> -Original Message-
> From: David Marchand
> Sent: Thursday, July 11, 2024 7:39 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Ali Alnubani ; Jiale, SongX
> ; Dmitry Kozlyuk
> Subject: [PATCH] buildtools: fix build with clang 17 and ASan
>
> ASan included in clang 17 and later suffixes sy
> -Original Message-
> From: Burakov, Anatoly
> Sent: Friday, July 12, 2024 12:10 AM
> To: Ye, MingjinX ; dev@dpdk.org
> Cc: sta...@dpdk.org
> Subject: Re: [PATCH 3/3] net/vdev: fix insert vdev core dump
>
> On 3/14/2024 10:36 AM, Mingjin Ye wrote:
> > Inserting a vdev device when the d
This commit improves DLB credit handling scenarios when
ports hold on to credits but can't release them due to insufficient
accumulation (less than 2 * credit quanta).
Worker ports now release all accumulated credits when back-to-back
zero poll count reaches preset threshold.
Producer ports relea
DLB has 64 LDB ports and 2048 HL entries. If all LDB ports are used,
possible HL entries per LDB port equals 2048 / 64 = 32. So, the
maximum CQ depth possible is 16, if all 64 LB ports are needed in a
high-performance setting.
In case all CQs are configured to have HL = 2* CQ Depth as a
performanc
In DLB 2.5, hardware assist is available, complementing the Delayed
token POP software implementation. When it is enabled, the feature
works as follows:
It stops CQ scheduling when the inflight limit associated with the CQ
is reached. So the feature is activated only if the core is
congested. If t
v6: Address review for main tree pmd options.
v5: Address reviews and update documentation.
v4: Fix CI Issues.
V3: Fix CI issues.
v2: Fix compilation issues.
v1: Initial commit.
Abdullah Sevincer (3):
event/dlb2: add support for HW delayed token
event/dlb2: add support for dynamic HL entries
To use independent enqueue capability applications need to set flag
RTE_EVENT_PORT_CFG_INDEPENDENT_ENQ during port setup only if the
capability RTE_EVENT_DEV_CAP_INDEPENDENT_ENQ exists. Hence, this
commit adds the capability of independent enqueue to the DSW driver.
Signed-off-by: Abdullah Sevince
This commit adds support for independent enqueue feature
and updates Event Device and PMD feature list.
A new capability RTE_EVENT_DEV_CAP_INDEPENDENT_ENQ is introduced
to support independent enqueue to support PMD to enqueue in any order
even the underlined hardware device needs enqueues in a str
DLB devices need events to be enqueued in the same order they are
dequeued. Applications are not suppose to change event order between
dequeue and to enqueue. Since Eventdev standard does not add such
restrictions independent enqueue support is needed for DLB PMD so that
it restores dequeue order o
v6: Update patch with more documentation
v5: Address build issues
v4: Address comments
v3: Fix CI/build issues
v2: Fix CI/build issues
v1: Initial patchset
Abdullah Sevincer (3):
event/dlb2: add support for independent enqueue
eventdev: add support for independent enqueue
event/dsw: add capa
Code all looked good to me, just a couple of documentation comments.
On Tue, Jul 2, 2024 at 3:25 PM Nicholas Pratte wrote:
>
> +class TestMacFilter(TestSuite):
> +"""Mac address allowlist filtering test suite.
> +
> +Configure mac address filtering on a given port, and test the port's
>
There were a few emails that came through for this series but this was
the most recent one so I went with reviewing this one, but it looks
like the more descriptive commit subject got lost in this iteration.
It would probably be worth adding that back.
Additionally, it looks like the functions you
I think this change makes sense, and I mentioned this on Dean's patch
that has the same change in it but I think we should have more
discussion about which route to take with this addressing problem. It
is definitely something that we have to address since it is required
for a suite like this or an
#
July 11, 2024
Attendees
1. Patrick Robb
2. Juraj Linkeš
3. Alex Chapman
4. Tomas Durovec
5. Jeremy Spewock
6. Adam Hassick
7. Manit Mahajan
8. Dean Marx
#
Minut
> On Thu, Jul 11, 2024 at 2:44 PM Srujana Challa wrote:
> >
> > This patch modifies the code to convert descriptor buffer IOVA
> > addresses to virtual addresses only when use_va flag is false.
> >
> > This patch resolves a segmentation fault with the vhost-user backend
> > that occurs during the
Hi Bruce,
Sure, will submit v2.
Thanks!
-Original Message-
From: Richardson, Bruce
Sent: Thursday, July 11, 2024 6:05 PM
To: Medvedkin, Vladimir
Cc: dev@dpdk.org; ting...@intel.com; Stolarchuk, Michael
Subject: Re: [PATCH] net/ice: fix possible memory leak
On Thu, Jul 11, 2024 at 0
On Thu, Jul 11, 2024 at 04:59:07PM +, Vladimir Medvedkin wrote:
> This patch fixes possible memleak inside the
> ice_hash_parse_raw_pattern().
> Additionally replaces using strlen() with more secure strnlen().
> Also replaces the returned inconsistent rte_errno
> with explicit error statuses.
>
This patch fixes possible memleak inside the
ice_hash_parse_raw_pattern().
Additionally replaces using strlen() with more secure strnlen().
Also replaces the returned inconsistent rte_errno
with explicit error statuses.
Fixes: 1b9c68120a1c ("net/ice: enable protocol agnostic flow offloading in RSS
The hash table used for managing the filter rules in the ice driver was
dimensioned to a hard-coded 16k, which is insufficient for holding all
the filters supported by E830 HW.
Rather than using a hard-coded value which may need updates for new
hardware support, we can query the NIC max filter sup
On 7/11/2024 1:07 PM, Serhii Iliushyk wrote:
> Add initial ntnic ethdev skeleton and register PCI probe functions
> Update documentation: Device description and feature list
>
> Signed-off-by: Serhii Iliushyk
>
Hi Serhii,
I put a few minor comments.
What is the latest testing status of the driv
On 7/11/2024 1:07 PM, Serhii Iliushyk wrote:
> +++ b/doc/guides/nics/features/ntnic.ini
> @@ -0,0 +1,12 @@
> +;
> +; Supported features of the 'ntnic' network poll mode driver.
> +;
> +; Refer to default.ini for the full list of available PMD features.
> +;
> +[Features]
> +Speed capabilities = Y
On 7/11/2024 1:07 PM, Serhii Iliushyk wrote:
> @@ -383,6 +658,8 @@ nthw_pci_dev_init(struct rte_pci_device *pci_dev)
> /* connect structs */
> internals->p_drv = p_drv;
> eth_dev->data->dev_private = internals;
> + eth_dev->data->mac_addrs = r
Tested-by: Dean Marx
From: Jeremy Spewock
The InteractiveShell class currently relies on being cleaned up and
shutdown at the time of garbage collection, but this cleanup of the class
does no verification that the session is still running prior to cleanup.
So, if a user were to call this method themselves prior to ga
From: Jeremy Spewock
Interactive shells are managed in a way currently where they are closed
and cleaned up at the time of garbage collection. Due to there being no
guarantee of when this garbage collection happens in Python, there is no
way to consistently know when an application will be closed
From: Jeremy Spewock
v2:
* addresses the comments from version 1, adjusting documentation
accordingly and condensing usage of the context manager.
Jeremy Spewock (2):
dts: add context manager for interactive shells
dts: improve starting and stopping interactive shells
dts/framework/rem
On 3/14/2024 10:36 AM, Mingjin Ye wrote:
Inserting a vdev device when the device arguments are already stored
in devargs_list, the rte_devargs_insert function replaces the supplied
new devargs with the found devargs and frees the new devargs. As a
result, the use of free devargs results in a core
On 3/14/2024 10:36 AM, Mingjin Ye wrote:
The asan tool detected a memory leak in the vdev driver alloc_devargs.
The previous commit does not insert device arguments into devargs_list
when attaching a device during a bus scan of a secondary process.
This resulted in an existing memory leak when re
On Wed, Jul 10, 2024 at 10:55 AM wrote:
>
> From: Vanshika Shukla
>
> This series includes fixes for NXP DPAA drivers.
>
> V5 changes:
> Added CID for "remove redundant file descriptor check" patch.
> Note: Other patches are bug fixes reported by our internal coverity tool.
>
> V4 changes:
> Remo
On Thu, Jul 11, 2024 at 10:53 AM Juraj Linkeš
wrote:
>
> With the docs changes below,
> Reviewed-by: Juraj Linkeš
>
> > diff --git a/dts/framework/remote_session/interactive_shell.py
> > b/dts/framework/remote_session/interactive_shell.py
> > index 11dc8a0643..fcaf1f6a5f 100644
> > --- a/dts/fra
On Thu, Jul 11, 2024 at 10:35 AM Juraj Linkeš
wrote:
>
>
>
> On 9. 7. 2024 18:31, jspew...@iol.unh.edu wrote:
> > From: Jeremy Spewock
> >
> > Interactive shells are managed in a way currently where they are closed
> > and cleaned up at the time of garbage collection. Due to there being no
> > gu
On Thu, Jul 11, 2024 at 2:44 PM Srujana Challa wrote:
>
> This patch modifies the code to convert descriptor buffer IOVA
> addresses to virtual addresses only when use_va flag is false.
>
> This patch resolves a segmentation fault with the vhost-user backend
> that occurs during the processing of
With the docs changes below,
Reviewed-by: Juraj Linkeš
diff --git a/dts/framework/remote_session/interactive_shell.py
b/dts/framework/remote_session/interactive_shell.py
index 11dc8a0643..fcaf1f6a5f 100644
--- a/dts/framework/remote_session/interactive_shell.py
+++ b/dts/framework/remote_sessi
This patch fixes a typo in the log message for error allocation.
The typo incorrectly spelled 'allocating' as 'allocationg' in the
log message for error allocation.
Additionally, this commit updates the .mailmap file to ensure proper
attribution and mapping of contributor email addresses.
Fixes:
On 9. 7. 2024 18:31, jspew...@iol.unh.edu wrote:
From: Jeremy Spewock
Interactive shells are managed in a way currently where they are closed
and cleaned up at the time of garbage collection. Due to there being no
guarantee of when this garbage collection happens in Python, there is no
way t
On Wed, Jul 10, 2024 at 12:01 PM Dean Marx wrote:
>
> added set promisc, set verbose, and port stop
> commands to testpmd shell.
>
> Signed-off-by: Dean Marx
> ---
> dts/framework/remote_session/testpmd_shell.py | 194 +-
> 1 file changed, 193 insertions(+), 1 deletion(-)
> -
On Wed, Jul 10, 2024 at 12:01 PM Dean Marx wrote:
>
> This suite tests the ability of the Poll Mode Driver to enable
> and disable Rx/Tx queues on a port.
>
> Signed-off-by: Dean Marx
> ---
Reviewed-by: Jeremy Spewock
https://bugs.dpdk.org/show_bug.cgi?id=1486
Bug ID: 1486
Summary: Port over rte_flow suite from old dts
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
This patch modifies the code to convert descriptor buffer IOVA
addresses to virtual addresses only when use_va flag is false.
This patch resolves a segmentation fault with the vhost-user backend
that occurs during the processing of the shadow control queue.
'Fixes: 67e9e504dae2 ("net/virtio_user:
When allocating memory for an ethdev, the rte_malloc_socket call used
only allocates memory on the NUMA node/socket local to the device. This
means that even if the user wanted to, they could never use a remote NIC
without also having memory on that NIC's socket.
For example, if we change examples
This patch modifies the code to convert descriptor buffer IOVA
addresses to virtual addresses only when use_va flag is false.
'Fixes: 67e9e504dae2 ("net/virtio_user: convert cq descriptor IOVA
address to Virtual address")'
Signed-off-by: Srujana Challa
---
.../net/virtio/virtio_user/virtio_user
Because the ntnic hardware supports multiple different FPGAs with different
pipelines and port speeds,
the clock profile is not hardcoded into the product,
and need to be initialized from software.
The clock profile itself is an array of integers
that was generated by Silicon Labs ClockBuilder.
S
Adds functionality to control the physical layer of the OSI model.
This takes the form of the module
MAC PCS (Media Access Control Physical Coding Sublayer).
The functionality is used by the 100G link functionality to establish link.
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/include/n
As the ntnic can support different port speeds, it also needs to support
different NIMs (Network Interface Module). This commit add the generic NIM
support for ntnic, such that the specific modules,
such as QSFP28 can be added later.
The communication with NIMs is in the form of I2C, so support fo
For NIM reset sequence GPIO communication is used.
After this commit the NIMs supported by ntnic
is able to start up and be controlled fully by the adapter.
Signed-off-by: Serhii Iliushyk
---
v5
* Fix Typo/Spelling
---
drivers/net/ntnic/include/nt4ga_link.h| 1 +
.../link_mgmt/link_100
Includes support for QSFP28
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/include/ntnic_nim.h| 21 ++
drivers/net/ntnic/link_mgmt/nt4ga_link.c | 25 +++
drivers/net/ntnic/nim/i2c_nim.c | 267 ++-
drivers/net/ntnic/nim/nim_defines.h | 1 +
4 files
Add functionality to read and control the link-state of the ntnic.
Note that must functions are not implemented yet.
Adds the following eth_dev_ops:
- dev_set_link_up
- dev_set_link_down
- link_update
- mac_addr_add
- mac_addr_set
- set_mc_addr_list
- promiscuous_enable
Signed-off-by: Serhii Iliu
Includes support for QSFP and QSFP+.
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/include/ntnic_nim.h | 10 +
.../link_mgmt/link_100g/nt4ga_link_100g.c | 12 +-
drivers/net/ntnic/nim/i2c_nim.c | 310 +-
drivers/net/ntnic/nim/i2c_nim.h
New ntnic FPGA modules:
- Host Interface (HIF): Basic FPGA info such as prod ID and build time.
- Inter-Integrated Circuit Controller (IIC): Use the FPGA to access
the other integrated circuits on the ntnic.
- PCI Express Gen3 (PCIE3): The FPGA part of PCIe3 initialization,
speed te
Add ntnic support for NT200A0X NIC
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/adapter/nt4ga_adapter.c | 6 +++
drivers/net/ntnic/meson.build | 1 +
.../net/ntnic/nthw/core/include/nthw_fpga.h | 7 +++
.../nthw/core/nt200a0x/nthw_fpga_nt200a0x.c | 54 +
Enable FPGA initialization and adds ethdev fw_version_get.
Signed-off-by: Serhii Iliushyk
---
v5
* Fix Typo/Spelling
---
drivers/net/ntnic/adapter/nt4ga_adapter.c | 52 +++-
drivers/net/ntnic/meson.build | 7 +
.../net/ntnic/nthw/core/include/nthw_core.h | 4 +
.../net
As the ntnic can support different speeds,
an abstraction layer for 100G speed is needed.
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/adapter/nt4ga_adapter.c | 12 -
.../link_mgmt/link_100g/nt4ga_link_100g.c | 49 +++
drivers/net/ntnic/meson.build
Adds reset (RST) module for FW 9563.
Also adds SDRAM Controller (SDC) module,
as it is part of the startup and reset sequence.
Signed-off-by: Serhii Iliushyk
---
v5
* Fix Typo/Spelling
---
.../include/ntnic_nthw_fpga_rst_nt200a0x.h| 81 +++
drivers/net/ntnic/meson.build |
Adds support for eth_dev configure, start, stop, close, and infos_get.
The internal structs of ntnic is also added and initialized.
Signed-off-by: Serhii Iliushyk
---
v6
* Replace if_index with n_intf_no
* Unnecessry resources free was fixed
* Fix typo
* Useless vars were removed
---
drivers/net
Add interfaces for initialize the adapter
Signed-off-by: Serhii Iliushyk
---
v6
* Function for global var clearing was removed
---
drivers/net/ntnic/adapter/nt4ga_adapter.c | 148 ++
drivers/net/ntnic/include/nt4ga_adapter.h | 40 ++
drivers/net/ntnic/include/ntdrv_4ga.h
Adds many of the high level structures needed by the ntnic FPGA modules
and adapter control.
This is considered the first part of the skeleton of ntnic FPGA support
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/include/ntos_drv.h | 2 +
drivers/net/ntnic/meson.build |
Adds VFIO functionality and the DMA it requires.
The VFIO context is initialized during ntnic ethdev startup.
Signed-off-by: Serhii Iliushyk
---
drivers/net/ntnic/meson.build | 1 +
drivers/net/ntnic/ntnic_ethdev.c | 23 +++
drivers/net/ntnic/ntnic_vfio.c | 235
Add ntnic utilities.
Signed-off-by: Serhii Iliushyk
---
v6
* NT utils header file was moved
---
drivers/net/ntnic/meson.build | 2 ++
drivers/net/ntnic/ntnic_ethdev.c | 2 ++
drivers/net/ntnic/ntutil/nt_util.c | 33 +++
drivers/net/ntnic/ntutil/nt_util.h | 43 +++
Add initial ntnic ethdev skeleton and register PCI probe functions
Update documentation: Device description and feature list
Signed-off-by: Serhii Iliushyk
---
v6
* Remove unused includes
* Remove EOF markers
* Remove unnecesarry commpiller flags
* Update documentation INI and RST files
*
add implementation for probe/init and remove/deinit of the PCI device
Signed-off-by: Serhii Iliushyk
---
v6
* Add driver deinitialization
* Add API rte_eth_dev_probing_finish
* Add correct API for finishing probing the device
* Remove duplicated calling 'rte_eth_copy_pci_info()'
Adds ntnic specific implementation for logging.
NT NIC uses this logging abstraction layer
to ensure that FPGA module implementations function both
within and outside in DPDK environment
Signed-off-by: Serhii Iliushyk
---
v6
* Logging header file was moved
* Default log type was set to NOTICE
---
Hello,
On Sat, Jul 6, 2024 at 5:39 AM Ferruh Yigit wrote:
> >>> Fix some bugs in txgbe/ngbe, and support new features in ngbe driver.
> >>>
> >>> Jiawen Wu (19):
> >>> net/txgbe: fix to parse tunnel packets
> >>> net/txgbe: fix flow filters in VT mode
> >>> net/txgbe: fix Tx hang on queue d
On Fri, Jul 5, 2024 at 10:49 AM Mingjin Ye wrote:
>
> In scenarios where a higher clang compiler is used and ASAN is enabled,
> the generated ELF file will additionally insert undefined debug symbols
> with the same prefix. This causes duplicate C code to be generated.
>
> This patch fixes this is
ASan included in clang 17 and later suffixes symbols.
$ nm build/drivers/libtmp_rte_net_null.a | grep this_pmd
r this_pmd_name3
n this_pmd_name3.f2cd16678ab09dba8fd23405d8d11fce
This breaks the detection of driver symbols in pmdinfogen which then
creates duplicate
> -Original Message-
> From: Mingjin Ye
> Sent: Friday, July 5, 2024 4:26 PM
> To: dev@dpdk.org
> Cc: Marchand, David ; alia...@nvidia.com;
> Ye, MingjinX ; sta...@dpdk.org; Dmitry Kozlyuk
>
> Subject: [PATCH v3] buildtools: fix invalid symbols
>
> In scenarios where a higher clang compi
> Hello Srujana, Jerin,
>
> On Wed, Jul 3, 2024 at 12:04 PM Srujana Challa
> wrote:
> >
> > This patch modifies the code to convert descriptor buffer IOVA
> > addresses to virtual addresses during the processing of shadow control
> > queue when IOVA mode is PA. This change enables Virtio-user to
66 matches
Mail list logo
