dpdk.git
5 years agodoc: update QoS scheduler guides
Jasvinder Singh [Tue, 26 Nov 2019 14:28:29 +0000 (14:28 +0000)]
doc: update QoS scheduler guides

Updates documentation to reflect the changes in the QoS scheduler
library and example.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
5 years agocrypto/ccp: fix digest size capabilities
Amaranath Somalapuram [Fri, 22 Nov 2019 06:47:47 +0000 (12:17 +0530)]
crypto/ccp: fix digest size capabilities

CCP can support varied digest sizes ranging from 1 to some max value
But the current code support only fixed max values.
This patch updates the minimum digest sizes to 1

Fixes: e0d88a394e ("crypto/ccp: support run-time CPU based auth")
Cc: stable@dpdk.org
Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
5 years agocrypto/ccp: fix scheduling of burst
Amaranath Somalapuram [Fri, 22 Nov 2019 06:47:17 +0000 (12:17 +0530)]
crypto/ccp: fix scheduling of burst

CCP driver was scheduling only one CCP in a single burst(enqueue).
Effective throughput was limited to 1 CCP performance.
Scheduling multiple ccp within one burst will increase the ccp performance.
this changes will divide the enqueue packets equally among the multiple CCP

Fixes: e0d88a394e ("crypto/ccp: support run-time CPU based auth")
Cc: stable@dpdk.org
Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
5 years agocrypto/ccp: fix CPU authentication crash
Amaranath Somalapuram [Fri, 22 Nov 2019 06:46:47 +0000 (12:16 +0530)]
crypto/ccp: fix CPU authentication crash

when ccp_auth_opt=1 is set and if authentication error occurred,
ccp driver crash. The enqueue referance count nb_ops miss match
with dequeue nb_ops on authentication error.

Fixes: e0d88a394e ("crypto/ccp: support run-time CPU based auth")
Cc: stable@dpdk.org
Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
5 years agocrypto/ccp: fix maximum queues and burst size
Amaranath Somalapuram [Fri, 22 Nov 2019 06:46:15 +0000 (12:16 +0530)]
crypto/ccp: fix maximum queues and burst size

CCP driver crash when running l2fwd
issue happen when changing MAX_PKT_BURST 32 to 256
increasing the CCP qp to 8 and CCP_MAX_BURST to 256

Fixes: ef4b04f87f ("crypto/ccp: support device init")
Cc: stable@dpdk.org
Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
5 years agodoc: fix l2fwd-crypto usage in CCP guide
Amaranath Somalapuram [Fri, 22 Nov 2019 06:45:45 +0000 (12:15 +0530)]
doc: fix l2fwd-crypto usage in CCP guide

Update the CCP doc for fixing description of some parameter,
Changing AES_CBC to aes-cbc, SHA1_HMAC to sha1-hmac.

Fixes: 4433ced9aa ("doc: add AMD CCP guide")
Cc: stable@dpdk.org
Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
5 years agoapp/crypto-perf: fix input of AEAD decrypt
Archana Muniganti [Thu, 21 Nov 2019 11:14:27 +0000 (16:44 +0530)]
app/crypto-perf: fix input of AEAD decrypt

In AEAD decrypt (verify mode), test data should point to
cipher text instead of plain text

Fixes: 5b2b0a740fba ("app/crypto-perf: overwrite mbuf when verifying")
Cc: stable@dpdk.org
Signed-off-by: Archana Muniganti <marchana@marvell.com>
Acked-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
5 years agotest/event_crypto: fix check for HW support
Hemant Agrawal [Thu, 7 Nov 2019 09:01:21 +0000 (14:31 +0530)]
test/event_crypto: fix check for HW support

When HW is available but the mode mismatches, it is better
to skip the test case.

Fixes: aa026482baf9 ("test/event_crypto: no service core when HW support available")

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
5 years agotest/event_crypto: fix to avail mempool entries
Hemant Agrawal [Thu, 7 Nov 2019 09:01:20 +0000 (14:31 +0530)]
test/event_crypto: fix to avail mempool entries

Added the missed code to avail the mempool entries before
pool free.

Fixes: 24054e3640a2 ("test/crypto: use separate session mempools")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
5 years agotest/event_crypto: fix missing IV for AES
Hemant Agrawal [Thu, 7 Nov 2019 09:01:19 +0000 (14:31 +0530)]
test/event_crypto: fix missing IV for AES

The IV was not set, which was causing HW based SEC on DPAA1
to fail.

Fixes: b2196237eb8a ("test/event_crypto: change cipher algo")

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
5 years agoapp/eventdev: check function errors
Pavan Nikhilesh [Thu, 21 Nov 2019 19:22:39 +0000 (00:52 +0530)]
app/eventdev: check function errors

Fix unchecked return values reported by coverity.

Coverity Issue: 336861
Coverity Issue: 349906
Fixes: 032a965a8f1d ("app/eventdev: support Tx adapter")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agoapp/eventdev: fix divide by zero
Pavan Nikhilesh [Thu, 21 Nov 2019 19:22:38 +0000 (00:52 +0530)]
app/eventdev: fix divide by zero

Fix possible divide by zero condition when calculating percentages.

Coverity issue: 277205
Coverity issue: 277234
Fixes: d008f20bce23 ("app/eventdev: add event timer adapter as a producer")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agodoc: fix command line in l2fwd-event guide
Pavan Nikhilesh [Thu, 21 Nov 2019 18:27:16 +0000 (23:57 +0530)]
doc: fix command line in l2fwd-event guide

Fix incorrect command line argument documented for l2fwd-event.

Bugzilla ID: 367
Bugzilla ID: 368
Fixes: 14d7ea259aeb ("doc: add application usage guide for l2fwd-event")

Reported-by: Vipin Varghese <vipin.varghese@intel.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoexamples/l2fwd-event: check function errors
Pavan Nikhilesh [Thu, 21 Nov 2019 18:27:15 +0000 (23:57 +0530)]
examples/l2fwd-event: check function errors

Validate `rte_eth_link_get_nowait`, `rte_service_map_lcore_set` and
`rte_eth_dev_info_get` return values.

Coverity issue: 350600
Coverity issue: 350601
Coverity issue: 350602

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoevent/octeontx2: update start timestamp periodically
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:30 +0000 (21:14 +0530)]
event/octeontx2: update start timestamp periodically

Update start timestamp periodically to prevent drift.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoevent/octeontx2: update SSO buffers based on timer count
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:29 +0000 (21:14 +0530)]
event/octeontx2: update SSO buffers based on timer count

Update SSO internal XAQ buffers based on number of timers in event timer
adapter.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoevent/octeontx2: improve chunk pool performance
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:28 +0000 (21:14 +0530)]
event/octeontx2: improve chunk pool performance

Enable mempool cache for internal mempool to improve alloc performance.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoevent/octeontx2: use opposite bucket for current chunk
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:27 +0000 (21:14 +0530)]
event/octeontx2: use opposite bucket for current chunk

Since TIM buckets are always aligned to 32B and our cache line size being
128B, we will always have a cache miss when reading current_chunk pointer.
Avoid the cache miss by storing the current_chunk pointer in the bucket
opposite to the current bucket.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoevent/octeontx2: fix HW timer race condition
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:26 +0000 (21:14 +0530)]
event/octeontx2: fix HW timer race condition

Fix HW race condition observed when timeout resolution is low (<5us).
When HW traverses a given TIM bucket it will clear chunk_remainder,
but since SW always decreases the chunk_remainder at the start of the
arm routine it might cause a race where SW updates chunk_remainder
after HW has cleared it that lead to nasty side effects.

Fixes: 95e4e4ec7469 ("event/octeontx2: add timer arm timeout burst")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agonet/octeontx: set queue offload capabilities
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:07 +0000 (09:18 +0530)]
net/octeontx: set queue offload capabilities

Make Rx queue specific offloads same as device Rx offloads.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agoevent/octeontx: add appication domain validation
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:06 +0000 (09:18 +0530)]
event/octeontx: add appication domain validation

Add applicaton domain validation for OCTEON TX TIM vfs aka Event timer.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agomempool/octeontx: add application domain validation
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:05 +0000 (09:18 +0530)]
mempool/octeontx: add application domain validation

Add application domain validation for OCTEON TX FPA vfs.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agonet/octeontx: cleanup redundant mbox structs
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:04 +0000 (09:18 +0530)]
net/octeontx: cleanup redundant mbox structs

Cleanup redudant mail box structures.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agonet/octeontx: add application domain validation
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:03 +0000 (09:18 +0530)]
net/octeontx: add application domain validation

Add domain validation for PKI and PKO vfs

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agocommon/octeontx: update mbox to version 1.1.3
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:02 +0000 (09:18 +0530)]
common/octeontx: update mbox to version 1.1.3

Sync mail box data structures to version 1.1.3.
Add mail box version verification and defer initializing octeontx
devices if mail box version mismatches.
Update OCTEON TX limitaion with max mempool size used.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
5 years agobuildtools: fix build with coverage
David Marchand [Mon, 25 Nov 2019 08:10:07 +0000 (09:10 +0100)]
buildtools: fix build with coverage

A compiler can reuse a variable name and prefix it when instrumenting
with coverage.

Example:
$ make defconfig T=x86_64-native-linux-gcc O=master
$ make EXTRA_CFLAGS='--coverage' O=master
[...]
    CC rte_flow.o
 rte_flow_dynf_metadata_offs is not flagged as experimental but is listed
 in version map
 Please add __rte_experimental to the definition of
 rte_flow_dynf_metadata_offs

$ objdump -t master/build/lib/librte_ethdev/rte_flow.o |grep _offs$
0000000000000000 l     F .text.startup 000000000000000a
  _GLOBAL__sub_I_65535_0_rte_flow_dynf_metadata_offs
0000000000000620 g     O .data 0000000000000004
  rte_flow_dynf_metadata_offs

Protect against this by adding a space character in the pattern.

Fixes: a4bcd61de82d ("buildtools: add script to check experimental API exports")
Cc: stable@dpdk.org
Reported-by: Andrew Rybchenko <arybchenko@solarflare.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Tested-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
5 years agobuildtools: fix experimental symbols listing
David Marchand [Mon, 25 Nov 2019 08:10:06 +0000 (09:10 +0100)]
buildtools: fix experimental symbols listing

The map-list-symbol.sh script displays the filename, section and symbol
names of map files.

Example:
$ buildtools/map-list-symbol.sh -S EXPERIMENTAL \
    lib/librte_ethdev/rte_ethdev_version.map |grep rte_mtr_create
lib/librte_ethdev/rte_ethdev_version.map EXPERIMENTAL rte_mtr_create

The experimental symbol check should only consider the symbol name.

Fixes: 3290ac14eb94 ("buildtools: detect discrepancies for experimental symbols")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Tested-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
5 years agomem: clarify documentation of virt2iova behaviour
Anatoly Burakov [Thu, 14 Nov 2019 14:13:06 +0000 (14:13 +0000)]
mem: clarify documentation of virt2iova behaviour

It may not be immediately clear that rte_mem_virt2iova does not actually
check the internal memseg table, and will instead either return VA (in
IOVA as VA mode), or will fall back to kernel page table walk (in IOVA
as PA mode).

Add a note to API documentation indicating the above.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
5 years agopower: fix error log on guest message polling
David Hunt [Mon, 25 Nov 2019 15:00:42 +0000 (15:00 +0000)]
power: fix error log on guest message polling

Should be passing errno rather than ret, which could be negative.

Coverity issue: 350362
Fixes: 9dc843eb273b ("power: extend guest channel API for reading")
Cc: stable@dpdk.org
Signed-off-by: David Hunt <david.hunt@intel.com>
5 years agoremove blank lines at end of file
Stephen Hemminger [Mon, 11 Nov 2019 19:50:15 +0000 (11:50 -0800)]
remove blank lines at end of file

Remove trailing blank lines. They serve no purpose and are just
editor leftovers.
These can cause git to complain about whitespace errors during merges.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
5 years agoexamples/multi_process: fix client crash with sparse ports
Stephen Hemminger [Mon, 5 Aug 2019 16:38:17 +0000 (09:38 -0700)]
examples/multi_process: fix client crash with sparse ports

The mp_client crashes if run on Azure or any system where ethdev
ports are owned. In that case, the tx_buffer and tx_stats for the
real port were initialized correctly, but the wrong port was used.

For example if the server has Ports 3 and 5. Then calling
rte_eth_tx_buffer_flush on any other buffer will dereference null
because the tx buffer for that port was not allocated.

Also:
   - the flush code is common enough that it should not be marked
     unlikely
   - combine conditions to reduce indentation
   - avoid unnecessary if() if sent is zero.

Fixes: e2366e74e029 ("examples: use buffered Tx")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Acked-by: Matan Azrad <matan@mellanox.com>
5 years agoexamples/multi_process: check server port validity
Stephen Hemminger [Mon, 5 Aug 2019 16:38:16 +0000 (09:38 -0700)]
examples/multi_process: check server port validity

The mp_server incorrectly allows a port mask that included hidden
ports and which later caused either lost packets or failed initialization.

This fixes explicitly checking that each bit in portmask is a
valid port before using it.

Fixes: 5b7ba31148a8 ("ethdev: add port ownership")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Acked-by: Matan Azrad <matan@mellanox.com>
5 years agotest/rcu: fix build for small number of cores
Gavin Hu [Mon, 25 Nov 2019 03:36:33 +0000 (11:36 +0800)]
test/rcu: fix build for small number of cores

If the RTE_MAX_LCORE is less than 10, a compilation error is generated:
app/test/test_rcu_qsbr.c:234:10: error: comparison of integer
expressions of different signedness: ‘unsigned int’ and ‘int’
[-Werror=sign-compare]

The cause is (RTE_MAX_LCORE - 10) results in a negative value.

To fix, use rte_rand() to find a number between 0 and RTE_MAX_LCORE.

Fixes: b87089b0bb19 ("test/rcu: add API and functional tests")
Cc: stable@dpdk.org
Signed-off-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Steve Capper <steve.capper@arm.com>
Reviewed-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
5 years agomk: remove library search path from binary
Ferruh Yigit [Fri, 22 Nov 2019 11:30:23 +0000 (11:30 +0000)]
mk: remove library search path from binary

This patch functionally reverts the patch in fixes line to not have any
hardcoded library path in the final binary for the security reasons, in
case this binary distributed to production environment.

RPATH only added in RTE_DEVEL_BUILD case and this binary shouldn't
distributed, but still removing it to be cautious.

Fixes: 8919f73bcbaa ("mk: add build directory to library search path")
Cc: stable@dpdk.org
Suggested-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agombuf: extend mbuf pool private structure
Shahaf Shuler [Mon, 25 Nov 2019 10:21:32 +0000 (10:21 +0000)]
mbuf: extend mbuf pool private structure

With the API and ABI freeze ahead, it will be good to reserve
some bits on the private structure for future use.

Otherwise we will potentially need to maintain two different
private structure during 2020 period.

There is already one use case for those reserved bits[1]

The reserved field should be set to 0 by the user.

[1] https://patches.dpdk.org/patch/63077/

Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
5 years agobuild: fix Windows build
Bruce Richardson [Sat, 23 Nov 2019 02:59:59 +0000 (18:59 -0800)]
build: fix Windows build

While most windows apps can handle both "\" and "/" as path separators,
"more" is treating the "/" as the start of a command-line flag in this
case, causing errors.

Fixes: cba806e07d6f ("build: change ABI versioning to global")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Ranjit Menon <ranjit.menon@intel.com>
Tested-by: Pallavi Kadam <pallavi.kadam@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
5 years agodevtools: disable automatic probing in null testing
Thomas Monjalon [Fri, 22 Nov 2019 13:38:28 +0000 (14:38 +0100)]
devtools: disable automatic probing in null testing

The script test-null.sh is supposed to do a quick and simple
run of testpmd with null PMD only, for sanity check.
As it is not supposed to test probing of any other PMD,
physical device probing is switched to whitelist mode
by using a fake PCI address (0:0.0).
It will also help to keep memory usage stable across platforms.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: David Marchand <david.marchand@redhat.com>
5 years agoversion: 19.11-rc3
Thomas Monjalon [Wed, 20 Nov 2019 23:56:30 +0000 (00:56 +0100)]
version: 19.11-rc3

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
5 years agopower: handle frequency increase with turbo disabled
Mattias Rönnblom [Thu, 14 Nov 2019 14:10:36 +0000 (15:10 +0100)]
power: handle frequency increase with turbo disabled

Calling pstate's or acpi's rte_power_freq_up() when on the highest
non-turbo frequency results in an error, if turbo is enabled in the BIOS,
but disabled via the power library.
The error is in the form of a return code and a RTE_LOG() entry
on the ERR level.

According to the API documentation, the frequency is scaled up
"according to the available frequencies". In case turbo is disabled,
that frequency is not available. This patch's rte_power_freq_up()
behaviour is also consistent with how rte_power_freq_max() is
implemented (i.e. the highest non-turbo frequency is set, in case
turbo is disabled).

Fixes: 445c6528b55f ("power: common interface for guest and host")
Fixes: e6c6dc0f96c8 ("power: add p-state driver compatibility")
Cc: stable@dpdk.org
Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
Reviewed-by: Liang Ma <liang.j.ma@intel.com>
5 years agoapp/testpmd: fix invalid port detaching
Matan Azrad [Tue, 12 Nov 2019 08:47:39 +0000 (08:47 +0000)]
app/testpmd: fix invalid port detaching

The port was not validated before detaching.

Ignore port detach operation when the port is not valid.

Fixes: f8e5baa2662d ("app/testpmd: check not detaching device twice")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
5 years agobus/pci: clear driver on detach
Matan Azrad [Tue, 12 Nov 2019 08:47:38 +0000 (08:47 +0000)]
bus/pci: clear driver on detach

When probing, a bus is responsible for filling the driver field to
indicate the device is bound.
Let's clear this field when detaching to keep a consistent behavior.

This is not a fix per se, since the device is freed when detaching.
But at least clearing the field has been added to remind that the
driver field has a meaning for the EAL.

Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agokni: increase kernel version requirement for VA
Ferruh Yigit [Wed, 20 Nov 2019 17:22:26 +0000 (17:22 +0000)]
kni: increase kernel version requirement for VA

A build error reported related to the selected 'get_user_pages_remote()'
kernel API:

.../kernel/linux/kni/kni_dev.h:113:8:
  error: too few arguments to function ‘get_user_pages_remote’
  ret = get_user_pages_remote(tsk, tsk->mm, iova, 1
        ^~~~~~~~~~~~~~~~~~~~~

Currently there are three versions of the 'get_user_pages_remote()'
supported, based on kernel version < 4.9, = 4.9, > 4.9.

These version based checks are not working fine with the distro kernels
which is the cause of reported build error. The error reported by the
kernel version 4.8, but it is using API defined in > 4.9.

To be able to take control of this, and possible more, related build
error, increasing the minimum supported kernel version for iova=va with
KNI to kernel version 4.9.

This leaves us with single version of the kernel API and more manageable.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
5 years agobpf/arm: fix clang build
Ruifeng Wang [Fri, 15 Nov 2019 06:47:54 +0000 (14:47 +0800)]
bpf/arm: fix clang build

Clang has different prototype for __builtin___clear_cache().
It requires 'char *' parameters while gcc requires 'void *'.

Clang version 8.0 was used.
Warning messages during build:
../lib/librte_bpf/bpf_jit_arm64.c:1438:26: warning: incompatible pointer
types passing 'uint32_t *' (aka 'unsigned int *') to parameter of type
'char *' [-Wincompatible-pointer-types]
        __builtin___clear_cache(ctx.ins, ctx.ins + ctx.idx);
                                ^~~~~~~
../lib/librte_bpf/bpf_jit_arm64.c:1438:35: warning: incompatible pointer
types passing 'uint32_t *' (aka 'unsigned int *') to parameter of type
'char *' [-Wincompatible-pointer-types]
        __builtin___clear_cache(ctx.ins, ctx.ins + ctx.idx);
                                         ^~~~~~~~~~~~~~~~~

Fixes: f3e516772464 ("bpf/arm: add prologue and epilogue")
Cc: jerinj@marvell.com
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agoconfig: add eMAG
Jerry Hao OS [Mon, 14 Oct 2019 09:34:18 +0000 (17:34 +0800)]
config: add eMAG

Add the make and meson based build infrastructure for the eMAG platform
from Ampere Computing corp., which is a 64-bit ARM processor with 32
Armv8 64-bit CPU cores. For more information, refer to:
https://amperecomputing.com/product/

Signed-off-by: Jerry Hao OS <jerryhao@os.amperecomputing.com>
Signed-off-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agoconfig: add cortex-a76 for meson
Gavin Hu [Mon, 11 Nov 2019 05:41:51 +0000 (13:41 +0800)]
config: add cortex-a76 for meson

To make the list complete and consistent, add cortex-a76 configuration.

Signed-off-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agoconfig: add Arm Neoverse N1 SDP
Gavin Hu [Mon, 11 Nov 2019 05:41:50 +0000 (13:41 +0800)]
config: add Arm Neoverse N1 SDP

Arm N1 SDP is an infrastructure segment development platform
based on armv8.2-a Neoverse N1 CPU. For more information, refer to:
https://community.arm.com/developer/tools-software/oss-platforms/w/
docs/440/neoverse-n1-sdp

Signed-off-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Steve Capper <steve.capper@arm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agodevtools: reduce list of Arm builds tested with meson
Thomas Monjalon [Wed, 20 Nov 2019 23:05:55 +0000 (00:05 +0100)]
devtools: reduce list of Arm builds tested with meson

The list of Arm configs is growing:
config/arm/arm64_armada_linux_gcc
config/arm/arm64_armv8_linux_gcc
config/arm/arm64_bluefield_linux_gcc
config/arm/arm64_dpaa_linux_gcc
config/arm/arm64_emag_linux_gcc
config/arm/arm64_n1sdp_linux_gcc
config/arm/arm64_octeontx2_linux_gcc
config/arm/arm64_thunderx2_linux_gcc
config/arm/arm64_thunderx_linux_gcc

In order to keep testing time reasonable,
and also because n1sdp is merged without a related fix in tests,
the list of configs is reduced in the script test-meson-builds.sh.

The list of tested Arm builds becomes:
build-arm64-host-clang (armv8a)
build-arm64-bluefield
build-arm64-dpaa
build-arm64-octeontx2

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: David Marchand <david.marchand@redhat.com>
5 years agobuildtools: add ABI version check script
Marcin Baran [Wed, 20 Nov 2019 17:23:39 +0000 (17:23 +0000)]
buildtools: add ABI version check script

Add a shell script that checks whether built libraries are
versioned with expected ABI (current ABI, current ABI + 1,
or EXPERIMENTAL).

The following command was used to verify current source tree
(assuming build directory is in ./build):

find ./build/lib ./build/drivers -name  \*.so \
-exec ./buildtools/check-abi-version.sh {} \; -print

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agobuild: align symbols with global ABI version
Pawel Modrak [Wed, 20 Nov 2019 17:23:38 +0000 (17:23 +0000)]
build: align symbols with global ABI version

Merge all versions in linker version script files to DPDK_20.0.

This commit was generated by running the following command:

:~/DPDK$ buildtools/update-abi.sh 20.0

Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agocommon/octeontx: add missing public symbol
Anatoly Burakov [Wed, 20 Nov 2019 17:23:37 +0000 (17:23 +0000)]
common/octeontx: add missing public symbol

The logtype symbol was missing from the .map file. Add it.

Fixes: d8dd31652cf4 ("common/octeontx: move mbox to common folder")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agodistributor: rename v2.0 ABI to _single suffix
Marcin Baran [Wed, 20 Nov 2019 17:23:36 +0000 (17:23 +0000)]
distributor: rename v2.0 ABI to _single suffix

The original ABI versioning was slightly misleading in that the
DPDK 2.0 ABI was really a single mode for the distributor, and is
used as such throughout the distributor code.

Fix this by renaming all _v20 API's to _single API's, and remove
symbol versioning.

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agodistributor: remove deprecated code
Marcin Baran [Wed, 20 Nov 2019 17:23:35 +0000 (17:23 +0000)]
distributor: remove deprecated code

Remove code for old ABI versions ahead of ABI version bump.

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agolpm: remove deprecated code
Marcin Baran [Wed, 20 Nov 2019 17:23:34 +0000 (17:23 +0000)]
lpm: remove deprecated code

Remove code for old ABI versions ahead of ABI version bump.

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agotimer: remove deprecated code
Marcin Baran [Wed, 20 Nov 2019 17:23:33 +0000 (17:23 +0000)]
timer: remove deprecated code

Remove code for old ABI versions ahead of ABI version bump.

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Erik Gabriel Carrillo <erik.g.carrillo@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agobuildtools: add ABI version update script
Anatoly Burakov [Wed, 20 Nov 2019 17:23:32 +0000 (17:23 +0000)]
buildtools: add ABI version update script

In order to facilitate mass updating of version files, add a shell
script that recurses into lib/ and drivers/ directories and calls
the ABI version update script.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agobuildtools: add symbols map update script
Pawel Modrak [Wed, 20 Nov 2019 17:23:31 +0000 (17:23 +0000)]
buildtools: add symbols map update script

Add a script that automatically merges all stable ABI's under one
ABI section with the new version, while leaving experimental
section exactly as it is.

Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agobuild: remove individual library versions
Anatoly Burakov [Wed, 20 Nov 2019 17:23:30 +0000 (17:23 +0000)]
build: remove individual library versions

Since the library versioning for both stable and experimental ABI's is
now managed globally, the LIBABIVER and version variables no longer
serve any useful purpose, and can be removed.

The replacement in Makefiles was done using the following regex:

^(#.*\n)?LIBABIVER\s*:=\s*\d+\n(\s*\n)?

(LIBABIVER := numbers, optionally preceded by a comment and optionally
succeeded by an empty line)

The replacement for meson files was done using the following regex:

^(#.*\n)?version\s*=\s*\d+\n(\s*\n)?

(version = numbers, optionally preceded by a comment and optionally
succeeded by an empty line)

[David]: those variables are manually removed for the files:
- drivers/common/qat/Makefile
- lib/librte_eal/meson.build
[David]: the LIBABIVER is restored for the external ethtool example
library.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agobuild: change ABI versioning to global
Marcin Baran [Wed, 20 Nov 2019 17:23:28 +0000 (17:23 +0000)]
build: change ABI versioning to global

As per new ABI policy [1], all of the libraries are now versioned using
one global ABI version. Stable libraries use the MAJOR.MINOR ABI
version for their shared objects, while experimental libraries
use the 0.MAJORMINOR convention for their versioning.
Experimental library versioning is managed globally. Changes in this
patch implement the necessary steps to enable that.

The CONFIG_RTE_MAJOR_ABI option was introduced to permit multiple
DPDK versions installed side by side. The problem is now addressed
through the new ABI policy, and thus can be removed.

[David] For external libraries relying on Makefile, LIBABIVER is
preserved to avoid using DPDK global ABI version.

[1] https://doc.dpdk.org/guides/contributing/abi_policy.html

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agoexamples: hide error for missing pkg-config path flag
Bruce Richardson [Fri, 15 Nov 2019 15:17:00 +0000 (15:17 +0000)]
examples: hide error for missing pkg-config path flag

Some versions of pkg-config don't support the --path flag, which is not a
fatal error when building the apps. Without the flag, the makefile just
cannot track the .pc file of DPDK as a dependency of the build. Therefore,
we can ignore the error and suppress it by redirecting to /dev/null the
stderr from that call to pkg-config.

Fixes: 22119c4591a0 ("examples: use pkg-config in makefiles")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoexamples: fix build with old pkg-config
Bruce Richardson [Fri, 15 Nov 2019 15:16:59 +0000 (15:16 +0000)]
examples: fix build with old pkg-config

Not all versions of pkg-config in distros have support for the
--define-prefix flag [1], causing errors when building examples manually or
with test-meson-builds.sh script [2].

For the former case, we need to remove the hard-coded use of the flag in
the Makefiles.

For the latter case, the flag is necessary for builds to succeed, so we
skip the tests when it's not present, passing it as part of the pkg-config
command if it is supported.

[1]
CentOS Linux release 7.7.1908 (Core)
pkg-config version 0.27.1

[2]
 ## Building cmdline
Unknown option --define-prefix
gmake: Entering directory
`...ild-x86-default/install-root/usr/local/share/dpdk/examples/cmdline'
rm -f build/cmdline build/cmdline-static build/cmdline-shared
test -d build && rmdir -p build || true
Unknown option --define-prefix
Unknown option --define-prefix
gcc -O3  main.c commands.c parse_obj_list.c -o build/cmdline-shared
main.c:14:28: fatal error: cmdline_rdline.h: No such file or directory

Fixes: ca9268529d2b ("examples: support relocated DPDK install")
Fixes: 7f80a2102bbb ("devtools: test pkg-config file")
Cc: stable@dpdk.org
Reported-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/mlx5: fix Rx queue release assertions
Matan Azrad [Wed, 20 Nov 2019 09:21:16 +0000 (09:21 +0000)]
net/mlx5: fix Rx queue release assertions

In debug mode, there is assertion to validate the CQ object before the
release.

Wrongly, the assertion is done for any type of RX queue even if it
doesn't use CQ at all, for example in hairpin Rx queue.

Ignore CQ assertion when hairpin queue is released.

Fixes: e79c9be91515 ("net/mlx5: support Rx hairpin queues")

Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
5 years agonet/hinic: fix secondary process
Xiaoyun Wang [Tue, 19 Nov 2019 12:31:34 +0000 (20:31 +0800)]
net/hinic: fix secondary process

The secondary process does not need to register interrupt handle,
remove rte_intr_callback_register from secondary process branch.

Fixes: 64727024d2fd ("net/hinic: add device initialization")
Cc: stable@dpdk.org
Signed-off-by: Xiaoyun Wang <cloud.wangxiaoyun@huawei.com>
5 years agoapp/testpmd: report invalid command line parameter
David Marchand [Mon, 18 Nov 2019 15:37:14 +0000 (16:37 +0100)]
app/testpmd: report invalid command line parameter

We currently do not check that a non option string has been passed to
testpmd.

Example:
$ ./master/app/testpmd --no-huge -m 512 --vdev net_null0 \
--vdev net_null1 -- -i nb-cores=2 --total-num-mbuf 2048
[...]
testpmd> show config fwd
io packet forwarding - ports=2 - cores=1 - streams=2 - NUMA support
enabled, MP allocation mode: native
Logical Core 1 (socket 0) forwards packets on 2 streams:
  RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
  RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00

Here nb-cores=2 is just ignored, while the (probably sleepy) user did not
notice this.

Validate that all strings passed to testpmd are part of a known option.

After this patch:
$ ./master/app/testpmd --no-huge -m 512 --vdev net_null0 \
--vdev net_null1 -- -i nb-cores=2 --total-num-mbuf 2048
[...]
Invalid parameter: nb-cores=2
EAL: Error - exiting with code: 1
  Cause: Command line incorrect

While at it, when passing an unknown option, print the string that gets
refused by getopt_long to help the user.

Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/mlx5: fix L3 encapsulation flow validation
Matan Azrad [Tue, 19 Nov 2019 15:32:28 +0000 (15:32 +0000)]
net/mlx5: fix L3 encapsulation flow validation

In order to configure L3 encapsulation\decapsulation flow to mlx5
devices, 2 actions should be added to the flow actions list:
RTE_FLOW_ACTION_TYPE_RAW_DECAP and RTE_FLOW_ACTION_TYPE_RAW_ENCAP.

One of the validations for this scenario is to check that modify actions
is not done before the L3 decapsulation, because it doesn't make sense
to decapsulate a modified data.

Wrongly, this check was done for the case of L3 encapsulation what
causes a validation failure in modify + L3 encapsulation flow.

Ignore this check in case of L3 encapsulation.

Fixes: 4bb14c83df95 ("net/mlx5: support modify header using Direct Verbs")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
5 years agonet/bnxt: fix flow check for null spec and mask
Ajit Khaparde [Wed, 20 Nov 2019 03:00:23 +0000 (19:00 -0800)]
net/bnxt: fix flow check for null spec and mask

bnxt_validate_and_parse_flow_type already has protocol specific NULL
checks for rte_flow_item spec and mask and take actions accordingly.
A check at the top of the loop is redundant and is preventing the
protocol specific checks from being executed.

Fixes: b7e01f386059 ("net/bnxt: handle cleanup if flow creation fails")

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agonet/bnxt: fix vector Rx selection
Lance Richardson [Tue, 19 Nov 2019 22:12:15 +0000 (17:12 -0500)]
net/bnxt: fix vector Rx selection

Take DEV_RX_OFFLOAD_RSS_HASH into account when deciding whether vector
mode receive can be enabled.

Fixes: 8b945a7f7dcb ("drivers/net: update Rx RSS hash offload capabilities")

Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
5 years agonet/bnxt: fix jumbo frame configuration in firmware
Rahul Gupta [Fri, 15 Nov 2019 09:40:09 +0000 (15:10 +0530)]
net/bnxt: fix jumbo frame configuration in firmware

In order to prevent reconfiguration of firmware resources for every
MTU change, configure FW with max MTU value using hwrm_func_cfg
to support all frame sizes. There is no need to overwrite the driver
level MTU variable data->mtu with the FW MTU.

Fixes: 905cd45ce30e ("net/bnxt: use configured MTU during load")
Cc: stable@dpdk.org
Signed-off-by: Rahul Gupta <rahul.gupta@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agonet/bnxt: fix resource qcaps with older FW
Ajit Khaparde [Mon, 18 Nov 2019 22:23:45 +0000 (14:23 -0800)]
net/bnxt: fix resource qcaps with older FW

On some old versions of FW, bnxt_hwrm_func_resc_qcaps can return an
error. This is because the command was not implemented completely
in FW till the subsequent version. Ignore the error and continue with
the driver initialization.

Fixes: edafb57ba4a1 ("net/bnxt: fix VF resource allocation")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agonet/ice: fix flow director rule duplication check
Yahui Cao [Tue, 19 Nov 2019 16:01:00 +0000 (00:01 +0800)]
net/ice: fix flow director rule duplication check

When FDIR filter detects duplicated rule and then returns EEXIST, ice
flow will capture this error and return immediately.

Fixes: 4e27d3ed02bd ("net/ice: fix flow API framework")

Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
5 years agonet/mlx5: fix hairpin split detection
Ori Kam [Tue, 19 Nov 2019 12:38:33 +0000 (12:38 +0000)]
net/mlx5: fix hairpin split detection

When creating a flow, the flow is checked if it should be split into
2 flows based on the queue/rss acton.

If the RSS action with given without any queues, it will result in crash
due to the fact that the function checks the queue type.

This commit fixes this issue by checking if the rss action is not empty,
and at least one queue.

Fixes: d85c7b5ea59f ("net/mlx5: split hairpin flows")

Signed-off-by: Ori Kam <orika@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agonet/mlx5: fix assert in Tx inline settings
Viacheslav Ovsiienko [Tue, 19 Nov 2019 13:13:09 +0000 (13:13 +0000)]
net/mlx5: fix assert in Tx inline settings

Assert condition is fixed to not alert for the case
when multi-packet write is not supported/engaged at all.

Fixes: b53cd86965a1 ("net/mlx5: adjust inline setting for large Tx queue sizes")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agonet/mlx5: fix port action for LAG
Viacheslav Ovsiienko [Tue, 19 Nov 2019 09:16:04 +0000 (09:16 +0000)]
net/mlx5: fix port action for LAG

To support LAG configurations the mlx5dv_dr_action_create_dest_ib_port()
should be called instead of mlx5dv_dr_action_create_dest_vport().

Fixes: f07341e7aed3 ("net/mlx5: update source and destination vport translations")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
5 years agoethdev: avoid undefined behaviour on configuration copy
Andrew Rybchenko [Tue, 19 Nov 2019 08:22:50 +0000 (08:22 +0000)]
ethdev: avoid undefined behaviour on configuration copy

memcpy() source and destination areas must not overlap and equal
pointers is the case which is really met, so handle it.

Fixes: 68b931bff287 ("ethdev: eliminate interim variable")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/ice: add flow mark hint
Qi Zhang [Tue, 19 Nov 2019 06:14:42 +0000 (14:14 +0800)]
net/ice: add flow mark hint

Since not all data paths support flow mark, the driver needs
a hint from application to select the correct data path if
flow mark is required. The patch introduces a devarg
"flow-mark-support" as a workaround solution, since a standard
way is still ongoing.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>
5 years agonet/ice: fix flow director rule after device stop
Yahui Cao [Mon, 18 Nov 2019 22:23:43 +0000 (06:23 +0800)]
net/ice: fix flow director rule after device stop

By moving irq enable/disable from device start/stop to FDIR
setup/teardown, FDIR queue irq config is independent of LAN queue irq
config. So device stop will not cause FDIR rule failure.

Fixes: 84dc7a95a2d3 ("net/ice: enable flow director engine")

Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Tested-by: Qi Zhang <qi.z.zhang@intel.com>
5 years agonet/ice: fix flow director programming status check
Yahui Cao [Mon, 18 Nov 2019 22:23:42 +0000 (06:23 +0800)]
net/ice: fix flow director programming status check

To make sure if FDIR programming succeed or fail, legacy programming
status descriptor WB format is enabled and FDIR queue irq is opened.

Fixes: 84dc7a95a2d3 ("net/ice: enable flow director engine")

Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Tested-by: Qi Zhang <qi.z.zhang@intel.com>
5 years agonet/ice: fix flow type selection for flow director
Qi Zhang [Tue, 19 Nov 2019 01:07:02 +0000 (09:07 +0800)]
net/ice: fix flow type selection for flow director

The FDIR parser will select ICE_FLTR_PTYPE_NONF_IPV4_OTHER as flow type
for an IPv4 UDP flow with empty l4 matching field which is not correct.
Same issues happens on all the combination between IPv4/IPv6 and
UDP/TCP/SCTP cases.

The patch fixes all the wrong flow ptype selections.

Fixes: f5cafa961fae ("net/ice: add flow director create and destroy")

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
5 years agonet/mlx5: fix Tx doorbell write memory barrier
Viacheslav Ovsiienko [Fri, 15 Nov 2019 11:35:06 +0000 (11:35 +0000)]
net/mlx5: fix Tx doorbell write memory barrier

As the result of testing it was found that some hosts have
the performance penalty imposed by required write memory barrier
after doorbell writing. Before 19.08 release there was some
heuristics to decide whether write memory barrier should be
performed. For the bursts of recommended size (or multiple)
it was supposed there were some extra ongoing packets in the
next burst and write memory barrier may be skipped (supposed
to be performed in the next burst, at least after descriptor
writing).

This patch restores that behaviour, the devargs tx_db_nc=2
must be specified to engage this performance tuning feature.

Fixes: 8409a28573d3 ("net/mlx5: control transmit doorbell register mapping")
Cc: stable@dpdk.org
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agonet/mlx5: change default flow engine to DV
Dekel Peled [Thu, 24 Oct 2019 12:52:53 +0000 (15:52 +0300)]
net/mlx5: change default flow engine to DV

The default flow engine is Verbs flow engine, for legacy reasons.
This patch changes the default to DV flow engine (dv_flow_en = 1).
Documentation is updated accordingly.

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agonet/af_packet: advertise Tx offload capabilities
Paul Atkins [Wed, 13 Nov 2019 12:08:00 +0000 (12:08 +0000)]
net/af_packet: advertise Tx offload capabilities

The af_packet pmd already supports MULTI_SEG tx packets, and tx
VLAN_INSERT so advertise these capabilities.

Signed-off-by: Paul Atkins <paul.atkins@intl.att.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/mlx5: fix vport index in port action
Viacheslav Ovsiienko [Sun, 17 Nov 2019 08:48:26 +0000 (08:48 +0000)]
net/mlx5: fix vport index in port action

The rdma_core routine mlx5dv_dr_create_flow_action_dest_vport()
requires the vport id parameter to create port action.
The register c[0] value was used to deduce the port id value
and it fails in bonding configuration. The correct way is
to apply vport_num value queried from the rdma_core library.

Fixes: f07341e7aed3 ("net/mlx5: update source and destination vport translations")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agonet/mlx5: fix flow table hash list conversion
Matan Azrad [Sun, 17 Nov 2019 12:14:54 +0000 (12:14 +0000)]
net/mlx5: fix flow table hash list conversion

For the case when DR is not supported and DV is supported:
multi-tables feature is off.
In this case, only table 0 is supported.
Table 0 structure wrongly was not created what prevented any
matcher object to be created and even caused crashes.

Create the table hash list in DV case too.
Create table zero empty structure for each domain when DR is not
supported.
Allow NULL DR internal table object to be used.

Fixes: 860897d2895a ("net/mlx5: reorganize flow tables with hash list")

Signed-off-by: Matan Azrad <matan@mellanox.com>
5 years agodoc: add matching component list for ice
Qi Zhang [Mon, 18 Nov 2019 04:03:34 +0000 (12:03 +0800)]
doc: add matching component list for ice

Add kernel driver, firmware and DDP package matching list
for ice PMD.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/ice/base: update version
Qi Zhang [Mon, 18 Nov 2019 04:14:43 +0000 (12:14 +0800)]
net/ice/base: update version

Update base code version info in readme.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/memif: support zero-copy slave
Jakub Grajciar [Mon, 4 Nov 2019 11:03:00 +0000 (12:03 +0100)]
net/memif: support zero-copy slave

Zero-copy slave support for memif PMD.
Slave interface exposes DPDK memory to
master interface. Only single file segments
are supported (EAL option --single-file-segments).

Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet: constify pointer to IPv6 header
Stephen Hemminger [Fri, 15 Nov 2019 00:02:59 +0000 (16:02 -0800)]
net: constify pointer to IPv6 header

The function rte_ipv6_get_next_ext does not modify
the header that is passed in.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoapp/testpmd: fix cleanup of Tx metadata offload
Dekel Peled [Thu, 14 Nov 2019 13:59:13 +0000 (15:59 +0200)]
app/testpmd: fix cleanup of Tx metadata offload

Commit in fixes tag removed the match_metadata Tx offload.
This patch removes the option to select this offload from testpmd
menu, help text and documentation.

It also modifies the cmd_show_tx_metadata_parsed() function, to
display the value correctly, and the dump_pkt_burst() function to
display the relevant (Tx/Rx) metadata only.

Fixes: 9bf26e1318e3 ("ethdev: move egress metadata to dynamic field")

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
5 years agonet/bonding: fix selection logic
Krzysztof Kanas [Wed, 13 Nov 2019 08:22:24 +0000 (09:22 +0100)]
net/bonding: fix selection logic

Arrays agg_count and agg_bandwidth should be indexed by slave_id not by
aggregator port_id.

The new_agg_id should be chosen as slave_id from slaves table in
different selection modes.

Fixes: 6d72657ce379 ("net/bonding: add other aggregator modes")
Cc: stable@dpdk.org
Signed-off-by: Krzysztof Kanas <kkanas@marvell.com>
Acked-by: Chas Williams <chas3@att.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet: use IPV4 VHL constant
Reshma Pattan [Fri, 8 Nov 2019 10:02:00 +0000 (10:02 +0000)]
net: use IPV4 VHL constant

Use new macro RTE_IPV4_VHL_DEF instead of IP_VHL_DEF
wherever applicable.

Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoethdev: improve message about not disabled offload
Andrew Rybchenko [Thu, 14 Nov 2019 16:40:52 +0000 (16:40 +0000)]
ethdev: improve message about not disabled offload

Avoid usaged of "failed" in the message about not requested but
enabled offload, since it is not a failure.

Fixes: 1daa33805824 ("ethdev: validate offloads set by PMD")

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
5 years agoethdev: decrease verbosity of not disabled offload logs
Andrew Rybchenko [Thu, 14 Nov 2019 16:40:51 +0000 (16:40 +0000)]
ethdev: decrease verbosity of not disabled offload logs

Right now a PMD decides if it is critical that an offload cannot
be disabled (i.e. not requested, but still enabled). If PMD treaks
it as OK, we should not spam logs with corresponding messages
by default. Default log level in ethdev is INFO, so change the
message level to DEBUG.

Fixes: 1daa33805824 ("ethdev: validate offloads set by PMD")

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
5 years agodrivers/net: fix RSS hash offload flag if no RSS
Andrew Rybchenko [Thu, 14 Nov 2019 16:40:50 +0000 (16:40 +0000)]
drivers/net: fix RSS hash offload flag if no RSS

By default RSS hash delivery (offload) is bound to RSS mode and
it is incorrect to advertise it as enabled if Rx multi-queue mode
has no RSS.

Fixes: 8b945a7f7dcb ("drivers/net: update Rx RSS hash offload capabilities")

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/octeontx2: support reduced set of packet types
Pavan Nikhilesh [Thu, 7 Nov 2019 02:52:54 +0000 (08:22 +0530)]
net/octeontx2: support reduced set of packet types

Add support to set supported ptypes for octeontx2 and
remove devarg scheme to disable ptype parsing support as
application can use rte_eth_dev_set_ptypes() normative API
to enable the same use case.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agonet/octeontx2: fix PTP configurations for VF
Harman Kalra [Mon, 11 Nov 2019 05:48:54 +0000 (05:48 +0000)]
net/octeontx2: fix PTP configurations for VF

Issue has been observed if PTP is already enabled on PF and
later VFs are configured. Since PTP requires mbuf data off
to be shifted by 8 bytes, due to this l3fwd/l2fwd was not
working with VFs.
Also some extra garbage bytes were observed in packet data
when ptp was enabled.

Fixes: b5dc3140448e ("net/octeontx2: support base PTP")
Cc: stable@dpdk.org
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agonet/octeontx2: fix error handling after CQ init
Anoob Joseph [Mon, 11 Nov 2019 13:21:14 +0000 (18:51 +0530)]
net/octeontx2: fix error handling after CQ init

After otx2_nix_register_cq_irqs() is called and the IRQs are setup,
otx2_nix_unregister_cq_irqs() need to be called in the subsequent error
exit paths.

Fixes: d34db5ccbf30 ("net/octeontx2: fix driver reconfiguration")
Cc: stable@dpdk.org
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agonet/octeontx2: add Rx/Tx burst mode info
Sunil Kumar Kori [Tue, 12 Nov 2019 09:02:27 +0000 (14:32 +0530)]
net/octeontx2: add Rx/Tx burst mode info

Retrieve burst mode information according to the selected Rx/Tx mode and
offloads.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
5 years agonet/bnxt: fix IP checksum error indication
Kalesh AP [Fri, 15 Nov 2019 04:50:28 +0000 (10:20 +0530)]
net/bnxt: fix IP checksum error indication

Update "mbuf->ol_flags" correctly for 'Checksum Unknown' errors
for both tunneled and non-tunneled IP packets.

Fixes: b875339622a3 ("net/bnxt: fix L4 checksum indication in non-vector Rx")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agoethdev: fix log line feed
Pavan Nikhilesh [Thu, 14 Nov 2019 19:31:32 +0000 (01:01 +0530)]
ethdev: fix log line feed

Fix missing new line token at the end of log.

Fixes: 5d308972954c ("ethdev: add mbuf RSS update as an offload")

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoraw/ifpga: introduce IRQ functions
Tianfei Zhang [Thu, 14 Nov 2019 09:03:07 +0000 (17:03 +0800)]
raw/ifpga: introduce IRQ functions

Introducing new register and unregister API for ifpga interrupt.
1. register FME and AFU interrupt
   ifpga_register_msix_irq()

2. unregister FME and AFU interrupt
   ifpga_unregister_msix_irq()

On PAC N3000 card, there is one PCIe MSIX interrupt for FME management,
like the error report, thermal management, we use this interrupt in
ifpga_rawdev device driver. on the other hand, there are about 4 PCIe
MSIX interrupts are reserved for AFU which end-user can use those
interrupts in their AFU logic design. End-user can use those APIs to
register interrupt handler in their AFU drivers.

Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Signed-off-by: Andy Pei <andy.pei@intel.com>
5 years agoraw/ifpga/base: support multiple cards
Tianfei Zhang [Thu, 14 Nov 2019 09:03:06 +0000 (17:03 +0800)]
raw/ifpga/base: support multiple cards

In PAC N3000 card, there is one MAX10 chip in each card, and
all of the sensors are connected to MAX10 chip. To support multiple
cards in one server, we introducing a sensor device list under
intel_max10_device instead of a global list. On the other hand, we
using separate intel_max10_device instance for each opae_adatper.

Add mutex lock on do_transaction() function for SPI driver to avoid
race condition.

Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Signed-off-by: Andy Pei <andy.pei@intel.com>