On 06/05, David Marchand wrote:
>Hello,
>
>On Tue, Jun 4, 2019 at 6:41 PM Xiaolong Ye wrote:
>
>> Fixes: af75078fece3 ("first public release")
>> Cc: sta...@dpdk.org
>>
>> Signed-off-by: Xiaolong Ye
>> ---
>> lib/librte_eal/linux/eal/eal.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
Hello,
On Tue, Jun 4, 2019 at 6:41 PM Xiaolong Ye wrote:
> Fixes: af75078fece3 ("first public release")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Xiaolong Ye
> ---
> lib/librte_eal/linux/eal/eal.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/linux/eal/
When a tbl8 group is getting attached to a tbl24 entry, lookup
might fail even though the entry is configured in the table.
For ex: consider a LPM table configured with 10.10.10.1/24.
When a new entry 10.10.10.32/28 is being added, a new tbl8
group is allocated and tbl24 entry is changed to point
When a tbl8 group is getting attached to a tbl24 entry, lookup
might fail even though the entry is configured in the table.
For ex: consider a LPM table configured with 10.10.10.1/24.
When a new entry 10.10.10.32/28 is being added, a new tbl8
group is allocated and tbl24 entry is changed to point
>From: Shahed Shaikh
>Sent: Tuesday, June 04, 2019 11:54 AM
>
>Patch "8bd31421c593 ("net/bnx2x: fix ramrod timeout")"
>introduced a regression where sc->scan_fp flags is set for unexpectedly long
>time. So the slow path completion handler flow is run unnecessarily which
>walks over receive descrip
Hi
> -Original Message-
> From: Ye, Xiaolong
> Sent: Tuesday, June 4, 2019 16:49
> To: Li, Xiaoyun
> Cc: Wu, Jingjing ; Wiles, Keith
> ;
> Liang, Cunming ; Maslekar, Omkar
> ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 4/6] examples/ntb: enable an example for ntb
> > [snip]
>
> >---
In some scenario, if the PF's running status is stopped, the VF link
status can be detected as down by checking the mailbox status. This
fully checking will introduce mailbox interrupt in PF, and it will be
bad if many VFs are running and the application is checking the VF's
link status quickly (wa
Hi,
Thanks for the patch!
On Mon, Jun 03, 2019 at 07:50:05PM +0100, Andrew Rybchenko wrote:
[...]
> uint16_t
> +virtio_xmit_pkts_prepare(void *tx_queue __rte_unused, struct rte_mbuf
> **tx_pkts,
> + uint16_t nb_pkts)
> +{
> + uint16_t nb_tx;
> + int error;
> +
> +
From: Bruce Richardson
When including the rte_ether.h header in applications with warnings
enabled, a warning was given because of the assumption of 2-byte alignment
of ethernet addresses when processing them.
.../include/rte_ether.h:149:2: warning: converting a packed ‘const
struct ether_addr
Now that there is a version of ether_aton in rte_ether, it can
be used by the cmdline ethernet address parser.
Signed-off-by: Stephen Hemminger
---
lib/Makefile | 2 +-
lib/librte_cmdline/Makefile | 3 +-
lib/librte_cmdline/cmdline_parse_etherad
Using bit operations like or and xor is faster than a loop
on all architectures. Really just explicit unrolling.
Similar cast to uint16 unaligned is already done in
other functions here.
Signed-off-by: Stephen Hemminger
---
lib/librte_net/rte_ether.h | 17 +++--
1 file changed, 7 in
Make a function that coresponds with eth_aton_r which can
be used to convert string to rte_ether_addr.
This also allows rte_ethdev to no longer depend on the
cmdline library.
Signed-off-by: Stephen Hemminger
---
lib/librte_net/rte_ether.c | 47 ++
lib/librte_
Use rte_eth_unformat_addr, so that ethdev can be built and work
without the cmdline library. The dependency on cmdline was
an arrangement of convenience anyway.
Signed-off-by: Stephen Hemminger
---
lib/Makefile | 1 -
lib/librte_ethdev/Makefile| 2 +-
lib/librte_ethd
This is a collection of patches around the ethernet address
manipulation routines in librte_net/rte_ether.
v3
rebase to use rte_ether prefix
drop aligning ethernet headers for now.
Bruce Richardson (1):
net/ether: mark ethernet addresses as being 2-byte aligned
Stephen Hemminger (5):
Formatting ethernet address and getting a random value are
not in critical path so they should not be inlined.
Signed-off-by: Stephen Hemminger
Acked-by: Rami Rosen
---
lib/librte_net/Makefile| 1 +
lib/librte_net/rte_ether.c | 29 +
lib/librte_n
There will be a newer version of this patch that removes this code.
Paul
> -Original Message-
> From: Venkataramanan, Anirudh
> Sent: Tuesday, June 4, 2019 12:17 PM
> To: Zhang, Qi Z ; Rong, Leyi
> Cc: Stillwell Jr, Paul M ; dev@dpdk.org
> Subject: Re: [PATCH 42/49] net/ice/base: change
> -Original Message-
> From: Maxime Coquelin
> Sent: Tuesday, June 4, 2019 10:15 AM
> To: Rong, Leyi ; Zhang, Qi Z
> Cc: dev@dpdk.org; Cao, Chinh T ; Ertman, David M
> ; Stillwell Jr, Paul M
>
> Subject: Re: [dpdk-dev] [PATCH 03/49] net/ice/base: add API to configure
> MIB
>
>
>
> On
On 2019-06-04 14:04, Thomas Monjalon wrote:
04/06/2019 22:25, dwilder:
On 2019-06-03 15:30, Thomas Monjalon wrote:
> 27/05/2019 11:41, Bruce Richardson:
>> On Fri, May 24, 2019 at 09:04:04AM -0700, David Wilder wrote:
>> > Setting RTE_MAX_LCORE to reflect the largest available configuration.
>>
04/06/2019 22:25, dwilder:
> On 2019-06-03 15:30, Thomas Monjalon wrote:
> > 27/05/2019 11:41, Bruce Richardson:
> >> On Fri, May 24, 2019 at 09:04:04AM -0700, David Wilder wrote:
> >> > Setting RTE_MAX_LCORE to reflect the largest available configuration.
> >> > Adding defines for missing RTE_MACH
On 2019-06-03 15:30, Thomas Monjalon wrote:
27/05/2019 11:41, Bruce Richardson:
On Fri, May 24, 2019 at 09:04:04AM -0700, David Wilder wrote:
> Setting RTE_MAX_LCORE to reflect the largest available configuration.
> Adding defines for missing RTE_MACHINE_CPUFLAGs.
>
> Signed-off-by: David Wilder
When using DPDK on Azure it is common to have one non-DPDK interface.
If that non-DPDK interface is present vdev_netvsc correctly skip it.
But if the non-DPDK has accelerated networking the Mellanox driver will
still get associated with DPDK (and break connectivity).
The current process is to tell
bnx2x_free_hsi_mem() does not free DMA memory.
Fix it here.
Fixes: 540a211084a7 ("bnx2x: driver core")
Cc: sta...@dpdk.org
Signed-off-by: Shahed Shaikh
---
drivers/net/bnx2x/bnx2x.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
i
Driver does not provide limit on number Rx and Tx descriptors per queue,
this may result in application configuring 64k descriptors (default set
by rte_eth_dev_info_get()) and further result in issues in PMD and HW
flows due to unsupported number.
Fixes: 540a211084a7 ("bnx2x: driver core")
Cc: sta
Patch "8bd31421c593 ("net/bnx2x: fix ramrod timeout")"
introduced a regression where sc->scan_fp flags is
set for unexpectedly long time. So the slow path completion
handler flow is run unnecessarily which walks over receive
descriptor ring of fast path and drops the data packets while looking
for
Don't call bnx2x_link_status_update() from bnx2x_link_update().
Actual use case of bnx2x_link_status_update() is to update the link
status in shared memory between driver and MFW, and not to get the
link status from HW.
So ideally, bnx2x_link_status_update() should be called when there
is an actua
PMD sets up and clears the slow path interrupt status block in dev_start
and dev_stop flow and slow path interrupt status block DMA memory for
device is allocated in dev_configure flow.
This situation creates a state where, after dev_stop is called, and if
there is a slow path interrupt from devic
On 6/4/19 7:42 AM, Leyi Rong wrote:
Add functions to support following admin queue commands:
1. 0x0208: allocate resource to hold a switch recipe. This is needed
when a new switch recipe needs to be created.
2. 0x0290: create a recipe with protocol header information and
other details that det
On 6/4/19 7:42 AM, Leyi Rong wrote:
Add ice_cfg_lldp_mib_change and treat DCBx state NOT_STARTED as valid.
Signed-off-by: Chinh T Cao
Signed-off-by: Dave Ertman
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Leyi Rong
---
drivers/net/ice/base/ice_dcb.c | 41 ++
On 6/4/19 7:42 AM, Leyi Rong wrote:
Once upon a time, the ice_flow_create_xtrct_seq() function in ice_flow.c
extracted only protocol fields explicitly specified by the caller of the
ice_flow_add_prof() function via its struct ice_flow_seg_info instances.
However, to support different ingress a
> -Original Message-
> From: Ruifeng Wang
> Sent: Monday, April 29, 2019 3:32 PM
> To: yipeng1.w...@intel.com; sameh.gobr...@intel.com;
> bruce.richard...@intel.com; pablo.de.lara.gua...@intel.com
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran ;
> honnappa.nagaraha...@arm.com; n...@arm.com;
Hi Leyi,
On 6/4/19 7:41 AM, Leyi Rong wrote:
Main changes:
1. Advanced switch rule support.
2. Add more APIs for tunnel management.
3. Add some minor features.
4. Code clean and bug fix.
In order to ease the review process, I think it would be much better
to split this series in multiple ones,
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Signed-off-by: Xiaolong Ye
---
lib/librte_eal/linux/eal/eal.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_eal/linux/eal/eal.c b/lib/librte_eal/linux/eal/eal.c
index 161399619..10e2887ca 100644
---
On Tue, Jun 04, 2019 at 04:59:47PM +0100, Ferruh Yigit wrote:
> On 5/16/2019 3:08 PM, Bruce Richardson wrote:
> > On Wed, May 15, 2019 at 12:13:46PM -0400, David Harton wrote:
> >> Use of weak symbols can hide makefile errors especially when
> >> custom makefiles are used. Removing the use of weak
> -Original Message-
> From: Ferruh Yigit
> Sent: Tuesday, June 04, 2019 12:00 PM
> To: Bruce Richardson ; David Harton (dharton)
>
> Cc: dev@dpdk.org; beilei.x...@intel.com; qi.z.zh...@intel.com
> Subject: Re: [dpdk-dev] [PATCH v2] net/i40e: Eliminate weak symbols in
> i40e_rxtx.c
>
>
On 5/29/2019 10:02 PM, Lance Richardson wrote:
> This patchset adds the following:
> 1) Support for vector mode TX and RX.
> 2) HWRM API update (split into multiple patches).
> 3) Fixes for RSS reta update and query.
>
> It also updates the release notes.
>
> v2:
> * Squashed patches 3 and
On 5/16/2019 3:08 PM, Bruce Richardson wrote:
> On Wed, May 15, 2019 at 12:13:46PM -0400, David Harton wrote:
>> Use of weak symbols can hide makefile errors especially when
>> custom makefiles are used. Removing the use of weak symbols
>> to avoid a stub function being linked in production code.
On Mon, 3 Jun 2019 20:57:36 +0800
Loc Nguyen wrote:
> Add a device index in front of the PCI ID for easy counting
>
> Network devices using DPDK-compatible driver
>
> 0: :07:00.0 ...
> 1: :07:00.1 ...
>
> Signed-off-by: Loc Nguyen
> ---
>
On 6/4/19 4:59 AM, David Marchand wrote:
This is a joint effort to make the unit tests ready for CI.
The first 8 patches are fixes that I had accumulated.
Then the second part of the series focuses on skipping tests when some
requirements are not fulfilled so that we can start them in a restraine
After having dequeued a burst of descriptors, there may be a
need to dequeue a few more if the last packet was segmented
and not complete.
When it happens, the extra segments were not properly attached
to the mbuf chain, and so were lost.
This patch fixes the mbuf chaining.
Fixes: bcac5aa207f8 (
This patch removes uses checks on 'prev' pointer, as it
is always set before with a valid value.
Signed-off-by: Maxime Coquelin
---
drivers/net/virtio/virtio_rxtx.c | 15 ---
1 file changed, 4 insertions(+), 11 deletions(-)
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net
After having dequeued a burst of descriptors, there may be a
need to dequeue a few more if the last packet was segmented
and not complete.
When it happens, the extra segments were not properly attached
to the mbuf chain, and so were lost.
This patch fixes the mbuf chaining.
Fixes: e5f456a98d3c (
This series fixes a bug reported by Yaroslav, where segmented
packets miss some segments.
Last patch is not a fix, but it removes some useless checks
in Rx paths.
Yaroslav, can you try the series and confirm it fixes the issue
you faced?
Maxime Coquelin (3):
net/virtio: fix segmented packet is
This patch changes some void functions to return a value,
so that the init sequence may tear down orderly
instead of calling panic.
Signed-off-by: Arnon Warshavsky
---
The calls for launching core messaging threads were left in tact
in all 3 eal implementations.
This should be addressed in a dif
On 04/06/2019 16:00, Ferruh Yigit wrote:
> On 5/29/2019 10:48 PM, Thomas Monjalon wrote:
>> 27/05/2019 06:02, michael@intel.com:
>>> From: Luo Gaoliang
>>>
>>> Build error was seen on RHEL8 when CONFIG_RTE_KNI_KMOD_ETHTOOL is enabled.
>>>
>>> Build error log:
>>> /root/kni/dpdk-19.05/kernel/li
> From: Stephen Hemminger
>
> Don't refer to lcore_config directly.
>
> Signed-off-by: Stephen Hemminger
> Signed-off-by: David Marchand
Acked-by: Gage Eads
On 5/29/2019 10:48 PM, Thomas Monjalon wrote:
> 27/05/2019 06:02, michael@intel.com:
>> From: Luo Gaoliang
>>
>> Build error was seen on RHEL8 when CONFIG_RTE_KNI_KMOD_ETHTOOL is enabled.
>>
>> Build error log:
>> /root/kni/dpdk-19.05/kernel/linux/kni/ethtool/igb/kcompat.h:3928:25:
>> error: '
On 5/28/2019 12:07 PM, Bruce Richardson wrote:
> By default on 32-bit systems, file offsets are given as 32-bit values
> which prevents support for large files. While this is unlikely to be
> a problem, enabling large file support globally makes "make" and
> "meson" builds consistent, since meson a
> -Original Message-
> From: dev On Behalf Of Ruifeng Wang
> Sent: Wednesday, May 15, 2019 2:11 PM
> To: bruce.richard...@intel.com; konstantin.anan...@intel.com
> Cc: dev@dpdk.org; honnappa.nagaraha...@arm.com; n...@arm.com; Ruifeng
> Wang
> Subject: [dpdk-dev] [PATCH v1] doc: update
On 5/2/2019 1:11 PM, Tom Barbette wrote:
> Add rte_eth_read_clock to read the raw clock of a device.
>
> The main use is to get the device clock conversion co-efficients to be
> able to translate the raw clock of the timestamp field of the pkt mbuf
> to a local synced time value.
>
> This functio
On Tue, Jun 4, 2019 at 3:29 PM Aaron Conole wrote:
> Hi David and Michael,
>
> David Marchand writes:
>
> > From: Michael Santana
> >
> > The eal_flags_autotest test currently fails due to a memory leak in the
> > timer library[1][2]. This failure occurs when the test calls one of its
> > subte
Hi Damian,
> -Original Message-
> From: Nowak, DamianX
> Sent: Monday, June 3, 2019 3:51 PM
> To: dev@dpdk.org
> Cc: Trahe, Fiona ; Kusztal, ArkadiuszX
> ; Nowak,
> DamianX
> Subject: [PATCH 7/9] crypto/qat: add digest encrypted feature flag
>
> From: Damian Nowak
>
> This patch adds
David Marchand writes:
> Caught in one Travis run:
> + --- +
> + Test Suite : Latency Stats Unit Test Suite
> + --- +
> + TestCase [ 0] : test_latency_init succeeded
> + TestCase [ 1] : te
Hi Bruce,
Thanks for the feedback. In my application, we map the NIC interfaces into port
numbers,
which are these indexes. For example, we have an config file saying we would
like to use
port 2, 3 and 5 for RX. Having the indexes printed together with the NIC
interfaces via this
script helps u
David Marchand writes:
> Fixes: 3c518ca41ffa ("test/hash: remove hash scaling unit test")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
Acked-by: Aaron Conole
Hi David and Michael,
David Marchand writes:
> From: Michael Santana
>
> The eal_flags_autotest test currently fails due to a memory leak in the
> timer library[1][2]. This failure occurs when the test calls one of its
> subtests test_file_prefix().
>
> Fixing the memory leak is not trivial, so
David Marchand writes:
> From: Michael Santana
>
> The eal flags unit test assumes that a certain number of cores are
> available (4 and 8 cores), however this may not always be the case.
> Individual developers may run the unit test on their local desktop
> which typically have 2 to 4 cores, in
On Wed, May 29, 2019 at 3:15 PM Stephen Hemminger
wrote:
>
> The device private pointer (dev_private) is of type void *
> therefore no cast is necessary in C.
>
> Signed-off-by: Stephen Hemminger
> ---
> drivers/net/bnxt/bnxt_ethdev.c | 86 -
> drivers/net/bnxt/b
Hi Damian,
> -Original Message-
> From: Nowak, DamianX
> Sent: Monday, June 3, 2019 3:51 PM
> To: dev@dpdk.org
> Cc: Trahe, Fiona ; Kusztal, ArkadiuszX
> ; Nowak,
> DamianX
> Subject: [PATCH 0/9] add QAT support for digest encrypted
>
> From: Damian Nowak
>
> This patchset adds digest
David Marchand writes:
> Setting a coremask was mandatory a long time ago but has been optional
> for a while.
> The checks on PCI whitelist/blacklist, vdev, memory rank, memory channel,
> HPET, memory size and other miscs options have no requirement wrt cores.
>
> Let's remove those coremasks so
David Marchand writes:
> The -n option is an optimisation configuration option that defaults to 0.
> Such a default value makes the mempool library distributes objects as if
> there was 4 memory channels, so -n 4 is the same as the default behavior.
>
> This parameter was mandatory a long time ag
David Marchand writes:
> Fixes: 0420378bbfc4 ("test/stack: check lock-free implementation")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
Acked-by: Aaron Conole
David Marchand writes:
> This subtest wants to start rwc_core_cnt[n] reader threads, while the
> master core is waiting for them to report.
>
> Fixes: c7eb0972e74b ("test/hash: add lock-free r/w concurrency")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
Acked-by: Aaron Conole
David Marchand writes:
> Same treatment than crypto tests, move the eventdev drivers tests in the
> driver list.
>
> While at it:
> - eventdev_octeontx_autotest has been renamed as
> eventdev_selftest_octeontx,
> - eventdev_sw_autotest has been renamed as
> eventdev_selftest_sw,
>
> Fixes: 50
David Marchand writes:
> For consistency, put all specific crypto driver tests in the dedicated
> list (in alphabetic order).
>
> While at it:
> - remove dead reference to cryptodev_sw_mrvl_autotest (renamed as
> cryptodev_sw_mvsam_autotest),
> - call the crypto scheduler test only when built,
David Marchand writes:
> Fixes: 3d20ffe6ddb1 ("test: reorder test cases in meson")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
Acked-by: Aaron Conole
This patch fixes Coverity Issue 328524:
Reliance on integer endianness (INCOMPATIBLE_CAST) in function
*parse_window_sz*.
Coverity issue: 328524
Fixes: e0b6287c035d ("app/compress-perf: add parser")
Signed-off-by: Artur Trybula
---
app/test-compress-perf/comp_perf_options_parse.c | 4 +++-
1 fi
v2:
* add 'fixes' line to the patch description
Artur Trybula (1):
app/test-compress-perf: fix reliance on integer endianness (CI:328524)
app/test-compress-perf/comp_perf_options_parse.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--
2.17.1
On 04-Jun-19 11:45 AM, Michał Krawczyk wrote:
wt., 4 cze 2019 o 12:28 Burakov, Anatoly napisał(a):
On 03-Jun-19 2:36 PM, Michał Krawczyk wrote:
On 03.06.2019 09:33, Michał Krawczyk wrote:
On 29.05.2019 18:31, Anatoly Burakov wrote:
The ENA driver calculates a ring's NUMA node affinity by di
On Mon, Jun 03, 2019 at 04:44:17PM +0200, Thomas Monjalon wrote:
> 30/05/2019 23:25, Bruce Richardson:
> > Add local APIs to trigger data copies, and retrieve handle values once
> > those copies are completed. Included are unit tests to validate the data
> > is copies correctly.
>
> Would be nice
From: David Marchand
Sent: Monday, June 3, 2019 4:19 PM
To: Ben Walker ; Jerin Jacob Kollanukkaran
Cc: dev ; Burakov, Anatoly
Subject: Re: [dpdk-dev] eal/pci: Improve automatic selection of IOVA mode
> - This leaves the --iova-va forcing option.
> Why do we need it?
> If we don't have acce
16/05/2019 09:58, Olivier Matz:
> On Thu, May 16, 2019 at 03:28:56PM +0800, Xiaolong Ye wrote:
> > As memzone.h is introduced by commit 38c9817ee1d8 ("mempool: adjust name
> > size in related data types"), forward declaration for rte_memzone is no
> > longer needed.
> >
> > Signed-off-by: Xiaolong
23/05/2019 16:29, David Marchand:
> On Thu, May 23, 2019 at 11:53 AM Nicolas Dichtel
> wrote:
>
> > move_pages() is only used to get the numa node id, but this function
> > is not allowed by default in docker (it needs CAP_SYS_NICE and an update of
> > the seccomp profile).
> > get_mempolicy() al
Previous solution was using memzones in invalid way in hope to assign
IO queue to the appropriate NUMA zone.
The right way is to use socket_id from the rx/tx queue setup function
and then pass it to the IO queue.
Fixes: 3d3edc265fc8 ("net/ena: make coherent memory allocation NUMA-aware")
Cc: sta.
On Tue, Jun 4, 2019 at 12:43 PM Michal Krawczyk wrote:
> Previous solution was using memzones in invalid way in hope to assign
> IO queue to the appropriate NUMA zone.
>
> The right way is to use socket_id from the rx/tx queue setup function
> and then pass it to the IO queue.
>
> Fixes: 3d3edc26
wt., 4 cze 2019 o 12:28 Burakov, Anatoly napisał(a):
>
> On 03-Jun-19 2:36 PM, Michał Krawczyk wrote:
> > On 03.06.2019 09:33, Michał Krawczyk wrote:
> >> On 29.05.2019 18:31, Anatoly Burakov wrote:
> >>> The ENA driver calculates a ring's NUMA node affinity by directly
> >>> accessing the memzone
06/05/2019 11:49, Ali Alnubani:
> The application tries to configure queue ids larger
> than the maximum allowed by MAX_TX_QUEUE_PER_PORT. This
> causes the startup error:
>
> "
> ...
> Initializing port 0 on lcore 0... Address:7C:FE:90:12:23:0D
> txq=0,0 txq=1,1 txq=2,2 txq=3,3 txq=4,4 txq=
Previous solution was using memzones in invalid way in hope to assign
IO queue to the appropriate NUMA zone.
The right way is to use socket_id from the rx/tx queue setup function
and then pass it to the IO queue.
Fixes: 3d3edc265fc8 ("net/ena: make coherent memory allocation NUMA-aware")
Cc: sta.
23/05/2019 20:42, Luca Boccassi:
> On Thu, 2019-05-23 at 17:40 +, Ali Alnubani wrote:
> > This enables building the example multiprocess applications in
> > the subdirectory multi_process.
> >
> > Signed-off-by: Ali Alnubani
>
> Acked-by: Luca Boccassi
Applied, thanks
Previous solution was using memzones in invalid way in hope to assign
IO queue to the appropriate NUMA zone.
The right way is to use socket_id from the rx/tx queue setup function
and then pass it to the IO queue.
Fixes: 3d3edc265fc8 ("net/ena: make coherent memory allocation NUMA-aware")
Change-
On 03-Jun-19 2:36 PM, Michał Krawczyk wrote:
On 03.06.2019 09:33, Michał Krawczyk wrote:
On 29.05.2019 18:31, Anatoly Burakov wrote:
The ENA driver calculates a ring's NUMA node affinity by directly
accessing the memzone list. Fix it to do it through the public
API's instead.
Signed-off-by: An
When ixgbe_crypto_add_sa() is called, it checks whether the ip type is
IPv6 or IPv4 to write correct addresses to the registers. Type itself
is never specified, and act as IPv4, which is the default value.
It causes lack of support for IPv6.
To fix that, ip type needs to be stored in device privat
Application ipsec-secgw is not working for IPv4 transport mode and for
IPv6 both transport and tunnel mode.
IPv6 tunnel mode is not working due to wrongly assigned fields of
security association patterns, as it was IPv4, during creation of
inline crypto session.
IPv6 and IPv4 transport mode is it
Fixes for ipsec-secgw application to work with inline crypto IPv4
transport mode and IPv6 tunnel/transport modes.
Mariusz Drost (2):
net/ixgbe: fix lack of ip type for crypto session
examples/ipsec-secgw: fix not working inline ipsec modes
drivers/net/ixgbe/ixgbe_ipsec.c | 6 +-
examples/i
On Mon, Jun 03, 2019 at 08:57:36PM +0800, Loc Nguyen wrote:
> Add a device index in front of the PCI ID for easy counting
>
> Network devices using DPDK-compatible driver
>
> 0: :07:00.0 ...
> 1: :07:00.1 ...
>
> Signed-off-by: Loc Nguyen
>
> > > -Original Message-
> > > From: Wang, Xiao W
> > > Sent: Wednesday, May 15, 2019 2:59 PM
> > > To: Singh, Jasvinder
> > > Cc: dev@dpdk.org; Dumitrescu, Cristian
> > > ; Wang, Xiao W
> > > ; sta...@dpdk.org
> > > Subject: [PATCH] net/softnic: fix pipeline time calculation
> > >
> > >
From: Krzysztof Kanas
Fixes: a753e53d517b ("eal: add device event monitor framework")
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Signed-off-by: Krzysztof Kanas
---
lib/librte_eal/common/include/rte_dev.h| 2 +-
lib/librte_eal/common/include/rte_interrupts.h | 2 +-
From: Krzysztof Kanas
The configuration for native-linux-* targets is done using symbolic
links, e.g
config/defconfig_x86_64-native-linux-gcc
which points to
config/defconfig_x86_64-native-linuxapp-gcc.
When configuring mk will output correct target based on
RTE_CONFIG_TEMPLATE.
When building m
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Nowak
> Sent: Monday, June 3, 2019 3:51 PM
> To: dev@dpdk.org
> Cc: Trahe, Fiona ; Kusztal, ArkadiuszX
> ; Nowak, DamianX
> Subject: [dpdk-dev] [PATCH 5/9] cryptodev: document usage of digest-
> appended operati
> -Original Message-
> From: pbhagavat...@marvell.com [mailto:pbhagavat...@marvell.com]
> Sent: Monday, June 3, 2019 11:34 AM
> To: jer...@marvell.com; Pavan Nikhilesh ;
> Mcnamara, John ; Kovacevic, Marko
>
> Cc: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH v2 6/6] doc: update Marvell OCT
> -Original Message-
> From: Ye, Xiaolong
> Sent: Tuesday, June 4, 2019 9:41 AM
> To: Shreyansh Jain ; Hemant Agrawal
> ; Mcnamara, John ;
> Kovacevic, Marko
> Cc: dev@dpdk.org; Ye, Xiaolong ; sta...@dpdk.org
> Subject: [PATCH] doc: fix a grammar mistake in rawdev guide
>
> Fixes: a9bb
> -Original Message-
> From: stable [mailto:stable-boun...@dpdk.org] On Behalf Of Ye Xiaolong
> Sent: Tuesday, June 4, 2019 8:27 AM
> To: Zhang, Tianfei
> Cc: dev@dpdk.org; Xu, Rosen ; sta...@dpdk.org;
> zh...@dpdk.org
> Subject: Re: [dpdk-stable] [dpdk-dev] [PATCH v1 2/2] raw/ifpga: fi
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Artur Trybula
> Sent: Monday, June 3, 2019 2:51 PM
> To: dev@dpdk.org; Trahe, Fiona ; Trybula, ArturX
> ; Dybkowski, AdamX
> Subject: [dpdk-dev] [PATCH] app/test-compress-perf: fix reliance on
> integer endianne
From: Michael Santana
The eal_flags_autotest test currently fails due to a memory leak in the
timer library[1][2]. This failure occurs when the test calls one of its
subtests test_file_prefix().
Fixing the memory leak is not trivial, so this patch is a workaround that
makes the eal_flags_autotes
From: Michael Santana
The eal flags unit test assumes that a certain number of cores are
available (4 and 8 cores), however this may not always be the case.
Individual developers may run the unit test on their local desktop
which typically have 2 to 4 cores, in said case the test is bound
to fail
Running the tests in parallel has two drawbacks:
- the tests are racing on the hugepages allocations,
- the tests are sharing the cores to run their checks which results in
undeterministic execution time,
This results in random failures.
For better reproducibility in CI, start them all in a seri
Let's mark as skipped the tests when they are missing some requirements
like a number of used cores or specific hardware availability, like
compress, crypto or eventdev devices.
Signed-off-by: David Marchand
---
app/test/test.c | 24
app/test/test_com
The -n option is an optimisation configuration option that defaults to 0.
Such a default value makes the mempool library distributes objects as if
there was 4 memory channels, so -n 4 is the same as the default behavior.
This parameter was mandatory a long time ago, but has been optional for
a whi
Setting a coremask was mandatory a long time ago but has been optional
for a while.
The checks on PCI whitelist/blacklist, vdev, memory rank, memory channel,
HPET, memory size and other miscs options have no requirement wrt cores.
Let's remove those coremasks so that we only care about it in the
d
Fixes: 0420378bbfc4 ("test/stack: check lock-free implementation")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
app/test/meson.build | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/app/test/meson.build b/app/test/meson.build
index cd46289..3c9706a 100644
--- a/ap
Same treatment than crypto tests, move the eventdev drivers tests in the
driver list.
While at it:
- eventdev_octeontx_autotest has been renamed as
eventdev_selftest_octeontx,
- eventdev_sw_autotest has been renamed as
eventdev_selftest_sw,
Fixes: 50fb749a3972 ("event/octeontx: move test to d
1 - 100 of 126 matches
Mail list logo