For i40evf, internal rx interrupt and adminq interrupt share the same
source, that cause a lot cpu cycles be wasted on interrupt handler
on rx path. This is complained by customers which require low latency
(when set I40E_ITR_INTERVAL to small value), but have to be sufferred by
tremendous interrupts handling that eat significant CPU resources.
The patch disable pci interrupt and remove the interrupt handler,
replace it with a low frequency (50ms) interrupt polling daemon
which is implemented by registering a alarm callback periodly, this
save CPU time significently: On a typical x86 server with 2.1GHz CPU,
with low latency configure (32us) we saw CPU usage from top commmand
reduced from 20% to 0% on management core in testpmd).
Also with the new method we can remove compile option: I40E_ITR_INTERVAL
which is used to balance between low latency and low CPU usage previously.
Now we don't need it since we can reach both at same time.
Suggested-by: Jingjing Wu <jingjing.wu@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
For all trace on the data-path move to dynamic logging.
Add new QAT_DP_LOG macro to separate from non-dp trace.
Use the overall RTE_LOG_DP_LEVEL to override and compile out
debug data-path trace.
Add "unlikely" in two appropriate error checks.
Signed-off-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com>
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
For all trace not on the data-path move to dynamic logging.
Signed-off-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com>
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Removed INIT_FUNC trace and other unused macros.
Signed-off-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com>
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Make the compiler switch name and document name consistent as ``ifc`` to
avoid confusion. Also rename the map file to standard name for meson
build in the process.
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
This patch removes current support of Memory Region (MR) in order to
accommodate the dynamic memory hotplug patch. This patch can be compiled
but traffic can't flow and HW will raise faults. Subsequent patches will
add new MR support.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
This patch removes current support of Memory Region (MR) in order to
accommodate the dynamic memory hotplug patch. This patch can be compiled
but traffic can't flow and HW will raise faults. Subsequent patches will
add new MR support.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
With this change, DPAA and DPAA2 configuration compiled applications
are interoperable without any impact on the functional behavior.
Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
When we check for libbsd we were just checking for the library itself.
However, on some distros, e.g. CentOS, libbsd will be found even though
it's not usable by DPDK as it's missing the string header. Therefore we
should check for both the library and the header file.
Fixes: 5364de644a ("eal: support strlcpy function")
Signed-off-by: Bruce Richarson <bruce.richardson@intel.com>
Reported-by: Vladimir Medvedkin <medvedkinv@gmail.com>
Acked-by: Vladimir Medvedkin <medvedkinv@gmail.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Introduce rte_bpf_elf_load() function to provide ability to
load eBPF program from ELF object file.
It also adds dependency on libelf.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
librte_bpf provides a framework to load and execute eBPF bytecode
inside user-space dpdk based applications.
It supports basic set of features from eBPF spec
(https://www.kernel.org/doc/Documentation/networking/filter.txt).
Not currently supported features:
- JIT
- cBPF
- tail-pointer call
- eBPF MAP
- skb
- function calls for 32-bit apps
- mbuf pointer as input parameter for 32-bit apps
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Defined FPGA-BUS for Acceleration Drivers of AFUs
1. FPGA PCI Scan (1st Scan) follows DPDK UIO/VFIO PCI Scan Process,
probe Intel FPGA Rawdev Driver, it will be covered in following patches.
2. AFU Scan(2nd Scan) bind DPDK driver to FPGA Partial-Bitstream.
This scan is trigged by hotplug of IFPGA Rawdev probe, in this scan
the AFUs will be created and their drivers are also probed.
This patch will introduce rte_afu_device which describe the AFU device
listed in the FPGA-BUS.
Signed-off-by: Rosen Xu <rosen.xu@intel.com>
Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>
CCP PMD supports authentication offload to either of CCP or CPU.
The earlier version of patch provides this option as compile time.
This patch changes this option from compile time to run time.
User can pass "ccp_auth_opt=1" as an additional arguments to vdev
parameter to enable authentication operations on CPU.
Signed-off-by: Ravi Kumar <ravi1.kumar@amd.com>
Adding basic skeleton of the ISA-L compression driver.
No compression functionality, but lays the foundation for
operations in the rest of the patchset.
The ISA-L compression driver utilizes Intel's ISA-L compression
library and compressdev API.
Signed-off-by: Lee Daly <lee.daly@intel.com>
Reviewed-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
New test created to measure offload cost.
Changes were introduced in API, turbo software driver
and test application
Signed-off-by: Kamil Chalupnik <kamilx.chalupnik@intel.com>
Acked-by: Amr Mokhtar <amr.mokhtar@intel.com>
This commit introduces the initial tests for compressdev,
performing basic compression and decompression operations
of sample test buffers, using the Zlib library in one direction
and compressdev in another direction, to make sure that
the library is compatible with Zlib.
Due to the use of Zlib API, the test is disabled by default,
to avoid adding a new dependency on DPDK.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Signed-off-by: Ashish Gupta <ashish.gupta@caviumnetworks.com>
Signed-off-by: Shally Verma <shally.verma@caviumnetworks.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>
Add basic functions to manage compress devices,
including driver and device allocation, and the basic
interface with compressdev PMDs.
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Signed-off-by: Shally Verma <shally.verma@caviumnetworks.com>
Signed-off-by: Ashish Gupta <ashish.gupta@caviumnetworks.com>
Picking a company stock ticker for a PMD name might not be a best approach
in a long run since name is too generic.
This patch addresses that and renames mrvl to mvsam.
Signed-off-by: Natalie Samsonov <nsamsono@marvell.com>
Signed-off-by: Tomasz Duszynski <tdu@semihalf.com>
This patch adds common code for the crypto adapter to support
SW and HW based transfer mechanisms. The adapter uses an EAL
service core function for SW based packet transfer and uses
the eventdev PMD functions to configure HW based packet
transfer between the crypto device and the event device.
This patch also adds adapter to the meson build system &
updates the necessary makefile & map file.
Signed-off-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Gage Eads <gage.eads@intel.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
DPAA2 QDMA driver uses MC DPDMAI object. This driver enables
the user (app) to perform data DMA without involving CPU in
the DMA process
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
IXGBE vector PMD was re-enabled for 32 bits, but only for gcc.
This commit enables it for icc too.
Fixes: e6672d2f0f ("net/ixgbe: enable ixgbe vector PMD for i686")
Cc: stable@dpdk.org
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
The IFCVF vDPA (vhost data path acceleration) driver provides support for
the Intel FPGA 100G VF (IFCVF). IFCVF's datapath is virtio ring compatible,
it works as a HW vhost backend which can send/receive packets to/from
virtio directly by DMA.
Different VF devices serve different virtio frontends which are in
different VMs, so each VF needs to have its own DMA address translation
service. During the driver probe a new container is created, with this
container vDPA driver can program DMA remapping table with the VM's memory
region information.
Key vDPA driver ops implemented:
- ifcvf_dev_config:
Enable VF data path with virtio information provided by vhost lib,
including IOMMU programming to enable VF DMA to VM's memory, VFIO
interrupt setup to route HW interrupt to virtio driver, create notify
relay thread to translate virtio driver's kick to a MMIO write onto HW,
HW queues configuration.
- ifcvf_dev_close:
Revoke all the setup in ifcvf_dev_config.
Live migration feature is supported by IFCVF and this driver enables
it. For the dirty page logging, VF helps to log for packet buffer write,
driver helps to make the used ring as dirty when device stops.
Because vDPA driver needs to set up MSI-X vector to interrupt the
guest, only vfio-pci is supported currently.
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Rosen Xu <rosen.xu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Currently eal vfio framework binds vfio group fd to the default
container fd during rte_vfio_setup_device, while in some cases,
e.g. vDPA (vhost data path acceleration), we want to put vfio group
to a separate container and program IOMMU via this container.
This patch extends the vfio_config structure to contain per-container
user_mem_maps and defines an array of vfio_config. The next patch will
base on this to add container API.
Signed-off-by: Junjie Chen <junjie.j.chen@intel.com>
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
The manager provides a way to allocate physically and virtually
contiguous set of objects.
Signed-off-by: Artem V. Andreev <artem.andreev@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
This patch adds the support for dynamic logging in dpaa_sec.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
The virtio crypto device is a virtual cryptography device
as well as a kind of virtual hardware accelerator for
virtual machines. The linux kernel virtio-crypto driver
has been merged, and this patch introduces virtio crypto
PMD to achieve better performance.
Signed-off-by: Jay Zhou <jianjay.zhou@huawei.com>
Reviewed-by: Fan Zhang <roy.fan.zhang@intel.com>
Auth operations can be performed on CPU without offloading
to CCP if CONFIG_RTE_LIBRTE_PMD_CCP_CPU_AUTH is enabled in
DPDK configuration. CCP PMD skip offloading auth operations
to hardware engines and perform them using openssl APIs.
Signed-off-by: Ravi Kumar <ravi1.kumar@amd.com>
The make based build system has crc+crypto instruction
support for the default arm64 build.
http://dpdk.org/browse/dpdk/tree/mk/machine/armv8a/rte.vars.mk#n31
This patch fixes the disparity with meson build flags for armv8.
As a bonus, This patch fixes the following errors with
ip_pipeline example application.
Assembler messages:
Error: selected processor does not support `crc32cx w3,w3,x0'
Fixes: c6e536e384 ("build: add more implementers IDs and PNs for ARM")
Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Add librte_rawdev to the meson build of DPDK.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Meson reports the toolchain using cc.get_id and we can set RTE_TOOLCHAIN,
RTE_TOOLCHAIN_X in dpdk_conf so that it can be used by both x86 and arm.
Suggested-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
This commit adds the logic that is shared by all event timer adapter
drivers; the common code handles instance allocation and some
initialization.
Signed-off-by: Erik Gabriel Carrillo <erik.g.carrillo@intel.com>
Acked-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
NFB cards employ multiple Ethernet ports.
Until now, Ethernet port-related operations were performed on all of them
(since the whole card was represented as a single port).
With new NFB-200G2QL card, this is no longer viable.
Since there is no fixed mapping between the queues and Ethernet ports,
and since a single card can be represented as two ports in DPDK,
there is no way of telling which (if any) physical ports should be
associated with individual ports in DPDK.
This is also described in documentation in more detail.
Signed-off-by: Matej Vido <vido@cesnet.cz>
Signed-off-by: Jan Remes <remes@netcope.com>
Before, we were aggregating multiple pages into one memseg, so the
number of memsegs was small. Now, each page gets its own memseg,
so the list of memsegs is huge. To accommodate the new memseg list
size and to keep the under-the-hood workings sane, the memseg list
is now not just a single list, but multiple lists. To be precise,
each hugepage size available on the system gets one or more memseg
lists, per socket.
In order to support dynamic memory allocation, we reserve all
memory in advance (unless we're in 32-bit legacy mode, in which
case we do not preallocate memory). As in, we do an anonymous
mmap() of the entire maximum size of memory per hugepage size, per
socket (which is limited to either RTE_MAX_MEMSEG_PER_TYPE pages or
RTE_MAX_MEM_MB_PER_TYPE megabytes worth of memory, whichever is the
smaller one), split over multiple lists (which are limited to
either RTE_MAX_MEMSEG_PER_LIST memsegs or RTE_MAX_MEM_MB_PER_LIST
megabytes per list, whichever is the smaller one). There is also
a global limit of CONFIG_RTE_MAX_MEM_MB megabytes, which is mainly
used for 32-bit targets to limit amounts of preallocated memory,
but can be used to place an upper limit on total amount of VA
memory that can be allocated by DPDK application.
So, for each hugepage size, we get (by default) up to 128G worth
of memory, per socket, split into chunks of up to 32G in size.
The address space is claimed at the start, in eal_common_memory.c.
The actual page allocation code is in eal_memalloc.c (Linux-only),
and largely consists of copied EAL memory init code.
Pages in the list are also indexed by address. That is, in order
to figure out where the page belongs, one can simply look at base
address for a memseg list. Similarly, figuring out IOVA address
of a memzone is a matter of finding the right memseg list, getting
offset and dividing by page size to get the appropriate memseg.
This commit also removes rte_eal_dump_physmem_layout() call,
according to deprecation notice [1], and removes that deprecation
notice as well.
On 32-bit targets due to limited VA space, DPDK will no longer
spread memory to different sockets like before. Instead, it will
(by default) allocate all of the memory on socket where master
lcore is. To override this behavior, --socket-mem must be used.
The rest of the changes are really ripple effects from the memseg
change - heap changes, compile fixes, and rewrites to support
fbarray-backed memseg lists. Due to earlier switch to _walk()
functions, most of the changes are simple fixes, however some
of the _walk() calls were switched to memseg list walk, where
it made sense to do so.
Additionally, we are also switching locks from flock() to fcntl().
Down the line, we will be introducing single-file segments option,
and we cannot use flock() locks to lock parts of the file. Therefore,
we will use fcntl() locks for legacy mem as well, in case someone is
unfortunate enough to accidentally start legacy mem primary process
alongside an already working non-legacy mem-based primary process.
[1] http://dpdk.org/dev/patchwork/patch/34002/
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Tested-by: Santosh Shukla <santosh.shukla@caviumnetworks.com>
Tested-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Tested-by: Gowrishankar Muthukrishnan <gowrishankar.m@linux.vnet.ibm.com>
Replace the BSD license header with the SPDX tag for files
with only an RehiveTech copyright on them.
Signed-off-by: Jan Viktorin <viktorin@rehivetech.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
The strncpy function is error prone for doing "safe" string copies, so
we generally try to use "snprintf" instead in the code. The function
"strlcpy" is a better alternative, since it better conveys the
intention of the programmer, and doesn't suffer from the non-null
terminating behaviour of it's n'ed brethern.
The downside of this function is that it is not available by default
on linux, though standard in the BSD's. It is available on most
distros by installing "libbsd" package.
This patch therefore provides the following in rte_string_fns.h to ensure
that strlcpy is available there:
* for BSD, include string.h as normal
* if RTE_USE_LIBBSD is set, include <bsd/string.h>
* if not set, fallback to snprintf for strlcpy
Using make build system, the RTE_USE_LIBBSD is a hard-coded value to "n",
but when using meson, it's automatically set based on what is available
on the platform.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
The old RTE_LOG_LEVEL has been replaced by RTE_DP_LOG_LEVEL and
is not used anywhere in the current source tree.
It is a merge leftover.
Fixes: 43e9f17ce7 ("log: remove log level config option")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Some comments are not relevant in a config which only overrides
the default config.
The option CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER is already disabled
by default so it can be removed from this file.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
The name "mrvl" for Marvell PMD driver for PPv2 Marvell PPv2
(Packet Processor v2) 1/10 Gbps adapter is too generic and causes
problem for adding new PMD drivers for other Marvell devices.
Changed to "mvpp2" for specific Marvell PPv2 PMD.
This patch doesn't introduce any change except renaming.
Signed-off-by: Natalie Samsonov <nsamsono@marvell.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
