Meson build on i686, part of DPDK CI, reports: (example report [1])
*Meson Failed Build #1:
OS: UB1604-32
Target: i686-native-linuxapp-gcc
../drivers/common/dpaax/dpaax_iova_table.c:121:3: note: in expansion of
macro ‘DPAAX_DEBUG’
DPAAX_DEBUG("Invalid memory node values o
On 5/28/19 9:23 PM, Lance Richardson wrote:
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 5b3932b7b..bccba445a 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -638,13 +638,73 @@ static int bnxt_scattered_rx(struct rte_eth
> -Original Message-
> From: Zhang, Roy Fan
> Sent: Tuesday, May 28, 2019 3:22 PM
> To: Akhil Goyal ; Thomas Monjalon
>
> Cc: dev@dpdk.org; Trahe, Fiona ; Kusztal, ArkadiuszX
>
> Subject: RE: [dpdk-dev] [PATCH v3] cryptodev: make xform key pointer constant
>
> Hi Akhil,
>
> This patc
From: Ankur Dwivedi
The format for printing float is incorrect for the following fields
in pmd-cyclecount test: Cycles/Op,Cycles/Enq,Cycles/Deq.
Currently, the format is %.f3. This format will round off the number to
the nearest integer and append a 3 after that.
This patch changes the format t
On Wed, 29 May 2019 11:50:36 +0800
Ziyang Xuan wrote:
> +Huawei hinic
> +M: Xiaoyun Wang
> +M: Ziyang Xuan
> +M: Guoyang Zhou
> +M: Rami Rosen
You should remove Rami, he will be missed.
> diff --git a/drivers/net/hinic/base/hinic_logs.c
> b/drivers/net/hinic/base/hinic_logs.c
> new file mo
Add code for RX module.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/Makefile | 2 +-
drivers/net/hinic/hinic_pmd_rx.c | 919 +++
2 files changed, 920 insertions(+), 1 deletion(-)
create mode 100644 drivers/net/hinic/hinic_pmd_rx.c
diff --git a/drivers/n
Add code for TX module
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/Makefile |2 +-
drivers/net/hinic/hinic_pmd_tx.c | 1070 ++
2 files changed, 1071 insertions(+), 1 deletion(-)
create mode 100644 drivers/net/hinic/hinic_pmd_tx.c
diff --git a/drivers/
Add hinic PMD initialization and ethernet operatioins code.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/Makefile |2 +-
drivers/net/hinic/hinic_pmd_ethdev.c| 2157 +++
drivers/net/hinic/rte_pmd_hinic_version.map |4 +
3 files changed, 2162 in
Add build and doc files, and common code files.
Signed-off-by: Ziyang Xuan
---
MAINTAINERS | 9 +
config/common_base | 5 +
doc/guides/nics/features/hinic.ini | 32 ++
doc/guides/nics/hinic.rst | 49 ++
drivers/net/Makefile
Add various headers that define mgmt commands, cmdq commands,
rx data structures, tx data structures and basic defines for
use in the code.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_compat.h | 252
drivers/net/hinic/base/hinic_port_cmd.h | 494 +++
Add code for nic business, including qps structures, qps configuration,
wqs configuration for qps, nic business configuration functionalities.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_pmd_nic.h| 85 ++
drivers/net/hinic/base/hinic_pmd_niccfg.c | 1408
Add code for eq. Aeq is a kind queue for mgmt asynchronous message
and mgmt command response message.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_ctx_def.h | 195 ++
drivers/net/hinic/base/hinic_pmd_dpdev.h | 149 +
drivers/net/hinic/base/hinic_pmd_eqs.c | 725 +++
Add code for hardware operation, including configuration,
query and so on.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_pmd_cfg.c | 269
drivers/net/hinic/base/hinic_pmd_cfg.h | 264
drivers/net/hinic/base/hinic_pmd_hw.h| 49 +
drivers/net/hinic/base/hinic_pm
Add the structures, functionalities for interaction with
mgmt module.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_pmd_hw_mgmt.h| 85 +++
drivers/net/hinic/base/hinic_pmd_mgmt.c | 619 ++
drivers/net/hinic/base/hinic_pmd_mgmt.h | 125
.../net/
Admin channels include api channel and command queue,
Api channel is for mgmt module. And command queue is
for ucode module.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_pmd_api_cmd.c | 1040
drivers/net/hinic/base/hinic_pmd_api_cmd.h | 271 +
drivers/net
Add the registers that comprise the Huawei Hi1822 NIC.
There is no functionality in this patch.
Signed-off-by: Ziyang Xuan
---
drivers/net/hinic/base/hinic_csr.h | 164 +
1 file changed, 164 insertions(+)
create mode 100644 drivers/net/hinic/base/hinic_csr.h
diff --
This patch set adds support of a new net PMD
for Huawei Intelligent nic. This patch provides supoort
for basic RX/TX and the contorl path needed for it.
Later on new features will be added like VLAN, VFs, etc.
Basic features:
1. Basic device operations: probe, initialization, start/stop,
config
On 3/4/2019 11:18 AM, David Marchand wrote:
> According to the api, the q_errors[] per queue statistic is for reception
> errors not transmit errors.
> This is a first cleanup on statistics before looking at oerrors.
>
The fix is correct according the documentation but it was waiting for a soluti
This driver was inserting its own stub for queue_stats_mapping which
did nothing but cause this device to return a different errno
than every other device driver. All devices that don't implement
queue stats mapping should return the same error.
Maybe the plan originally was to implement somethin
From: Ajit Khaparde
Update HWRM API to version 1.10.0.74
Signed-off-by: Ajit Khaparde
Reviewed-by: Lance Richardson
Signed-off-by: Lance Richardson
---
v3:
* Fixed headline in commit log.
doc/guides/rel_notes/release_19_08.rst |1 +
drivers/net/bnxt/hsi_struct_def_dpdk.h | 1399
From: Ajit Khaparde
Update HWRM version to 1.10.0.48
Signed-off-by: Ajit Khaparde
Reviewed-by: Lance Richardson
Signed-off-by: Lance Richardson
---
v3:
* Fixed headline in commit log.
drivers/net/bnxt/hsi_struct_def_dpdk.h | 3156 +---
1 file changed, 2874 insertions(+),
From: Ajit Khaparde
Update HWRM API to version 1.10.0.19
Signed-off-by: Ajit Khaparde
Reviewed-by: Lance Richardson
Signed-off-by: Lance Richardson
---
v3:
* Fixed headline in commit log.
drivers/net/bnxt/hsi_struct_def_dpdk.h | 3112 +++-
1 file changed, 2514 insertions
From: Ajit Khaparde
We are trying to update the indirection table for all the VNICs.
We should update the table only for the default vnic0.
Fix the reta update function to only update table entries that are
selected by the update mask. Translate queue number to firmware
group ID when updating an
From: Ajit Khaparde
Update release doc briefly describing updates to bnxt PMD,
including transmit optimization changes in the following
commits:
commit 5c26b1706862 ("net/bnxt: support bulk free of Tx mbufs")
commit f8dea49ef89b ("net/bnxt: optimize Tx batching code")
Signed-off-by: Ajit Kh
In preparation for a bnxt vector-mode driver, compute and store
scattered_rx status for the device when started.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
Reviewed-by: Maxime Coquelin
---
drivers/net/bnxt/bnxt_ethdev.c | 23 +++
1 file changed, 23 insertion
Introduce SSE vector mode support for the bnxt pmd.
Signed-off-by: Lance Richardson
Signed-off-by: Ajit Khaparde
---
v2:
* Squashed with v1 patch 4 ("fix double counting VLAN tags").
* Dropped two unnecessary coding style changes from bnxt_txr.h.
v3:
* Eliminated vector mode configration flag.
To allow sharing of tx_bds_in_hw() and bnxt_tx_avail() between
vector-mode and non-vector transmit functions, move these functions
into bnxt_txr.h.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
Reviewed-by: Maxime Coquelin
---
drivers/net/bnxt/bnxt_txr.c | 15 ---
driv
This patchset bsae on top of the previous submission 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 4 from v1 patchset.
* Adde
On Tue, May 28, 2019 at 5:19 AM Ferruh Yigit wrote:
...
> There are warnings from
> "./devtools/checkpatches.sh" & "./devtools/check-git-log.sh"
> Can you please address them in next version?
>
> It is acceptable to have some level of "checkpatches.sh" in the base code
> update
> (HWRM update cod
WIll do.
Thanks,
Lance
On Tue, May 28, 2019 at 5:12 AM Ferruh Yigit wrote:
>
> On 5/24/2019 3:49 PM, Lance Richardson wrote:
> > From: Ajit Khaparde
> >
> > Update HWRM API to version 1.10.0.74
>
> There are three commits related:
> "update HWRM API" (to version 1.10.0.19)
> "update HWRM v
The hairpin feature acts as "bump on the wire", meaning that a packet that is
received from the wire
can be modified using offloaded action and then sent back to the wire without
application intervention which
can save CPU cycles.
The hairpin is the inverse function of loopback in which applicat
> On May 27, 2019, at 5:15 AM, David Marchand wrote:
>
> Hello Dharmik,
Hi David,
>
> On Wed, May 22, 2019 at 9:38 PM Dharmik Thakkar
> wrote:
> > On May 22, 2019, at 12:17 PM, David Marchand
> > wrote:
> > On Wed, May 22, 2019 at 6:27 PM Dharmik Thakkar
> > wrote:
> > > @@ -738,10 +736
On Tue, 28 May 2019 14:05:33 +0200
Lukasz Krakowiak wrote:
>
> static int
> pipe_profile_check(struct rte_sched_pipe_params *params,
> - uint32_t rate)
> + uint32_t rate, uint16_t *qsize)
> {
> uint32_t i;
>
> /* Pipe parameters */
> if (params == NULL)
> -
Someone kindly approved it, and saved me sending it again from the right
email a/c, thank you.
On 28/05/2019 13:08, Bruce Richardson wrote:
> On Tue, May 28, 2019 at 12:51:56PM +0100, Ray Kinsella wrote:
>> This patchset adds ABI Version Testing functionality to the app/test unit
>> test framework
On 5/10/19 11:28 AM, Burakov, Anatoly wrote:
On 10-May-19 3:53 PM, Michael Santana wrote:
snprintf guarantees to always correctly place a null terminator in
the buffer
string. So manually placing a null terminator in a buffer right after
a call
to snprintf is redundant code.
Additionally, the
Should we have a page with an up-to-date status of the work?
Would be good to have an overview of the progress,
even if it's not as simple as a progress bar :)
> From: Ranjit Menon
> > 15/03/2019 00:04, Thomas Monjalon:
> >> Hi,
> >>
> >> Below is a list of directories, files or functions which
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
This second patch adds the LPM ABI Version Unit Tests, comprised of
1. Registering DPDK v2.0 and DPDK v16.04 ABI Versions with the
infrastructure.
2. Forward Porting the DPDK v2.0 and DPDK v16.04 LPM Unit Test
cases, remapping the LPM Library symbols to the appropriate versions.
3. Refactori
This patchset adds ABI Version Testing functionality to the app/test
unit test framework, comprised of
1. The TEST_DPDK_ABI_VERSION_* and REISTER_TEST_ABI_VERSION macros to
register abi versions with infrastructure.
2. The MAP_ABI_SYMBOL_VERSION macro to remap symbols based on their ABI
Vers
This patchset adds ABI Version Testing functionality to the app/test unit
test framework.
The patchset is intended to address two issues previously raised during ML
conversations on ABI Stability;
1. How do we unit test still supported previous ABI Versions.
2. How to we unit test inline functions
Setup event when the Rx queue is added to the
adapter in place of generating the event when it is
being enqueued to the event device.
Signed-off-by: Nikhil Rao
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 39 ++
1 file changed, 15 insertions(+), 24 deletions(-)
T
Hi Bruce,
There was a bit of a misfire on the patch submission - it came from the
wrong email a/c and the ML (rightly) rejected it.
Let me submit the patch properly and the feedback can begin in earnest then.
Ray K
On 28/05/2019 13:08, Bruce Richardson wrote:
> On Tue, May 28, 2019 at 12:51:56P
From: Jasvinder Singh
Update packet enqueue api implementation of the scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krako
From: Jasvinder Singh
Update port memory footprint api implementation of scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Kr
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
From: Jasvinder Singh
Remove redundant macros and fields from the data structures.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 53
lib/librte_sched/rte_sched.h | 18 ---
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
From: Jasvinder Singh
Update softnic tm function to allow configuration flexiblity for
pipe traffic classes and queues, and subport level configuration
of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
---
drivers/net/softnic/
From: Jasvinder Singh
Update qos sched sample app to allow configuration flexibility for
pipe traffic classes and queues, and subport level configuration
of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
---
examples/qos_sched
From: Jasvinder Singh
Update ip pipeline sample app to allow configuration flexiblity
for pipe traffic classes and queues, and subport level configuration
of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
---
examples/ip_pipel
From: Jasvinder Singh
Update unit test to allow configuration flexiblity for pipe traffic
classes and queues, and subport level configuration of the pipe
parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
---
app/test/test_sched.c | 37 +++
From: Jasvinder Singh
Update grinder handle function implementation of the scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz
From: Jasvinder Singh
Modify credits update function of the scheduler grinder to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowi
From: Jasvinder Singh
Update grinder schedule function of the scheduler to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
--
From: Jasvinder Singh
Update queue stats read api implementation of the scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Kra
From: Jasvinder Singh
Update mbuf prefetch function of the scheduler grinder to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowia
From: Jasvinder Singh
Update packet dequeue api implementation of scheduler to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
From: Jasvinder Singh
Update implementation of grinder next pipe and tc functions to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Kra
From: Jasvinder Singh
Update pipe and tc queues prefetch functions of scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters..
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krak
From: Jasvinder Singh
Update grinder pipe and tc cache population of scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and
subport level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakow
From: Jasvinder Singh
Update weighted round robin function for best-effort traffic class
queues of the scheduler to allow configuration flexiblity for pipe traffic
classes and queues, and subport level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham T
From: Jasvinder Singh
Update subport and tc queue stats implementation of scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz K
From: Jasvinder Singh
Update the pipe configuration api implementation of the scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Luka
From: Jasvinder Singh
Update run time packet read and write api implementation of the scheduler
to allow configuration flexiblity for pipe traffic classes and queues, and
subport level configuration of the pipe parameters..
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off
From: Jasvinder Singh
Update the pipe profile add api implementation of the scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz
From: Jasvinder Singh
Update the scheduler port free api implementation to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
--
From: Jasvinder Singh
Update suport configuration api implementation of the scheduler to allow
configuration flexiblity for pipe traffic classes and queues, and subport
level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz
From: Jasvinder Singh
Update internal data structures of the scheduler to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
---
From: Jasvinder Singh
Update public data structures for subport and pipe to allow configuration
flexiblity for pipe traffic classes and queues, and subport level
configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Krakowiak
-
On Tue, May 28, 2019 at 12:51:56PM +0100, Ray Kinsella wrote:
> This patchset adds ABI Version Testing functionality to the app/test unit
> test framework.
>
> The patchset is intended to address two issues previously raised during ML
> conversations on ABI Stability;
> 1. How do we unit test stil
From: Jasvinder Singh
Update the scheduler port configuration api implementation to allow
configuration flexiblity for pipe traffic classes and queues, and
subport level configuration of the pipe parameters..
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-off-by: Lukasz Kra
From: Jasvinder Singh
Update existing macros and add new one for best-effort traffic class
queues to allow configuration flexiblity for pipe traffic classes and
queues, and subport level configuration of the pipe parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Abraham Tovar
Signed-of
This patchset refactors the dpdk qos sched library to add
following features to enhance the scheduler functionality.
1. flexibile configuration of the pipe traffic classes and queues;
Currently, each pipe has 16 queues hardwired into 4 TCs scheduled with
strict priority, and each TC has exa
On Tue, 2019-05-28 at 12:07 +0100, Bruce Richardson wrote:
> When compiling with clang on 32-bit platforms, we are missing copies
> of 64-bit atomic functions. We can solve this by linking against
> libatomic for the drivers and libs which need those atomic ops.
>
> Signed-off-by: Bruce Richardson
On Tue, 2019-05-28 at 12:07 +0100, Bruce Richardson wrote:
> The get_socket_mem_size() function is only used in 64-bit builds,
> causing clang to warn about it for 32-bit builds. Add the
> __rte_unused
> attribute to the function to silence the warning.
>
> Signed-off-by: Bruce Richardson <
> bruc
On Tue, May 28, 2019 at 5:08 AM Ferruh Yigit wrote:
>
> On 5/24/2019 3:49 PM, Lance Richardson wrote:
> > @@ -1597,6 +1659,8 @@
> >
> > bnxt_txq_info_get_op(struct rte_eth_dev *dev, uint16_t queue_id,
> >
> > static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
> > {
> > + ui
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation comman
The get_socket_mem_size() function is only used in 64-bit builds,
causing clang to warn about it for 32-bit builds. Add the __rte_unused
attribute to the function to silence the warning.
Signed-off-by: Bruce Richardson
---
lib/librte_eal/linux/eal/eal_memory.c | 1 +
1 file changed, 1 insertion(
When compiling with clang on 32-bit platforms, we are missing copies
of 64-bit atomic functions. We can solve this by linking against
libatomic for the drivers and libs which need those atomic ops.
Signed-off-by: Bruce Richardson
---
drivers/event/octeontx/meson.build | 5 +
drivers/event/op
This set fixes some issues seen on the automated CI system with building
on 32-bit Linux distro's using meson and ninja. The fixes are to disable
unsupported parts of the build, and switch the 32-bit builds to always
having large file support, so that make and meson are consistent in that
regard.
Since we now always use _FILE_OFFSET_BITS=64 flag when building
DPDK, we can remove the Makefile and C-file #defines setting it
individually for parts of the build.
Signed-off-by: Bruce Richardson
Acked-by: Luca Boccassi
---
examples/vhost_crypto/meson.build | 1 -
examples/vhost_scsi/mes
>From comment in the build config for the make build-system:
"NFP PMD is not supported on 32-bit"
Therefore disable NFP for 32-bit meson builds as well as for 32-bit
make ones.
Signed-off-by: Bruce Richardson
Acked-by: Luca Boccassi
---
drivers/net/nfp/meson.build | 2 +-
1 file changed, 1 i
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 always enables large file
support, and without this chan
> -Original Message-
> From: Bakota, Gosia
> Sent: Tuesday, May 28, 2019 11:07 AM
> To: dev@dpdk.org
> Cc: akhil.go...@nxp.com; Trahe, Fiona ; Bakota, Gosia
>
> Subject: [PATCH] crypto/qat: set message field to zero in sym SGL case
>
> Signed-off-by: Gosia Bakota
Nack
Checkpatch issu
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Signed-off-by: Gosia Bakota
---
drivers/crypto/qat/qat_sym.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
mode change 100644 => 100755 drivers/crypto/qat/qat_sym.c
diff --git a/drivers/crypto/qat/qat_sym.c b/drivers/crypto/qat/qat_sym.c
old mode 100644
new mode 100755
index 8801c
Hi Akhil,
This patch was sent in 19.05 but I just found it can be applied to 19.08 RC0
cleanly.
Do you think I need to send a v4 (with no change) for 19.08?
Regards,
Fan
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Friday, March 29, 2019 4:02 PM
> To: Th
On Mon, 2019-05-27 at 17:15 +0100, Bruce Richardson wrote:
> This set fixes some issues seen on the automated CI system with
> building
> on 32-bit Linux distro's using meson and ninja. The fixes are to
> disable
> unsupported parts of the build, and switch the 32-bit builds to
> always
> having la
On 25-May-19 1:45 AM, Yutaro Hayakawa wrote:
Hello,
In current AF_XDP PMD implementation, we cannot use it with software NICs like
VETH. Because it allocates umem with RTE_MEMZONE_IOVA_CONTIG.
Since AF_XDP itself is usable with any network devices, I think it is better to
have option to take
On 5/24/2019 3:49 PM, Lance Richardson wrote:
> This patchset bsae on top of the previous submission 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
On 5/24/2019 3:49 PM, Lance Richardson wrote:
> From: Ajit Khaparde
>
> Update release doc briefly describing updates to bnxt PMD.
It is better to keep the doc update with the feature patch, instead of a
separate doc patch. Can you please distribute content of this patch into
relevant patches?
On 5/24/2019 3:49 PM, Lance Richardson wrote:
> From: Ajit Khaparde
>
> Update HWRM API to version 1.10.0.74
There are three commits related:
"update HWRM API" (to version 1.10.0.19)
"update HWRM version" (to 1.10.0.48)
"HWRM version update" (to version 1.10.0.74)
Is "HWRM API" & "HWRM version"
Hi folks,
I’m sorry to inform the DPDK Community that Rami Rosen has passed away.
Rami was well known in Software Defined Networking circles were he
literally wrote the book on Linux Kernel Networking [1].
He was a frequent contributor to DPDK documentation submitting and
acknowledging patches,
On 5/24/2019 3:49 PM, Lance Richardson wrote:
> @@ -1597,6 +1659,8 @@
>
> bnxt_txq_info_get_op(struct rte_eth_dev *dev, uint16_t queue_id,
>
> static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
> {
> + uint32_t new_pkt_size = new_mtu + ETHER_HDR_LEN + ETHER_CRC_LEN +
> + V
On 5/24/2019 3:49 PM, Lance Richardson wrote:
> Introduce vector mode support for the bnxt pmd.
>
> Signed-off-by: Lance Richardson
> Signed-off-by: Ajit Khaparde
> ---
> v2:
> * Squashed with v1 patch 4 ("fix double counting VLAN tags").
> * Dropped two unnecessary coding style changes from bnx
Added info about socket id used by compression perf test process and by the
driver
Signed-off-by: Tomasz Cel
Signed-off-by: Fiona Trahe
Acked-by: Artur Trybula < arturx.tryb...@intel.com>
---
Rx checksum flags and input errors shouldn't be updated on Tx, as it
would work only for packets forwarding.
The ierrors statistic should be updated on Rx, right after checking
Rx checksum flags if the Rx checksum offload is enabled.
Signed-off-by: Michal Krawczyk
---
drivers/net/ena/ena_ethdev
1 - 100 of 101 matches
Mail list logo
