Prefer using simple structure assignment instead of memcpy.
Using a structure assignment preserves type information and
compiler checks types already.
Signed-off-by: Stephen Hemminger
---
Reran cocci script against -rc1 and did minor replacement of "*&"
drivers/common/sfc_efx/base/efx_mae.c | 2
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v25.07-rc1
There are 542 new patches in this snapshot.
Release notes:
https://doc.dpdk.org/guides/rel_notes/release_25_07.html
Highlights of 25.07-rc1:
- Mucse RNP driver
- ZTE St
12/06/2025 03:42, Jiawen Wu:
> From: Thomas Monjalon
> > 06/06/2025 10:01, Jiawen Wu:
> > > There were some defects in the original configuration for flow director
> > > filter. Now make the following improvements:
> > >
> > > 1) Fix incorrect parsing to ntuple filter when set the pattern likes:
>
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, June 12, 2025 8:30 AM
> To: Jiawen Wu
> Cc: dev@dpdk.org; zaiyuw...@trustnetic.com; Stephen Hemminger
>
> Subject: Re: [PATCH 02/12] net/txgbe: refactor FDIR filter to improve
> functionality
>
> 06/06/2025 10:01, Jiawen W
>> modify some mac config issues,
>> which were discovered during the testing process.
>>
>> Signed-off-by: Junlong Wang
> That's not how we fix issues in general in DPDK.
> We want to provide tracking history, so:
> - we split the fixes per issue
> - we provide the cause origin (with "Fixes:" l
05/05/2025 22:40, Andre Muezerie:
> Existing DPDK code uses getline(), which is a POSIX function and is
> not available in the Windows APIs.
>
> Instead of rewriting it or coming up with some other replacement, this
> patch makes use of the implementation provided by NetBSD to make it
> possible t
06/05/2025 03:43, Andre Muezerie:
> The asprintf function is not part of the C standard library but is a
> GNU extension commonly available in Unix-like systems. It dynamically
> allocates memory to store the formatted output string, similar to
> sprintf, but avoids buffer overflow issues by automa
06/06/2025 10:01, Jiawen Wu:
> There were some defects in the original configuration for flow director
> filter. Now make the following improvements:
>
> 1) Fix incorrect parsing to ntuple filter when set the pattern likes:
>flow create ... ipv4 / udp dst is ... / raw ... / end actions ... / e
Hello,
07/06/2025 04:30, Junlong Wang:
> modify some mac config issues,
> which were discovered during the testing process.
>
> Signed-off-by: Junlong Wang
That's not how we fix issues in general in DPDK.
We want to provide tracking history, so:
- we split the fixes per issue
- we provide the c
From: Pavan Nikhilesh
Update event vector adapter test to depend on service
core cycles instead of system cycles when waiting for
events.
Reorder event_type check to make sure event received
is a vector before dereferencing vector structure.
Bugzilla ID: 1720
Fixes: de09387f1313 ("eventdev/vecto
10/06/2025 16:50, Bruce Richardson:
> Add an entry in the release notes about the "-l" flag now being the
> same as the "--lcores" flag. Refer to the GSG for examples.
>
> Fixes: 1ea3833e9c4e ("eal: merge corelist and core mapping options")
>
> Signed-off-by: Bruce Richardson
Applied, thanks.
08/02/2025 21:21, Stephen Hemminger:
> This patch series reduces use of memcpy and rte_memcpy
> in DPDK. It changes memcpy to a simple assignment operator
> which easier to read, generates the same code, and
> keeps type safety.
>
> Historically memcpy was used in textbooks because there
> was no
> Stephen Hemminger (7):
> latencystats: handle fractional cycles per ns
> latencystats: do not use floating point
> latencystats: fix log messages
> latencystats: update include files
> latencystats: enforce that unused callback function is NULL
> latencystats: add metric for number of
10/06/2025 16:12, Stephen Hemminger:
> The cycle counts do not need to be stored as floating point.
> Instead keep track of latency in cycles, and convert to
> nanoseconds when read.
>
> Use scaled math with exponential Weighted Moving Average weight
> of .25 to avoid use of floating point for tha
Loongarch is spelled with 2 o's :)
Recheck-request: loongarch-unit-testing
Recheck-request: rebase=main, loongarch-compilation
I don't actually need a compile recheck, I'm submitting this recheck solely
as a means of testing/demonstrating the recheck framework rebase function.
On Wed, 11 Jun 2025 12:40:31 +0530
vanshika.shu...@nxp.com wrote:
> From: Jun Yang
>
> Add FMan node(s) and associate FMan to it's interface(port).
> This method describes FMan attributes and avoid accessing FMan from
> port directly.
> Logically, something like IEEE 1588 is FMan global resource
On Wed, 11 Jun 2025 12:40:39 +0530
vanshika.shu...@nxp.com wrote:
> From: Hemant Agrawal
>
> This patch improves data access during qman enequeue ring check.
>
> Signed-off-by: Jun Yang
> Signed-off-by: Hemant Agrawal
> ---
Recheck-request: longarch-unit-testing
> Bruce Richardson (3):
> argparse: add support for string and boolean args
> argparse: make argparse EAL-args compatible
> argparse: use enums to remove max-value defines in lists
Applied, thanks.
11/06/2025 15:37, David Marchand:
> This test triggers random failures in the CI.
> Disable it until we have a fix.
>
> Bugzilla ID: 1720
> Fixes: de09387f1313 ("eventdev/vector: add default software vector adapter")
>
> Signed-off-by: David Marchand
Applied, thanks.
On Wed, 11 Jun 2025 16:21:45 +0530
Gagandeep Singh wrote:
> From: Apeksha Gupta
>
> Added all mac counters including Tx stats buckets for
> packet sizes.
>
> Signed-off-by: Apeksha Gupta
> Signed-off-by: Vanshika Shukla
Spelling error, should get fixed.
### [PATCH] net/dpaa2: support dpmac
On Wed, 11 Jun 2025 07:56:21 -0700
Mario Limonciello wrote:
> On 6/11/2025 7:14 AM, Nicolas Dichtel wrote:
> > Le 06/05/2025 à 01:06, Bjorn Helgaas a écrit :
> >> On Wed, Apr 23, 2025 at 11:31:32PM -0500, Mario Limonciello wrote:
> >>> From: Mario Limonciello
> >>>
> >>> AMD BIOS team has ro
11/06/2025 16:57, Stephen Hemminger:
> On Wed, 11 Jun 2025 14:34:49 +0200
> Thomas Monjalon wrote:
>
> > 20/02/2025 17:27, Stephen Hemminger:
> > > --- a/lib/eal/include/rte_string_fns.h
> > > +++ b/lib/eal/include/rte_string_fns.h
> > > +__rte_experimental
> > > +void
> > > +rte_memzero_explicit
Thanks, they have been updated in v5.
> -原始邮件-
> 发件人: "Medvedkin, Vladimir"
> 发送时间: 2025-06-11 19:52:57 (星期三)
> 收件人: u...@foxmail.com, dev@dpdk.org
> 抄送: "Sun Yuechi" , "Thomas Monjalon"
, "Bruce Richardson" ,
"Stanislaw Kardach"
> 主题: Re: [PATCH v4 2/3] lib/lpm: R-V V rte_lpm_lookupx
From: Sun Yuechi
Support using -Dcpu_instruction_set=rv64gcv to enable V extension.
Signed-off-by: Sun Yuechi
---
config/riscv/meson.build | 2 ++
1 file changed, 2 insertions(+)
diff --git a/config/riscv/meson.build b/config/riscv/meson.build
index e3694cf2e6..1036a86d05 100644
--- a/config/
From: Sun Yuechi
This patch is derived from "config/riscv: detect presence of Zbc
extension with modifications".
The RISC-V C api defines architecture extension test macros
These let us detect whether the V extension is supported on the
compiler and -march we're building with. The C api also def
From: Sun Yuechi
This patch series adds support for the RISC-V Vector (V) extension and
provides an optimized implementation of `rte_lpm_lookupx4` using RVV.
The initialization of vtbl_entry is not fully vectorized here because
doing so would require __riscv_vluxei32_v_u32m1, which is slower
tha
From: Sun Yuechi
Implement LPM lookupx4 function for RISC-V architecture using RISC-V
Vector Extension instruction set
Signed-off-by: Sun Yuechi
---
MAINTAINERS | 2 ++
lib/lpm/meson.build | 1 +
lib/lpm/rte_lpm.h | 2 ++
lib/lpm/rte_lpm_rvv.h | 62 ++
On Mon, Jun 09, 2025 at 04:37:16PM +0100, Anatoly Burakov wrote:
> Add a function to check DD bit status, and use it everywhere we do these
> checks.
>
> Signed-off-by: Anatoly Burakov
> ---
Acked-by: Bruce Richardson
> Stephen Hemminger (15):
> eal: introduce new secure memory zero
> app/test: remove unused variable
> eal: add new secure free function
> app/test: use unit test runner for malloc tests
> app/test: add test for rte_free_sensitive
> common/cnxk: remove unused variable
> crypto/qat: fo
Le 06/05/2025 à 01:06, Bjorn Helgaas a écrit :
> On Wed, Apr 23, 2025 at 11:31:32PM -0500, Mario Limonciello wrote:
>> From: Mario Limonciello
>>
>> AMD BIOS team has root caused an issue that NVME storage failed to come
>> back from suspend to a lack of a call to _REG when NVME device was probed.
Previously, `capture_name` defaulted to a "dynamic" value.
This caused non-deterministic documentation and could lead
to overwriting capture files if the method was called
multiple times within a single Python process.
Bugzilla ID: 1718
Signed-off-by: Clemens Famulla-Conrad
---
.mailmap
This test triggers random failures in the CI.
Disable it until we have a fix.
Bugzilla ID: 1720
Fixes: de09387f1313 ("eventdev/vector: add default software vector adapter")
Signed-off-by: David Marchand
---
app/test/test_event_vector_adapter.c | 8 +---
1 file changed, 5 insertions(+), 3 de
On Wed, Jun 11, 2025 at 11:45:31AM +0200, David Marchand wrote:
> From: Andre Muezerie
>
> The archiver tool from the MSVC toolset is lib.exe. It has different
> parameters then it's GNU counterpart "ar".
>
> buildtools\meson.build was updated to use lib.exe when MSVC compiler is
> used. This is
On Wed, Jun 11, 2025 at 11:45:30AM +0200, David Marchand wrote:
> From: Andre Muezerie
>
> DPDK uses GCC attribute "used" through macro __rte_used to indicate
> that a variable not referenced in the code should be assumed being
> used and therefore not be optimized away. This technique is used to
On Wed, Jun 11, 2025 at 11:45:29AM +0200, David Marchand wrote:
> __COUNTER__ is not supported by MSVC.
> Plus the this_pmd_name* symbols will become global (again for MSVC) in a
> next commit.
> Prefer the driver name to generate a unique symbol.
>
> Signed-off-by: David Marchand
> ---
Acked-by
On Wed, Jun 11, 2025 at 11:45:28AM +0200, David Marchand wrote:
> The macros for tagging/exporting informations about a driver do not need
> to be exported in the public API.
> Move this to driver only header.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Andre Muezerie
__COUNTER__ is not supported by MSVC.
Plus the this_pmd_name* symbols will become global (again for MSVC) in a
next commit.
Prefer the driver name to generate a unique symbol.
Signed-off-by: David Marchand
---
buildtools/pmdinfogen.py | 2 +-
drivers/bus/auxiliary/bus_auxilia
On Wed, Apr 2, 2025 at 10:54 AM Jie Hai wrote:
>
> I am moving on to other things and dengdui is going to
> take over the role of hns3 maintainer. Update the
> MAINTAINERS accordingly.
>
> Signed-off-by: Jie Hai
Thanks for all your contributions.
Applied.
--
David Marchand
Hello,
On Wed, Mar 26, 2025 at 2:37 PM Bruce Richardson
wrote:
>
> On Wed, Mar 26, 2025 at 01:30:10PM +, Ian Stokes wrote:
> > Removing myself as maintainer for iavf and i40e.
> >
> > Signed-off-by: Ian Stokes
> > ---
> Acked-by: Bruce Richardson
>
Thanks for all your contributions Ian.
Ap
Sometimes we don't want to parse the string at all, when doing arg
parsing, and just save it off for later. Add support for that.
Also, rather than assuming boolean values have to be the same size as
uint8 (or some other size), add an explicitly type for that - which also
allows checking for true/
The use of lists of #defines with _MAX entries at the end causes issues
for ABI compatibility as those MAX values often leak through to
applications and can cause issues when changed.
We can rework the code to increase type safety by splitting the flags
field and using enums for each set of values
The argparse library was missing two key features which made it
unsuitable for use by EAL or any program wanting similar behaviour.
1. It didn't stop parsing arguments when it hit a "--" character
2. It never returned the number of arguments parsed
Fix both these issues - the latter is a change t
This patchset is based off the work to adjust how we do argument parsing
inside EAL. To enable argparse to be effectively used for EAL, we have
new features and some changes in the first two patches, which are
relatively small - though are ABI/API affecting.
These add support for saving off string
20/02/2025 17:27, Stephen Hemminger:
> --- a/lib/eal/include/rte_string_fns.h
> +++ b/lib/eal/include/rte_string_fns.h
> +__rte_experimental
> +void
> +rte_memzero_explicit(void *dst, size_t sz);
This function is not about strings.
Better to move it to rte_memory.h (even if not ideal).
I'll try to
20/02/2025 17:27, Stephen Hemminger:
> The buffer tmp is set but never used. This leads to warning
> since the memset could be eliminated by the compiler.
>
> Signed-off-by: Stephen Hemminger
> ---
> app/test/test_cmdline_cirbuf.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/app/te
Hi Sun,
You did not address my previous comments regarding commit message. You
can put everything you've wrote in this commit as a note and add
meaningful description about what commit generally does, like (please
correct if needed):
"Implement LPM lookupx4 routine for RISC-V architecture us
28/04/2025 12:37, Nitin Saxena:
> Currently each rte_node registers separate mbuf dynamic fields for their
> own purpose. This leads to wastage of mbuf space as once mbuf get passed
> a particular node, the registered dynamic field(by that node) is no
> longer used.
>
> This patch series adds a gl
24/02/2025 17:40, Ariel Otilibili:
> rte_pcapng_close() might dereference a null pointer; as example,
> PVS-Studio gives its usage in test_pcapng.c: indeed, that call to
> rte_pcapng_close() might receive a null pointer.
>
> Link: https://pvs-studio.com/en/docs/warnings/v522/
> Link:
> https://gi
On Tue, May 13, 2025 at 11:52 AM Mingjin Ye wrote:
>
> This patch reverts commit f5b2eff0847d ("bus/vdev: fix devargs after
> multi-process bus scan")
>
> With current code, we do not add devargs to devargs list when we add a
> vdev in secondary process (because `init` flag is set to `false`).
>
>
From: Chengwen Feng
The DCB-related fields span in multiple structures, this patch moves
them into struct hns3_dcb_info.
Signed-off-by: Chengwen Feng
Signed-off-by: Dengdui Huang
---
drivers/net/hns3/hns3_dcb.c | 13 +--
drivers/net/hns3/hns3_ethdev.c| 38 +++
When doing a build for a target that already has the instruction sets
for AVX2/AVX512 enabled, skip emitting the AVX compiler flags, or the
x86-64-v4 '-march' flags, as they are unnecessary. Instead, when
the default flags produce the desired output, just use them unmodified,
and don't bother addin
From: Jun Yang
1) The max size of extract buffer should be size of
"struct dpni_ext_set_rx_tc_dist"(488), which is greater than 256.
2) dpkg_prepare_key_cfg is responsible to clear
extract buffer before preparing.
Fixes: 25d0ae624245 ("net/dpaa2: check IOVA before sending MC command")
Cc:
https://bugs.dpdk.org/show_bug.cgi?id=1722
Kevin Traynor (ktray...@redhat.com) changed:
What|Removed |Added
Resolution|--- |WONTFIX
Status|
From: Apeksha Gupta
Add support of dpmac counters in xstats.
Signed-off-by: Apeksha Gupta
Signed-off-by: Vanshika Shukla
---
drivers/net/dpaa2/dpaa2_ethdev.c| 118 ++--
drivers/net/dpaa2/dpaa2_ethdev.h| 12 +++
drivers/net/dpaa2/mc/dpni.c | 29 ++-
This patch adds support for software taildrop on ordered queues
in the DPAA2 driver.
It also enables congestion notification by default on
traffic management (TM) queues, which is a prerequisite for
software taildrop functionality.
Signed-off-by: Gagandeep Singh
---
drivers/net/dpaa2/dpaa2_rxtx
From: Ioana Ciornei
The list of availabla DPNI APIs depends on the DPNI API version
available on a system. Retrieve the API version in the init stage and
store it in the private structure so that it can be used at a later
stage.
Signed-off-by: Ioana Ciornei
---
.mailmap
From: Ioana Ciornei
Previously, the speed_capa field from struct rte_eth_dev_info was
populated with all possible speeds that the MACs on the system could
sustain. What this meant is that the bitmap of speed capability did not
reflect what the MAC could do in the current circumstance.
Fix this b
From: Apeksha Gupta
Added all mac counters including Tx stats buckets for
packet sizes in xstats.
Signed-off-by: Apeksha Gupta
Signed-off-by: Vanshika Shukla
---
drivers/net/dpaa2/dpaa2_ethdev.c | 20 +++-
drivers/net/dpaa2/dpaa2_ethdev.h | 4 ++--
2 files changed, 21 inserti
From: Apeksha Gupta
Added all mac counters including Tx stats buckets for
packet sizes.
Signed-off-by: Apeksha Gupta
Signed-off-by: Vanshika Shukla
---
drivers/net/dpaa2/mc/fsl_dpmac.h | 48 +++-
1 file changed, 47 insertions(+), 1 deletion(-)
diff --git a/drivers
From: Apeksha Gupta
This patch upgrades the MC firmware release API compatibility
to 10.39.0. Added mc/fsl_dpmac.h file.
Signed-off-by: Apeksha Gupta
---
drivers/net/dpaa2/mc/fsl_dpmac.h | 465 +++
1 file changed, 465 insertions(+)
create mode 100644 drivers/net/dp
From: Jun Yang
Use DPAA2_VADDR_TO_IOVA to convert va to iova for both VA and PA modes.
Signed-off-by: Jun Yang
---
drivers/mempool/dpaa2/dpaa2_hw_mempool.c | 222 +++
1 file changed, 142 insertions(+), 80 deletions(-)
diff --git a/drivers/mempool/dpaa2/dpaa2_hw_mempool.c
From: Vanshika Shukla
This patch adds DPBP APIs for setting and getting the depletion/surplus
thresholds for controlling pause frame generation.
By default, depletion_entry is set as 128 by MC. This API would help user
configure the buffer in depletion state as per requirement.
Signed-off-by: V
This patch fixes the shaper rate by configuring the
user given rate in bytes. Earlier driver was considering
the user given rate value in bits.
Fixes: ac624068ee25 ("net/dpaa2: support traffic management")
Cc: sta...@dpdk.org
Signed-off-by: Gagandeep Singh
---
drivers/net/dpaa2/dpaa2_tm.c | 8 +
V1-logs:
This patch series introduces enhancements and fixes to the
NXP DPAA2 Ethernet driver.
It includes support for
- software taildrop on ordered queues.
- setup speed capabilities.
- DPAA2 resource version.
- MAC level statistics.
- improve PA-VA conversion.
- add buffer pool depletion s
https://bugs.dpdk.org/show_bug.cgi?id=1672
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
Resolution|--- |FIXED
St
https://bugs.dpdk.org/show_bug.cgi?id=1683
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Res
> Ankur Dwivedi (14):
> fib: move macro to header file
> node: add IP4 lookup FIB node
> node: add IP4 FIB route add
> node: add process callback for IP4 FIB
> node: move next nodes to public header file
> node: add next node in packet classification
> app/graph: add IP4 lookup mode c
https://bugs.dpdk.org/show_bug.cgi?id=1668
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Res
Get rss hash configuration, rss_hf is always the default value.
The driver does nothing if the rss key is invalid during the rss
hash update.
This patch is get the current configuration of rss_hf. Extract
the update rss key code from ice_dcf_dev_rss_hash_update to
ice_dcf_set_rss_key and make it c
The macros for tagging/exporting informations about a driver do not need
to be exported in the public API.
Move this to driver only header.
Signed-off-by: David Marchand
---
lib/eal/include/dev_driver.h | 41
lib/eal/include/rte_dev.h| 41
From: Andre Muezerie
DPDK uses GCC attribute "used" through macro __rte_used to indicate
that a variable not referenced in the code should be assumed being
used and therefore not be optimized away. This technique is used to embed
information in the binaries, by having crafted information stored i
From: Andre Muezerie
Script usertools\dpdk-pmdinfo.py was enhanced to also be able to parse
symbols from sections in PE images.
Signed-off-by: Andre Muezerie
Acked-by: Robin Jarry
---
usertools/dpdk-pmdinfo.py | 53 +++
1 file changed, 43 insertions(+), 10
From: Andre Muezerie
The archiver tool from the MSVC toolset is lib.exe. It has different
parameters then it's GNU counterpart "ar".
buildtools\meson.build was updated to use lib.exe when MSVC compiler is
used. This is to allow the code to be built without requiring GNU "ar"
to be installed in t
This is a rework of Andre v5, trying to hide the ugly details in a
driver only macro not exposed publicly.
--
David Marchand
Changes since v6:
- changed this_pmd_name symbols construction,
Changes since v5:
- moved details in dev_driver.h,
- renamed RTE_INCLUDE as RTE_PMD_EXPORT_SYMBOL,
Andre
https://bugs.dpdk.org/show_bug.cgi?id=1664
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
Resolution|--- |FIXED
https://bugs.dpdk.org/show_bug.cgi?id=1722
Bug ID: 1722
Summary: Kernel crash with WARNING: CPU: 17 PID: 4510 at
lib/refcount.c:28 refcount_warn_saturate+0xd8/0xe0
Product: DPDK
Version: 20.11
Hardware: x86
On Tue, Jun 10, 2025 at 10:37 AM David Marchand
wrote:
>
> MSVC warns about atomic qualifier:
> ../app/test/test_per_lcore.c(101): error C2220: the following warning is
> treated as an error
> ../app/test/test_per_lcore.c(101): warning C4090: 'function': different
> '_Atomic' quali
From: Chengwen Feng
The VF multi-TCs feature depends on firmware and PF driver, the
capability was set when:
1) Firmware report VF multi-TCs flag.
2) PF driver report VF multi-TCs flag.
3) PF driver support query multi-TCs info mailbox message.
Signed-off-by: Chengwen Feng
Signed-off-by: Dengdu
From: Chengwen Feng
The mailbox message HNS3_MBX_GET_BASIC_INFO can obtain the maximum
number of TCs of the device. The VF does not support multiple TCs,
therefore, this field is not saved.
Now the VF needs to support multiple TCs, therefore, this field needs
to be saved.
This commit also suppo
From: Chengwen Feng
If VF has the multi-TCs capability, then application could configure the
multi-TCs feature through the DCB interface. Because VF does not have
its own ETS and PFC components, the constraints are as follows:
1. The DCB configuration (struct rte_eth_dcb_rx_conf and
rte_eth_d
From: Chengwen Feng
The VF cannot configure the mapping of queue to TC by directly writing
the register. Instead, the mapping must be modified by using firmware
command.
Fixes: bba636698316 ("net/hns3: support Rx/Tx and related operations")
Cc: sta...@dpdk.org
Signed-off-by: Chengwen Feng
Sign
This patchset adds the VF multi-TCs feature.
Chengwen Feng (6):
net/hns3: fix VF fail to config queue TC
net/hns3: remove duplicate struct field
net/hns3: refactor DCB module code
net/hns3: VF support parse max TC number
net/hns3: VF support discover multi-TCs capability
net/hns3: VF s
From: Chengwen Feng
The struct hns3_hw and hns3_hw.dcb_info both has num_tc field, their
meanings are the same, to ensure code readability, remove the num_tc
field of struct hns3_hw.
Signed-off-by: Chengwen Feng
Signed-off-by: Dengdui Huang
---
drivers/net/hns3/hns3_dcb.c | 44 +
From: Gagandeep Singh
This patch addresses DPAA driver issues with the introduction of
rte_eal_cleanup, which caused driver-specific destructors to fail
due to memory cleanup.
To resolve this, we remove the driver destructor and relocate the
code to the bus cleanup function.
So, this patch also
From: Vanshika Shukla
Currently, error queue is mapped to the Rx queue and enabled by default.
This patch adds the devargs to control the err packets on main queue.
Also, in VSP mode the error queue should be disabled because the error
packets from kernel are diverted to the Rx queue/err queue ca
05/05/2025 08:24, kirankum...@marvell.com:
> From: Kiran Kumar K
>
> The node id is determined based on a global variable that is
> incremented every time a node is created. Adding changes to
> remove the global counter. Make sure that the node list is always
> ordered by increasing node ids. Whe
From: Jun Yang
1) Reduce byte swap between big endian and little endian.
2) Reduce ci(cache invalid) access by 128bit R/W instructions.
These methods improve ~10% buffer acquire/release performance.
Signed-off-by: Jun Yang
---
drivers/bus/dpaa/base/qbman/bman.c | 149 -
From: Jun Yang
Adjust every element of pool by populate callback.
1) Make sure start DMA address is aligned with 16B.
2) For buffer across 4KB boundary, make sure start DMA address is
aligned with 256B.
Signed-off-by: Jun Yang
---
drivers/mempool/dpaa/dpaa_mempool.c | 145 ++
From: Vanshika Shukla
This series includes fixes for NXP DPAA drivers.
V4:
Fixed compilation errors in the "bus/dpaa: optimize qman enqueue check" patch
on Ubuntu 22.04.
- Ensured buffer bounds are respected and added validation for buffer count
extracted from mcr->verb.
V3:
Fixed compilation e
From: Jun Yang
Add FMan node(s) and associate FMan to it's interface(port).
This method describes FMan attributes and avoid accessing FMan from
port directly.
Logically, something like IEEE 1588 is FMan global resource,
which is in range of 0xF_E000–0xF_EFFF.
Port specific resource is in range of
From: Hemant Agrawal
This patch improves data access during qman enequeue ring check.
Signed-off-by: Jun Yang
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/base/fman/fman.c | 3 ++-
drivers/bus/dpaa/base/qbman/bman.c | 6 ++---
drivers/bus/dpaa/base/qbman/qman.c | 41 ++
From: Vinod Pullabhatla
Add support to set Tx rate on DPAA platform through PMD APIs
Signed-off-by: Vinod Pullabhatla
Signed-off-by: Vanshika Shukla
---
.mailmap | 1 +
drivers/net/dpaa/dpaa_flow.c | 87 +++-
drivers/net/dpaa/fmlib/
From: Jun Yang
Use new BMan APIs to improve performance and support burst release.
Improve release performance ~90% by burst release.
Signed-off-by: Jun Yang
---
drivers/mempool/dpaa/dpaa_mempool.c | 85 -
drivers/mempool/dpaa/dpaa_mempool.h | 2 +-
2 files changed
From: Hemant Agrawal
rte_dpaa_bus was being used as structure and variable name both.
Signed-off-by: Jun Yang
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/dpaa_bus.c | 56 ++---
1 file changed, 28 insertions(+), 28 deletions(-)
diff --git a/drivers/bus/d
From: Jun Yang
Provide internal API to identify DPAA1 SoC version instead of accessing
global variable directly.
Signed-off-by: Jun Yang
---
drivers/bus/dpaa/base/qbman/qman.c | 9 +++---
drivers/bus/dpaa/bus_dpaa_driver.h | 9 +++---
drivers/bus/dpaa/dpaa_bus.c| 48 +
96 matches
Mail list logo
