02/10/2025 09:37, Morten Brørup:
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > --- a/config/meson.build
> > +++ b/config/meson.build
> > @@ -379,6 +379,7 @@ if get_option('mbuf_refcnt_atomic')
> > dpdk_conf.set('RTE_MBUF_REFCNT_ATOM
01/10/2025 12:24, Shreesh Adiga:
> On Wed, Oct 1, 2025 at 1:25 PM Thomas Monjalon wrote:
>
> > 29/09/2025 18:28, Shreesh Adiga:
> > > On Wed, Sep 24, 2025 at 8:28 PM Thomas Monjalon
> > wrote:
> > >
> > > > Hello,
> > > >
> > &
From: Shani Peretz
Dump the mbuf history to the console or to a file.
The dump will contain:
- Operation history for each mbuf
- Summary and statistics about all mbufs
Dump the history of all mbuf pools:
testpmd> dump_mbuf_history_all
Dump the history of one mbuf pool:
testpmd> dump_mbuf_histo
From: Shani Peretz
Mark the Rx/Tx steps in mbufs for debugging.
It has no performance impact if mbuf history is disabled (by default).
Signed-off-by: Shani Peretz
---
lib/ethdev/rte_ethdev.h | 15 +++
1 file changed, 15 insertions(+)
diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethd
detailed marks in the mlx5 PMD is removed from this series.
Shani Peretz (4):
mbuf: record mbuf operations history
ethdev: mark mbufs in burst functions
app/testpmd: add commands to dump mbuf history
usertools/mbuf: parse mbuf history dump
Thomas Monjalon (1):
mbuf: move header include for
23/09/2025 15:28, Bruce Richardson:
> On Tue, Sep 23, 2025 at 02:08:35PM +0100, Kevin Traynor wrote:
> > Yes, that is a good point for discussion. Seen as support for the legacy
> > names were already dropped and I wasn't aware of any ABI like policy
> > about it, I thought there may be a preferenc
26/09/2025 02:04, Stephen Hemminger:
> On Wed, 27 Aug 2025 16:32:15 -0600
> Schneide wrote:
>
> >
> > +* pcapng: Changed the API for adding interfaces to include a link type
> > argument.
> > + The link type was previously hardcoded to the ethernet link type in the
> > API.
> > + This argum
25/09/2025 12:33, Akhil Goyal:
> > +/**
> > + * Constant-time memory comparison for cryptographic use.
> > + * Returns 0 if the memory regions are equal, nonzero otherwise.
> > + * Runs in constant time with respect to the length to prevent timing
> > attacks.
> > + *
> > + * @param a
> > + * Po
24/09/2025 17:48, Bruce Richardson:
> The file "rte_thash.c" uses the rte_cpu_get_flag_enabled() API but does
> not include "rte_cpuflags.h". Add the include to fix build issues this
> causes:
>
> ../lib/hash/rte_thash.c: In function ‘rte_thash_gfni_supported’:
> ../lib/hash/rte_thash.c:79:13: err
ime checks:
- efd: remove AVX2 build-time check
- member: remove AVX2 build-time checks
because I'm not sure losing inlining for cleanup is worth.
Thomas Monjalon (2):
efd: fix AVX2 support
member: hide internal macro
lib/efd/rte_efd.c | 3 ++-
lib/member/member.h | 9 ++
The hash function used by the library is not supposed
to be exposed and be part of the API.
Signed-off-by: Thomas Monjalon
Acked-by: Bruce Richardson
---
lib/member/member.h | 9 +
lib/member/rte_member.h | 9 -
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a
checking AVX2 availability,
so the obsolete build-time check for AVX2 can be simply removed,
and the missing include added back.
Fixes: 5b9656b157d3 ("lib: build with meson")
Fixes: 30a1de105a5f ("lib: remove unneeded header includes")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Mo
22/09/2025 09:40, Bruce Richardson:
> On Fri, Sep 19, 2025 at 09:02:26PM +0200, Thomas Monjalon wrote:
> > rte_cpuflags.h is not used in rte_spinlock.h,
> > so it is removed, and added back in .c files where needed.
> >
> > Signed-off-by: Thomas Monjalon
> >
18/09/2025 14:17, Thomas Monjalon:
> Fix the namespace of include guards by prefixing with "RTE",
> and add a missing include guard in rte_eal_paging.h.
>
> Signed-off-by: Thomas Monjalon
These are minor changes.
Applied
Hello,
16/07/2025 12:34, Shreesh Adiga:
> Replace the clearing of lower 32 bits of XMM register with blend of
> zero register.
> Replace the clearing of upper 64 bits of XMM register with _mm_move_epi64.
> Clang is able to optimize away the AND + memory operand with the
> above sequence, however G
24/07/2025 13:05, Jerin Jacob:
> From: David Christensen
> >
> > Current code in eal_trace_init() specifies 8 byte alignment for CTF
> > generation,
> > but fallback code in __rte_trace_mem_per_thread() does not enforce similar
> > requirements when calling malloc(). Modify fallback heap request
31/07/2025 20:31, Tofig Aliev:
> Added immediate return from ntb_dev_intr_handler()
> if no doorbells.
Please could you explain the consequence of this bug?
Should it be backported?
[...
> @@ -158,8 +158,10 @@ ntb_dev_intr_handler(void *param)
> }
>
> db_bits = (*hw->ntb_ops->db_re
23/07/2025 20:53, Wathsala Vithanage:
> Add Microsoft Azure Cobalt-100 to arm build.
>
> Signed-off-by: Wathsala Vithanage
> Reviewed-by: Dhruv Tripathi
> ---
> +soc_cobalt100 = {
> +'description' : 'Microsot Azure Cobalt 100',
typo: Microsoft
Applied with fixed typo, thanks.
22/09/2025 13:07, Kevin Traynor:
> Up until DPDK 25.03 'net/*' could be used with meson options
> enable_drivers or disable_drivers to explicitly enable or
> disable all net drivers.
>
> In DPDK 25.03 commit
> c1d145834f28 ("net/intel: move Intel drivers to a subdirectory")
> moved Intel drivers t
22/09/2025 13:07, Kevin Traynor:
> --- a/drivers/meson.build
> +++ b/drivers/meson.build
> -# add cmdline disabled drivers and meson disabled drivers together
> -disable_drivers += ',' + get_option('disable_drivers')
> +# map legacy driver names
> +driver_map = {
> +'net/e1000': 'net/intel/e100
03/09/2025 15:54, Maxime Coquelin:
> On 8/27/25 10:21 AM, Thomas Monjalon wrote:
> > Most messages from the Linux uAPI script are not useful
> > when running the script in a regular fashion.
> > The quiet mode (option -q) is added to output only error messages if any.
&
18/09/2025 09:28, Thomas Monjalon:
> These are fixes for AVX2 in efd and member libraries.
> While at it, I've hidden a macro which was wrongly exported in the API
> without having a correct prefix.
>
> Thomas Monjalon (3):
> efd: fix AVX2 support
> member: rem
08/09/2025 23:17, Thomas Monjalon:
> After an upgrade to MinGW version 13, some compilation errors appear:
>
> drivers/bus/pci/windows/pci.c:362:58:
> error: 'GUID_DEVCLASS_NETUIO' undeclared
> drivers/bus/pci/windows/pci_netuio.c:57:39:
> error: 'GUI
rte_cpuflags.h is not used in rte_spinlock.h,
so it is removed, and added back in .c files where needed.
Signed-off-by: Thomas Monjalon
---
drivers/crypto/ipsec_mb/ipsec_mb_private.c | 1 +
lib/eal/x86/include/rte_spinlock.h | 1 -
lib/eal/x86/rte_power_intrinsics.c | 1 +
lib
18/09/2025 16:37, 孙越池:
> > 发件人: "Thomas Monjalon"
> > 27/11/2024 04:26, Eric Long:
> > > On 27/11/2024 04:29, David Marchand wrote:
> > > > You can extend the timeout via the multiplier option (default
> timeout
> > > > of 10s * mul
19/09/2025 20:52, 孙越池:
> > Please could you add a cross file in config/riscv/ for rv64gcv
> > and use it in devtools/test-meson-builds.sh so we can test compiling
> > with the vector extension?
>
> Sure, I plan to submit this next week.
>
> > Another question about RISC-V maintenance in general,
19/09/2025 18:33, u...@foxmail.com:
> 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
> compile
19/09/2025 18:33, u...@foxmail.com:
> From: Sun Yuechi
> +#ifndef _DIR248_RVV_H_
> +#define _DIR248_RVV_H_
> +
> +#include "rte_cpuflags.h"
It should be
#include
and it should be only where really needed (.c probably).
Don't bother doing another version, I'm working on it,
and will merg
01/07/2025 20:20, u...@foxmail.com:
> From: Sun Yuechi
>
> Same as the x86 style, include "rte_cpuflags.h" in rte_spinlock.h
> so that files like lib/fib/dir24_8.c don’t need to include it.
>
> Signed-off-by: Sun Yuechi
I disagree.
I prefer removing useless includes like in x86 spinlock.
Since all supported compilers can generate AVX2 code,
it is possible to force AVX2 compilation on some specific functions
and remove the checks for AVX2 support when x86 arch is already checked.
The functions have to be moved in a .c file, losing inlining.
Signed-off-by: Thomas Monjalon
18/09/2025 14:36, kumaraparameshwaran rathinavel:
> Reviewed by : @kumaraparameshwaran rathinavel
>
> On Tue, Sep 2, 2025 at 4:14 PM Tathagat Priyadarshi
> wrote:
>
> > In the current implementation when cmp is larger than 0,
> > the new packet is appended to the original packet. However
> > th
18/09/2025 09:55, Shani Peretz:
> From: Shani Peretz
> >
> > When --iova-mode is explicitly specified in command line, DMA mask
> > constraints were not being validated, leading to potential runtime failures
> > when device DMA capabilities are exceeded.
> >
> > The issue occurred because rte_bu
The hash function used by the library is not supposed
to be exposed and be part of the API.
Signed-off-by: Thomas Monjalon
---
lib/member/member.h | 9 +
lib/member/rte_member.h | 9 -
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/lib/member/member.h b/lib
06/09/2025 06:11, Stephen Hemminger:
> On Wed, 3 Sep 2025 17:17:58 +0200
> David Marchand wrote:
>
> > Import VFIO header (from v6.16) to be included in many parts of DPDK.
> >
> > Signed-off-by: David Marchand
> > ---
> > kernel/linux/uapi/linux/vfio.h | 1836
Fix the namespace of include guards by prefixing with "RTE",
and add a missing include guard in rte_eal_paging.h.
Signed-off-by: Thomas Monjalon
---
lib/eal/include/rte_eal_paging.h | 11 ---
lib/eal/include/rte_keepalive.h | 6 +++---
lib/jobstats/rte_jobstats.h | 6 +
These are fixes for AVX2 in efd and member libraries.
While at it, I've hidden a macro which was wrongly exported in the API
without having a correct prefix.
Thomas Monjalon (3):
efd: fix AVX2 support
member: remove AVX2 build-time checks
member: hide internal macro
lib/efd/rte_
27/11/2024 04:26, Eric Long:
> On 27/11/2024 04:29, David Marchand wrote:
> > You can extend the timeout via the multiplier option (default timeout
> > of 10s * multiplier).
> > So in your case:
> > $ meson test -C --suite fast-tests -t 6
>
> I hope the RISC-V specific extended timeout could be u
11/09/2025 04:32, Junlong Wang:
> modify meson.build for support compiling on riscv-64 systems
[...]
> -if not is_linux
> +if not is_linux or not dpdk_conf.get('RTE_ARCH_64')
> build = false
> -reason = 'only supported on Linux'
> -subdir_done()
> -endif
> -
> -if arch_subdir != 'x86'
18/09/2025 11:40, Thomas Monjalon:
> 18/09/2025 11:08, Thomas Monjalon:
> > When switching to Meson build, the compilation check on CC_SUPPORT_AVX2
> > became obsolete, thus the case EFD_LOOKUP_AVX2 became dead.
> > The function efd_lookup_internal_avx2() was never calle
18/09/2025 11:08, Thomas Monjalon:
> When switching to Meson build, the compilation check on CC_SUPPORT_AVX2
> became obsolete, thus the case EFD_LOOKUP_AVX2 became dead.
> The function efd_lookup_internal_avx2() was never called,
> and its header include rte_efd_x86.h has been r
The hash function used by the library is not supposed
to be exposed and be part of the API.
Signed-off-by: Thomas Monjalon
Acked-by: Bruce Richardson
---
lib/member/member.h | 9 +
lib/member/rte_member.h | 9 -
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a
Since all supported compilers can generate AVX2 code,
it is possible to force AVX2 compilation on the specific function
and remove the check for AVX2 support.
The function has to be moved in a .c file, losing inlining.
Signed-off-by: Thomas Monjalon
---
lib/efd/{rte_efd_x86.h => efd_avx
with supported compilers,
so the check for AVX2 is simply removed, and the include is added back.
Fixes: 5b9656b157d3 ("lib: build with meson")
Fixes: 30a1de105a5f ("lib: remove unneeded header includes")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
---
lib/efd/rte_e
I've found the AVX2 function is not used in efd library.
Then I tried to make AVX2 compilation simpler in efd and member libraries.
While at it, I've hidden a macro which was wrongly exported in the API
without having a correct prefix.
Thomas Monjalon (4):
efd: fix AVX2 support
e
Hello,
17/09/2025 16:07, Arik Alexander:
> At Cato Networks we are currently working on implementing LACP LAG (bonding
> device, mode 4) using the DPDK library.
> During our evaluation, we identified some shortcomings in the implementation
> — for example, the lack of support for short timeout a
18/09/2025 09:48, Bruce Richardson:
> On Thu, Sep 18, 2025 at 09:28:03AM +0200, Thomas Monjalon wrote:
> > -#if defined(RTE_ARCH_X86) && defined(CC_SUPPORT_AVX2)
> > +#if defined(RTE_ARCH_X86)
>
> Ok to remove this, because indeed all supported compilers have AVX2.
Since all supported compilers can generate AVX2 code,
no need to check for AVX2 support when x86 arch is already checked.
Signed-off-by: Thomas Monjalon
---
lib/member/rte_member_ht.c | 14 +++---
lib/member/rte_member_x86.h | 3 ---
2 files changed, 7 insertions(+), 10 deletions
with supported compilers,
so the checks for AVX2 are simply removed, and the include is added back.
Fixes: 5b9656b157d3 ("lib: build with meson")
Fixes: 30a1de105a5f ("lib: remove unneeded header includes")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
---
lib/ef
15/09/2025 16:41, Gregory Etelson:
> From: Bing Zhao
>
> The API name and prototype have been changed in the new CUDA, this
> commit will ensure the compatibility with different versions.
[...]
> /* CUDA Driver functions loaded with cuGetProcAddress for versioning */
> +#if defined(CUDA_VERSION)
Hello,
The title can be simply "ethdev: add link type".
08/09/2025 10:51, sk...@marvell.com:
> +* **Added ethdev API in library.*
Please be more specific, we don't need to group all ethdev changes together.
Each new API feature deserves a separate bullet.
> +
> + * Added API to report type of
25/08/2025 09:23, Shani Peretz:
> v3:
> add comma in the meson.build file to fix build error
>
> v2:
> sent mistakely as RFC - remove RFC tag, no functional changes
>
> v1:
> This patch series adds port affinity matching for bonded ports and
> refactors queue management to support the new port af
15/09/2025 16:41, Gregory Etelson:
> From: eagostini
>
> Add new NVIDIA GPU IDs taken from public driver list.
> https://github.com/NVIDIA/open-gpu-kernel-modules/blob/e45d91de0224e17a6bf7c971d8f165090b3c07a6/src/nvidia/generated/g_nv_name_released.h
It is not synchronized.
For instance, there a
> > ARM cortex-a78ae configuration (part number 0xd42) is being built without
> > the
> > necessary crypto extensions. This results in infinite loops in the crypto
> > performance tests applications because the hardware crypto features are not
> > properly enabled.
> > This fix enables the crypto
Recheck-request: rebase=main, iol-mellanox-Functional,
iol-compile-amd64-testing, iol-compile-arm64-testing, iol-unit-amd64-testing,
iol-unit-arm64-testing
16/09/2025 23:17, Stephen Hemminger:
> On Tue, 16 Sep 2025 18:12:04 +0300
> Shani Peretz wrote:
>
> > @@ -607,6 +608,9 @@ static inline struct rte_mbuf
> > *rte_mbuf_raw_alloc(struct rte_mempool *mp)
> > if (rte_mempool_get(mp, &ret.ptr) < 0)
> > return NULL;
> > __rte_mbuf_r
16/09/2025 23:14, Stephen Hemminger:
> On Tue, 16 Sep 2025 18:12:05 +0300
> Shani Peretz wrote:
>
> > record operations on mbufs when it is allocated
> > and released inside the mlx5 PMD.
> >
> > Signed-off-by: Shani Peretz
> > ---
>
> If you are adding this to one driver, it means it should b
15/09/2025 09:54, Dengdui Huang:
> I am moving on to other things, Chengwen and Xingui will take on
> the role of hns3 PMD maintainers. Update the MAINTAINERS accordingly.
>
> Signed-off-by: Dengdui Huang
> ---
> Hisilicon hns3
> -M: Dengdui Huang
> +M: Xingui Yang
> +M: Chengwen Feng
> F: d
04/09/2025 17:06, Harry van Haaren:
> Remove myself from DPDK maintainership, as I will no longer be
> working at Intel or directly with DPDK.
>
> Signed-off-by: Harry van Haaren
>
> ---
>
> I'll take this opportunity to thank Intel & DPDK community!
> Its been great to be involved in DPDK, bui
07/08/2025 16:30, Luca Vizzarro:
> The MAINTAINERS file was only updated to replace Paul with Luca for the
> next-dts tree only. This commit adds Luca as a maintainer for the dts
> folder on the main tree as well.
>
> Signed-off-by: Luca Vizzarro
> ---
> DTS
> -M: Paul Szczepanek
> +M: Luca Viz
03/09/2025 13:00, Morten Brørup:
> Add myself as maintainer for the packet buffer (mbuf) library.
>
> Signed-off-by: Morten Brørup
> ---
> Packet buffer
> +M: Morten Brørup
> F: lib/mbuf/
Acked-by: Thomas Monjalon
Applied, thanks for volunteering.
12/09/2025 13:35, Kerem Aksu:
> dumpcap crashes when trying to capture from af_packet devices. This is
> caused by allocating interface name with
> strdup (i.e. malloc). Interface name is not accessible from secondary
> process and causes segmentation fault. Use rte_malloc instead of
> strdup to fi
12/09/2025 10:47, Kerem Aksu:
> dumpcap crashes when trying to capture from af_packet devices. This is
> caused by allocating interface name with
> strdup (i.e. malloc). Interface name is not accessible from secondary
> process and causes segmentation fault. Use rte_malloc instead of
> strdup to fi
08/09/2025 11:35, Thomas Monjalon:
> 27/08/2025 17:16, Thomas Monjalon:
> > After an upgrade to MinGW version 13 on Arch Linux,
> > some compilation errors appear in PCI, mlx5 and bbdev code.
> > This series fixes them all.
> >
> > Bruce doesn't see them
).
Also it does not make sense to check this value to return the update status
as it does not change.
Fixes: 6fbd73709ee4 ("net/mlx5: support link update on Windows")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
Acked-by: Dariusz Sosnowski
Acked-by: Bruce Richardson
---
driv
10/09/2025 02:50, Stephen Hemminger:
> On Mon, 8 Sep 2025 14:21:48 +0530
> wrote:
>
> > From: Sunil Kumar Kori
> >
> > Adding link type parameter to provide the type
> > of port like twisted pair, fibre etc.
> >
> > Also added an API to convert the RTE_ETH_LINK_CONNECTOR_XXX
> > to a readable
08/09/2025 23:17, Thomas Monjalon:
> + const char *status_str = rte_bbdev_enqueue_status_str(i);
> + if (status_str == NULL)
> + continue;
> if (q_data->queue_stats.enqueue_status_count[i] > 0)
>
09/09/2025 10:08, Renyong Wan:
> Thanks David for catching this issue.
> We'll address it in the 25.11 release.
I don't understand your answer.
Do you ack this change?
We want to merge it today because it is breaking our CI on the main branch
(next-net has been pulled yesterday).
> On 2025/9/9 1
18/08/2025 08:32, Raslan Darawsheh:
> On 31/07/2025 9:08 AM, Gregory Etelson wrote:
> > mlx5_ext_rxq_get() and mlx5_ext_txq_get() functions did not return
> > NULL value if query index was not referencing external queue.
> >
> > As a result, calling functions did not expect the NULL on return.
It
After an upgrade to MinGW version 13, some compilation errors appear
in PCI, mlx5 and bbdev code.
This series fixes them all.
v2: improve bbdev patch
v3: try to fix Windows build of PCI patch
Thomas Monjalon (3):
bus/pci: fix build with MinGW 13
net/mlx5: fix build with MinGW 13
bbdev: fix
debug dump")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
Acked-by: Bruce Richardson
Acked-by: Hemant Agrawal
---
v2: make status_str variable local
---
lib/bbdev/rte_bbdev.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/lib/bbdev/rte_bbdev.c b/lib/bbd
.20253")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
Acked-by: Bruce Richardson
---
v3: try to fix Windows build by moving includes
---
drivers/bus/pci/windows/pci.c| 11 ++-
drivers/bus/pci/windows/pci_netuio.c | 6 --
drivers/bus/pci/windows/pci_netuio.h | 13 +
08/09/2025 21:58, Stephen Hemminger:
> On Mon, 08 Sep 2025 13:53:38 +0200
> Thomas Monjalon wrote:
>
> > 13/08/2025 23:06, Stephen Hemminger:
> > > On Wed, 13 Aug 2025 15:33:15 +0800
> > > Dengdui Huang wrote:
> > >
> > > > The hns3 dr
13/08/2025 23:06, Stephen Hemminger:
> On Wed, 13 Aug 2025 15:33:15 +0800
> Dengdui Huang wrote:
>
> > The hns3 driver supports configuring RSS through both ops API and
> > rte_flow API. The ops API uses spink lock, while the rte_flow API uses
> > pthread mutex lock. When concurrent calls occur,
27/08/2025 10:21, Thomas Monjalon:
> The main goal of this series is to check Linux uAPI headers quietly.
> While doing so, few minor improvements are done.
>
> Thomas Monjalon (6):
> devtools: remove superfluous curly braces in uAPI script
> devtools: remove superfluous
27/08/2025 17:16, Thomas Monjalon:
> After an upgrade to MinGW version 13 on Arch Linux,
> some compilation errors appear in PCI, mlx5 and bbdev code.
> This series fixes them all.
>
> Bruce doesn't see them with MinGW 13.2 on OpenSUSE Tumbleweed,
> but perhaps just mi
03/09/2025 17:09, Matthew McGovern (LINUX):
> No reason other than to keep the patch small. I'll add some additional
> patches for the other apps.
>
> -Original Message-----
> From: Thomas Monjalon
> Sent: Wednesday, July 9, 2025 7:30 AM
> To: Matthew McGov
30/08/2025 19:17, Vladimir Medvedkin:
> Vladimir Medvedkin (6):
> ethdev: extend and refactor DCB configuration
> ethdev: remove nb_tcs from rte_eth_dcb_conf structure
> ethdev: decouple VMDq and DCB cofiguration
> ethdev: extend VMDq/DCB configuration with queue mapping
> ethdev: remove
28/08/2025 06:49, rongwei liu:
>
> On 2025/8/28 12:40, Stephen Hemminger wrote:
> > On Thu, 28 Aug 2025 06:21:34 +0300
> > Rongwei Liu wrote:
> >
> >> With dedicated RDMA link monitor, there are two threads
> >> which can update the IB device port information.
> >>
> >> Add a new flag to avoid t
27/08/2025 17:08, Morten Brørup:
> > From: Burakov, Anatoly [mailto:anatoly.bura...@intel.com]
> > Sent: Wednesday, 27 August 2025 16.14
> >
> > On 8/20/2025 8:42 AM, Morten Brørup wrote:
> > >> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> > >> Sent: Monday, 18 August 2025 18.34
>
debug dump")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
Acked-by: Bruce Richardson
---
v2: make status_str variable local
---
lib/bbdev/rte_bbdev.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/lib/bbdev/rte_bbdev.c b/lib/bbdev/rte_bbdev.c
i
).
Also it does not make sense to check this value to return the update status
as it does not change.
Fixes: 6fbd73709ee4 ("net/mlx5: support link update on Windows")
Cc: sta...@dpdk.org
Signed-off-by: Thomas Monjalon
Acked-by: Dariusz Sosnowski
Acked-by: Bruce Richardson
---
driv
as set NTDDI_VERSION to the highest version
without defining the expected NETUIO constants.
It is safer to not rely on Windows headers version,
and instead define what is not already defined.
Fixes: 6605c7f02e24 ("bus/pci: fix build with Windows SDK >= 10.0.20253")
Cc: sta...@dpdk.or
After an upgrade to MinGW version 13 on Arch Linux,
some compilation errors appear in PCI, mlx5 and bbdev code.
This series fixes them all.
Bruce doesn't see them with MinGW 13.2 on OpenSUSE Tumbleweed,
but perhaps just missing some dependencies.
Thomas Monjalon (3):
bus/pci: fix build
ieved in all cases,
so it is moved at the beginning of any processing.
If the check option (-c) is used with an import (-i) or an upgrade (-u),
an error is raised.
Signed-off-by: Thomas Monjalon
---
devtools/linux-uapi.sh | 22 --
1 file changed, 12 insertions(+), 10 dele
Most messages from the Linux uAPI script are not useful
when running the script in a regular fashion.
The quiet mode (option -q) is added to output only error messages if any.
Signed-off-by: Thomas Monjalon
---
devtools/linux-uapi.sh | 26 +++---
1 file changed, 15
Some local variables were not used in the function update_headers()
of the Linux uAPI script.
Signed-off-by: Thomas Monjalon
---
devtools/linux-uapi.sh | 2 --
1 file changed, 2 deletions(-)
diff --git a/devtools/linux-uapi.sh b/devtools/linux-uapi.sh
index 05f3ddf87c..654c4a8c25 100755
--- a
There are two main parts in the Linux uAPI script:
- update the headers
- check the headers
In order to facilitate reading and rework,
these two parts are move in functions.
There is no change in the behavior.
Signed-off-by: Thomas Monjalon
---
devtools/linux-uapi.sh | 96
The option -e was set on the script shebang (first line),
so the script will stop if an error is not caught.
Hence the explicit return or exit commands in case of error
were superfluous and can be removed.
Signed-off-by: Thomas Monjalon
---
devtools/linux-uapi.sh | 16
1 file
The use of curly braces for variables is inconsistent
in the Linux uAPI script.
It is chosen to remove all of them if not required.
Signed-off-by: Thomas Monjalon
---
devtools/linux-uapi.sh | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/devtools
The main goal of this series is to check Linux uAPI headers quietly.
While doing so, few minor improvements are done.
Thomas Monjalon (6):
devtools: remove superfluous curly braces in uAPI script
devtools: remove superfluous fallbacks in uAPI script
devtools: move uAPI processing into
25/08/2025 16:13, Gavin Li:
> Previously, mac_own was a bit-field indicating the MAC addresses owned by
> the PMD, specifically marking those added to VFs by the PMD and allowing
> them to be flushed during exit.
>
> With this commit, the bit-field now applies to all MAC addresses added by
> the P
25/08/2025 16:13, Gavin Li:
> Rules for multicast MAC addresses are intended to filter multicast traffic
> and are managed through multicast MAC add/remove APIs.
You should name the functions to be explicit.
Also it is not only managed via DPDK API,
as you say below we also retrieve kernel address
20/08/2025 11:50, Dariusz Sosnowski:
> There were 2 changes introduced in checkpatch.pl script [1][2]
> which verify if Reported-by tag is followed by Link or Closes tag.
> Example of the warning output:
>
> WARNING:BAD_REPORTED_BY_LINK:
> Reported-by: should be immediately followed by C
triggers a abidiff warning in the RHEL CI when this enum gets updated.
>
> Signed-off-by: David Marchand
Acked-by: Thomas Monjalon
19/08/2025 17:24, Stephen Hemminger:
> On Tue, 19 Aug 2025 10:37:27 +0200
> Thomas Monjalon wrote:
>
> > 19/08/2025 01:27, Stephen Hemminger:
> > > +++ b/lib/eal/include/bsd_queue.h
> > > @@ -0,0 +1,1075 @@
> > > +/*-
> > > + * SPDX-License-Ide
Comments inline
12/08/2025 04:33, Pravin M Bathija:
> - add protocol feature support add and removal of memory
> regions/slots dynamically.
> - this patch set has been tested with multiple instances of fio
What is "fio"?
> connected via libblkio to spdk target backed unix domain socket.
> - plea
12/08/2025 04:33, Pravin M Bathija:
> add new user to mailmap
>
> Signed-off-by: Pravin M Bathija
This patch can be squashed with the next one.
> --- a/.mailmap
> +++ b/.mailmap
> +Pravin M Bathija
If you intend to use the first address for your contributions,
no need to fill the gmail one h
19/08/2025 01:27, Stephen Hemminger:
> +++ b/lib/eal/include/bsd_queue.h
> @@ -0,0 +1,1075 @@
> +/*-
> + * SPDX-License-Identifier: BSD-3-Clause
> + *
> + * Copyright (c) 1991, 1993
> + * The Regents of the University of California. All rights reserved.
> + *
> + * This is a copy of sys/queue.
13/08/2025 16:16, Dariusz Sosnowski:
> Hi,
>
> Thank you for the patch. Please see comments inline.
>
> On Tue, Aug 12, 2025 at 08:46:30AM -0400, Khadem Ullah wrote:
> > This patch validate a connection tracking state when matching
> > 'conntrack is' in rte_flow rules. Since conntrack item flags
15/08/2025 03:55, Minggang(Gavin) Li:
>
> On 8/14/2025 10:05 PM, Thomas Monjalon wrote:
> > 14/08/2025 12:08, Gavin Li:
> >> Rules for multicast MAC addresses are intended to filter multicast traffic
> >> and are managed through multicast MAC add/remove APIs. I
1 - 100 of 1822 matches
Mail list logo
