dpdk.git
5 years agonet: add rte prefix to ip defines
Olivier Matz [Mon, 18 Mar 2019 17:07:33 +0000 (18:07 +0100)]
net: add rte prefix to ip defines

Add 'RTE_' prefix to defines:
- rename IPv4( as RTE_IPv4(.
- rename IPV4_MAX_PKT_LEN as RTE_IPV4_MAX_PKT_LEN.
- rename IPV4_HDR_IHL_MASK as RTE_IPV4_HDR_IHL_MASK.
- rename IPV4_IHL_MULTIPLIER as RTE_IPV4_IHL_MULTIPLIER.
- rename IPV4_HDR_DF_SHIFT as RTE_IPV4_HDR_DF_SHIFT.
- rename IPV4_HDR_MF_SHIFT as RTE_IPV4_HDR_MF_SHIFT.
- rename IPV4_HDR_FO_SHIFT as RTE_IPV4_HDR_FO_SHIFT.
- rename IPV4_HDR_DF_FLAG as RTE_IPV4_HDR_DF_FLAG.
- rename IPV4_HDR_MF_FLAG as RTE_IPV4_HDR_MF_FLAG.
- rename IPV4_HDR_OFFSET_MASK as RTE_IPV4_HDR_OFFSET_MASK.
- rename IPV4_HDR_OFFSET_UNITS as RTE_IPV4_HDR_OFFSET_UNITS.
- rename IPV4_ANY as RTE_IPV4_ANY.
- rename IPV4_LOOPBACK as RTE_IPV4_LOOPBACK.
- rename IPV4_BROADCAST as RTE_IPV4_BROADCAST.
- rename IPV4_ALLHOSTS_GROUP as RTE_IPV4_ALLHOSTS_GROUP.
- rename IPV4_ALLRTRS_GROUP as RTE_IPV4_ALLRTRS_GROUP.
- rename IPV4_MAX_LOCAL_GROUP as RTE_IPV4_MAX_LOCAL_GROUP.
- rename IPV4_MIN_MCAST as RTE_IPV4_MIN_MCAST.
- rename IPV4_MAX_MCAST as RTE_IPV4_MAX_MCAST.
- rename IS_IPV4_MCAST as RTE_IS_IPV4_MCAST.
- rename IPV6_HDR_FL_SHIFT as RTE_IPV6_HDR_FL_SHIFT.
- rename IPV6_HDR_TC_SHIFT as RTE_IPV6_HDR_TC_SHIFT.
- rename IPV6_HDR_FL_MASK as RTE_IPV6_HDR_FL_MASK.
- rename IPV6_HDR_TC_MASK as RTE_IPV6_HDR_TC_MASK.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to ip structure
Olivier Matz [Mon, 18 Mar 2019 17:07:30 +0000 (18:07 +0100)]
net: add rte prefix to ip structure

Add 'rte_' prefix to structures:
- rename struct ipv4_hdr as struct rte_ipv4_hdr.
- rename struct ipv6_hdr as struct rte_ipv6_hdr.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to icmp defines
Olivier Matz [Mon, 18 Mar 2019 17:07:29 +0000 (18:07 +0100)]
net: add rte prefix to icmp defines

Add 'RTE_' prefix to defines:
- rename IP_ICMP_ECHO_REPLY as RTE_IP_ICMP_ECHO_REPLY.
- rename IP_ICMP_ECHO_REQUEST as RTE_IP_ICMP_ECHO_REQUEST.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to icmp structure
Olivier Matz [Mon, 18 Mar 2019 17:07:28 +0000 (18:07 +0100)]
net: add rte prefix to icmp structure

Add 'rte_' prefix to structures:
- rename struct icmp_hdr as struct rte_icmp_hdr.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to gre structure
Olivier Matz [Mon, 18 Mar 2019 17:07:28 +0000 (18:07 +0100)]
net: add rte prefix to gre structure

Add 'rte_' prefix to structures:
- rename struct gre_hdr as struct rte_gre_hdr.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to esp structure
Olivier Matz [Mon, 18 Mar 2019 17:07:28 +0000 (18:07 +0100)]
net: add rte prefix to esp structure

Add 'rte_' prefix to structures:
- rename struct esp_hdr as struct rte_esp_hdr.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to ether defines
Olivier Matz [Mon, 18 Mar 2019 17:07:27 +0000 (18:07 +0100)]
net: add rte prefix to ether defines

Add 'RTE_' prefix to defines:
- rename ETHER_ADDR_LEN as RTE_ETHER_ADDR_LEN.
- rename ETHER_TYPE_LEN as RTE_ETHER_TYPE_LEN.
- rename ETHER_CRC_LEN as RTE_ETHER_CRC_LEN.
- rename ETHER_HDR_LEN as RTE_ETHER_HDR_LEN.
- rename ETHER_MIN_LEN as RTE_ETHER_MIN_LEN.
- rename ETHER_MAX_LEN as RTE_ETHER_MAX_LEN.
- rename ETHER_MTU as RTE_ETHER_MTU.
- rename ETHER_MAX_VLAN_FRAME_LEN as RTE_ETHER_MAX_VLAN_FRAME_LEN.
- rename ETHER_MAX_VLAN_ID as RTE_ETHER_MAX_VLAN_ID.
- rename ETHER_MAX_JUMBO_FRAME_LEN as RTE_ETHER_MAX_JUMBO_FRAME_LEN.
- rename ETHER_MIN_MTU as RTE_ETHER_MIN_MTU.
- rename ETHER_LOCAL_ADMIN_ADDR as RTE_ETHER_LOCAL_ADMIN_ADDR.
- rename ETHER_GROUP_ADDR as RTE_ETHER_GROUP_ADDR.
- rename ETHER_TYPE_IPv4 as RTE_ETHER_TYPE_IPv4.
- rename ETHER_TYPE_IPv6 as RTE_ETHER_TYPE_IPv6.
- rename ETHER_TYPE_ARP as RTE_ETHER_TYPE_ARP.
- rename ETHER_TYPE_VLAN as RTE_ETHER_TYPE_VLAN.
- rename ETHER_TYPE_RARP as RTE_ETHER_TYPE_RARP.
- rename ETHER_TYPE_QINQ as RTE_ETHER_TYPE_QINQ.
- rename ETHER_TYPE_ETAG as RTE_ETHER_TYPE_ETAG.
- rename ETHER_TYPE_1588 as RTE_ETHER_TYPE_1588.
- rename ETHER_TYPE_SLOW as RTE_ETHER_TYPE_SLOW.
- rename ETHER_TYPE_TEB as RTE_ETHER_TYPE_TEB.
- rename ETHER_TYPE_LLDP as RTE_ETHER_TYPE_LLDP.
- rename ETHER_TYPE_MPLS as RTE_ETHER_TYPE_MPLS.
- rename ETHER_TYPE_MPLSM as RTE_ETHER_TYPE_MPLSM.
- rename ETHER_VXLAN_HLEN as RTE_ETHER_VXLAN_HLEN.
- rename ETHER_ADDR_FMT_SIZE as RTE_ETHER_ADDR_FMT_SIZE.
- rename VXLAN_GPE_TYPE_IPV4 as RTE_VXLAN_GPE_TYPE_IPV4.
- rename VXLAN_GPE_TYPE_IPV6 as RTE_VXLAN_GPE_TYPE_IPV6.
- rename VXLAN_GPE_TYPE_ETH as RTE_VXLAN_GPE_TYPE_ETH.
- rename VXLAN_GPE_TYPE_NSH as RTE_VXLAN_GPE_TYPE_NSH.
- rename VXLAN_GPE_TYPE_MPLS as RTE_VXLAN_GPE_TYPE_MPLS.
- rename VXLAN_GPE_TYPE_GBP as RTE_VXLAN_GPE_TYPE_GBP.
- rename VXLAN_GPE_TYPE_VBNG as RTE_VXLAN_GPE_TYPE_VBNG.
- rename ETHER_VXLAN_GPE_HLEN as RTE_ETHER_VXLAN_GPE_HLEN.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to ether functions
Olivier Matz [Mon, 18 Mar 2019 17:07:21 +0000 (18:07 +0100)]
net: add rte prefix to ether functions

Add 'rte_' prefix to functions:
- rename is_same_ether_addr() as rte_is_same_ether_addr().
- rename is_zero_ether_addr() as rte_is_zero_ether_addr().
- rename is_unicast_ether_addr() as rte_is_unicast_ether_addr().
- rename is_multicast_ether_addr() as rte_is_multicast_ether_addr().
- rename is_broadcast_ether_addr() as rte_is_broadcast_ether_addr().
- rename is_universal_ether_addr() as rte_is_universal_ether_addr().
- rename is_local_admin_ether_addr() as rte_is_local_admin_ether_addr().
- rename is_valid_assigned_ether_addr() as rte_is_valid_assigned_ether_addr().
- rename eth_random_addr() as rte_eth_random_addr().
- rename ether_addr_copy() as rte_ether_addr_copy().
- rename ether_format_addr() as rte_ether_format_addr().

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to ether structures
Olivier Matz [Mon, 18 Mar 2019 17:07:17 +0000 (18:07 +0100)]
net: add rte prefix to ether structures

Add 'rte_' prefix to structures:
- rename struct ether_addr as struct rte_ether_addr.
- rename struct ether_hdr as struct rte_ether_hdr.
- rename struct vlan_hdr as struct rte_vlan_hdr.
- rename struct vxlan_hdr as struct rte_vxlan_hdr.
- rename struct vxlan_gpe_hdr as struct rte_vxlan_gpe_hdr.

Additionally, adapt the command line library to use the rte
structures and defines.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to arp defines
Olivier Matz [Mon, 18 Mar 2019 17:07:15 +0000 (18:07 +0100)]
net: add rte prefix to arp defines

Add 'RTE_' prefix to defines:
- rename ARP_HRD_ETHER as RTE_ARP_HRD_ETHER.
- rename ARP_OP_REQUEST as RTE_ARP_OP_REQUEST.
- rename ARP_OP_REPLY as RTE_ARP_OP_REPLY.
- rename ARP_OP_REVREQUEST as RTE_ARP_OP_REVREQUEST.
- rename ARP_OP_REVREPLY as RTE_ARP_OP_REVREPLY.
- rename ARP_OP_INVREQUEST as RTE_ARP_OP_INVREQUEST.
- rename ARP_OP_INVREPLY as RTE_ARP_OP_INVREPLY.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agonet: add rte prefix to arp structures
Olivier Matz [Mon, 18 Mar 2019 17:07:14 +0000 (18:07 +0100)]
net: add rte prefix to arp structures

Add 'rte_' prefix to structures:
- rename struct arp_hdr as struct rte_arp_hdr.
- rename struct arp_ipv4 as struct rte_arp_ipv4.

Also rename arp_hrd, arp_pro, arp_hln, arp_pln and arp_op fields
to avoid conflict with the #defines in gnu libc.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agodoc: announce API change for net definitions, structures, functions
Olivier Matz [Mon, 18 Mar 2019 17:00:20 +0000 (18:00 +0100)]
doc: announce API change for net definitions, structures, functions

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
5 years agogit: ignore build directories
Bruce Richardson [Mon, 11 Mar 2019 10:57:32 +0000 (10:57 +0000)]
git: ignore build directories

test-meson-build.sh generates multiple build directories for various
targets. As these follow a known pattern, and since they don't need
to be tracked in git, we can add them to the gitignore file,
along with the default build directory "build".

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Rami Rosen <ramirose@gmail.com>
5 years agogit: ignore hidden files
Bruce Richardson [Mon, 11 Mar 2019 10:57:31 +0000 (10:57 +0000)]
git: ignore hidden files

Generally hidden files are hidden for good reason and we don't want to
track them in git. They can always be manually added to git tracking
individually if needed.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agogit: ignore python bytecode files
Bruce Richardson [Mon, 11 Mar 2019 10:57:30 +0000 (10:57 +0000)]
git: ignore python bytecode files

After you run a python script, a .pyc file is often left behind,
which we don't want to track in git.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agogit: add comments for ignored files
Bruce Richardson [Mon, 11 Mar 2019 10:57:29 +0000 (10:57 +0000)]
git: add comments for ignored files

Split the ignored file list into section based on logical groups of files,
putting a comment at the top of each section clarifying what it is.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agodevtools: fix meson build test to exit on failure
Bruce Richardson [Tue, 12 Mar 2019 10:18:28 +0000 (10:18 +0000)]
devtools: fix meson build test to exit on failure

When piping the ninja command through cat, we lose the error value from
the call to ninja in the case of failure. This prevents the script from
exiting at the first broken build. Fix this by setting the "pipefail"
shell option.

Fixes: 4bcb9b768604 ("devtools: add verbose option to meson build test")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agomk: use linux and freebsd in config names
Bruce Richardson [Wed, 6 Mar 2019 16:22:42 +0000 (16:22 +0000)]
mk: use linux and freebsd in config names

Rather than using linuxapp and bsdapp everywhere, we can change things to
use the, more readable, terms "linux" and "freebsd" in our build configs.
Rather than renaming the configs we can just duplicate the existing ones
with the new names using symlinks, and use the new names exclusively
internally. ["make showconfigs" also only shows the new names to keep the
list short] The result is that backward compatibility is kept fully but any
new builds or development can be done using the newer names, i.e.  both
"make config T=x86_64-native-linuxapp-gcc" and "T=x86_64-native-linux-gcc"
work.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agobuild: rename linuxapp to linux in meson cross files
Bruce Richardson [Wed, 6 Mar 2019 16:22:41 +0000 (16:22 +0000)]
build: rename linuxapp to linux in meson cross files

Rename the cross files for meson compilation from having linuxapp
in the name to just linux in the name.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agobuild/freebsd: rename macro BSDPAPP to FREEBSD
Bruce Richardson [Wed, 6 Mar 2019 16:22:40 +0000 (16:22 +0000)]
build/freebsd: rename macro BSDPAPP to FREEBSD

Rename the macro and all instances in DPDK code, but keep a copy of
the old macro defined for legacy code linking against DPDK

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agobuild/linux: rename macro from LINUXAPP to LINUX
Bruce Richardson [Wed, 6 Mar 2019 16:22:39 +0000 (16:22 +0000)]
build/linux: rename macro from LINUXAPP to LINUX

Rename the macro to make things shorter and more comprehensible. For
both meson and make builds, keep the old macro around for backward
compatibility.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agoeal/linux: rename linuxapp to linux
Bruce Richardson [Wed, 6 Mar 2019 16:22:38 +0000 (16:22 +0000)]
eal/linux: rename linuxapp to linux

The term "linuxapp" is a legacy one, but just calling the subdirectory
"linux" is just clearer for all concerned.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agoeal/bsd: rename bsdapp to freebsd
Bruce Richardson [Wed, 6 Mar 2019 16:22:37 +0000 (16:22 +0000)]
eal/bsd: rename bsdapp to freebsd

The term "bsdapp" is a legacy one, but just calling the subdirectory
"freebsd" is just clearer for all concerned.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agotest/crypto: fix duplicate id used by CCP device
Hemant Agrawal [Wed, 6 Mar 2019 16:40:34 +0000 (22:10 +0530)]
test/crypto: fix duplicate id used by CCP device

These duplicate device id is causing incorrect mapping
for DPAA_SEC for test case execution on the basis of
capabilities.

Fixes: e155ca055e84 ("test/crypto: add tests for AMD CCP")
Cc: stable@dpdk.org
Reported-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
5 years agocryptodev: rework modexp and modinv comments
Arek Kusztal [Wed, 6 Feb 2019 10:34:26 +0000 (11:34 +0100)]
cryptodev: rework modexp and modinv comments

This patch changes modular exponentiation and modular multiplicative
inverse API comments to make it more precise.

Signed-off-by: Arek Kusztal <arkadiuszx.kusztal@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Shally Verma <shallyv@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
5 years agocrypto/openssl: fix modexp
Arek Kusztal [Tue, 5 Feb 2019 09:13:19 +0000 (10:13 +0100)]
crypto/openssl: fix modexp

Fixes bad reference of modinv struct in openssl pmd

Fixes: 3e9d6bd447fb ("crypto/openssl: add RSA and mod asym operations")
Cc: stable@dpdk.org
Signed-off-by: Arek Kusztal <arkadiuszx.kusztal@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Shally Verma <shallyv@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
5 years agocrypto/openssl: fix big numbers after computations
Arek Kusztal [Thu, 7 Feb 2019 10:54:39 +0000 (11:54 +0100)]
crypto/openssl: fix big numbers after computations

After performing mod exp and mod inv big numbers (BIGNUM) should
be cleared as data already is copied into op fields and this BNs would
very likely contain private information for unspecified amount of time
(duration of the session).

Fixes: 3e9d6bd447fb ("crypto/openssl: add RSA and mod asym operations")
Cc: stable@dpdk.org
Signed-off-by: Arek Kusztal <arkadiuszx.kusztal@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Shally Verma <shallyv@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
5 years agodoc: fix table of kernel drivers in qat guide
Fiona Trahe [Thu, 7 Feb 2019 18:46:27 +0000 (18:46 +0000)]
doc: fix table of kernel drivers in qat guide

Added missing line informing which kernel driver can
be used for device DH895xcc for compression service.
Moved service columns to start of table for better visibility
and to prepare for future asymmetric crypto service.

Fixes: e2e35849ea78 ("compress/qat: add compression on DH895x")
Cc: stable@dpdk.org
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com>
5 years agocommon/cpt: fix null auth only
Anoob Joseph [Wed, 13 Feb 2019 09:22:50 +0000 (09:22 +0000)]
common/cpt: fix null auth only

Fixes: 351fbee21986 ("common/cpt: support hash")
Cc: stable@dpdk.org
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
5 years agonet/ixgbe: support VF promiscuous by PF driver
Wei Zhao [Fri, 8 Mar 2019 02:46:17 +0000 (10:46 +0800)]
net/ixgbe: support VF promiscuous by PF driver

The patch adds the PF counterpart changes to support VF promiscuous
mode by DPDK PF driver.

For ixgbe, in order to support VF VLAN promiscuous or unicast
promiscuous, PF need to set register PFVML2FLT of bit UPE and VPE.
The patch aligned to kernel driver's implementation.

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
5 years agonet/ixgbevf: enable promiscuous mode
Wei Zhao [Fri, 8 Mar 2019 02:46:16 +0000 (10:46 +0800)]
net/ixgbevf: enable promiscuous mode

Add promiscuous mode support on VF

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
5 years agonet/mlx5: fix sync when handling Tx completions
Dekel Peled [Thu, 28 Feb 2019 15:20:30 +0000 (17:20 +0200)]
net/mlx5: fix sync when handling Tx completions

Function mlx5_tx_complete() reads completion entry information
from Tx queue.
For some processors not having strongly-ordered memory model,
there has to be a memory barrier between reading the entry index
and the entry fields, in order to guarantee data is valid.

Fixes: 54d3fe948dba ("net/mlx5: poll completion queue once per a call")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
5 years agonet/mlx5: fix hex dump of error completion
Dekel Peled [Thu, 28 Feb 2019 15:18:45 +0000 (17:18 +0200)]
net/mlx5: fix hex dump of error completion

struct mlx5_cqe is defined in MLX5 PMD code (mlx5_prm.h).
It includes 64 bytes padding in case of (RTE_CACHE_LINE_SIZE == 128).

struct mlx5_err_cqe is defined in kernel, and doesn't include padding.

When running in debug mode, in case an error CQE is detected
it is printed using rte_hexdump().

The size of data to print should be sizeof(*cqe) instead of
sizeof(*err_cqe), to handle the case of (RTE_CACHE_LINE_SIZE == 128),
and print the full data in any case.

Fixes: c7714992092f ("net/mlx5: extend debug logs verbosity")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
5 years agonet/mlx5: call generic strlcpy
Thomas Monjalon [Sun, 24 Feb 2019 22:42:40 +0000 (23:42 +0100)]
net/mlx5: call generic strlcpy

The call to strlcpy uses either libc, libbsd or internal rte_strlcpy.
No need to call the DPDK flavor explicitly.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
5 years agonet/mlx4: fix default flow rule create
Dekel Peled [Sun, 24 Feb 2019 09:41:09 +0000 (11:41 +0200)]
net/mlx4: fix default flow rule create

Original patch changed logic of function mlx4_flow_merge_eth().
The setting of flow->promisc was wrongly removed.
This patch adds the removed setting of flow->promisc, to restore
the required behavior.

Fixes: c0d239263156 ("net/mlx4: support flow w/o ETH spec and with VLAN")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
5 years agonet/mlx5: fix flow priorities probing error path
Viacheslav Ovsiienko [Thu, 21 Feb 2019 09:02:16 +0000 (09:02 +0000)]
net/mlx5: fix flow priorities probing error path

The mlx5 PMD probes the Verbs flow priorities supported with
ibv_create_flow() function. If rdma-core or kernel fails for
some reason, the returned error causes the drop queue is not
destroyed, and pd is locked by not freed resource.

Also the mlx5_flow_discover_priorities() returned negative value
as error, and this code was reported "as is", without sign
changing (eventually causing assert(err > 0)).

Fixes: 2815702baea7 ("net/mlx5: replace verbs priorities by flow")
Cc: stable@dpdk.org
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
5 years agonet/i40e: fix negative check on unsigned queue pairs
Kevin Traynor [Tue, 5 Mar 2019 16:30:38 +0000 (16:30 +0000)]
net/i40e: fix negative check on unsigned queue pairs

Fix the check and associated log. Also, fix a typo in other log.

Fixes: 03d478e9609d ("net/i40e: support PF respond VF request more queues")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
5 years agonet/ixgbe: fix crash on remove
Yunjian Wang [Wed, 13 Feb 2019 02:48:52 +0000 (10:48 +0800)]
net/ixgbe: fix crash on remove

The NIC's interrupt source has some active handler when the
port removed. We should cancel the delay handler before removing
dev to prevent executing the delay handler.

Call Trace:
  #0  ixgbe_disable_intr (hw=0x0, hw=0x0)
      at /usr/src/debug/dpdk-18.11/drivers/net/ixgbe/ixgbe_ethdev.c:852
  #1  ixgbe_dev_interrupt_delayed_handler (param=0xadb9c0
      <rte_eth_devices@@DPDK_2.2+33024>)
      at /usr/src/debug/dpdk-18.11/drivers/net/ixgbe/ixgbe_ethdev.c:4386
  #2  0x00007f05782147af in eal_alarm_callback (arg=<optimized out>)
      at /usr/src/debug/dpdk-18.11/lib/librte_eal/linuxapp/eal/
      eal_alarm.c:90
  #3  0x00007f057821320a in eal_intr_process_interrupts (nfds=1,
      events=0x7f056cbf3e88) at /usr/src/debug/dpdk-18.11/lib/
      librte_eal/linuxapp/eal/eal_interrupts.c:838
  #4  eal_intr_handle_interrupts (totalfds=<optimized out>, pfd=18)
      at /usr/src/debug/dpdk-18.11/lib/librte_eal/linuxapp/eal/
      eal_interrupts.c:885
  #5  eal_intr_thread_main (arg=<optimized out>)
      at /usr/src/debug/dpdk-18.11/lib/librte_eal/linuxapp/eal/
      eal_interrupts.c:965
  #6  0x00007f05708a0e45 in start_thread () from /usr/lib64/libpthread.so.0
  #7  0x00007f056eb4ab5d in clone () from /usr/lib64/libc.so.6

Fixes: 2866c5f1b87e ("ixgbe: support port hotplug")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
5 years agodoc: fix tag for inner RSS feature
Rami Rosen [Fri, 1 Mar 2019 11:10:10 +0000 (13:10 +0200)]
doc: fix tag for inner RSS feature

This patch fixes a wrong tag in guides/nics/features.rst.
The features tags should be, according to the
"Features Overview" section in this doc, one of the following:
"uses", "implements", "provides", or "related".
Hence in Inner RSS section, it should be "uses"
instead of "users".

Fixes: d0a87d9aa8de ("doc: update mlx5 guide on tunnel offloading")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/enic: fix inner packet matching
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:51 +0000 (02:42 -0800)]
net/enic: fix inner packet matching

Inner packet matching is currently buggy in many cases.

1. Mishandling null spec ("match any").
The copy_item functions do nothing if spec is null. This is incorrect,
as all patterns should be appended to the L5 pattern buffer even for
null spec (treated as all zeros).

2. Accessing null spec causing segfault.

3. Not setting protocol fields.
The NIC filter API currently has no flags for "match inner IPv4, IPv6,
UDP, TCP, and so on". So, the driver needs to explicitly set EtherType
and IP protocol fields in the L5 pattern buffer to avoid false
positives (e.g. reporting IPv6 as IPv4).

Instead of keep adding "if inner, do something differently" cases to
the existing copy_item functions, introduce separate functions for
inner packet patterns and address the above issues in those
functions. The changes to the previous outer-packet copy_item
functions are mechanical, due to reduced indentation.

Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters enabled")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: fix endianness in VLAN match
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:50 +0000 (02:42 -0800)]
net/enic: fix endianness in VLAN match

The VLAN fields in the NIC filter use little endian. The VLAN item is
in big endian, so swap bytes.

Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters enabled")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: fix VXLAN match
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:49 +0000 (02:42 -0800)]
net/enic: fix VXLAN match

The filter API does not have flags for "match VXLAN". Explicitly set
the UDP destination port and mask in the L4 pattern. Otherwise, UDP
packets with non-VXLAN ports may be falsely reported as VXLAN.

1400 series VIC adapters have hardware VXLAN parsing. The L5 buffer on
the NIC starts with the inner Ethernet header, and the VXLAN header is
now in the L4 buffer following the UDP header. So the VXLAN spec/mask
needs to be in the L4 pattern, not L5. Older models still expect the
VXLAN spec/mask in the L5 pattern. Fix up the L4/L5 patterns
accordingly.

Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters enabled")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: reset VXLAN port regardless of overlay offload
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:48 +0000 (02:42 -0800)]
net/enic: reset VXLAN port regardless of overlay offload

Currently, the driver resets the vxlan port register only if overlay
offload is enabled. But, the register is actually tied to hardware
vxlan parsing, which is an independent feature and is always enabled
even if overlay offload is disabled. If left uninitialized, it can
affect flow rules that match vxlan. So always reset the port number
when HW vxlan parsing is available.

Fixes: 8a4efd17410c ("net/enic: add handlers to add/delete vxlan port number")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: enable limited support for raw flow item
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:47 +0000 (02:42 -0800)]
net/enic: enable limited support for raw flow item

Some apps like VPP use a raw item to match UDP tunnel headers like
VXLAN or GENEVE. The NIC hardware supports such usage via L5 match,
which does pattern match on packet data immediately following the
outer L4 header. Accept raw items for these limited use cases.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: move arguments into struct
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:46 +0000 (02:42 -0800)]
net/enic: move arguments into struct

There are many copy_item functions, all with the same arguments, which
makes it difficult to add/change arguments. Move the arguments into a
struct to help subsequent commits that will add/fix features. Also
remove self-explanatory verbose comments for these local functions.

These changes are purely mechanical and have no impact on
functionalities.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: enable limited passthru flow action
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:45 +0000 (02:42 -0800)]
net/enic: enable limited passthru flow action

Some apps like VPP use PASSTHRU+MARK flow rules to offload packet
matching to the NIC. Just like MARK+RSS used by OVS-DPDK and others,
PASSTHRU+MARK is used to "mark and then receive normally". Recent VIC
adapters support such flow rules, so enable PASSTHRU for this limited
use case.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
5 years agonet/enic: enable limited RSS flow action
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:44 +0000 (02:42 -0800)]
net/enic: enable limited RSS flow action

Some apps like OVS-DPDK use MARK+RSS flow rules in order to offload
packet matching to the NIC. The RSS action in such flow rules simply
indicates "receive packet normally", not trying to override the port
wide RSS. The action is included in the flow rules simply to terminate
them, as MARK is not a fate-deciding action. And, the RSS action has a
most basic config: default hash, level, types, null key, and identity
queue mapping.

Recent VIC adapters can support these "mark and receive" flow
rules. So, enable support for RSS action for this limited use case.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
5 years agonet/enic: check for unsupported flow item types
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:43 +0000 (02:42 -0800)]
net/enic: check for unsupported flow item types

Currently a pattern with an unsupported item type causes segfault,
because the flow handler is using the type as an array index without
checking bounds. Add an explicit check for unsupported item types and
avoid out-of-bound accesses.

Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters enabled")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
5 years agonet/enic: allow flow mark ID 0
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:42 +0000 (02:42 -0800)]
net/enic: allow flow mark ID 0

The driver currently accepts mark ID 0 but does not report it in
matching packet's mbuf. For example, the following testpmd command
succeeds. But, the mbuf of a matching IPv4 UDP packet does not have
PKT_RX_FDIR_ID set.

flow create 0 ingress pattern ... actions mark id 0 / queue index 0 / end

The problem has to do with mapping mark IDs (32-bit) to NIC filter
IDs. Filter ID is currently 16-bit, so values greater than 0xffff are
rejected. The firmware reserves filter ID 0 for filters that do not
mark (e.g. steer w/o mark). And, the driver reserves 0xffff for the
flag action. This leaves 1...0xfffe for app use.

It is possible to simply reject mark ID 0 as unsupported. But, 0 is
commonly used (e.g. OVS-DPDK and VPP). So, when adding a filter, set
filter ID = mark ID + 1 to support mark ID 0. The receive handler
subtracts 1 from filter ID to get back the original mark ID.

Fixes: dfbd6a9cb504 ("net/enic: extend flow director support for 1300 series")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
5 years agonet/enic: fix SCTP match for flow API
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:41 +0000 (02:42 -0800)]
net/enic: fix SCTP match for flow API

The driver needs to explicitly set the protocol number (132) in the IP
header pattern, as the current firmware filter API lacks "match SCTP
packet" flag. Otherwise, the resulting NIC filter may lead to false
positives (i.e. NIC reporting non-SCTP packets as SCTP packets). The
flow director handler does the same (enic_clsf.c).

Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters enabled")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
5 years agonet/enic: fix flow director SCTP matching
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:40 +0000 (02:42 -0800)]
net/enic: fix flow director SCTP matching

The firmware filter API does not have flags indicating "match SCTP
packet". Instead, the driver needs to explicitly add an IP match and
set the protocol number (132 for SCTP) in the IP header.

The existing code (copy_fltr_v2) has two bugs.

1. It sets the protocol number (132) in the match value, but not the
mask. The mask remains 0, so the match becomes a wildcard match. The
NIC ends up matching all protocol numbers (i.e. thinks non-SCTP
packets are SCTP).

2. It modifies the input argument (rte_eth_fdir_input). The driver
tracks filters using rte_hash_{add,del}_key(input). So, addding
(RTE_ETH_FILTER_ADD) and deleting (RTE_ETH_FILTER_DELETE) must use the
same input argument for the same filter. But, overwriting the protocol
number while adding the filter breaks this assumption, and causes
delete operation to fail.

So, set the mask as well as protocol value. Do not modify the input
argument, and use const in function signatures to make the intention
clear. Also move a couple function declarations to enic_clsf.c from
enic.h as they are strictly local.

Fixes: dfbd6a9cb504 ("net/enic: extend flow director support for 1300 series")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
5 years agonet/enic: remove unused functions
Hyong Youb Kim [Sat, 2 Mar 2019 10:42:39 +0000 (02:42 -0800)]
net/enic: remove unused functions

Remove unused functions. Specifically, vnic_set_rss_key() is
obsolete. enic_{add,del}_vlan() has never been supported in the
firmware. And, remove vnic_rss.c altogether as it becomes empty. These
were discovered by cppcheck.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
5 years agoeal: fix core list validation with disabled cores
David Marchand [Wed, 13 Feb 2019 20:06:59 +0000 (21:06 +0100)]
eal: fix core list validation with disabled cores

-l and -c options are two ways to select the cores used by DPDK.
Their format differs, but the checks on the selected cores are the same.
Use an intermediate array to separate the specific parsing checks from
the common consistency checks.
The parsing functions now concentrate on validating the passed string
and do nothing more.

We can report all invalid core indexes rather than only the first error.
In the error log message, reporting [0, cfg->lcore_count - 1] as a valid
range is then wrong when the core list is not continuous.

Example on my 8 cpus laptop with core 2 and 6 disabled.
echo 0 > /sys/devices/system/cpu/cpu2/online
echo 0 > /sys/devices/system/cpu/cpu6/online

Before:
./master/app/testpmd -l 0-7 --no-huge -m 512 -- --total-num-mbufs 2048
EAL: Detected 6 lcore(s)
EAL: Detected 1 NUMA nodes
EAL: invalid core list, please check core numbers are in [0, 5] range
...

After:
./master/app/testpmd -l 0-7 --no-huge -m 512 -- --total-num-mbufs 2048
EAL: Detected 6 lcore(s)
EAL: Detected 1 NUMA nodes
EAL: lcore 2 unavailable
EAL: lcore 6 unavailable
EAL: invalid core list, please check specified cores are part of 0-1,3-5,7
...

Fixes: d888cb8b9613 ("eal: add core list input format")
Fixes: b38693b612b4 ("eal: fix core number validation")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
5 years agoeal: remove dead code in core list parsing
David Marchand [Wed, 13 Feb 2019 20:06:58 +0000 (21:06 +0100)]
eal: remove dead code in core list parsing

We don't need to look for trailing spaces.
This is a copy/paste block from eal_parse_coremask().
Remove it and the associated comment.

Fixes: d888cb8b9613 ("eal: add core list input format")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
5 years agoeal: restrict control threads to startup CPU affinity
David Marchand [Tue, 19 Feb 2019 20:41:11 +0000 (21:41 +0100)]
eal: restrict control threads to startup CPU affinity

Spawning the ctrl threads on anything that is not part of the eal
coremask is not that polite to the rest of the system, especially
when you took good care to pin your processes on cpu resources with
tools like taskset (linux) / cpuset (freebsd).

Rather than introduce yet another eal options to control on which cpu
those ctrl threads are created, let's take the startup cpu affinity
as a reference and remove the eal coremask from it.
If no cpu is left, then we default to the master core.

The cpuset is computed once at init before the original cpu affinity
is lost.

Introduced a RTE_CPU_AND macro to abstract the differences between linux
and freebsd respective macros.

Examples in a 4 cores FreeBSD vm:

$ ./build/app/testpmd -l 2,3 --no-huge --no-pci -m 512 \
 -- -i --total-num-mbufs=2048

$ procstat -S 1057
  PID    TID COMM                TDNAME              CPU CSID CPU MASK
 1057 100131 testpmd             -                     2    1 2
 1057 100140 testpmd             eal-intr-thread       1    1 0-1
 1057 100141 testpmd             rte_mp_handle         1    1 0-1
 1057 100142 testpmd             lcore-slave-3         3    1 3

$ cpuset -l 1,2,3 ./build/app/testpmd -l 2,3 --no-huge --no-pci -m 512 \
 -- -i --total-num-mbufs=2048

$ procstat -S 1061
  PID    TID COMM                TDNAME              CPU CSID CPU MASK
 1061 100131 testpmd             -                     2    2 2
 1061 100144 testpmd             eal-intr-thread       1    2 1
 1061 100145 testpmd             rte_mp_handle         1    2 1
 1061 100147 testpmd             lcore-slave-3         3    2 3

$ cpuset -l 2,3 ./build/app/testpmd -l 2,3 --no-huge --no-pci -m 512 \
 -- -i --total-num-mbufs=2048

$ procstat -S 1065
  PID    TID COMM                TDNAME              CPU CSID CPU MASK
 1065 100131 testpmd             -                     2    2 2
 1065 100148 testpmd             eal-intr-thread       2    2 2
 1065 100149 testpmd             rte_mp_handle         2    2 2
 1065 100150 testpmd             lcore-slave-3         3    2 3

Fixes: d651ee4919cd ("eal: set affinity for control threads")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
5 years agoeal: fix control threads pinnning
David Marchand [Tue, 19 Feb 2019 20:41:10 +0000 (21:41 +0100)]
eal: fix control threads pinnning

pthread_setaffinity_np returns a >0 value on error.
We could end up letting the ctrl threads on the current process cpu
affinity.

Fixes: d651ee4919cd ("eal: set affinity for control threads")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
5 years agoeal: fix check when retrieving current CPU affinity
David Marchand [Tue, 19 Feb 2019 20:38:13 +0000 (21:38 +0100)]
eal: fix check when retrieving current CPU affinity

pthread_getaffinity_np returns a >0 value when failing.

This is mainly for the sake of correctness.
The only case where it could fail is when passing an incorrect cpuset
size wrt to the kernel.

Fixes: 2eba8d21f3c9 ("eal: restrict cores auto detection")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
5 years agobus/vdev: fix hotplug twice
Raslan Darawsheh [Thu, 21 Feb 2019 19:01:25 +0000 (20:01 +0100)]
bus/vdev: fix hotplug twice

In case vdev was already probed, it shouldn't be probed again,
and it should return -EEXIST as error.
There are some checks in vdev_probe() and insert_vdev(),
but a check was missing in vdev_plug().
The check is moved in vdev_probe_all_drivers() which is called
in all code paths.

Fixes: e9d159c3d534 ("eal: allow probing a device again")
Cc: stable@dpdk.org
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
5 years agobus/vdev: fix debug message on probing
Thomas Monjalon [Thu, 21 Feb 2019 19:01:24 +0000 (20:01 +0100)]
bus/vdev: fix debug message on probing

The log was printing the device name two times,
first one being supposed to be the driver name.
As we don't know yet the driver name, the log is simplified.

Fixes: 9bf4901d1a11 ("bus/vdev: remove probe with driver name option")
Cc: stable@dpdk.org
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
5 years agomk: fix AVX512 disabled warning on non x86
Jerin Jacob [Sun, 24 Feb 2019 18:27:10 +0000 (18:27 +0000)]
mk: fix AVX512 disabled warning on non x86

AVX512 is a x86 specific feature, So, enable AVX512
warning only on x86.

Fixes: a32ca9a4ebc1 ("mk: fix scope of disabling AVX512F support")
Cc: stable@dpdk.org
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agodoc: fix PCI whitelist typo in prog guide
Thomas Monjalon [Sun, 24 Feb 2019 22:32:59 +0000 (23:32 +0100)]
doc: fix PCI whitelist typo in prog guide

The placeholder for PCI address should be named DBDF
which stands for Domain/Bus/Device/Function.

Fixes: 33af337773ac ("ethdev: add common devargs parser")
Cc: stable@dpdk.org
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
5 years agodoc: fix link in Linux getting started guide
Rami Rosen [Fri, 1 Mar 2019 11:50:13 +0000 (13:50 +0200)]
doc: fix link in Linux getting started guide

This patch fixes a wrong link in gsg. The
Documentation/kernel-parameters.txt file from the kernel
source tree was moved quite a time ago to
Documentation/admin-guide/kernel-parameters.txt.

Fixes: 1ab07743b21b ("doc: getting started guide for linux")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoexamples/ethtool: fix two typos
Rami Rosen [Mon, 25 Feb 2019 19:21:37 +0000 (21:21 +0200)]
examples/ethtool: fix two typos

This patch fixes 2 typos in examples/ethtool:

There is no such thing as ethtool_ops::get_driverinfo
It should be get_drvinfo:
see include/linux/ethtool.h in the kernel tree.

rte_net_change_mtu should be ndo_change_mtu:
see include/linux/netdevice.h in the kernel tree.

Fixes: bda68ab9d1e7 ("examples/ethtool: add user-space ethtool sample application")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Acked-by: Remy Horton <remy.horton@intel.com>
5 years agodoc: remove reference to rte.doc.mk in programmers guide
Rami Rosen [Mon, 25 Feb 2019 15:37:03 +0000 (17:37 +0200)]
doc: remove reference to rte.doc.mk in programmers guide

This patch removes the reference to rte.doc.mk in
DPDK programmers guide.

Fixes: ee801f6cc7b8 ("mk: clean dead doc rules")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agodevtools: fix index generation
Ferruh Yigit [Mon, 4 Mar 2019 11:04:19 +0000 (11:04 +0000)]
devtools: fix index generation

build-tags.sh is broken because of removed 'test' folder, this breaks
helper make targets like 'make cscope', 'make tags', etc...

Fixing it by removing 'test' from source directories list.

Fixes: a9de470cc7c0 ("test: move to app directory")

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
5 years agodevtools: fix result of svg include check
Michael Santana [Mon, 4 Mar 2019 19:07:16 +0000 (14:07 -0500)]
devtools: fix result of svg include check

Fix trivial bug. In sh shell, 'foo = 1' is not the same as
'foo=1'. Using 'foo = 1' makes the shell attempt to interpret foo
as a command, rather than a simple variable assignment.

Fixes: dafc04c15174 ("devtools: fix return of forbidden addition checks")
Cc: stable@dpdk.org
Signed-off-by: Michael Santana <msantana@redhat.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
5 years agodevtools: enable codespell in checkpatch
Michael Santana [Mon, 4 Mar 2019 19:07:15 +0000 (14:07 -0500)]
devtools: enable codespell in checkpatch

Enable codespell by default.
codespell is a feature by checkpatch.pl that
checks for common spelling mistakes in patches.

This feature is disabled by default. To enable it one must add
the '--codespell' flag to the $options variable in
checkpatches.sh. With this change codespell is enabled by default.
The user can decide to turn off codespell from a one of the config
files read by checkpatches.sh.

Signed-off-by: Michael Santana <msantana@redhat.com>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agodrivers/net: fix shifting 32-bit signed variable 31 times
Andrius Sirvys [Tue, 19 Feb 2019 13:24:50 +0000 (13:24 +0000)]
drivers/net: fix shifting 32-bit signed variable 31 times

Shifting signed 32-bit values by 31-bits has the potential for
unexpected  outcomes as compiler can overwrite a bit.
Specified that values are unsigned.

Errors are observed from running cppcheck.

Bugzilla ID: 58
Fixes: 69e209be5464 ("net/axgbe: add register map and related macros")
Fixes: b5bf7719221d ("bnx2x: driver support routines")
Fixes: ed2ced6fe927 ("net/bnxt: check initialization before accessing stats")
Fixes: 6fda3f0ddda9 ("net/cxgbe: add API to program hardware MPS table")
Fixes: bdb244b96920 ("e1000: whitespace changes")
Fixes: 5a32a257f957 ("e1000: more NICs in base driver")
Fixes: 2fe669f4bcd2 ("net/nfp: support MAC address change")
Fixes: defb9a5dd156 ("nfp: introduce driver initialization")
Fixes: ec94dbc57362 ("qede: add base driver")
Fixes: d2e7d931d0ad ("net/qede/base: formatting changes")
Fixes: cdc07e83bb24 ("net/tap: add eBPF program file")
Cc: stable@dpdk.org
Signed-off-by: Andrius Sirvys <andrius.sirvys@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/softnic: fix possible buffer overflow
Pallantla Poornima [Mon, 4 Feb 2019 07:23:48 +0000 (07:23 +0000)]
net/softnic: fix possible buffer overflow

sprintf function is not secure as it doesn't check the length of string.
More secure function snprintf is used.

Fixes: daabf2fb949b ("net/softnic: map flow action to table action")
Cc: stable@dpdk.org
Signed-off-by: Pallantla Poornima <pallantlax.poornima@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/ixgbe: support loopback for X540/X550
Julien Meunier [Wed, 20 Feb 2019 21:05:30 +0000 (23:05 +0200)]
net/ixgbe: support loopback for X540/X550

Loopback mode is also supported on X540 and X550 NICs, according to
their datasheet (section 15.2). The way to set it up is a little
different of the 82599.

Signed-off-by: Julien Meunier <julien.meunier@nokia.com>
Acked-by: Wei Zhao <wei.zhao1@intel.com>
5 years agonet/ixgbe: do not start on unsupported loopback mode
Julien Meunier [Wed, 20 Feb 2019 21:05:29 +0000 (23:05 +0200)]
net/ixgbe: do not start on unsupported loopback mode

Only TX->RX loopback is supported currently on 82599EB. If a user wants
to apply an another loopback configuration (!= IXGBE_LPBK_82599_TX_RX),
ixgbe PMD ignores it and continues the configuration without raising
any error.

Let's increase robustness of this part by checking if the requested
loopback mode is correct for the current device, before starting it.
If it is not valid, PMD will refuse to start.

Signed-off-by: Julien Meunier <julien.meunier@nokia.com>
Acked-by: Wei Zhao <wei.zhao1@intel.com>
5 years agonet/sfc: do not use PMD log type
Stephen Hemminger [Wed, 27 Feb 2019 16:08:03 +0000 (16:08 +0000)]
net/sfc: do not use PMD log type

The sfc driver was still using RTE_LOGTYPE_PMD which was superseded
by local logging.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
5 years agoeal: remove legacy PMD log macro
Stephen Hemminger [Tue, 26 Feb 2019 21:34:23 +0000 (13:34 -0800)]
eal: remove legacy PMD log macro

The RTE_PMD_DEBUG_TRACE was only enabled for EVENTDEV_DEBUG and
that configuration is now handled by RTE_EDEV_LOG macros.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoeventdev: use same log macro for all unsupported calls
Stephen Hemminger [Tue, 26 Feb 2019 21:34:22 +0000 (13:34 -0800)]
eventdev: use same log macro for all unsupported calls

The driver already has RTE_EDEV_XXX log macros so use
them in two more places.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agocrypto/virtio: use local log type
Stephen Hemminger [Tue, 26 Feb 2019 21:34:21 +0000 (13:34 -0800)]
crypto/virtio: use local log type

The virtio crypto driver was using PMD log type and it should
be using the local log type.

Fixes: 25500d4b8076 ("crypto/virtio: support device init")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoeal: drop unused macros for primary process check
Stephen Hemminger [Tue, 26 Feb 2019 21:34:20 +0000 (13:34 -0800)]
eal: drop unused macros for primary process check

No usage in current DPDK code base.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoapp/pdump: remove only created vdevs
Gage Eads [Tue, 26 Feb 2019 16:04:37 +0000 (10:04 -0600)]
app/pdump: remove only created vdevs

This commit fixes a bug in which a unidirectional pdump could attempt to
remove devices it didn't create.

Fixes: 35cb223ab7be ("app/pdump: fix vdev cleanup")
Cc: stable@dpdk.org
Signed-off-by: Gage Eads <gage.eads@intel.com>
Acked-by: Reshma Pattan <reshma.pattan@intel.com>
5 years agovhost: remove vhost-net requirements from generic APIs
Darek Stojaczyk [Sun, 24 Feb 2019 23:35:55 +0000 (00:35 +0100)]
vhost: remove vhost-net requirements from generic APIs

The rte_vhost API to put data into virtqueues operates
on mbufs and hence it is strictly vhost-net specific.
External backends need to implement virtqueue handling
from scratch and that's just not possible without APIs
to get/set vring base addresses.

Those relevant APIs are there, but they have a check that
prevents them from working with any non-vhost-net device.
This patch removes those checks.

rte_vhost_get_log_base() is not necessarily needed for
external backends, as other, higher level vhost APIs for
live migration are available and could be used instead.
We remove the extra check from it anyway for consistency.

Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
5 years agodoc: improve vhost zero copy guide
Tiwei Bie [Fri, 22 Feb 2019 02:42:09 +0000 (10:42 +0800)]
doc: improve vhost zero copy guide

Highlight that vhost zero copy mbufs should be consumed
as soon as possible.

Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: fix potential use-after-free for memory region
Tiwei Bie [Fri, 22 Feb 2019 02:42:08 +0000 (10:42 +0800)]
vhost: fix potential use-after-free for memory region

Reclaim outstanding zmbufs first before freeing memory regions,
otherwise there could be use-after-free.

Fixes: b0a985d1f340 ("vhost: add dequeue zero copy")
Cc: stable@dpdk.org
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: fix potential use-after-free for zero copy mbuf
Tiwei Bie [Fri, 22 Feb 2019 02:42:07 +0000 (10:42 +0800)]
vhost: fix potential use-after-free for zero copy mbuf

Don't free the zero copy mbufs before they have been consumed,
otherwise there could be use-after-free.

Fixes: b0a985d1f340 ("vhost: add dequeue zero copy")
Cc: stable@dpdk.org
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: restore mbuf first when freeing zmbuf
Tiwei Bie [Fri, 22 Feb 2019 02:42:06 +0000 (10:42 +0800)]
vhost: restore mbuf first when freeing zmbuf

The mbufs should also be restored in free_zmbufs().

Fixes: b0a985d1f340 ("vhost: add dequeue zero copy")
Fixes: 3ebd930588b7 ("vhost: fix mbuf free")
Cc: stable@dpdk.org
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agonet/dpaa2: support generic flow
Sunil Kumar Kori [Fri, 22 Feb 2019 11:16:05 +0000 (11:16 +0000)]
net/dpaa2: support generic flow

Signed-off-by: Sunil Kumar Kori <sunil.kori@nxp.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
5 years agonet/dpaa2: support low level loopback tester
Hemant Agrawal [Fri, 22 Feb 2019 11:16:04 +0000 (11:16 +0000)]
net/dpaa2: support low level loopback tester

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
5 years agonet/dpaa2: support 16 Rx queues per traffic class
Ashish Jain [Fri, 22 Feb 2019 11:16:02 +0000 (11:16 +0000)]
net/dpaa2: support 16 Rx queues per traffic class

Adding support for 16 queues per TC per DPNI port
which is required for LX2 platform.

Signed-off-by: Ashish Jain <ashish.jain@nxp.com>
5 years agobus/fslmc: add enqueue response read in qbman
Nipun Gupta [Fri, 22 Feb 2019 11:16:01 +0000 (11:16 +0000)]
bus/fslmc: add enqueue response read in qbman

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
5 years agomempool/dpaa2: fix continuous print on empty pool
Hemant Agrawal [Fri, 22 Feb 2019 11:15:59 +0000 (11:15 +0000)]
mempool/dpaa2: fix continuous print on empty pool

Changing the print to DP_DEBUG to avoid continuous prints when
buffer pools runs out of buffers

Fixes: 3646ccf0b036 ("mempool/dpaa2: support dynamic logging")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
5 years agonet/dpaa2: support VLAN TPID config
Hemant Agrawal [Fri, 22 Feb 2019 11:15:58 +0000 (11:15 +0000)]
net/dpaa2: support VLAN TPID config

This patch add support to config custom tpid in dpni.
i.e. value other than 0x8100 and 0x88A8

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
5 years agonet/tap: fix multi-process request
Raslan Darawsheh [Tue, 26 Feb 2019 09:50:56 +0000 (09:50 +0000)]
net/tap: fix multi-process request

The structure was not initialized.

Fixes: c9aa56edec8e ("net/tap: access primary process queues from secondary")
Cc: stable@dpdk.org
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
5 years agoapp/compress-perf: call generic strlcpy
Thomas Monjalon [Sun, 24 Feb 2019 22:42:01 +0000 (23:42 +0100)]
app/compress-perf: call generic strlcpy

The call to strlcpy uses either libc, libbsd or internal rte_strlcpy.
No need to call the DPDK flavor explicitly.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rami Rosen <ramirose@gmail.com>
5 years agonet/bnx2x: fix segfaults due to stale interrupt status
Shahed Shaikh [Thu, 21 Feb 2019 19:24:44 +0000 (11:24 -0800)]
net/bnx2x: fix segfaults due to stale interrupt status

Previous ungraceful exit may leave behind un-acked stale
interrupts for slowpath and fastpath.

Interrupt status polling function is started before FLR is
initiated, so we don't have a real way to protect this polling
function invoking an interrupt handler caused due to stale
interrupt status from previous ungraceful exit.

So, check uninitialized status block variables in interrupt
handling path which may lead to segfault.

Fixes: 540a211084a7 ("bnx2x: driver core")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shshaikh@marvell.com>
5 years agonet/mlx: prefix private structure
Thomas Monjalon [Thu, 21 Feb 2019 09:29:14 +0000 (10:29 +0100)]
net/mlx: prefix private structure

The private structure stored in rte_eth_dev->data->dev_private
was named "struct priv".
In order to ease code browsing, the structure is renamed
"struct mlx[45]_priv".

Cc: stable@dpdk.org
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
5 years agonet/iavf: rename remaining avf strings
Leyi Rong [Mon, 25 Feb 2019 17:18:52 +0000 (01:18 +0800)]
net/iavf: rename remaining avf strings

This is the main patch which renames the macros, functions,
structs and any remaining strings in the iavf code.

Signed-off-by: Leyi Rong <leyi.rong@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/iavf: rename avf to iavf
Leyi Rong [Mon, 25 Feb 2019 17:18:51 +0000 (01:18 +0800)]
net/iavf: rename avf to iavf

Rename Intel Ethernet Adaptive Virtual Function driver avf to iavf.

This is the first patch which will only renames the directory name,
lib name, filenames and updates the new name in makefile and meson
files. Also updates the #include files in source files.

Signed-off-by: Leyi Rong <leyi.rong@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agodevtools: add verbose option to meson build test
Bruce Richardson [Tue, 26 Feb 2019 14:15:33 +0000 (14:15 +0000)]
devtools: add verbose option to meson build test

When running ninja, the commands are, by default, always printed on top of
each other. For those who want more detail in the output, two levels of
verbose output has been added to the test-meson-builds script. When "-v" is
passed, or the "TEST_MESON_BUILD_VERBOSE" flag is set in the environment,
then the output of ninja is passed through "cat" to prevent each line
overwriting the next. If "-vv" is passed, or
"TEST_MESON_BUILD_VERY_VERBOSE" is set in the environment, then ninja is
called with the "-v" flag to print out each command in full as it is
executing.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agobuild: support Cortex a55 and a76 platforms
Phil Yang [Thu, 31 Jan 2019 10:11:52 +0000 (18:11 +0800)]
build: support Cortex a55 and a76 platforms

Enable native/cross compiling for Cortex-A55 and Cortex-A76 on meson.

Signed-off-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ola Liljedahl <ola.liljedahl@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
5 years agobuild: improve pcap dependency handling
Luca Boccassi [Tue, 26 Feb 2019 17:46:37 +0000 (17:46 +0000)]
build: improve pcap dependency handling

pcap has historically shipped a custom pcap-config binary tool which
does the job of pkg-config. It was never compatible with cross
compilation.
Meson uses it when using dependency(), which then means cross
compilation fails.
Set pcap-config to empty in the meson cross compilation files so
that Meson will not use it, and add a fallback in case
dependency() fails.
libpcap 1.9.0 finally ships a pkg-config file so everything will
work out of the box in the future.

Signed-off-by: Luca Boccassi <luca.boccassi@microsoft.com>
5 years agomaintainers: resign from keepalive, metrics and ethtool
Remy Horton [Fri, 1 Mar 2019 07:15:17 +0000 (07:15 +0000)]
maintainers: resign from keepalive, metrics and ethtool

After today I will no longer be with Intel, so I am
removing my name from all related maintainer roles.

Signed-off-by: Remy Horton <remy.horton@intel.com>
5 years agoexamples/vhost_scsi: disable build if missing dependency
Aaron Conole [Thu, 7 Feb 2019 22:01:13 +0000 (17:01 -0500)]
examples/vhost_scsi: disable build if missing dependency

The vhost_scsi example code is set to build, even if the requisite header
file virtio_scsi.h isn't available.  This happens on some Ubuntu systems
when some versions of the libc-dev package aren't available.

Check whether the virtio_scsi.h file exists, and if not, set the build
flag to false.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
5 years agodoc: add meson build to contributing guide
Vipin Varghese [Wed, 13 Feb 2019 05:41:06 +0000 (11:11 +0530)]
doc: add meson build to contributing guide

Patches has to be validated for meson devtool script for
code and document changes. Updating documentation for meson
build steps in checking Compilation category.

Signed-off-by: Vipin Varghese <vipin.varghese@intel.com>
Tested-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>