> -Original Message-
> From: Luca Boccassi
> Sent: Thursday, July 28, 2022 8:34 PM
> To: sta...@dpdk.org
> Cc: dev@dpdk.org; Abhishek Marathe ;
> Ali Alnubani ; Walker, Benjamin
> ; David Christensen
> ; Hemant Agrawal ;
> Stokes, Ian ; Jerin Jacob ;
> Mcnamara, John ; Ju-Hyoung Lee
> ; Ke
Hi,
> -Original Message-
> From: David Marchand
> Sent: Thursday, July 28, 2022 23:27
> To: dev@dpdk.org
> Cc: Chautru, Nicolas ; Ray Kinsella
> ; Nithin Dabilpuram ; Kiran
> Kumar K ; Sunil Kumar Kori
> ; Satha Rao ; Matan Azrad
> ; Viacheslav Ovsiienko ;
> Zhang, Roy Fan ; Andrew Rybche
Hi,
> -Original Message-
> From: David Marchand
> Sent: Thursday, July 28, 2022 23:27
> To: dev@dpdk.org
> Cc: Gagandeep Singh ; Hemant Agrawal
> ; Chautru, Nicolas ;
> Ray Kinsella ; Daly, Lee ; Zhang, Roy
> Fan ; Ashish Gupta ;
> Sunila Sahu ; Ruifeng Wang ;
> Ajit Khaparde ; Raveendra
Hi,
> -Original Message-
> From: David Marchand
> Sent: Thursday, July 28, 2022 23:26
> To: dev@dpdk.org
> Cc: Xu, Rosen ; Ray Kinsella ; Zhang,
> Tianfei
> Subject: [RFC v3 15/26] bus/ifpga: make driver-only headers private
>
> The ifpga bus interface is for drivers only.
> Mark as int
Hi,
> -Original Message-
> From: David Marchand
> Sent: Thursday, July 28, 2022 23:26
> To: dev@dpdk.org
> Cc: Xu, Rosen ; Ray Kinsella ; Zhang,
> Tianfei
> Subject: [RFC v3 14/26] bus/ifpga: cleanup exported symbols
>
> Remove unused symbols (exposed only in an internal header which
>
Hi,
> -Original Message-
> From: David Marchand
> Sent: Thursday, July 28, 2022 23:26
> To: dev@dpdk.org
> Cc: Ray Kinsella ; Xu, Rosen ; Zhang,
> Tianfei
> Subject: [RFC v3 03/26] raw/ifpga: remove PCI bus accessor
>
> There is no in-tree user for this accessor that returns the PCI bus
On Thu, Jul 28, 2022 at 7:03 PM Bruce Richardson
wrote:
>
> On Thu, Jul 28, 2022 at 05:26:39PM +0200, David Marchand wrote:
> > For diagnostic, it may be useful to provide a description of the device
> > with Bus specific information like, for PCI, the vendor and device ID.
> >
> > Signed-off-by:
On Thu, Jul 28, 2022 at 6:38 PM Bruce Richardson
wrote:
>
> On Thu, Jul 28, 2022 at 05:26:22PM +0200, David Marchand wrote:
> > RTE_FUNC_PTR_OR_* macros have nothing to do with the rte_device object
> > and associated API.
> > Move them to rte_common.h and include it where needed.
> >
> > Signed-o
On Thu, Jul 28, 2022 at 6:32 PM Bruce Richardson
wrote:
>
> On Thu, Jul 28, 2022 at 05:26:20PM +0200, David Marchand wrote:
> > Rely on the generic device name rather than restrict to only supporting
> > PCI devices.
> >
> > Signed-off-by: David Marchand
> > ---
> > examples/ethtool/lib/rte_etht
On Thu, Jul 28, 2022 at 6:57 PM Bruce Richardson
wrote:
>
> On Thu, Jul 28, 2022 at 05:26:35PM +0200, David Marchand wrote:
> > Make rte_bus opaque for non internal users.
> > This will make extending this object possible without breaking the ABI.
> >
> > Introduce a new driver header and move rte
2022-07-28 14:41 (UTC+0500), Fidaullah Noonari:
> The amount of memory to allocate from the system for heap expansion
> was calculated in a way that may yield one page more than needed.
> This could hit the allocation limit from the system or EAL.
> The allocation would fail despite enough memory b
On Thu, Jul 28, 2022 at 05:26:40PM +0200, David Marchand wrote:
> Make rte_device opaque for non internal users.
> This will make extending this object possible without breaking the ABI.
>
> Some applications may have been dereferencing rte_device objects, mark
> this object's accessors as stable.
On Thu, Jul 28, 2022 at 05:26:39PM +0200, David Marchand wrote:
> For diagnostic, it may be useful to provide a description of the device
> with Bus specific information like, for PCI, the vendor and device ID.
>
> Signed-off-by: David Marchand
> ---
Good idea.
Acked-by: Bruce Richardson
On Thu, Jul 28, 2022 at 05:26:38PM +0200, David Marchand wrote:
> Prepare for making the device object opaque by adding accessors.
> Update existing "external" users.
>
> Signed-off-by: David Marchand
> ---
> Changes since RFC v2:
> - added rte_dev_devargs,
> - updated testpmd,
>
Acked-by: Bruce
On Thu, Jul 28, 2022 at 05:26:37PM +0200, David Marchand wrote:
> Make rte_driver opaque for non internal users.
> This will make extending this object possible without breaking the ABI.
>
> Introduce a new driver header and move rte_driver definition.
> Update drivers and library to use the inter
On Thu, Jul 28, 2022 at 05:26:36PM +0200, David Marchand wrote:
> Prepare for making the driver object opaque by adding accessors.
> Update existing "external" users.
> Internal users may still dereference a rte_driver object.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Bruce Richardson
On Thu, Jul 28, 2022 at 05:26:35PM +0200, David Marchand wrote:
> Make rte_bus opaque for non internal users.
> This will make extending this object possible without breaking the ABI.
>
> Introduce a new driver header and move rte_bus definition and helpers.
> Update drivers and library to use the
On Thu, Jul 28, 2022 at 9:47 AM Bruce Richardson
wrote:
>
> On Thu, Jul 28, 2022 at 05:26:30PM +0200, David Marchand wrote:
> > The pci bus interface is for drivers only.
> > Mark as internal and move the header in the driver headers list.
> >
> > While at it, cleanup the code:
> > - fix indentati
On Thu, Jul 28, 2022 at 05:26:34PM +0200, David Marchand wrote:
> Add helpers to get a rte_bus object details.
> This will be used externally.
> Internal users may still dereference a rte_bus object.
>
> Signed-off-by: David Marchand
> ---
> Changes since RFC v1:
> - changed approach: only extern
On Thu, Jul 28, 2022 at 05:26:33PM +0200, David Marchand wrote:
> iova enum definition does not need to be defined as part of the bus API.
> Move it to rte_eal.h.
> With this step, rte_eal.h does not depend on rte_bus.h and rte_dev.h.
> Fix existing code that was relying on these implicit inclusion
On Thu, Jul 28, 2022 at 05:26:30PM +0200, David Marchand wrote:
> The pci bus interface is for drivers only.
> Mark as internal and move the header in the driver headers list.
>
> While at it, cleanup the code:
> - fix indentation,
> - remove unneeded reference to bus specific singleton object,
>
On Thu, Jul 28, 2022 at 05:26:24PM +0200, David Marchand wrote:
> Same as for device classes, external DPDK users may need to include some
> bus headers for their out of tree drivers.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Bruce Richardson
On Thu, Jul 28, 2022 at 05:26:23PM +0200, David Marchand wrote:
> We don't need to include rte_bus.h in rte_devargs.h.
> Only a forward declaration of rte_bus and an inclusion of rte_dev.h are
> needed.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Bruce Richardson
On Thu, Jul 28, 2022 at 05:26:22PM +0200, David Marchand wrote:
> RTE_FUNC_PTR_OR_* macros have nothing to do with the rte_device object
> and associated API.
> Move them to rte_common.h and include it where needed.
>
> Signed-off-by: David Marchand
> ---
Personally, I really don't like these ma
On Thu, Jul 28, 2022 at 05:26:21PM +0200, David Marchand wrote:
> For any bus that does not support device iteration, rte_dev_iterator_init
> both returned an error code and logged an error message.
> An application (like testpmd) that only wants to list devices, would have
> no choice but to inspe
2022-07-28 11:41 (UTC-0400), Don Wallwork:
> On 7/28/2022 10:46 AM, Don Wallwork wrote:
> Let me try to give a trivial example of how it would work
> to make sure we're on the same page and then we can get
> back to details..
> >>> Thanks you, Don.
> >>> Now it's perfectly clear wh
On Thu, Jul 28, 2022 at 05:26:20PM +0200, David Marchand wrote:
> Rely on the generic device name rather than restrict to only supporting
> PCI devices.
>
> Signed-off-by: David Marchand
> ---
> examples/ethtool/lib/rte_ethtool.c | 17 ++---
> 1 file changed, 2 insertions(+), 15 dele
On Thu, Jul 28, 2022 at 05:26:19PM +0200, David Marchand wrote:
> addr and id fields are deprecated and are not used in the kni library.
> Stop populating them in the in-tree examples.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Bruce Richardson
On Thu, Jul 28, 2022 at 05:26:18PM +0200, David Marchand wrote:
> Those commands date back to the early stages of DPDK when only PCI
> devices were supported.
> At the time, developers may have used those commands to help in
> debugging their buggy^Wwork in progress drivers.
>
> Removing them, we
On Thu, Jul 28, 2022 at 05:26:15PM +0200, David Marchand wrote:
> Let's be good citizens and enforce not using driver only headers in our
> own examples.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Bruce Richardson
On 7/28/2022 10:46 AM, Don Wallwork wrote:
Let me try to give a trivial example of how it would work
to make sure we're on the same page and then we can get
back to details..
Thanks you, Don.
Now it's perfectly clear what EAL should do and we can discuss API.
I think this RFC is an excellent pr
Make rte_device opaque for non internal users.
This will make extending this object possible without breaking the ABI.
Some applications may have been dereferencing rte_device objects, mark
this object's accessors as stable.
Signed-off-by: David Marchand
---
Changes since RFC v2:
- updated relea
For diagnostic, it may be useful to provide a description of the device
with Bus specific information like, for PCI, the vendor and device ID.
Signed-off-by: David Marchand
---
app/test-pmd/config.c| 2 ++
drivers/bus/pci/bsd/pci.c| 8
drivers/bus/pci/bus_pci_drive
Prepare for making the device object opaque by adding accessors.
Update existing "external" users.
Signed-off-by: David Marchand
---
Changes since RFC v2:
- added rte_dev_devargs,
- updated testpmd,
---
app/proc-info/main.c | 6 +--
app/test-pmd/config.c | 18 +++
Make rte_driver opaque for non internal users.
This will make extending this object possible without breaking the ABI.
Introduce a new driver header and move rte_driver definition.
Update drivers and library to use the internal header.
Some applications may have been dereferencing rte_driver obje
Prepare for making the driver object opaque by adding accessors.
Update existing "external" users.
Internal users may still dereference a rte_driver object.
Signed-off-by: David Marchand
---
app/test-pmd/config.c | 2 +-
lib/eal/common/eal_common_dev.c | 6 ++
lib/eal/include/rte
Make rte_bus opaque for non internal users.
This will make extending this object possible without breaking the ABI.
Introduce a new driver header and move rte_bus definition and helpers.
Update drivers and library to use the internal header.
Some applications may have been dereferencing rte_bus o
Add helpers to get a rte_bus object details.
This will be used externally.
Internal users may still dereference a rte_bus object.
Signed-off-by: David Marchand
---
Changes since RFC v1:
- changed approach: only external users are updated,
---
app/test-pmd/config.c| 4 +-
iova enum definition does not need to be defined as part of the bus API.
Move it to rte_eal.h.
With this step, rte_eal.h does not depend on rte_bus.h and rte_dev.h.
Fix existing code that was relying on these implicit inclusions.
Signed-off-by: David Marchand
---
app/test-compress-perf/comp_perf
The vmbus bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup the code:
- fix indentation,
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions and ma
The vdev bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup the code:
- fix indentation,
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions and mac
The pci bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup the code:
- fix indentation,
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions and macr
The ifpga bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup the code:
- remove unneeded list head structure type,
- reorder the definitions and macro manipulating the bus singleton object,
- remove inclusion of rte_bus.h and fi
Remove unused symbols (exposed only in an internal header which
guarantees that no application out there relied on them).
Remove rte_ prefix and inline the rest to avoid having to expose them as
global symbols for a relatively small added value.
Signed-off-by: David Marchand
---
drivers/bus/ifp
The fslmc bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup internal structures:
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions and macro mani
The dpaa bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup internal structures:
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions and macro manip
The auxiliary bus interface is for drivers only.
Mark as internal and move the header in the driver headers list.
While at it, cleanup the code:
- fix indentation,
- remove unneeded reference to bus specific singleton object,
- remove unneeded list head structure type,
- reorder the definitions an
Same as for device classes, external DPDK users may need to include some
bus headers for their out of tree drivers.
Signed-off-by: David Marchand
---
drivers/meson.build | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/meson.build b/drivers/meson.build
index b22c2adda7..7449643fa9
RTE_FUNC_PTR_OR_* macros have nothing to do with the rte_device object
and associated API.
Move them to rte_common.h and include it where needed.
Signed-off-by: David Marchand
---
drivers/common/qat/qat_device.c| 1 +
drivers/compress/qat/qat_comp_pmd.c| 1 +
We don't need to include rte_bus.h in rte_devargs.h.
Only a forward declaration of rte_bus and an inclusion of rte_dev.h are
needed.
Signed-off-by: David Marchand
---
app/test/test_vdev.c | 1 +
lib/eal/include/rte_devargs.h | 4 +++-
2 files changed, 4 insertions(+), 1 deletion(-)
dif
For any bus that does not support device iteration, rte_dev_iterator_init
both returned an error code and logged an error message.
An application (like testpmd) that only wants to list devices, would have
no choice but to inspect a bus object to avoid spewing error logs.
Make those log messages de
Rely on the generic device name rather than restrict to only supporting
PCI devices.
Signed-off-by: David Marchand
---
examples/ethtool/lib/rte_ethtool.c | 17 ++---
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/examples/ethtool/lib/rte_ethtool.c
b/examples/ethtool/
Those commands date back to the early stages of DPDK when only PCI
devices were supported.
At the time, developers may have used those commands to help in
debugging their buggy^Wwork in progress drivers.
Removing them, we can drop the dependency on the PCI bus and library and
make testpmd bus agno
addr and id fields are deprecated and are not used in the kni library.
Stop populating them in the in-tree examples.
Signed-off-by: David Marchand
---
app/test/test_kni.c| 30 --
examples/ip_pipeline/kni.c | 10 --
2 files changed, 40 deletions(-)
dif
There is no in-tree user for this accessor that returns the PCI bus
object.
On the other hand, a bus object can be retrieved by name using
rte_bus_find_by_name.
We can remove this driver specific API.
Signed-off-by: David Marchand
---
Changes since RFC v2:
- updated release notes,
---
doc/guide
Rely on a local flag rather than dereference a bus object.
This will help next commits.
Signed-off-by: David Marchand
---
drivers/common/mlx5/linux/mlx5_common_auxiliary.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/common/mlx5/linux/mlx5_common_auxilia
Let's be good citizens and enforce not using driver only headers in our
own examples.
Signed-off-by: David Marchand
---
devtools/checkpatches.sh | 8
1 file changed, 8 insertions(+)
diff --git a/devtools/checkpatches.sh b/devtools/checkpatches.sh
index 1edc5810ad..0e82c01947 100755
---
This is a PoC for hiding the rte_bus, rte_driver and rte_device objects.
And mark associated driver only API as internal.
A good amount of the patches are preparation work on rte_bus.h,
rte_dev.h, rte_devargs.h and rte_eal.h headers, removing dependencies
between them.
PCI bus specific handling a
> -Original Message-
> From: Richardson, Bruce
> Sent: Thursday, July 28, 2022 9:22 AM
> To: Dumitrescu, Cristian
> Cc: dev@dpdk.org; R, Kamalakannan
> Subject: Re: [PATCH V5 01/17] pipeline: add pipeline name
>
> On Wed, Jul 27, 2022 at 11:01:16PM +, Cristian Dumitrescu wrote:
>
Update the example CLI scripts with the commands for code generation
and shared object library build.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/examples/ethdev.io | 27 +
examples/pipeline/examples/fib.cli| 44 +++---
Remove the tap CLI command, as the file descriptor I/O ports of the
pipeline are now configured trough the I/O specification file.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 37 -
examples/pipeline/obj.c | 89 -
Rework the link CLI command for better alignment with the naming
conventions used in the pipeline I/O specification file. Use the
library linked list of devices and remove the application list.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 81 ++
Convert the CLI commands to use the pipeline name query API and remove
the linked list of pipeline objects previously maintained by the
application.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c| 268 -
exam
The pipeline mirroring configuration is done through the I/O
specification file, so this CLI command is no longer needed.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 76 -
1 file changed, 76 deletions(-)
The pipeline I/O ports configuration is now done through the I/O
specification file, hence these CLI commands are no longer needed.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 465
1 file changed, 465 d
The pipeline configuration is now done through the I/O specification
file, hence this CLI command is no longer needed.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 47 -
1 file changed, 47 deletions(-)
d
Add CLI command for the shared object library build operation.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 157 +++-
1 file changed, 153 insertions(+), 4 deletions(-)
diff --git a/examples/pipeline/cli.c b/
Add CLI command for the pipeline code generation operation.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/cli.c | 61 +
1 file changed, 61 insertions(+)
diff --git a/examples/pipeline/cli.c b/examples/pipeline/cl
Previously, the pipeline build operation was done based on the
specification file (typically produced by the P4 compiler), then the C
code with optimized functions for the pipeline actions and
instructions was generated, built into a shared object library, loaded
and installed into the pipeline in
Previously, the C code generation for the pipeline was hidden under
the hood; now, we make this an explicit API operation. Besides the
functions for the pipeline actions and the pipeline instructions,
the generated C source code now includes the pipeline specification
structure required for the pip
Add specification data structure and API for the pipeline I/O ports
and related pipeline configuration such as packet mirroring.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
lib/pipeline/rte_swx_pipeline_spec.c | 852 +++
lib/pipeline/rte_swx_pip
Add support to export the pipeline specification data structure to a C
source code file.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
lib/pipeline/rte_swx_pipeline_spec.c | 622 +++
lib/pipeline/rte_swx_pipeline_spec.h | 5 +
2 files changed, 6
Rework the specification file-based pipeline build operation to first
parse the specification file into the previously introduced pipeline
specification data structure, then use this structure to configure
and build the pipeline.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
Add specification data structure for the entire pipeline.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
lib/pipeline/rte_swx_pipeline_spec.c | 21 ++
lib/pipeline/rte_swx_pipeline_spec.h | 32
2 files changed, 53 insertions(+)
Move all the pipeline object specification data structures to an
internal header file.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
lib/pipeline/rte_swx_pipeline_spec.c | 126 +--
lib/pipeline/rte_swx_pipeline_spec.h | 176 +++
2
Add an unique name to every pipeline. This enables the library to
maintain a list of the existing pipeline objects, which can be
queried by the application.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Kamalakannan R.
---
examples/pipeline/obj.c | 2 +-
lib/pipeline/rte_
The sequence of steps to build a SWX pipeline object are:
1. Create the specification file (pipeline.p4 -> pipeline.spec): This text file
represents the
pipeline program that is typically generated by the P4 compiler or sometimes
manually created.
2. Generate the C source code file (pipeline
On 7/28/2022 7:29 AM, Morten Brørup wrote:
Let me try to give a trivial example of how it would work
to make sure we're on the same page and then we can get
back to details..
Thanks you, Don.
Now it's perfectly clear what EAL should do and we can discuss API.
I think this RFC is an excellent pr
ESP is one of IPSec protocols over both IPv4 and IPv6 and is considered
a tunnel layer that cannot be followed by any other layer. Taking that
into consideration, esp is considered as a 4 layer.
Not defining ESP's priority will make it match with the same priority as
its prior IP layer, which has
Hi Eli,
In RTE flow there is support for group_id attribute(u32).
Similar to the example you gave-
testpmd> flow create 0 group 0x1234 ingress pattern eth / end actions
count / drop / end
Please check if it fits the requirement.
Regards
Aman
On 7/20/2022 2:14 AM, Thomas Monjalon wrote:
+Cc
I know it's too late for this patch [1], but I am afraid it was performance
tested using test-pmd with RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE [2], so the test
didn't involve the function fixed by this patch: rte_pktmbuf_prefree_seg().
If so, the performance test results were completely irrelevant. Th
On 7/14/22 16:48, abhimanyu.sa...@xilinx.com wrote:
From: Abhimanyu Saini
In SW assisted live migration, vDPA driver will stop all virtqueues
and setup up SW vrings to relay the communication between the
virtio driver and the vDPA device using an event driven relay thread
This will allow vDPA
> -Original Message-
> From: Luca Boccassi
> Sent: Thursday, July 28, 2022 3:34 PM
> To: sta...@dpdk.org
> Cc: dev@dpdk.org; Abhishek Marathe ;
> Ali Alnubani ; benjamin.wal...@intel.com; David
> Christensen ; Hemant Agrawal
> ; Ian Stokes ; Jerin
> Jacob ; John McNamara ;
> Ju-Hyoung Lee
On Mon, 2022-07-18 at 10:58 +0100, luca.bocca...@gmail.com wrote:
> Hi all,
>
> Here is a list of patches targeted for stable release 21.11.2.
>
> The planned date for the final release is August 29th.
>
> Please help with testing and validation of your use cases and report
> any issues/results
On Thu, Jul 28, 2022 at 11:36 AM Andrew Rybchenko
wrote:
>
> Remove deprecated macros.
>
> Andrew Rybchenko (3):
> mempool: remove deprecated macro to calculate header size
> mempool: remove deprecated macro to register mempool driver
> mempool: remove deprecated unused defines
>
> doc/guid
On 7/14/22 16:48, abhimanyu.sa...@xilinx.com wrote:
From: Abhimanyu Saini
Please, try to stick to one E-mail address.
Change cidx and pidx definition to mean used queue and avail
queue index respectively.
Signed-off-by: Abhimanyu Saini
Acked-by: Andrew Rybchenko
On 7/14/22 16:48, abhimanyu.sa...@xilinx.com wrote:
From: Abhimanyu Saini
Regenerate MCDI headers from smartnic_registry:72940ad
since smartnic_registry is not publicly available, it does not make
sense to refer to its changeset in the patch description.
I think you should put update goal her
> From: Dmitry Kozlyuk [mailto:dmitry.kozl...@gmail.com]
> Sent: Thursday, 28 July 2022 09.26
>
> 2022-07-27 17:43 (UTC-0400), Don Wallwork:
> > On 7/27/2022 4:36 PM, Dmitry Kozlyuk wrote:
> > > I now understand more about_why_ you want this feature
> > > but became less confident_what_ do you'r
On 7/14/22 16:48, abhimanyu.sa...@xilinx.com wrote:
From: Abhimanyu Saini
Increase the number to default RX/TX queue pairs to 8,
and add MQ feature flag to vDPA protocol features.
Signed-off-by: Abhimanyu Saini
Acked-by: Andrew Rybchenko
I don't think the stack(software) cares if the length of a packet is less than
60, especially when receiving it.
In the linux kernel, if the packet length does not match the IP total-length,
the packet is flushed to the stack instead of GRO. The previous GRO cache
packets are also flushed into t
From: Stanisław Kardach [mailto:k...@semihalf.com]
Sent: Thursday, 28 July 2022 00.02
> On Wed, 27 Jul 2022, 21:53 Honnappa Nagarahalli,
> wrote:
>
> > > > > > Yes, x86 needs 16B alignment for NT load/stores But that's
> > > supposed
> > > > > to be arch
> > > > > > specific limitation, that we
The script is an interface to run DTS.
Signed-off-by: Owen Hilyard
Signed-off-by: Juraj Linkeš
---
dts/main.py | 24
1 file changed, 24 insertions(+)
create mode 100755 dts/main.py
diff --git a/dts/main.py b/dts/main.py
new file mode 100755
index 00..a70070765
The module implements methods needed to run DTS. It handles the creation
of objects and eventually the whole DTS workflow, such as running node
setups, test gathering, setup and execution and various cleanups.
Signed-off-by: Owen Hilyard
Signed-off-by: Juraj Linkeš
---
dts/framework/dts.py
The base class implements basic node management methods - connect and
execute commands.
Signed-off-by: Owen Hilyard
Signed-off-by: Juraj Linkeš
---
dts/framework/node.py | 99 +++
dts/framework/settings.py | 11 +
2 files changed, 110 insertions(+)
c
From: Owen Hilyard
The configuration is split into two parts, one defining the parameters
of the test run and the other defining the topology to be used.
The format of the configuration is YAML. It is validated according to a
json schema which also servers as detailed documentation of the variou
The class adds logging and history records to existing pexpect methods.
Signed-off-by: Owen Hilyard
Signed-off-by: Juraj Linkeš
---
dts/framework/ssh_connection.py | 70 +
1 file changed, 70 insertions(+)
create mode 100644 dts/framework/ssh_connection.py
diff
The library uses the pexpect python library and implements connection to
a node and two ways to interact with the node:
1. Send a string with specified prompt which will be matched after
the string has been sent to the node.
2. Send a command to be executed. No prompt is specified here.
Signed-
The logging module provides loggers distinguished by two attributes,
a custom format and a verbosity switch. The loggers log to both console
and more verbosely to files.
Signed-off-by: Owen Hilyard
Signed-off-by: Juraj Linkeš
---
dts/framework/__init__.py | 3 +
dts/framework/logger.py | 12
The Dockerfile contains basic image for CI and developers. There's also
an integration of the Dockerfile with Visual Studio.
The formatter script uses Black and Isort to format the Python code.
Signed-off-by: Owen Hilyard
Signed-off-by: Juraj Linkeš
---
dts/.devcontainer/devcontainer.json | 30
.gitignore contains standard Python-related files.
Apart from that, add configuration for Python tools used in DTS:
Poetry, dependency and package manager
Black, formatter
Pylama, static analysis
Isort, import sorting
.editorconfig modifies the line length to 88, which is the default Black
uses.
1 - 100 of 111 matches
Mail list logo
