Reshma Pattan [Mon, 10 Oct 2016 21:33:13 +0000 (22:33 +0100)]
mk: use -march option with recent Intel processors names
The GCC 4.9 -march option supports the intel code names for processors,
for example -march=silvermont, -march=broadwell.
The RTE_MACHINE config flag can be used to pass code name to
the compiler as -march flag.
Release notes is updated.
Linux and FreeBSD getting started guides are updated with recommended
gcc version as 4.9 and above.
Some of the gmake command examples in sample application guide and driver
guides are updated with gcc version as 4.9.
Olivier Matz [Wed, 12 Oct 2016 15:39:47 +0000 (17:39 +0200)]
app/testpmd: do not change IP addrs in checksum engine
The csum forward engine was updated to change the IP addresses in the
packet data in
commit 51f694dd40f5 ("app/testpmd: rework checksum forward engine")
This was done to ensure that the checksum is correctly reprocessed when
using hardware checksum offload. But the functions
process_inner_cksums() and process_outer_cksums() already reset the
checksum field to 0, so this is not necessary.
Moreover, this makes the engine more complex than needed, and prevents
to easily use it to forward traffic (like iperf) as it modifies the
packets.
This patch drops this behavior.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Olivier Matz [Wed, 12 Oct 2016 15:39:46 +0000 (17:39 +0200)]
app/testpmd: add option to enable LRO
Introduce a new argument '--enable-lro' to ask testpmd to enable the LRO
feature on enabled ports, like it's done for '--enable-rx-cksum' for
instance.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Olivier Matz [Wed, 12 Oct 2016 15:39:43 +0000 (17:39 +0200)]
mbuf: add functions to dump offload flags
The functions rte_get_rx_ol_flag_name() and rte_get_tx_ol_flag_name()
can dump one flag, or set of flag that are part of the same mask (ex:
PKT_TX_UDP_CKSUM, part of PKT_TX_L4_MASK). But they are not designed to
dump the list of flags contained in mbuf->ol_flags.
This commit introduce new functions to do that. Similarly to the packet
type dump functions, the goal is to factorize the code that could be
used in several applications and reduce the risk of desynchronization
between the flags and the dump functions.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Olivier Matz [Mon, 3 Oct 2016 08:38:57 +0000 (10:38 +0200)]
app/testpmd: display software packet type
In addition to the packet type returned by the PMD, also display the
packet type calculated by parsing the packet in software. This is
particularly useful to compare the 2 values.
Note: it does not mean that both hw and sw always have to provide the
same value, since it depends on what hardware supports.
Olivier Matz [Mon, 3 Oct 2016 08:38:54 +0000 (10:38 +0200)]
mbuf: add functions to dump packet type
Dumping the packet type is useful for debug purposes. Instead
of having each application providing its function to do that,
introduce functions to do it.
It factorizes the code and reduces the risk of desynchronization between
the new packet types and the dump function.
Olivier Matz [Mon, 3 Oct 2016 08:38:52 +0000 (10:38 +0200)]
net: support NVGRE in software packet type parser
Add support of Nvgre tunnels in rte_net_get_ptype(). At the same
time, as Nvgre transports Ethernet, we need to add the support for inner
Vlan, QinQ, and Mpls.
Signed-off-by: Jean Dao <jean.dao@6wind.com> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Olivier Matz [Mon, 3 Oct 2016 08:38:46 +0000 (10:38 +0200)]
net: add function to get packet type from data
Introduce the function rte_net_get_ptype() that parses a mbuf and
returns its packet type. For now, the following packet types are parsed:
L2: Ether
L3: IPv4, IPv6
L4: TCP, UDP, SCTP
The goal here is to provide a reference implementation for packet type
parsing. This function will be used by testpmd in next commits, allowing
to compare its result with the value given by the hardware.
This function will also be useful when implementing Rx offload support
in virtio pmd. Indeed, the virtio protocol gives the csum start and
offset, but it does not give the L4 protocol nor it tells if the
checksum is relevant for inner or outer. This information has to be
known to properly set the ol_flags in mbuf.
Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Signed-off-by: Jean Dao <jean.dao@6wind.com> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Olivier Matz [Mon, 3 Oct 2016 08:38:45 +0000 (10:38 +0200)]
net: introduce net library
Previously, librte_net only contained header files. Add a C file
(empty for now) and generate a library. It will contain network helpers
like checksum calculation, software packet type parser, ...
Olivier Matz [Mon, 3 Oct 2016 08:38:43 +0000 (10:38 +0200)]
net: move ethernet definitions to the net library
The proper place for rte_ether.h is in librte_net because it defines
network headers.
Moving it will also prevent to have circular references in the following
patches that will require the Ethernet header definition in rte_mbuf.c.
By the way, fix minor checkpatch issues.
Olivier Matz [Mon, 3 Oct 2016 08:38:42 +0000 (10:38 +0200)]
mbuf: add function to read packet data
Introduce a new function to read the packet data from an mbuf chain. It
linearizes the data if required, and also ensures that the mbuf is large
enough.
This function is used in next commits that add a software parser to
retrieve the packet type.
David Marchand [Fri, 7 Oct 2016 13:01:15 +0000 (15:01 +0200)]
ethdev: fix hotplug attach
If a pci probe operation creates a port but, for any reason, fails to
finish this operation and decides to delete the newly created port, then
the last created port id can not be trusted anymore and any subsequent
attach operations will fail.
This problem was noticed while working on a vm that had a virtio-net
management interface bound to the virtio-net kernel driver and no port
whitelisted in the commandline:
root@ubuntu1404:~/dpdk# ./build/app/testpmd -c 0x6 --
-i --total-num-mbufs=2049
EAL: Detected 3 lcore(s)
EAL: Probing VFIO support...
EAL: Debug logs available - lower performance
EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using
unreliable clock cycles !
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL: probe driver: 1af4:1000 (null)
rte_eth_dev_pci_probe: driver (null): eth_dev_init(vendor_id=0x6900
device_id=0x1000) failed
EAL: No probed ethernet devices
^
|
Here, rte_eth_dev_pci_probe() fails since vtpci_init() reports an
error. This results in a rte_eth_dev_release_port() right after a
rte_eth_dev_allocate().
Then, if we try to attach a port using rte_eth_dev_attach:
testpmd> port attach net_ring0
Attaching a new port...
PMD: Initializing pmd_ring for net_ring0
PMD: Creating rings-backed ethdev on numa socket 0
Two solutions:
- either update the last created port index to something invalid
(when freeing a ethdev port),
- or rely on the port count, before and after the eal attach.
The latter solution seems (well not really more robust but at least)
less fragile than the former.
We still have some issues with drivers that create multiple ethdev
ports with a single probe operation, but this was already the case.
Fixes: b0fb26685570 ("ethdev: convert to EAL hotplug") Reported-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com> Signed-off-by: David Marchand <david.marchand@6wind.com>
Jianfeng Tan [Mon, 26 Sep 2016 13:48:34 +0000 (13:48 +0000)]
app/testpmd: support tunneled TSO in checksum engine
Add a new command "tunnel_tso set <tso_segsz> <port>" to enable
segmentation offload and set MSS to tso_segsz. Another command,
"tunnel_tso show <port>" is added to show tunneled packet MSS.
Result 0 means tunnel_tso is disabled.
The original commands, "tso set <tso_segsz> <port>" and "tso show
<port>" are only reponsible for non-tunneled packets. And the new
commands are for tunneled packets.
Below conditions are needed to make it work:
a. tunnel TSO is supported by the NIC;
b. "csum parse_tunnel" must be set so that tunneled pkts are
recognized;
c. for tunneled pkts with outer L3 is IPv4, "csum set outer-ip"
must be set to hw, because after tso, total_len of outer IP
header is changed, and the checksum of outer IP header calculated
by sw should be wrong; that is not necessary for IPv6 tunneled
pkts because there's no checksum field to be filled anymore.
Suggested-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Signed-off-by: Zhe Tao <zhe.tao@intel.com> Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Jianfeng Tan [Mon, 1 Aug 2016 03:56:54 +0000 (03:56 +0000)]
net/i40e: support TSO on tunneling packet
To enable Tx side offload on tunneling packet, driver should set
correct tunneling parameters: (1) EIPT, External IP header type;
(2) EIPLEN, External IP; (3) L4TUNT; (4) L4TUNLEN. This parsing
behavior is based on (ol_flag & PKT_TX_TUNNEL_MASK). And when
it's a tunneling packet, MACLEN defines the outer L2 header.
Also, we define TSO on each kind of tunneling type as a capabilities.
Now only i40e declares to support them.
Signed-off-by: Zhe Tao <zhe.tao@intel.com> Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Jianfeng Tan [Mon, 1 Aug 2016 03:56:53 +0000 (03:56 +0000)]
mbuf: add Tx side tunneling type
To support tunneling packet offload capabilities on Tx side, PMDs
(e.g., i40e) need to know what kind of tunneling type of this packet.
Instead of analyzing the packet itself, we depend on applications to
correctly set the tunneling type. These flags are defined inside
rte_mbuf.ol_flags.
Signed-off-by: Zhe Tao <zhe.tao@intel.com> Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Fiona Trahe [Thu, 6 Oct 2016 17:34:29 +0000 (18:34 +0100)]
app/test: remove crypto queue number hard-coding
ts_params->conf.nb_queue_pairs should not be hard coded with device
specific number. It should be retrieved from the device info.
Any test which changes it should restore it to orig value.
Fiona Trahe [Thu, 6 Oct 2016 17:34:28 +0000 (18:34 +0100)]
app/test: cleanup unnecessary crypto ring size setup
Removed obsolete comments re inability to free and re-allocate
queue memory and obsolete workaround for it
which used to create maximum size queues first, then later
create smaller queues.
Pablo de Lara [Thu, 6 Oct 2016 20:54:59 +0000 (21:54 +0100)]
doc: fix typo in SNOW 3G guide
Fixes: 1d0c90e6cf0b ("doc: update build instructions for libsso_snow3g") Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com>
Pablo de Lara [Wed, 5 Oct 2016 02:45:51 +0000 (03:45 +0100)]
cryptodev: fix build on Suse 11 SP2
This commit fixes following build error, which happens in SUSE 11 SP2,
with gcc 4.5.1:
In file included from lib/librte_cryptodev/rte_cryptodev.c:70:0:
lib/librte_cryptodev/rte_cryptodev.h:772:7:
error: flexible array member in otherwise empty struct
Fixes: 347a1e037fd3 ("lib: use C99 syntax for zero-size arrays") Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
This patch initializes the salt value used by the following cipher
algorithms:
- CBC: random salt
- GCM/CTR: the key required is 20B, and the last 4B are used as salt.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Application will segfault if there is IPv4 or IPv6 and no SP/ACL rules
for IPv4 or IPv6 respectively.
Avoid checking the ACL/SP in such cases.
Fixes: 906257e965b7 ("examples/ipsec-secgw: support IPv6") Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
NIST SP800-38A recommends two methods to generate unpredictable IVs
(Initilisation Vector) for CBC mode:
1) Apply the forward function to a nonce (ie. counter)
2) Use a FIPS-approved random number generator
This patch implements the first recommended method by using the forward
function to generate the IV.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This patch contains unit tests for libcrypto PMD. User can
use app/test application to check how to use this pmd and to
verify crypto processing.
Test name is cryptodev_libcrypto_autotest.
For performance test cryptodev_libcrypto_perftest can be used.
Signed-off-by: Piotr Azarewicz <piotrx.t.azarewicz@intel.com> Signed-off-by: Marcin Kerlin <marcinx.kerlin@intel.com> Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This patch rework AES tests .
In general - rename AES-named functions to blockcipher functions pattern.
Signed-off-by: Piotr Azarewicz <piotrx.t.azarewicz@intel.com> Signed-off-by: Fiona Trahe <fiona.trahe@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This code provides the initial implementation of the libcrypto
poll mode driver. All cryptography operations are using Openssl
library crypto API. Each algorithm uses EVP_ interface from
openssl API - which is recommended by Openssl maintainers.
This patch adds libcrypto poll mode driver support to librte_cryptodev
library.
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com> Signed-off-by: Michal Kobylinski <michalx.kobylinski@intel.com> Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Pablo de Lara [Thu, 29 Sep 2016 02:59:48 +0000 (03:59 +0100)]
app/test: rename some SNOW 3G functions
Before adding the new ZUC tests, since they will use
the existing common functions for SNOW3G and KASUMI,
these functions are renamed to *_wireless_algo_*,
instead of *_snow3g_kasumi_*, as they are common
functions for all the three wireless algorithms.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Pablo de Lara [Thu, 29 Sep 2016 02:59:47 +0000 (03:59 +0100)]
crypto/zuc: add driver for ZUC library
Added new SW PMD which makes use of the libsso SW library,
which provides wireless algorithms ZUC EEA3 and EIA3
in software.
This PMD supports cipher-only, hash-only and chained operations
("cipher then hash" and "hash then cipher") of the following
algorithms:
- RTE_CRYPTO_SYM_CIPHER_ZUC_EEA3
- RTE_CRYPTO_SYM_AUTH_ZUC_EIA3
The ZUC hash and cipher algorithms, which are enabled
by this crypto PMD are implemented by Intel's libsso software
library.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
This patch adds AES Galois Counter Mode performance test case
for cryptodev QAT and AESNI GCM. Test is performed with different
buffer sizes, burst size of 32 and 128b key. Test vectors
are placed in app/test/test_cryptodev_perf_vectors.h file.
Pablo de Lara [Wed, 28 Sep 2016 00:31:27 +0000 (01:31 +0100)]
crypto: fix build with icc
This commit fixes a compilation error on icc,
due to unallowed conversion from int to enum:
drivers/crypto/snow3g/rte_snow3g_pmd.c(155):
error #188: enumerated type mixed with another type
sess->op = mode;
^
drivers/crypto/kasumi/rte_kasumi_pmd.c(155):
error #188: enumerated type mixed with another type
sess->op = mode;
^
Fixes: 3aafc423cf4d ("snow3g: add driver for SNOW 3G library") Fixes: 2773c86d061a ("crypto/kasumi: add driver for KASUMI library") Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch adds two sample configuration files to ipsec-secgw sample
application. The sample configuration files show how to setup
back-to-back systems that would forward traffic through an IPsec
tunnel.
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Fan Zhang [Wed, 21 Sep 2016 12:05:18 +0000 (13:05 +0100)]
examples/ipsec-secgw: support configuration file
This patch adds the configuration file support to ipsec_secgw
sample application. Instead of hard-coded rules, the users can
specify their own SP, SA, and routing rules in the configuration
file. A command line option "-f" is added to pass the
configuration file location to the application.
Cleanup of unused code.
Rename and simplify a badly named struct element, was aes, but
used for all types of ciphers.
Print correct error msg (Unsupported rather than Undefined)
for all ciphers not supported by QAT PMD.
Pablo de Lara [Wed, 21 Sep 2016 01:45:18 +0000 (02:45 +0100)]
crypto: rename some SNOW 3G references
SNOW 3G algorithm has all uppercase letters in its name
and a space between SNOW and 3G, but some references of it
had some lowercase letters or no space.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
In file rte_crypto_sym.h, GMAC API comments need to be changed
to comply with the GMAC specification. Main areas of change are
aad pointer and aad len, which now will be used to
provide plaintext.
Added Galois Message Authentication Code (GMAC) capability to
QuickAssist Technology symmetric cryptographic driver.
GMAC is authentication only variant of Galois Counter Mode (GCM)
where all plaintext is provided with AAD pointer only.
John Griffin [Thu, 4 Aug 2016 15:46:23 +0000 (16:46 +0100)]
crypto/qat: make the session struct variable in size
This patch changes the qat firmware session data structure from a fixed
size to a variable size which is dependent on the size of the chosen
algorithm.
This reduces the amount of bytes which are transferred across
PCIe and thus helps to increase qat performance when the
accelerator is bound by PCIe.
Signed-off-by: John Griffin <john.griffin@intel.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Shreyansh Jain [Thu, 6 Oct 2016 13:54:03 +0000 (19:24 +0530)]
vdev: rename init/uninit ops to probe/remove
Inline with PCI probe and remove, VDEV probe and remove hooks provide
a uniform naming.
PCI probe represents scan and driver initialization. For VDEV, it will
represent argument parsing and initialization.
When compiled with EXTRA_CFLAGS="-O1", the compiler is not
able to detect that size is always initialized when used, and
issues a wrong warning:
eal_memory.c: In function ‘rte_eal_hugepage_attach’:
eal_memory.c:1684:3: error: ‘size’ may be used uninitialized in this
function [-Werror=maybe-uninitialized]
munmap(hp, size);
^
Workaround this issue by initializing size to 0.
Seen on gcc (Debian 5.4.1-1) 5.4.1 20160803.
net/bnxt: update struct definitions for 1.5.1 HWRM API
Update the PMD to use the 1.5.1 HWRM API.
Most of the changes in the patch are white spaces and rearrangement of the
lines - a onetime change owing to the usage of a different auto generated
file.
Update the PMD to use packet filtering related structures as per the
1.5.1 version of the HWRM API. Most of the changes in the patch are
white spaces and rearrangement of the lines - a onetime change owing to
the usage of a different auto generated file.
Update the PMD to use HWRM ring related structures as per the 1.5.1
version of the HWRM API. Most of the changes in the patch are white
spaces and rearrangement of the lines - a onetime change owing to the
usage of a different auto generated file.
Update the PMD to use VNIC related structures as per the 1.5.1 HWRM API.
Most of the changes in the patch are white spaces and rearrangement of the
lines - a onetime change owing to the usage of a different
auto generated file.
Update the PMD to use the 1.5.1 HWRM API.
Most of the changes in the patch are white spaces and rearrangement of the
lines - a onetime change owing to the usage of a different
auto generated file.
Update the PMD to use the 1.5.1 HWRM API.
Most of the changes in the patch are white spaces and rearrangement of the
lines - a onetime change owing to the usage of a different
auto generated file.
Structures being updated in this patch:
input, output, hwrm_ver_get_input, hwrm_ver_get_output,
hwrm_func_reset_input, hwrm_func_reset_output, hwrm_func_qcaps_input,
hwrm_func_qcaps_output
Update the PMD to use structures as per the 1.5.1 HWRM API.
Most of the changes in the patch are white spaces and rearrangement of the
lines - hopefully a onetime change owing to the usage of a different
auto generated file.
Update the PMD to use the 1.5.1 HWRM API.
Most of the changes in the patch are white spaces and rearrangement of the
lines - hopefully a onetime change owing to the usage of a different
auto generated file.
Structures updated in this patch:
tx_bd_short, tx_bd_long, tx_bd_long_hi, rx_prod_pkt_bd
Update the PMD to use the 1.5.1 HWRM API.
Most of the changes in this patch and the following patches are
white spaces and rearrangement of the lines - hopefully a onetime change
owing to the usage of a different auto generated file.
Other than that, the following fields have been renamed:
1) rx_err_pkts and tx_err_pkts are now rx_discard_pkts and tx_discard_pkts
in struct ctx_hw_stats64
2) the perm_mac_addr field in the response of bnxt_hwrm_func_qcaps has
changed to mac_addr.
Add support to the bnxt PMD to load on a PCI VF.
1) VF cannot change parameters like - speed, autoneg and pause
2) If the VF MAC address shows up as all 0's it has to be provisioned
by the PF in the hypervisor.
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com> Reviewed-by: David Christensen <david.christensen@broadcom.com>
git.droids-corp.org / dpdk.git / log