On Tue, Sep 03, 2024 at 05:54:22PM +, Wathsala Wathawana Vithanage wrote:
> > Some SOCs may only show upper-level caches here, therefore cannot
> > be use blindly without knowing the SOC.
> >
> > Can you please help us understand
> >
>
> For instance, in Neoverse N1 can disable the use
On 2024-09-02 02:39, Varghese, Vipin wrote:
Thank you Mattias for the comments and question, please let me try to
explain the same below
We shouldn't have a separate CPU/cache hierarchy API instead?
Based on the intention to bring in CPU lcores which share same L3 (for
better cache hits
Hemant, Sachin,
dpaa_event_dequeue_wait() uses select(), which doesn't work if qman_thread_fd()
returns a file descriptor number above FD_SETSIZE.
You should not use select(), but a modern alternative.
-Morten
To simplify building 32-bit binaries on 64-bit system, we can supply a
cross-compilation file which provides the relevant compiler flags and
settings needed - '-m32' compile/link flag, and appropriate
PKG_CONFIG_LIBDIR value. This latter setting will depend upon the layout
format of the particular
When testing the 32-bit x86 build on debian or ubuntu linux systems, use
the cross-file rather than using args and pkgconfig environment
variable. The advantage of using the cross-file is that the paths are
saved across runs. While the '-m32' args settings are preserved in the
current setup, the PK
Hi all,
Here is a new stable release:
https://fast.dpdk.org/rel/dpdk-23.11.2.tar.xz
The git tree is at:
https://dpdk.org/browse/dpdk-stable/?h=23.11
Xueming Li
---
.mailmap | 24 +-
VERSION|
On Wed, 4 Sep 2024 11:30:59 +0200
Mattias Rönnblom wrote:
> On 2024-09-02 02:39, Varghese, Vipin wrote:
> >
> >
> > Thank you Mattias for the comments and question, please let me try to
> > explain the same below
> >
> >> We shouldn't have a separate CPU/cache hierarchy API instead?
> >
From: Bruce Richardson
Rather than the single monolithic graph that would be output from the
deps.dot file in a build directory, we can post-process that to generate
simpler graphs for different tasks. This new "draw_dependency_graphs"
script takes the "deps.dot" as input and generates an output
From: Bruce Richardson
Remove any unnecessary dependencies from the driver dependency lists.
This will give each driver a near-minimum set of required dependencies.
Signed-off-by: Bruce Richardson
---
drivers/baseband/fpga_5gnr_fec/meson.build | 2 +-
drivers/baseband/fpga_lte_fec/meson.build
As part of the meson build, we can record the dependencies for each
component as we process it, logging them to a file. This file can be
used as input to a number of other scripts and tools, for example, to
graph the dependencies, or to allow higher-level build-config tools to
automatically enable
From: Bruce Richardson
As meson processes our DPDK source tree it manages dependencies
specified by each individual driver. To enable future analysis of the
dependency links between components, log the dependencies of each DPDK
component as it gets processed. This could potentially allow other to
From: Bruce Richardson
While not a serious problem, DPDK components often list more
dependencies than are actually necessary to build, due to the use of
recursive dependencies. In extreme cases, such as with core libraries,
this can lead to longer configuration times due to meson having to
dedupl
Allow specifying dependencies as either mandatory or optional. This does
not change anything about the build, but it is useful for tooling to know
if a dependency is required or not.
Signed-off-by: Anatoly Burakov
---
app/meson.build | 3 ++-
app/proc-info/meson.build
From: Bruce Richardson
The kvargs library is used by EAL, and therefore is implicitly a
dependency of every DPDK driver. Remove it from the minimum set of
dependencies for each driver class as it's unnecessary to call it out
there.
Signed-off-by: Bruce Richardson
---
drivers/event/meson.build
From: Bruce Richardson
Rather than having each library depend up on EAL + any extra libs, we
can take advantage of recursive dependency support in meson and
just assign the dependencies of each directory directly, rather than
appending to the array. For libraries which only depend upon EAL, keep
From: Bruce Richardson
Remove any unnecessary dependencies from the app dependency lists.
This will give each app a near-minimum set of required dependencies.
Signed-off-by: Bruce Richardson
---
app/dumpcap/meson.build | 2 +-
app/graph/meson.build| 2 +-
app/pdump/meson.b
Note: this patch depends upon Bruce's v3 patchset:
https://patches.dpdk.org/project/dpdk/list/?series=32891
This patch is based on initial script for VSCode configuration:
https://patches.dpdk.org/project/dpdk/patch/6a6b20c037cffcc5f68a341c4b4e4f21990ae991.1721997016.git.anatoly.bura...@intel.co
Currently, the only way to set up a build directory for DPDK development
is through running Meson directly. This has a number of drawbacks.
For one, the default configuration is very "fat", meaning everything gets
enabled and built (aside from examples, which have to be enabled
manually), so while
From: Jeremy Spewock
Currently the only method provided in the test suite class for sending
packets sends a single packet and then captures the results. There is,
in some cases, a need to send multiple packets at once while not really
needing to capture any traffic received back. The method to do
From: Jeremy Spewock
This patch was originally part of the dynamic queue test suite series,
but since other patches require it this series creates an independent
patch to allow it to be prioritized.
This patch is slightly different than the one in dynamic queue as this
version supports address u
The TAP device does have per-queue stats and handles multi-process.
Signed-off-by: Stephen Hemminger
---
doc/guides/nics/features/tap.ini | 2 ++
1 file changed, 2 insertions(+)
diff --git a/doc/guides/nics/features/tap.ini b/doc/guides/nics/features/tap.ini
index f26355e57f..f2ea5cd833 100644
From: Jeremy Spewock
This patch adds methods for querying and modifying port queue state and
configuration. In addition to this, it also adds the ability to capture
the forwarding statistics that get outputted when you send the "stop"
command in testpmd. Querying of port queue information is hand
From: Jeremy Spewock
v4:
* split patch for updating packet addressing into its own series and
added a dependency in this one.
* squash commit for adding test suite to the yaml schema into the last
commit.
Jeremy Spewock (2):
dts: add port queue modification and forwarding stats to test
From: Jeremy Spewock
This patch adds a new test suite that is designed to test the stopping
and modification of port queues at runtime. Specifically, there are
test cases that display the ports ability to stop some queues but still
send and receive traffic on others, as well as the ability to con
Acked-by: Vladimir Medvedkin
On 26/07/2024 15:54, Niall Meade wrote:
Separated name, entries and key_len parameter checks in
rte_hash_create(). Also made the error messages more
informative/verbose to help with debugging. Also added myself to the
mailing list.
Signed-off-by: Niall Meade
---
This patch series is a continuation of the patch set that supports configuring
speed lanes.
https://patchwork.dpdk.org/project/dpdk/patch/20240708232351.491529-1-damodharam.ammepa...@broadcom.com/
The patchset consists
1) rtelib/testpmd changes (Addressing the comments). Earlier comments are
ava
Update the eth_dev_ops structure with new function vectors
to get, get capabilities and set ethernet link speed lanes.
Update the testpmd to provide required config and information
display infrastructure.
The supporting ethernet controller driver will register callbacks
to avail link speed lanes c
Broadcom Thor2 NICs support link mode settings where user
can configure fixed speed and associated supported number of
lanes. This patch does code-refactoring to address proposed
poll mode library design updates.
Signed-off-by: Damodharam Ammepalli
---
drivers/net/bnxt/bnxt.h| 3 +
dri
The API's in ethtool from before 23.11 should be marked stable.
Should probably include the trace api's but that is more complex change.
Stephen Hemminger (3):
eal: make rte_cpu_get_intrinsics_support stable
eal: mark rte_lcore_register_usage_cb stable
eal: mark rte_memzone_max_get/set stabl
This API was added in 20.11, after four years it should be stable.
Signed-off-by: Stephen Hemminger
---
lib/eal/include/generic/rte_cpuflags.h | 4
lib/eal/version.map| 2 +-
2 files changed, 1 insertion(+), 5 deletions(-)
diff --git a/lib/eal/include/generic/rte_cpufla
This API was added back in 23.03, can be marked stable now.
Signed-off-by: Stephen Hemminger
---
lib/eal/include/rte_lcore.h | 4
lib/eal/version.map | 2 +-
2 files changed, 1 insertion(+), 5 deletions(-)
diff --git a/lib/eal/include/rte_lcore.h b/lib/eal/include/rte_lcore.h
index
These were added in 23.03
Signed-off-by: Stephen Hemminger
---
lib/eal/include/rte_memzone.h | 8
lib/eal/version.map | 6 ++
2 files changed, 2 insertions(+), 12 deletions(-)
diff --git a/lib/eal/include/rte_memzone.h b/lib/eal/include/rte_memzone.h
index 931497f37c..e15
On Tue, Aug 6, 2024 at 8:53 AM Luca Vizzarro wrote:
> Add a basic L2 forwarding test suite which tests the correct
> functionality of the forwarding facility built-in in the DPDK.
>
> The tests are performed with different queues numbers per port.
>
> Bugzilla ID: 1481
>
> Signed-off-by: Luca Viz
On Wed, Aug 21, 2024 at 2:43 PM Nicholas Pratte wrote:
>
> In order to leverage the usability of unique identifiers on ports, the
> testbed_model Port object needs some refactoring/trimming of obsolete or
> needless attributes.
>
> Bugzilla ID: 1478
>
> Signed-off-by: Nicholas Pratte
Reviewed-by
On Wed, Aug 21, 2024 at 2:43 PM Nicholas Pratte wrote:
> diff --git a/dts/framework/config/__init__.py
> b/dts/framework/config/__init__.py
> index df60a5030e..534821ed22 100644
> --- a/dts/framework/config/__init__.py
> +++ b/dts/framework/config/__init__.py
> @@ -151,11 +151,10 @@ class PortCo
On Wed, Aug 21, 2024 at 2:43 PM Nicholas Pratte wrote:
> diff --git a/dts/framework/test_suite.py b/dts/framework/test_suite.py
> index 694b2eba65..fd51796a06 100644
> --- a/dts/framework/test_suite.py
> +++ b/dts/framework/test_suite.py
> @@ -20,6 +20,7 @@
> from scapy.layers.l2 import Ether #
Just a few comments on doc-strings, otherwise:
Reviewed-by: Jeremy Spewock
On Fri, Aug 23, 2024 at 4:22 PM Dean Marx wrote:
>
> Add rx_vxlan_port add/rm method to testpmd shell for adding
> or removing a vxlan id to the specified port filter list.
>
> Signed-off-by: Dean Marx
> ---
> dts/fram
I just left some general formatting comments and a few idea that I
thought might be helpful, but it looks good to me in general:
Reviewed-by: Jeremy Spewock
On Fri, Aug 23, 2024 at 4:22 PM Dean Marx wrote:
>
> Port over unified packet testing suite from old DTS. This suite
> tests the ability o
All looks great to me, thank you for the patch and addressing
comments, there is just one comment that I left here about something
that I missed pointing out on the previous version, but once that is
addressed I think this will be ready.
On Fri, Aug 23, 2024 at 5:16 PM Dean Marx wrote:
> +de
On Fri, Aug 23, 2024 at 5:16 PM Dean Marx wrote:
>
> added the following methods to testpmd shell class:
> vlan set filter on/off, rx vlan add/rm,
> vlan set strip on/off, port stop/start all/port,
> tx vlan set/reset, set promisc/verbose
>
> fixed bug in vlan_offload for
> show port info, removed
>> +__rte_experimental
>> +struct rte_flow *
>> +rte_flow_async_create_by_index_with_pattern(uint16_t port_id,
>> + uint32_t queue_id,
>> + const struct rte_flow_op_attr
>> *op_attr,
>> +
> I think we already discussed this same patch in previous emails
> (Aug-Oct 2023) at
> https://mails.dpdk.org/archives/dev/2023-October/279390.html and
> concluded that it is not needed ?
> Did anything change from then ?
Yes, Nithin, we found a way to distinguish the modes by looking into the ne
>
>
> +if should_receive:
> +self.verify(len(received_packets) == 1, "Expected packet not
> received")
> +else:
> +self.verify(len(received_packets) == 0, "Expected packet
> received")
>
Side note, didn't notice until I tested it but "Expected packet receiv
Hey Dean,
This patch is looking good, I like the changes from the previous
version, there were just a few comments that I left but they should be
pretty easy fixes.
On Mon, Aug 26, 2024 at 5:22 PM Dean Marx wrote:
> +def csum_set_hw(self, layer: ChecksumOffloadOptions, port_id: int,
> veri
Just one comment here but it is a super small doc-string thing that
still passes the formatting script and everything, but whenever we add
Ruff for checking format in the future, it would get caught so it
might be better to just add it now. Otherwise:
Reviewed-by: Jeremy Spewock
On Mon, Aug 26,
[1/3] ethdev: add frequency adjustment API
[2/3] net/ice: add frequency adjustment support for PTP
[3/3] examples/ptpclient: add frequency adjustment support
Mingjin Ye (3):
ethdev: add frequency adjustment API
net/ice: add frequency adjustment support for PTP
examples/ptpclient: add frequen
This patch adds freq adjustment API for PTP high accuracy.
Signed-off-by: Simei Su
Signed-off-by: Mingjin Ye
---
doc/guides/nics/features.rst | 3 ++-
lib/ethdev/ethdev_driver.h | 5 +
lib/ethdev/ethdev_trace.h| 9 +
lib/ethdev/ethdev_trace_points.c | 3 +++
li
Add ice support for new ethdev API to adjust frequency for IEEE1588
PTP. Also, this patch reworks code for converting software update
to hardware update.
Signed-off-by: Simei Su
Signed-off-by: Mingjin Ye
---
doc/guides/nics/ice.rst | 15 +++
drivers/net/ice/ice_ethdev.c | 177
This patch adds PI servo algorithm to support frequency
adjustment API for IEEE1588 PTP.
For example, the command for starting ptpclient with PI algorithm is:
./build/examples/dpdk-ptpclient -a :81:00.0 -c 1 -n 3 -- -T 0 -p 0x1
--controller=pi
Signed-off-by: Simei Su
Signed-off-by: Wenjun Wu
Hi, all maintainers,
Kindly ping for review.
Thanks,
Jie Hai
On 2024/7/19 17:04, Jie Hai wrote:
For hns3 NIC, when KEEP_CRC offload is enabled, the CRC data is
still be stripped in rare scenarios. Some users of hns3 are
already using this feature. So driver has to recaculate packet CRC.
In add
Hi Amit,
It indeed provide more flexible configuration.
There is a small comment below, with that fixed,
Acked-by: Chengwen Feng
Thanks
On 2024/8/5 21:51, Amit Prakash Shukla wrote:
> Add support to configure device specific config parameters for a
> testcase. Example:
>
> lcore_dma0=lcore=11
[...]
> > From: David Marchand [mailto:david.march...@redhat.com]
> > Sent: Thursday, 21 September 2023 10.27
> >
> > Hello,
> >
> > On Mon, Aug 7, 2023 at 10:12 AM Tomasz Duszynski
> >
> > wrote:
> > > >Ping for update
> > > >What is the status of this feature?
> > >
> > > I'll re-spin the serie
> From: Anatoly Burakov [mailto:anatoly.bura...@intel.com]
>
> As part of the meson build, we can record the dependencies for each
> component as we process it, logging them to a file. This file can be
> used as input to a number of other scripts and tools, for example, to
> graph the dependencies
> From: Anatoly Burakov [mailto:anatoly.bura...@intel.com]
>
> Currently, the only way to set up a build directory for DPDK development
> is through running Meson directly. This has a number of drawbacks.
>
> For one, the default configuration is very "fat", meaning everything gets
> enabled and
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Wednesday, 4 September 2024 20.09
>
> The API's in ethtool from before 23.11 should be marked stable.
> Should probably include the trace api's but that is more complex change.
For the series,
Acked-by: Morten Brørup
This patch series aims to fix some problems in the representor port link
status and speed update logic, also do some refactors to the related
logics.
Qin Ke (8):
net/nfp: fix incorrect type declaration of some variables
net/nfp: add help function to check link speed
net/nfp: add help functio
From: Qin Ke
The type declaration of variable 'speed' and 'i' in
'nfp_net_link_speed_rte2nfp()' is not correct, fix it.
Fixes: 36a9abd4b679 ("net/nfp: write link speed to control BAR")
Cc: james.hers...@corigine.com
Cc: sta...@dpdk.org
Signed-off-by: Qin Ke
Reviewed-by: Chaoyong He
Reviewed-b
From: Qin Ke
Add help function to check link speed.
Signed-off-by: Qin Ke
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
drivers/net/nfp/nfp_net_common.c | 23 ++-
1 file changed, 14 insertions(+), 9 deletions(-)
diff --git a/drivers/net/nfp/n
From: Qin Ke
Add help function to encapsulate logic of updating vf link speed.
Signed-off-by: Qin Ke
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
drivers/net/nfp/nfp_net_common.c | 33
1 file changed, 21 insertions(+), 12 deletio
From: Qin Ke
Rename PF speed update function of 'nfp_net_speed_aneg_update()'
to 'nfp_net_pf_speed_update()', which make codes more readable.
Also use maroc to replace the hard coded value.
Signed-off-by: Qin Ke
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
driv
From: Qin Ke
Add a new data field 'idx' into 'nfp_flower_representor' structure
to indicate the sequential physical port number of representor
devices, also initialize it for all representor ports.
Signed-off-by: Qin Ke
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
--
From: Qin Ke
For representor port link speed, the original logic finally calls
'nfp_net_speed_aneg_update()' to update it. But the reference of
'hw->idx' in this function is invalid for representor port devices,
the logic has problem.
Fix it by getting correct 'idx' for all type of deives includ
From: Qin Ke
Standardize the use of 'idx' in some functions which could be
used for both flower and coreNIC firmware.
Signed-off-by: Qin Ke
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
drivers/net/nfp/nfp_ethdev.c | 6 --
drivers/net/nfp/nfp_net_common.
From: Qin Ke
The link status of representor port is reported by the flower
firmware through control message and it already parsed and
stored in the 'link' field of representor port structure.
The original logic read link status from the control BAR again,
and use it rather then the 'link' field
On 7/19/24 12:04, Jie Hai wrote:
From: Dengdui Huang
The data exceeds the pkt_len in mbuf is inavailable for user.
When KEEP CRC offload is enabled, CRC field length should be
included in the pkt_len in mbuf. However, almost of drivers
supported KEEP CRC feature didn't add the CRC data length t
The registers can be dumped through the API rte_eth_dev_get_reg_info.
However, only register values are exported, which is inconvenient for
users to interpret. Therefore, an extension of the structure
"rte_dev_reg_info" and a new API rte_eth_dev_get_reg_info_ext is added
to support the capability o
For security reasons, some address registers are not suitable
to be exposed, remove them.
Cc: sta...@dpdk.org
Signed-off-by: Jie Hai
Acked-by: Huisong Li
---
drivers/net/hns3/hns3_regs.c | 12 ++--
1 file changed, 2 insertions(+), 10 deletions(-)
diff --git a/drivers/net/hns3/hns3_regs
This patch adds a telemetry command for registers dump,
and supports obtaining the registers of a specified module.
In one way, the number of registers that can be exported
is limited by the number of elements carried by dict and
container. In another way, the length of the string
exported by tele
This patch adds "filter" and "names" fields to "rte_dev_reg_info"
structure. Names of registers in data fields can be reported and
the registers can be filtered by their module names.
The new API rte_eth_dev_get_reg_info_ext() is added to support
reporting names and filtering by modules. And the o
Since the driver dumps the queue interrupt registers according
to the intr_tqps_num, the counter should be the same.
Fixes: acb3260fac5c ("net/hns3: fix dump register out of range")
Fixes: 936eda25e8da ("net/hns3: support dump register")
Cc: sta...@dpdk.org
Signed-off-by: Jie Hai
Acked-by: Huiso
This patch refactors codes dumping registers from firmware.
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_regs.c | 203 ---
1 file changed, 115 insertions(+), 88 deletions(-)
diff --git a/drivers/net/hns3/hns3_regs.c b/drivers/net/hns3/hns3_regs.c
index c8e3fb
Since the driver is going to support reporting names of
all registers, remove the counter and insert of separators
between different register modules.
Signed-off-by: Jie Hai
Reviewed-by: Huisong Li
Acked-by: Chengwen Feng
---
drivers/net/hns3/hns3_regs.c | 68 ++
This patch support dumping registers which name contains the
`filter` string. The module names are in lower case and so is
the `filter`. Available module names are cmdq, common_pf,
common_vf, ring, tqp_intr, 32_bit_dfx, 64_bit_dfx, bios, igu_egu,
ssu, ppp, rpu, ncsi, rtc, rcb, etc.
Signed-off-by:
This patch adds names for register lists, and support report
names of registers. Some registers has different names on
different platform, use names of HIP08 as default names.
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_regs.c | 1090 +-
1 file changed, 957 i
This patch set make clean codes.
Chengwen Feng (4):
ethdev: verify queue ID when Tx done cleanup
net/hns3: verify reset type from firmware
dmadev: fix potential null pointer access
dmadev: clean code for verify parameter
drivers/net/hns3/hns3_intr.c | 6 ++
lib/dmadev/rte_dmadev.c
From: Chengwen Feng
Verify queue_id for rte_eth_tx_done_cleanup API.
Fixes: 44a718c457b5 ("ethdev: add API to free consumed buffers in Tx ring")
Cc: sta...@dpdk.org
Signed-off-by: Chengwen Feng
---
lib/ethdev/rte_ethdev.c | 4
1 file changed, 4 insertions(+)
diff --git a/lib/ethdev/rte_
From: Chengwen Feng
Verify reset-type which get from firmware.
Fixes: 1c1eb759e9d7 ("net/hns3: support RAS process in Kunpeng 930")
Cc: sta...@dpdk.org
Signed-off-by: Chengwen Feng
---
drivers/net/hns3/hns3_intr.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/hns3/hns3
From: Chengwen Feng
When rte_dma_vchan_status(dev_id, vchan, NULL) is called, a null pointer
access is triggered.
This patch adds the null pointer checker.
Fixes: 5e0f85912754 ("dmadev: add channel status check for testing use")
Cc: sta...@dpdk.org
Signed-off-by: Chengwen Feng
---
lib/dmadev/
From: Chengwen Feng
Make sure to verify the 'dev_id' parameter before using them.
Note: there is no actual problem with the current code, but it is
recommended to clean it up.
Cc: sta...@dpdk.org
Signed-off-by: Chengwen Feng
---
lib/dmadev/rte_dmadev.c | 30 --
1 f
From: Dengdui Huang
Add dump the head and tail pointer of RxTx queue.
-- Rx queue head and tail info:
qid sw_head sw_hold hw_head hw_tail
0288 32 256 320
1248 56 192 280
2264 72 192 296
3256 64
80 matches
Mail list logo
