This patch fixes compiling problems on IBM Power architecture and turn
on the test-pmd compiling option in configuration file. Actually, this
is an big endian compiling fix.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Curent implementation in test_memzone.c has bugs in finding the
second smallest memory segment. It's the last smallest memory segment,
but it's not the second smallest memory segment. This bug may cause test
failure in some cases. This patch fixes this bug.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
IBM Power architecture has different huge page sizes (16MB, 16GB) than
x86.This patch defines RTE_PGSIZE_16M and RTE_PGSIZE_16G in the
rte_page_sizes enum variable and adds huge page size support of DPDK
for IBM Power architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
IBM Power architecture has different cache line size (128 bytes) than
x86 (64 bytes). This patch defines CACHE_LINE_SIZE to 128 bytes to
override the default value 64 bytes to support IBM Power Architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
IBM Power processor doesn't have CPU flag hardware registers. This patch
uses aux vector software register to get CPU flags and add CPU flag
checking support for IBM Power architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
If the user specifies 'set verbose 1' in testpmd command line,
the csum forward engine will dump some informations about received
and transmitted packets, especially which flags are set and what
values are assigned to l2_len, l3_len, l4_len and tso_segsz.
This can help someone implementing TSO or hardware checksum offload to
understand how to configure the mbufs.
Example of output for one packet:
--------------
rx: l2_len=14 ethertype=800 l3_len=20 l4_proto=6 l4_len=20
tx: m->l2_len=14 m->l3_len=20 m->l4_len=20
tx: m->tso_segsz=800
tx: flags=PKT_TX_IP_CKSUM PKT_TX_TCP_SEG
--------------
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Add two new commands in testpmd:
- tso set <segsize> <portid>
- tso show <portid>
These commands can be used enable TSO when transmitting TCP packets in
the csum forward engine. Ex:
set fwd csum
tx_checksum set ip hw 0
tso set 800 0
start
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Some of the NICs supported by DPDK have a possibility to accelerate TCP
traffic by using segmentation offload. The application prepares a packet
with valid TCP header with size up to 64K and deleguates the
segmentation to the NIC.
Implement the generic part of TCP segmentation offload in rte_mbuf. It
introduces 2 new fields in rte_mbuf: l4_len (length of L4 header in bytes)
and tso_segsz (MSS of packets).
To delegate the TCP segmentation to the hardware, the user has to:
- set the PKT_TX_TCP_SEG flag in mbuf->ol_flags (this flag implies
PKT_TX_TCP_CKSUM)
- set the flag PKT_TX_IPV4 or PKT_TX_IPV6
- set PKT_TX_IP_CKSUM if it's IPv4, and set the IP checksum to 0 in
the packet
- fill the mbuf offload information: l2_len, l3_len, l4_len, tso_segsz
- calculate the pseudo header checksum without taking ip_len in account,
and set it in the TCP header, for instance by using
rte_ipv4_phdr_cksum(ip_hdr, ol_flags)
The API is inspired from ixgbe hardware (the next commit adds the
support for ixgbe), but it seems generic enough to be used for other
hw/drivers in the future.
This commit also reworks the way l2_len and l3_len are used in igb
and ixgbe drivers as the l2_l3_len is not available anymore in mbuf.
Signed-off-by: Mirek Walukiewicz <miroslaw.walukiewicz@intel.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Introduce new functions to calculate checksums. These new functions
are derivated from the ones provided csumonly.c but slightly reworked.
There is still some room for future optimization of these functions
(maybe SSE/AVX, ...).
This API will be modified in tbe next commits by the introduction of
TSO that requires a different pseudo header checksum to be set in the
packet.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The csum forward engine was becoming too complex to be used and
extended (the next commits want to add the support of TSO):
- no explaination about what the code does
- code is not factorized, lots of code duplicated, especially between
ipv4/ipv6
- user command line api: use of bitmasks that need to be calculated by
the user
- the user flags don't have the same semantic:
- for legacy IP/UDP/TCP/SCTP, it selects software or hardware checksum
- for other (vxlan), it selects between hardware checksum or no
checksum
- the code relies too much on flags set by the driver without software
alternative (ex: PKT_RX_TUNNEL_IPV4_HDR). It is nice to be able to
compare a software implementation with the hardware offload.
This commit tries to fix these issues, and provide a simple definition
of what is done by the forward engine:
* Receive a burst of packets, and for supported packet types:
* - modify the IPs
* - reprocess the checksum in SW or HW, depending on testpmd command line
* configuration
* Then packets are transmitted on the output port.
*
* Supported packets are:
* Ether / (vlan) / IP|IP6 / UDP|TCP|SCTP .
* Ether / (vlan) / IP|IP6 / UDP / VxLAN / Ether / IP|IP6 / UDP|TCP|SCTP
*
* The network parser supposes that the packet is contiguous, which may
* not be the case in real life.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
In testpmd the rte_port->tx_ol_flags flag was used in 2 incompatible
manners:
- sometimes used with testpmd specific flags (0xff for checksums, and
bit 11 for vlan)
- sometimes assigned to m->ol_flags directly, which is wrong in case
of checksum flags
This commit replaces the hardcoded values by named definitions, which
are not compatible with mbuf flags. The testpmd forward engines are
fixed to use the flags properly.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
In test-pmd (rxonly.c), the code is able to dump the list of ol_flags.
The issue is that the list of flags in the application has to be
synchronized with the flags defined in rte_mbuf.h.
This patch introduces 2 new functions rte_get_rx_ol_flag_name()
and rte_get_tx_ol_flag_name() that returns the name of a flag from
its mask. It also fixes rxonly.c to use this new functions and to
display the proper flags.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Updated the unit tests to cover both librte_power implementations as well as
the external API.
Signed-off-by: Alan Carew <alan.carew@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Enable users to specify the lcore id that is used as master lcore.
Signed-off-by: Simon Kuenzer <simon.kuenzer@neclab.eu>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
In current version, used cores can only be specified using a bitmask.
It will now be possible to specify cores in 2 different ways:
- Using a bitmask (-c [0x]nnn): bitmask must be in hex format
- Using a list in following format: -l <c1>[-c2][,c3[-c4],...]
The letter -l can stand for lcore or list.
-l 0-7,16-23,31 being equivalent to -c 0x80FF00FF
Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
No need to have different headers for Linux and BSD.
These files are identicals with exception of internal config which has
uio and vfio fields only useful for Linux.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Test command is added to configure flexible payload
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
test command added to configure flexible mask
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Test command is added to flush flow director table
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Extended fdir info is printed in rxonly fwd engine when fdir match.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Commands are added to test adding or deleting flow director filters.
10 flow types in flow director are supported: ipv4, ipv4-frag, tcpv4, udpv4, sctpv4, ipv6, ipv6-frag, tcpv6, udpv6, sctpv6
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
As 40G NIC supports different sizes (128/512/64 entries) of
redirection table from that (128 entries) of 1G and 10G NICs,
support of multiple sizes of redirection table is needed.
It includes,
* Redefine 'struct rte_eth_rss_reta' in ethdev.
- To 'struct rte_eth_rss_reta_entry64' which contains 64
entries and 64 bits mask.
- Array of above new structure can be used for any number of
redirection table entries, as long as the number is multiple
of 64. This is quite flexible for the future expanding of
redirection table.
* Redefinition of relevant interfaces in ethdev.
- Interface of reta update has been redefined with new parameters.
- Interface of reta query has been redefined with new parameters.
* Rework of 1G PMD in igb.
- reta update has been reworked.
- reta query has been reworked.
* Rework of 10G PMD in ixgbe.
- reta update has been reworked.
- reta query has been reworked.
* Rework of 40G PMD (PF only) in i40e.
- reta update has been reworked.
- reta query has been reworked.
* Implement relevant commands in testpmd.
Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Tested-by: Erlu Chen <erlu.chen@intel.com>
Adding support for lsc interrupt from bonded device to link
bonding library with supporting unit tests in the test application.
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: SunX Jiajia <sunx.jiajia@intel.com>
set link-up and set link-down were not included
in the help command.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
This patch fixes two occurrences where a call to strncmp had the closing
brace in the wrong place. Changing this form:
if (strncmp(X,Y,sizeof(X) != 0))
which does a comparison of length 1, to
if (strncmp(X,Y,sizeof(X)) != 0)
which does the correct length comparison and then compares the result to
zero in the "if" part, as the author presumably originally intended.
Reported-by: Larry Wang <liang-min.wang@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
eal_flags and multiprocess unit tests use --file-prefix option
which is not supported in FreeBSD, so it has been removed
if compiled for this OS.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This field is added for librte_distributor. User of librte_distributor
is advocated to set value of mbuf->hash.usr before calling
rte_distributor_process. The value of usr is the tag which stands as
identifier of flow.
Signed-off-by: Qinglai Xiao <jigsaw@gmail.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The unit test can be used to measure cycles per packet in different rx/tx routines.
The NIC works in loopback mode. So it doesn't require test equipment to measure throughput.
As result, the unit test shows the average cycles per packet consuming.
When doing the test, make sure the link is UP.
Usage Example:
1. Run unit test app in interactive mode
app/test -c f -n 4 -- -i
2. Run and wait for the result
pmd_perf_autotest
There's option to choose rx/tx pair, default is vector.
set_rxtx_mode [vector|scalar|full|hybrid]
Note: To get acurate scalar fast, please choose 'vector' or 'hybrid' without INC_VEC=y in config
It supports to measure standalone rx or tx.
Usage Example:
Choose rx or tx standalone, default is both
set_rxtx_anchor [rxtx|rxonly|txonly]
It also supports to measure standalone RX burst cycles.
In this way, it won't repeat re-send received packets.
Now it measures two situations, poll before/after xmit(w or w/o desc. cache conflict)
Usage Example:
Set stream control mode, by default is continuous
set_rxtx_sc [continuous|poll_before_xmit|poll_after_xmit]
Test report: http://dpdk.org/ml/archives/dev/2014-October/007145.html
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Tested-by: Yong Liu <yong.liu@intel.com>
Add support to allow packet burst generator to create packets
in different sizes.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
These references to drivers break the layering isolation between
application and drivers.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Since commit a155d43011 ("support link bonding device initialization"),
rte_eal_pci_probe() is called in rte_eal_init().
pci_autotest called it to bind devices to the test_driver and test_driver2.
Therefore, the function is called twice and devices already allocated
will cause the test fail.
This patch solves that issue, unregistering all previous drivers before
calling rte_eal_pci_probe() for the first time, so DPDK does not try
to allocate data for the devices, binding them to their previous
drivers again.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
As a result of moving tailq's into local memory, some tailq data
is now reserved in rte_malloc heaps (because it needs to be
shared across DPDK processes). The first thing DPDK initializes
is a log mempool, and since it creates a tailq, it reserves
space in rte_malloc heap before allocating the mempool itself.
By default, rte_malloc allocates way more space than is necessary,
so under some conditions (namely, overall memory available is low)
this results in malloc heap eating up so much memory that log
mempool is not able to allocate its memzone.
This patch fixes the unit tests to account for that change.
Reported-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Add a test command in testpmd to test VF MAC filter feature.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
Add test cases in testpmd to test VxLAN Tx checksum offload, which includes
- IPv4 and IPv6 packet
- outer L3, inner L3 and L4 checksum offload for Tx side.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
Add the "tunnel_filter" command in testpmd to test the API of VxLAN
packet filter.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
Add two commands to test VXLAN packet identification.
The test steps are as follows:
1> use commands to add/delete VxLAN UDP port.
2> use rxonly mode to receive VxLAN packet.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
Link bonding broadcast mode requires refcnt parameter in the mbuf struct to
allow efficient transmission of duplicated mbufs on slave ports.
This patch disables broadcast mode when the complication option RTE_MBUF_REFCNT
is disabled to allow clean building of the bonding library.
A warning message notify user of disabling of broadcast mode.
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The previous implementation of rte_kni_alloc() was allocating memzones with a
name composed of a fixed string and the interface name. When an application was
allocating and deallocating multiple interfaces with different names, memzones
were quickly exhausted, even though memzones from deallocated interfaces were
never used anymore (unless an interface with the same name was re-allocated).
As a result, the application was unable to allocate more KNI interfaces with
different names.
This patch implements the KNI memzone pool in order to prevent memzone
exhaustion when allocating/deallocating KNI interfaces. It adds a new API call,
rte_kni_init(max_kni_ifaces) that shall be called before any call to
rte_kni_alloc() if KNI is used. The memzones are pre-allocated with interface-
independent names so that they can be reused.
Signed-off-by: Marc Sune <marc.sune@bisdn.de>
Acked-by: Helin Zhang <helin.zhang@intel.com>
The four-byte-key (fbk) autotest was allocating the keys to be used for
the test on the stack. When the number of entries in the table was
increased significantly, for example, to test larger hashes by increase the
value of ENTRIES, this array of keys was greater than that
allowed on the stack, and so caused problems, i.e. crashes and core dumps.
The solution is to have the keys dynamically allocated on the heap using
malloc. Now if ENTRIES is increased and we run out of memory we get an
error message instead of a crash.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Print an error message to the user when trying to start/stop a rx/tx queue and
this function is not supported by the PMD driver. The patch does not check if
the return value is -EINVAL because testpmd is already validating the port and
queue id.
Signed-off-by: Nicolás Pernas Maradei <nico@emutex.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
To improve performance by using bulk alloc or vectored RX routines, we
need to set rx free threshold (rxfreet) value to 32, so make this the
testpmd default.
Thirty-two is the minimum setting needed to enable either the
bulk alloc or vector RX routines inside the ixgbe driver, so it's
best made the default for that reason. Please see
"check_rx_burst_bulk_alloc_preconditions()" in ixgbe_rxtx.c, and
RX function assignment logic in "ixgbe_dev_rx_queue_setup()" in
the same file.
The difference in IO performance for testpmd when called without any
optional parameters, and using 10G NICs using the ixgbe driver, can be
significant - approx 25% or more.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
The function rte_snprintf() was deprecated in version 1.7.0
(commit 6f41fe75e2).
It's now totally removed.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
