that application
and driver executing on the same core.
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
It can be useful to use pcap files for some rudimental performance
testing. This patch enables this functionality in the pcap driver.
At a high level, this works by creating a ring of sufficient size to
store the packets in the pcap file passed to the application. When the
rx function for this mode is called, packets are dequeued from the ring
for use by the application and also enqueued back on to the ring to be
"received" again.
A tx_drop mode is also added since transmitting to a tx_pcap file isn't
desirable at a high traffic rate.
Jumbo frames are not supported in this mode. When filling the ring at rx
queue setup time, the presence of multi segment mbufs is checked for.
The PMD will exit on detection of these multi segment mbufs.
Signed-off-by: Cian Ferriter <cian.ferriter@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
The argument lists on some of the device creation functions are quite
large. Using a struct to hold the user options parsed in
'pmd_pcap_probe' will allow for cleaner function calls and definitions.
Adding user options will also be easier.
Signed-off-by: Cian Ferriter <cian.ferriter@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
The id values for VF stats were not being offset correctly.
And getting xstats for VF device only worked if VF device supported
it; it did not support the generic stats.
Fixes: dc7680e8597c ("net/netvsc: support integrated VF")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The xstats_get was not setting id correctly for each entry.
Fixes: 4e9c73e96e83 ("net/netvsc: add Hyper-V network device")
Cc: stable@dpdk.org
Signed-off-by: Mohsin Mazhar Shaikh <mohsinmazhar_shaikh@trendmicro.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Use the result from querying host to show the RSS info.
Fixes: 4e9c73e96e83 ("net/netvsc: add Hyper-V network device")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
This is to fix the error:
```
drivers/net/mlx5/mlx5_defs.h:14:26:
error: format '%lx' expects argument of type 'long unsigned int',
but argument 5 has type 'off_t {aka long long int}' [-Werror=format=]
drivers/net/mlx5/mlx5_txq.c:569:48: note: format string is defined here
DRV_LOG(DEBUG, "port %u: uar_mmap_offset 0x%lx"
~~^
%llx
```
Which reproduces with gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0.
Fixes: 6bf10ab69be0 ("net/mlx5: support 32-bit systems")
Cc: stable@dpdk.org
Signed-off-by: Ali Alnubani <alialnu@mellanox.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch add package sending and receiving function codes.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add build and doc files along with hinic_pmd_ethdev.c
which just includes PMD register and log initialization
for compilation.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add various headers that define mgmt commands, cmdq commands
and basic defines for use in the code.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Work queue is used for cmdq and Rx/Tx buff description.
NIC business needs to configure cmdq context and txq/rxq
context. This patch adds data structures and function codes
for work queue and context.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
The items of configurations and queries for NIC business include
MAC, VLAN, MTU, RSS and so on. These configurations and queries
are handled by mgmt module. This patch introduces related
data structures and function codes.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add code for hardware operation, including configuration,
query and so on.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Mgmt module is a kind of administration module for the chip.
It is responsible for handling administration command from host.
It uses api channel. This patch adds related data structures,
packaged interfaces and function codes.
Signed-off-by: Ziyang Xuan <xuanziyang2@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
