This doc describes how to enable DPDK on Openwrt in both virtual and
physical x86 environment.
Signed-off-by: Xiaolong Ye
---
V5 changes:
1. improve the doc's grammar and wording according to John's
suggestions.
V4 changes:
1. add release notes
V3 changes:
1. emphasize target select in `Ope
Hi, John
Thanks for your review.
On 01/17, Mcnamara, John wrote:
>Hi,
>
>Overall looks good. Some notes below:
>
>> +
>> +.. note::
>> +
>> +For compiling the NUMA lib, run libtool --version to ensure the
>> +libtool version >= 2.2, otherwise the compilation will fail with errors.
>
>
>This conti
I think this is a good feature to have.
>
> 2 thoughts on this, and adding dev list to discussion.
>
> 1) This would only apply to make builds, I think. Any internal headers should
> not be passed to the "headers = " line in meson which tracks headers for
> installation only (all headers are fo
17/01/2020 16:48, Honnappa Nagarahalli:
> > > > > On 1/16/20 2:11 PM, Jerin Jacob wrote:
> > > > > > On Thu, Dec 12, 2019 at 6:42 PM wrote:
> > > > > >> --- a/config/common_linux
> > > > > >> +++ b/config/common_linux
> > > > > >> @@ -8,9 +8,7 @@ CONFIG_RTE_EXEC_ENV_LINUX=y
> > > > > >> CONFIG_RTE
15/01/2020 16:50, Akhil Goyal:
> http://dpdk.org/git/next/dpdk-next-crypto
Pulled, except "compress test vector replacement", thanks
>-Original Message-
>> > Not sure how users could make use of this.
>> > But the abi check flags this as a breakage since this type was exported.
>> I think this is a false positive.
>>
>> Users include 'rte_hash.h' file which does not define the structure. It just
>> has the declaration '
15/01/2020 16:54, Akhil Goyal:
> > >
> > > This patch replaces existing test vector with a new
> > > one containing C code to fix license issue.
> > >
> > > Signed-off-by: Artur Trybula
> > Acked-by: Fiona Trahe
> Applied to dpdk-next-crypto
This is a terrible idea.
Adding C code in a C project
On Fri, Jan 17, 2020 at 8:02 PM Neil Horman wrote:
> HmmmI wonder if its worth orchestrating the build system to use a git
> submodule storing whatever our CI system can produce and using it as a cache,
> and falling back to a local build if the appropriate arch isn't found? That
> might offe
On Fri, Jan 17, 2020 at 9:54 PM Honnappa Nagarahalli
wrote:
>
>
>
> >
> > On Thu, Jan 16, 2020 at 6:25 AM Honnappa Nagarahalli
> > wrote:
> > >
> > > The freelist and external bucket indices are 32b. Using rings that use
> > > 32b element sizes will save memory.
> > >
> > > Signed-off-by: Honnap
>
> On Thu, Jan 16, 2020 at 6:25 AM Honnappa Nagarahalli
> wrote:
> >
> > The freelist and external bucket indices are 32b. Using rings that use
> > 32b element sizes will save memory.
> >
> > Signed-off-by: Honnappa Nagarahalli
> > Reviewed-by: Gavin Hu
> > Reviewed-by: Ola Liljedahl
> > --
On Fri, Jan 17, 2020 at 3:58 PM Olivier Matz wrote:
>
> Starting from v20.05, rte_mempool_populate_iova() will return
> -ENOBUFS. The ABI will be preserved through symbol versioning until
This deprecation notice describes a 0 return value, not -ENOBUFS.
> 20.11.
>
> Signed-off-by: Olivier Matz
On Fri, Jan 17, 2020 at 3:58 PM Olivier Matz wrote:
>
> Change rte_mempool_populate_iova() and rte_mempool_populate_iova() to
> return 0 instead of -EINVAL when there is not enough room to store one
> object, as it can be helpful for applications to distinguish this
> specific case.
>
> As this is
On Thu, Jan 16, 2020 at 6:25 AM Honnappa Nagarahalli
wrote:
>
> The freelist and external bucket indices are 32b. Using rings
> that use 32b element sizes will save memory.
>
> Signed-off-by: Honnappa Nagarahalli
> Reviewed-by: Gavin Hu
> Reviewed-by: Ola Liljedahl
> ---
> lib/librte_hash/rte_
The following changes since commit 46090d658ebfc8b17bd2ccc7d1513cd6d81b5544:
event/opdl: use new API to save cycles on aarch64 (2020-01-17 12:02:21 +0100)
are available in the Git repository at:
http://dpdk.org/git/next/dpdk-next-net for-main-repo
for you to fetch changes up to 8c5f9f273712
On 1/17/2020 11:55 AM, Xiaoyu Min wrote:
> Introduce an API which dump the device's internal representation
> information of rte flows in hardware.
>
> Signed-off-by: Xiaoyu Min
> Acked-by: Ori Kam
Reviewed-by: Ferruh Yigit
On 1/17/2020 11:55 AM, Xiaoyu Min wrote:
> This serial patchs is based on RFC [1] and the comments on it.
>
> A new rte flow API is added to dump device internal representation
> information for the offloaded rte flows.
>
> This is very helpful for user and developer to debug flow offloading
> st
From: Jerin Jacob
Based on the techboard meeting held on 2019-11-06,
It's been decided to disable all kmods by default from v20.02.
http://mails.dpdk.org/archives/dev/2019-November/151763.html
Signed-off-by: Jerin Jacob
---
v2:
- Updated meson config file to disable the kmods by default(Maxime
On Thu, Jan 16, 2020 at 09:01:52PM +0100, Thomas Monjalon wrote:
> 16/01/2020 19:49, Neil Horman:
> > On Thu, Jan 16, 2020 at 03:20:48PM +0100, Thomas Monjalon wrote:
> > > 16/01/2020 12:52, Neil Horman:
> > > > On Wed, Jan 15, 2020 at 01:38:17PM +0100, Thomas Monjalon wrote:
> > > > > 15/01/2020 1
+static __rte_always_inline void
+enqueue_elems_128(struct rte_ring *r, uint32_t prod_head,
+ const void *obj_table, uint32_t n)
+{
+ unsigned int i;
+ const uint32_t size = r->size;
+ uint32_t idx = prod_head & r->mask;
+ rte_int128_t *ring = (rte_int128_t *)
Bruce Richardson writes:
> On Fri, Jan 17, 2020 at 08:34:01AM -0500, Aaron Conole wrote:
>> Bruce Richardson writes:
>>
>> > On Fri, Jan 17, 2020 at 08:16:55AM -0500, Aaron Conole wrote:
>> >> Bruce Richardson writes:
>> >>
>> >> > When werror is set for the build, we should pass that flag th
On 1/17/2020 8:33 AM, Andrew Rybchenko wrote:
> On 1/17/20 5:08 AM, Fang TongHao wrote:
>
> Summary does not comply with [1].
>
>> Fixes the secondary process changed shared memory
>> in "rte_eth_copy_pci_info" function.In that function
>> only primary can update the value of "eth_dev->data"
>> w
On Fri, Jan 17, 2020 at 08:34:01AM -0500, Aaron Conole wrote:
> Bruce Richardson writes:
>
> > On Fri, Jan 17, 2020 at 08:16:55AM -0500, Aaron Conole wrote:
> >> Bruce Richardson writes:
> >>
> >> > When werror is set for the build, we should pass that flag through to
> >> > sphinx so that it c
On Thu, Jan 16, 2020 at 08:16:55AM +0100, Thomas Monjalon wrote:
> If a dependency is required for a driver build,
> but should not be exposed to the application (via pkg-config),
> it can be declared in the array hidden_deps.
>
> The hidden_deps are used as internal dependencies,
> when building
Hi,
I observed that DPDK kni interface is not able to receive solicited
node multicast addressed (derived from the IPv6 address configured on
vEth0 interface) when the kni sample app is being run in Ubuntu VM on
ESXI server. The only way to have the frames reach the kni interface
is by setting t
On 1/17/2020 2:59 PM, Viacheslav Ovsiienko wrote:
> The mlx5 datapath does not implement any endianness conversions
> for the metadata being sent and received to provide the better
> performance (because these conversions would be performed for
> each packet). These metadata are also involved into
On 1/17/2020 11:01 AM, Viacheslav Ovsiienko wrote:
> Matcher is flow table related structure providing the flow pattern
> to be translated directly in hardware controlling data. Some fields
> in this structure might be split (by software) between multiple items.
>
> For example, the metadata regis
On 1/17/2020 11:16 AM, Viacheslav Ovsiienko wrote:
> Matcher is flow table related structure providing the flow pattern
> to be translated directly in hardware controlling data. This structure
> includes the metadata register c0 field, that might be engaged to
> support META and MARK related flow i
On Wed, Jan 15, 2020 at 11:25:05PM -0600, Honnappa Nagarahalli wrote:
> The current rte_ring hard-codes the type of the ring element to 'void *',
> hence the size of the element is hard-coded to 32b/64b. Since the ring
> element type is not an input to rte_ring APIs, it results in couple
> of issue
On Wed, Jan 15, 2020 at 11:25:09PM -0600, Honnappa Nagarahalli wrote:
> Adjust the performance test cases to test rte_ring_xxx_elem APIs.
>
> Signed-off-by: Honnappa Nagarahalli
> Reviewed-by: Gavin Hu
> ---
> app/test/test_ring_perf.c | 454 +++---
> 1 file chan
Hi Akhil,
Indeed with our tests we also seeing ~15% perf drop for small packets
(~90B) and ~3-4% drop for 1KB packets. While I am looking on a ways to
minimize the drop, I think it would be hard, if possible at all to
eliminate it completely.
Reason for that: current SAD implementation is comp
On Wed, Jan 15, 2020 at 11:25:08PM -0600, Honnappa Nagarahalli wrote:
> Add basic infrastructure to test rte_ring_xxx_elem APIs.
> Adjust the existing test cases to test for various ring
> element sizes.
>
> Signed-off-by: Honnappa Nagarahalli
> Reviewed-by: Gavin Hu
> ---
> app/test/test_ring.
>
> Hi Honnappa,
Thanks Olivier for your review, appreciate your feedback.
>
> On Wed, Jan 15, 2020 at 11:25:07PM -0600, Honnappa Nagarahalli wrote:
> > Current APIs assume ring elements to be pointers. However, in many use
> > cases, the size can be different. Add new APIs to support configur
On 12/22/2019 5:55 PM, Stephen Hemminger wrote:
> This fixes a deadlock when using KNI with bifurcated drivers.
> Bringing kni device up always times out when using Mellanox
> devices.
>
> The kernel KNI driver sends message to userspace to complete
> the request. For the case of bifurcated driver
Hi Thomas,
Can this patch be merged?
> On Nov 25, 2019, at 5:14 PM, Thomas Monjalon wrote:
>
> 25/11/2019 23:55, Dharmik Thakkar:
>>
>>> On Nov 25, 2019, at 4:44 PM, Thomas Monjalon wrote:
>>>
>>> 25/11/2019 23:02, Wang, Yipeng1:
From: Thomas Monjalon [mailto:tho...@monjalon.net]
> 25/1
On 2020-01-17 14:34, Jerin Jacob wrote:
> On Fri, Jan 17, 2020 at 5:44 PM David Marchand
> wrote:
>> On Thu, Jan 16, 2020 at 5:36 PM Honnappa Nagarahalli
>> wrote:
>>> I see that the none of the CIs (except Travis) have run on this patch.
>>> Intel CI has reported a compilation error and I fixed
Hi Honnappa,
On Wed, Jan 15, 2020 at 11:25:07PM -0600, Honnappa Nagarahalli wrote:
> Current APIs assume ring elements to be pointers. However, in many
> use cases, the size can be different. Add new APIs to support
> configurable ring element sizes.
>
> Signed-off-by: Honnappa Nagarahalli
> Rev
> On Fri, Jan 17, 2020 at 3:29 PM Honnappa Nagarahalli
> wrote:
> > > - Pushed the series
> > > https://patchwork.dpdk.org/project/dpdk/list/?series=8147 to a
> > > branch of mine for checks.
> > > Travis reports:
> > > "
> > > [2155/2156] Compiling C object 'app/te...st@@dpdk-
> > > test@exe/te
15/01/2020 16:54, Akhil Goyal:
> > >
> > > Fixes: c0f87eb5252b ("cryptodev: change burst API to be crypto op
> > > oriented")
> > > Fixes: 77a217a19bb7 ("test/crypto: add AES-CCM tests")
> > >
> > > Signed-off-by: Adam Dybkowski
> > > ---
> > > app/test/test_cryptodev.c | 96
On Fri, Jan 17, 2020 at 3:29 PM Honnappa Nagarahalli
wrote:
> > - Pushed the series
> > https://patchwork.dpdk.org/project/dpdk/list/?series=8147 to a branch of
> > mine for checks.
> > Travis reports:
> > "
> > [2155/2156] Compiling C object 'app/te...st@@dpdk-
> > test@exe/test_ring_perf.c.o'.
>
On Fri, Jan 17, 2020 at 3:42 PM Jerin Jacob wrote:
>
> On Thu, Jan 16, 2020 at 10:56 AM Honnappa Nagarahalli
> wrote:
> >
> > Use custom element size ring APIs to replace event ring
> > implementation. This avoids code duplication.
> >
> > Signed-off-by: Honnappa Nagarahalli
> > Reviewed-by: Gav
>
> Minutes 16 January 2020
> ---
>
> Agenda:
> * Release Dates
> * Subtrees
> * OvS
>
> Participants:
> * Debian/Microsoft
> * Intel
> * Marvell
> * Mellanox
> * NXP
> * Red Hat
>
>
> Release Dates
> -
>
> * v20.02 dates:
> * Integration deadline passed, it
> >
> >
> > > >
> > > > On 1/16/20 2:11 PM, Jerin Jacob wrote:
> > > > > Ping.
> > > > >
> > > > > On Thu, Dec 12, 2019 at 6:42 PM wrote:
> > > > >>
> > > > >> From: Jerin Jacob
> > > > >>
> > > > >> Based on the techboard meeting held on 2019-11-06, It's been
> > > > >> decided to disable all k
Raslan, Ferruh
In these patches (merged in mlx5 tree) just the commit messages are updated:
http://patchwork.dpdk.org/patch/64831/
http://patchwork.dpdk.org/patch/64832/
http://patchwork.dpdk.org/patch/64852/
With best regards, Slava
> -Original Message-
> From: Ferruh Yigit
> Sent: W
The mlx5 datapath does not implement any endianness conversions
for the metadata being sent and received to provide the better
performance (because these conversions would be performed for
each packet). These metadata are also involved into flow processing
(there might be some flows matching on met
Starting from v20.05, rte_mempool_populate_iova() will return
-ENOBUFS. The ABI will be preserved through symbol versioning until
20.11.
Signed-off-by: Olivier Matz
---
doc/guides/rel_notes/deprecation.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/rel_notes/deprecation.
Change rte_mempool_populate_iova() and rte_mempool_populate_iova() to
return 0 instead of -EINVAL when there is not enough room to store one
object, as it can be helpful for applications to distinguish this
specific case.
As this is an ABI change, use symbol versioning to preserve old
behavior for
rte_mempool_populate_virt() sometimes fail, when it calls
rte_mempool_populate_iova() with an area which is too small to store one
object. This should not be an error.
I prepared a v2 which implements an ABI compatibility through symbol
versioning, as suggested [1]. It looks a bit overkill to me,
To populate a mempool with a virtual area, the mempool code calls
rte_mempool_populate_iova() for each iova-contiguous area. It happens
(rarely) that this area is too small to store one object. In this case,
rte_mempool_populate_iova() returns an error, which is forwarded by
rte_mempool_populate_vi
> -Original Message-
> From: Dybkowski, AdamX
> Sent: Friday, January 17, 2020 2:47 PM
> To: dev@dpdk.org; Trahe, Fiona ; akhil.go...@nxp.com
> Cc: Dybkowski, AdamX
> Subject: [PATCH v6 0/3] Refactor crypto unit tests.
>
> This patch set is a first step to refactor the overly complex
This patch refactors most of unit tests to be contained in one
combined array, and run depending on the PMD capabilities instead of
providing multiple array with tests for individual PMDs.
Only a subset of unit tests was merged into one array - it combines
all tests originally meant to be run on th
This patch set is a first step to refactor the overly complex symmetric
crypto unit tests. It merges many separate arrays of the tests
for these PMDs: null, aesni_mb, aesni_gcm, openssl, qat, sw_snow3g,
sw_kasumi, sw_zuc into one big array that's then used when running
unit tests on these PMDs.
In
This patch gets rid of individual functions that all call
test_blockcipher_all_tests separately for every PMD and instead
provides just one set universal for all PMDs that's basing on the
driver id from the global variable gbl_driver_id.
Signed-off-by: Adam Dybkowski
---
app/test/test_cryptodev.
This patch adds capability checks to many tests meant to be run
in the future on various PMDs. This way the code is prepared for
more thorough refactoring in order to create one big central
unit tests array.
Signed-off-by: Adam Dybkowski
---
app/test/test_cryptodev.c | 635 ++
On Thu, Jan 16, 2020 at 10:56 AM Honnappa Nagarahalli
wrote:
>
> Use custom element size ring APIs to replace event ring
> implementation. This avoids code duplication.
>
> Signed-off-by: Honnappa Nagarahalli
> Reviewed-by: Gavin Hu
> Reviewed-by: Ola Liljedahl
Please change the subject to eve
>
> >
> > On Thu, Jan 16, 2020 at 5:36 PM Honnappa Nagarahalli
> > wrote:
> > >
> > > I see that the none of the CIs (except Travis) have run on this
> > > patch. Intel CI
> > has reported a compilation error and I fixed it in this version. Does
> > anyone know if/when the CI will run on the pat
On 1/17/2020 2:31 PM, Ferruh Yigit wrote:
> On 1/17/2020 11:49 AM, Wei Hu (Xavier) wrote:
>> From: "Wei Hu (Xavier)"
>>
>> Currently, there is a potential problem that changing the content of
>> dev->data->dev_conf.rxmode.offloads even when there is no vlan_offload_set
>> driver callback.
>>
>> It
The documentation says that a negative errno is returned on error, but
in most places that's not the case.
Fix the documentation and the exceptions in code. The second one
(return from populate_virt) also fixes a memory leak.
Note that testpmd was using the function correctly.
Fixes: aa10457eb4c
On 1/17/2020 11:49 AM, Wei Hu (Xavier) wrote:
> From: "Wei Hu (Xavier)"
>
> Currently, there is a potential problem that changing the content of
> dev->data->dev_conf.rxmode.offloads even when there is no vlan_offload_set
> driver callback.
>
> It is a good idea that prevent the side effect and
>
> On Thu, Jan 16, 2020 at 5:36 PM Honnappa Nagarahalli
> wrote:
> >
> > I see that the none of the CIs (except Travis) have run on this patch.
> > Intel CI
> has reported a compilation error and I fixed it in this version. Does anyone
> know if/when the CI will run on the patches?
>
> - Pus
> -Original Message-
> From: Akhil Goyal
> Sent: Friday, January 17, 2020 20:12
> To: Ruifeng Wang ; Gavin Hu
>
> Cc: dev@dpdk.org; jer...@marvell.com; Honnappa Nagarahalli
> ; nd
> Subject: RE: [dpdk-dev] [RFC PATCH 0/3] armv8 crypto PMD update
>
> Hi Ruifeng,
>
> > Maintainance of
Hi,
Overall looks good. Some notes below:
> +
> +.. note::
> +
> + For compiling the NUMA lib, run libtool --version to ensure the
> +libtool version >= 2.2, otherwise the compilation will fail with errors.
This continuation line isn't indented and raised a build error:
$ make doc-guid
On Fri, Jan 17, 2020 at 5:44 PM David Marchand
wrote:
>
> On Thu, Jan 16, 2020 at 5:36 PM Honnappa Nagarahalli
> wrote:
> >
> > I see that the none of the CIs (except Travis) have run on this patch.
> > Intel CI has reported a compilation error and I fixed it in this version.
> > Does anyone kn
Bruce Richardson writes:
> On Fri, Jan 17, 2020 at 08:16:55AM -0500, Aaron Conole wrote:
>> Bruce Richardson writes:
>>
>> > When werror is set for the build, we should pass that flag through to
>> > sphinx so that it can flag warnings as errors too.
>> >
>> > Signed-off-by: Bruce Richardson
>
2 thoughts on this, and adding dev list to discussion.
1) This would only apply to make builds, I think. Any internal headers should
not be passed to the "headers = " line in meson which tracks headers for
installation only (all headers are found at build time in their original source
locations
On Fri, Jan 17, 2020 at 08:16:55AM -0500, Aaron Conole wrote:
> Bruce Richardson writes:
>
> > When werror is set for the build, we should pass that flag through to
> > sphinx so that it can flag warnings as errors too.
> >
> > Signed-off-by: Bruce Richardson
> > ---
>
> I see that this actuall
On 1/16/2020 6:49 PM, Dekel Peled wrote:
> This patch adds CLI option to enter the msg_type value for GTP
> flow pattern item.
>
> Signed-off-by: Dekel Peled
> Acked-by: Ori Kam
Applied to dpdk-next-net/master, thanks.
On 1/16/2020 7:38 PM, Ferruh Yigit wrote:
> On 1/16/2020 4:19 PM, Viacheslav Ovsiienko wrote:
>> The maximum amount of unique swutching domain is supposed
>> to be equal RTE_MAX_ETHPORTS. Current implementation allows
>> to allocate only RTE_MAX_ETHPORTS-1 domains.
>>
>> The definition of RTE_ETH_D
Bruce Richardson writes:
> When werror is set for the build, we should pass that flag through to
> sphinx so that it can flag warnings as errors too.
>
> Signed-off-by: Bruce Richardson
> ---
I see that this actually works to generate the errors... BUT
if we merge this it will break the build.
On Wed, Dec 18, 2019 at 7:00 AM Joyce Kong wrote:
>
> There are a lot functions of bit operations scattered in PMDs, consolidate
> them into a common API family and applied in different PMDs to reduce code
> duplication.
>
> v6:
> Trim 'unsigned long' in PMDs down to 'uint32_t', as on mainstream
Hi Vladimir,
The lookup logic for SAD has been brought more closer to real use case, but it
looks very high on CPU and should be optimized. We cannot have 10-15% drop
because of this change in SA lookup for small packet(82B) sizes where CPU is
bottleneck. For large packet sizes it will not impa
On Thu, Jan 16, 2020 at 5:36 PM Honnappa Nagarahalli
wrote:
>
> I see that the none of the CIs (except Travis) have run on this patch. Intel
> CI has reported a compilation error and I fixed it in this version. Does
> anyone know if/when the CI will run on the patches?
- Pushed the series
https
Hi Ruifeng,
> Maintainance of armv8_crypto library created by Marvell/Cavium will
> be discontinued. Going forward, Armv8 crypto PMD will link to AArch64
> crypto library hosted by Arm.
>
> Patch 1/3, 2/3 update source code and document respectively to reflect
> the change.
> Patch 3/3 fixed Clan
From: Xueming Li
Guide of mlx5 is updated on how to dump HW flows.
Signed-off-by: Xueming Li
Signed-off-by: Xiaoyu Min
---
doc/guides/nics/mlx5.rst | 28
1 file changed, 28 insertions(+)
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 92228
From: Xueming Li
Add pmd unix socket server to enable external tool applications to
trigger flow dump.
Socket path:
/var/tmp/dpdk_mlx5_
Socket format:
io_raw: port_id of uint16
file: file descriptor of int
Signed-off-by: Xueming Li
Signed-off-by: Xiaoyu Min
Acked-by: V
From: Xueming Li
New flow dump CLI to dump device internal representation information
of flows into screen.
Signed-off-by: Xueming Li
Signed-off-by: Xiaoyu Min
Acked-by: Ori Kam
---
app/test-pmd/cmdline_flow.c | 91 +
app/test-pmd/config.c | 27 +
Dump fdb/nic_rx/nic_tx raw flow data into specified file.
Signed-off-by: Xueming Li
Signed-off-by: Xiaoyu Min
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/Makefile | 7 ++-
drivers/net/mlx5/meson.build | 4 +++-
drivers/net/mlx5/mlx5.h | 4
drivers/net
Introduce an API which dump the device's internal representation
information of rte flows in hardware.
Signed-off-by: Xiaoyu Min
Acked-by: Ori Kam
---
lib/librte_ethdev/rte_ethdev_version.map | 3 +++
lib/librte_ethdev/rte_flow.c | 16
lib/librte_ethdev/rte_flow.h
This serial patchs is based on RFC [1] and the comments on it.
A new rte flow API is added to dump device internal representation
information for the offloaded rte flows.
This is very helpful for user and developer to debug flow offloading
stuff, i.e, to check whether PMD offloads the rte flow in
On Fri, 2020-01-17 at 11:22 +, Bruce Richardson wrote:
> For the doxygen API for the examples folder, we can generate a build
> dependency file when we generate the examples.dox file. This allows
> correct rebuilds if the files in examples change.
>
> Signed-off-by: Bruce Richardson <
> bruce.
On Fri, 2020-01-17 at 11:22 +, Bruce Richardson wrote:
> Generate a dependency file for the header files used in the API guide
> so that the docs can be rebuilt if a header changes.
>
> Signed-off-by: Bruce Richardson <
> bruce.richard...@intel.com
> >
> ---
> doc/api/generate_doxygen.sh | 5
From: "Wei Hu (Xavier)"
Currently, there is a potential problem that changing the content of
dev->data->dev_conf.rxmode.offloads even when there is no vlan_offload_set
driver callback.
It is a good idea that prevent the side effect and make the API return
success if no change requested. This pat
Hi,Andrew Rybchenko
On 2020/1/17 15:57, Andrew Rybchenko wrote:
On 1/17/20 9:23 AM, Wei Hu (Xavier) wrote:
From: "Wei Hu (Xavier)"
I think summary is misleading. It is too generic and does not
mention VLAN offloads. Consider something like:
ethdev: fix VLAN offloads set if no driver callback
Hi lads,
> > * next-net-crypto
> > * Pull request sent
> > * There is a performance concern on some ipsec-gw patches,
> > they can go in -rc2 if the issue is solved
> > * CPU crypto from last release may be breaking ABI, need to confirm
>
> AFAIK, there is n
When building doxygen docs as part of the build, there was no tracking
being done of the files processed, so that doxygen can be re-run if needed.
This patchset adds that tracking support, by generating a gcc-style
".d" file for the custom targets used in the doc/api build.
In doing so, we remove
For the doxygen API for the examples folder, we can generate a build
dependency file when we generate the examples.dox file. This allows
correct rebuilds if the files in examples change.
Signed-off-by: Bruce Richardson
---
doc/api/generate_examples.sh | 4
doc/api/meson.build | 5 +
Generate a dependency file for the header files used in the API guide
so that the docs can be rebuilt if a header changes.
Signed-off-by: Bruce Richardson
---
doc/api/generate_doxygen.sh | 5 -
doc/api/meson.build | 3 ++-
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git
On 1/14/2020 11:49 AM, Kumar, Ravi1 wrote:
<...>
>
> +[CAUTION: External Email]
> +
> +From: Girish Nandibasappa
> +
> +Added support for 1Gbps and 2.5Gbps in axgbe dpdk driver
> +
> +Signed-off-by: Girish Nandibasappa
>
> Acked-by: Ravi Kumar
Series applied to dpdk-next-net/master, thanks
On Thu, Nov 7, 2019 at 10:35 PM David Marchand
wrote:
>
> DPDK has multiple use cases where the core repeatedly polls a location in
> memory. This polling results in many cache and memory transactions.
>
> Arm architecture provides WFE (Wait For Event) instruction, which allows
> the cpu core to e
Matcher is flow table related structure providing the flow pattern
to be translated directly in hardware controlling data. This structure
includes the metadata register c0 field, that might be engaged to
support META and MARK related flow items and actions. Also, this
register might be used by kern
On Fri, Jan 17, 2020 at 4:14 PM David Marchand
wrote:
>
> On Fri, Jan 17, 2020 at 10:52 AM Jerin Jacob wrote:
> > > > If someone needs more time to think through or any clarification is
> > > > required then please discuss.
> > >
> > > I did not find the time to look at this.
> > > Some quick que
> -Original Message-
> From: David Marchand
> Sent: Tuesday, January 14, 2020 3:59 PM
> To: Richardson, Bruce
> Cc: Mcnamara, John ; Luca Boccassi
> ; dev ; Aaron Conole ;
> Thomas Monjalon
> Subject: Re: [PATCH v4 5/6] doc/guides: rebuild with meson whenever a file
> changes
>
> On F
Matcher is flow table related structure providing the flow pattern
to be translated directly in hardware controlling data. Some fields
in this structure might be split (by software) between multiple items.
For example, the metadata register c0 field in the matcher might be
split into two independe
On Fri, Jan 17, 2020 at 4:00 PM Mattias Rönnblom
wrote:
>
> > LTTng kernel tracing only needs kmod support.
> > For the userspace tracing at minium following libraries are required.
> >
> > a) LTTng-UST
> > b) LTTng-tools
> > c) liburcu
> > d) libpopt-dev
>
> This "DPDK CTF trace emitter" would m
When werror is set for the build, we should pass that flag through to
sphinx so that it can flag warnings as errors too.
Signed-off-by: Bruce Richardson
---
buildtools/call-sphinx-build.py | 5 ++---
doc/guides/meson.build | 4
2 files changed, 6 insertions(+), 3 deletions(-)
diff
When sphinx-build reports its version information to stderr rather
than stdout, the wrapper script misses it, and then fails to run.
We can fix this by redirecting stderr to stdout for the version
query call.
Fixes: f5ab2074cfba ("doc: rebuild with meson whenever a file changes")
Signed-off-by: B
On Fri, 2020-01-17 at 09:27 +0100, David Marchand wrote:
> On Thu, Jan 16, 2020 at 3:38 PM Thomas Monjalon <
> tho...@monjalon.net
> > wrote:
> > 16/01/2020 13:42, Ferruh Yigit:
> > > On 1/16/2020 11:54 AM, Neil Horman wrote:
> > > > On Thu, Jan 16, 2020 at 12:25:06PM +0100, David Marchand wrote:
>
On Thu, 2020-01-16 at 21:52 -0500, Pei Zhang wrote:
> Hi Luca,
>
> Testing with dpdk 17.11.10-rc1 from Red Hat looks good.
>
> We cover below 13 scenarios and and all get PASS:
>
> (1)Guest with device assignment(PF) throughput testing(1G hugepage
> size): PASS
> (2)Guest with device assignment
On Fri, Jan 17, 2020 at 10:52 AM Jerin Jacob wrote:
> > > If someone needs more time to think through or any clarification is
> > > required then please discuss.
> >
> > I did not find the time to look at this.
> > Some quick questions:
> > - is LTTng coming with out-of-tree kmod? making it hard t
> -Original Message-
> From: Ferruh Yigit
> Sent: Thursday, January 9, 2020 17:44
> To: Slava Ovsiienko ; dev@dpdk.org
> Cc: Matan Azrad ; Raslan Darawsheh
> ; Ori Kam ; Thomas
> Monjalon
> Subject: Re: [dpdk-dev] [PATCH v2 4/4] net/mlx5: engage free on completion
> queue
>
> On 1/9/2020
On 1/17/2020 6:55 AM, Xiaolong Ye wrote:
> Add release notes for Intel ice/iavf/i40e PMDs.
>
> Signed-off-by: Xiaolong Ye
Reviewed-by: Ferruh Yigit
Applied to dpdk-next-net/master, thanks.
1 - 100 of 120 matches
Mail list logo