18/02/2023 02:58, Honnappa Nagarahalli:
> The size generic atomic intrinsics generate the same
> code as the size specific intrinsics for gcc. Use size
> generic intrinsics for both gcc and clang.
>
> Fixes: 7bdccb93078e ("eal: fix ARM build with clang")
> Cc: sta...@dpdk.org
> Cc: pbhagavat...@ma
16/02/2023 09:36, Ruifeng Wang:
> From: Chengwen Feng
> > Subject: [PATCH] net/hns3: support disable IOVA as PA mode
Could we change the title to "support IOVA as VA" ?
> >
> > Claim PMD supports pmd_supports_disable_iova_as_pa.
> >
> > Signed-off-by: Chengwen Feng
> > ---
> > drivers/net/hn
https://bugs.dpdk.org/show_bug.cgi?id=1159
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
Assignee|dev@dpdk.org|david.march...@redhat
Sorry resend, because forget one line.
On 2023/2/20 14:08, Ashok Kaladi wrote:
> If ethdev enqueue or dequeue function is called during
> eth_dev_fp_ops_setup(), it may get pre-empted after setting
> the function pointers, but before setting the pointer to port data.
> In this case the newly regis
On 2023/2/20 14:08, Ashok Kaladi wrote:
> If ethdev enqueue or dequeue function is called during
> eth_dev_fp_ops_setup(), it may get pre-empted after setting
> the function pointers, but before setting the pointer to port data.
> In this case the newly registered enqueue/dequeue function will use
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Tuesday, February 7, 2023 9:20 PM
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Jerin Jacob Kollanukkaran
> ; Anup Prabhu ; Prince Takkar
> ; Parijat Shukla
> Subject: [PATCH v4 06/12] app/m
If ethdev enqueue or dequeue function is called during
eth_dev_fp_ops_setup(), it may get pre-empted after setting
the function pointers, but before setting the pointer to port data.
In this case the newly registered enqueue/dequeue function will use
dummy port data and end up in seg fault.
This p
If eventdev enqueue or dequeue function is called during
event_dev_fp_ops_set(), it may get pre-empted after setting
the function pointers, but before setting the pointer to port data.
In this case the newly registered enqueue/dequeue function will use
dummy port data and end up in seg fault.
This
> -Original Message-
> From: He, ShiyangX
> Sent: 2023年2月8日 14:39
> To: Zhang, Yuying
> Cc: dev@dpdk.org; Zhou, YidingX ;
> sta...@dpdk.org; Singh, Aman Deep ;
> Burakov, Anatoly ; Li, Xiaoyun
> ; Alvin Zhang
> Subject: RE: [PATCH] app/testpmd: fix secondary process not forwarding
>
> -Original Message-
> From: Raslan Darawsheh
> Sent: Thursday, February 16, 2023 10:04 PM
> To: tho...@monjalon.net
> Cc: dev@dpdk.org; Ruifeng Wang ; Bruce Richardson
>
> Subject: [PATCH v2 2/4] config/arm: add NVIDIA BlueField-3 platform
>
> Add meson build configuration for NVIDIA Bl
> -邮件原件-
> 发件人: Stephen Hemminger
> 发送时间: Saturday, February 18, 2023 12:23 AM
> 收件人: Feifei Wang
> 抄送: Ruifeng Wang ; dev@dpdk.org; nd
> ; David Hunt
> 主题: Re: [PATCH v5 2/2] eal: add power mgmt support on Arm
>
> On Wed, 14 Dec 2022 16:14:30 +0800
> Feifei Wang wrote:
>
> > +__che
https://bugs.dpdk.org/show_bug.cgi?id=1161
lingwei (weix.l...@intel.com) changed:
What|Removed |Added
Status|IN_PROGRESS |RESOLVED
Resolution|---
> On 2/17/2023 2:45 AM, Chaoyong He wrote:
> > Register the own RX/TX debug log level type, and get rid of the usage
> > of RTE_LOGTYPE_*. Then we can control the log by a independent switch.
> >
> > Signed-off-by: Chaoyong He
> > Reviewed-by: Niklas Söderlund
> > ---
> > drivers/net/nfp/nfp_log
07/02/2023 11:48, Conor Walsh:
> rte_log should be used instead of printf within the drivers. This patch
> changes a printf within the ioat dma driver's start function to an rte log.
> The printfs within ioat's dump function will be retained
>
> Signed-off-by: Conor Walsh
> Acked-by: Bruce Richar
16/02/2023 08:53, Zhang, Ke1X:
> From: Thomas Monjalon
> > 13/02/2023 08:16, Ke Zhang:
> > > --- a/lib/eal/windows/include/pthread.h
> > > +++ b/lib/eal/windows/include/pthread.h
> > > +static inline void
> > > +pthread_exit(__rte_unused void *__retval) { }
> >
> > Please don't add more shim laye
This series doesn't look reviewed.
What is the status?
17/11/2022 06:09, Zhirun Yan:
> Currently, rte_graph supports RTC (Run-To-Completion) model within each
> of a single core.
> RTC is one of the typical model of packet processing. Others like
> Pipeline or Hybrid are lack of support.
>
> The
08/02/2023 19:08, Tyler Retzlaff:
> On Tue, Feb 07, 2023 at 03:22:00PM +, Bruce Richardson wrote:
> > When using "icx" (Intel(R) oneAPI DPC++/C++ Compiler) to build DPDK,
> > meson reports the toolchain as "intel-llvm"[1]. This value is used
> > directly to define the RTE_TOOLCHAIN macros, whic
08/02/2023 11:48, David Hunt:
> The empty poll mechanism was an experimental API added in
> commit id 450f0791312c. It aimed to allow power saving depending
> on the traffic profile. However, it required a training phase
> and required the user to adjust magic numbers depending on
> their workload.
20/01/2023 11:21, Volodymyr Fialko:
> Add new test cases to cover `rte_reorder_drain_up_to_seqn` and
> `rte_reorder_min_seqn_set`.
Instead of doing a separate patch, please squash the tests
when adding the new API functions in patches 1 & 2.
We are waiting for a detailed review.
The series could be merged without review.
Please fix cosmetic comments below and send a new version to check CI result.
20/01/2023 11:21, Volodymyr Fialko:
> Introduce new reorder drain API:
> `rte_reorder_drain_up_to_seqn` - exhaustively drain all inserted
07/01/2023 16:19, Volodymyr Fialko:
> This patch address issues with reorder drain/free,
> discovered with enabled `RTE_LIBRTE_MEMPOOL_DEBUG`.
>
> Volodymyr Fialko (2):
> reorder: invalidate buf from ready queue in drain
> test/reorder: fix double free of drained buffers
Applied, thanks.
There doesn't seem to be anything keeping this from building on Windows.
Signed-off-by: Stephen Hemminger
---
lib/table/meson.build | 6 --
1 file changed, 6 deletions(-)
diff --git a/lib/table/meson.build b/lib/table/meson.build
index 4f826b4b27e8..f8cef24b5918 100644
--- a/lib/table/meson
Allow building on Windows, need to provide some comparability
wrappers for writev() and if_nametoindex.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
lib/pcapng/meson.build | 6 -
lib/pcapng/rte_pcapng.c | 59 +++--
2 files changed, 57 i
The FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_fib.c | 22 +-
app/test/test_fib6.c | 24 ++--
app/test/test_fib6_perf.c | 15 +--
app/
The RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_rib.c | 19 ---
app/test/test_rib6.c | 24 ++--
lib/rib/meson.build | 6 --
lib/rib/rte_rib.c| 1 +
li
This build works on Windows if sys/queue.h is included.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_ipfrag.c | 12
lib/ip_frag/ip_frag_common.h | 2 ++
lib/ip_frag/meson.build | 6 --
3 files changed, 2 insertions(+), 18 deletions(-)
This builds on Windows if sys/queue.h is included.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_reorder.c | 11 ---
lib/reorder/meson.build | 6 --
lib/reorder/rte_reorder.c | 1 +
3 files changed, 1 insertion(+), 17 deletions(-)
diff --git a/ap
This builds on Windows without changes.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_lpm.c | 15 +--
app/test/test_lpm6.c | 12
app/test/test_lpm6_data.h | 3 ++-
app/test/test_lpm6_perf.c | 14 +-
app/test/test_lpm
The command line arguments are stored in node_action_pair
and the name[] was sized to PATH_MAX which does not exist on Windows.
Since the name is either "CREATE" or "ATTACH" it is not
related to PATH_MAX (4096).
With this fix driver builds ok on windows, but need to modify the
test meson build to
Builds fine with current code, no changes needed.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
drivers/net/null/meson.build | 6 --
1 file changed, 6 deletions(-)
diff --git a/drivers/net/null/meson.build b/drivers/net/null/meson.build
index 4a483955a7a9..a51f8f5211b0 1006
While diagnosing some Windows cross build errors;
noticed that lots of important DPDK libraries are not
being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on
> Bruce Richardson (6):
> dma/ioat: fix device stop if no copies done
> dma/ioat: fix incorrectly set indexes after restart
> dma/ioat: fix incorrect error reporting on restart
> test/dmadev: check result for device stop
> test/dmadev: create separate function for single copy test
> tes
On Sun, 19 Feb 2023 23:43:08 +0300
Levend Sayar wrote:
> LS: “unlikely” can be added. You’re right. Code already tries to make a bulk
> allocation first.
> If that bulk allocation does not work, it tries to allocate one my one.
That seems like a unnecessary step and unlikely to help.
Unless the
From: Tanzeel Ahmed
This patch is new version of [PATCH] lib/net: added push MPLS header API.
I have also added the MPLS strip functionality in the same patch to
address the question asked in last patch.
> You should explain why you add this function.
None of the foundational NICs currently supp
19/02/2023 23:40, Thomas Monjalon:
> 07/02/2023 07:10, Megha Ajmera:
> > Big structures like bitmap, pipes and queues in subport are addressed
> > using offset of 'memory' field in subport structures. This means no other
> > variable should be added after 'memory' variable or else addressing of suc
07/02/2023 07:10, Megha Ajmera:
> Big structures like bitmap, pipes and queues in subport are addressed
> using offset of 'memory' field in subport structures. This means no other
> variable should be added after 'memory' variable or else addressing of such
> structs like bitmap etc. become incorre
16/02/2023 10:22, Bruce Richardson:
> On Thu, Feb 16, 2023 at 09:28:17AM +0800, fengchengwen wrote:
> > add cc maintainer
> >
> > On 2023/2/13 20:44, Chengwen Feng wrote:
> > > The spinlock should unlock when clock_gettime() failed. This patch
> > > fixes it by invoking clock_gettime() before lock
10/02/2023 20:51, Cristian Dumitrescu:
> While the small fields (size <= 64 bits) can be in either network byte
> order (header fields) or host byte order (meta-data fields), the big
> fields (size > 64 bits) are always expected to be in network byte
> order (big endian), both header and metadata f
07/02/2023 17:41, Cristian Dumitrescu:
> Add pipeline support for the Receive Side Scaling (RSS) hashing. While
> the pipeline already supports the stateless hashing schemes, the RSS
> scheme uses a key configured by the control plane and preserved
> between successive RSS hash invocations.
>
> Si
07/02/2023 19:15, Cristian Dumitrescu:
> In case of rte_ipsec_session_prepare() error, the crypto session was
> not freed, leading to its leakage.
>
> Fixes: 20777eb5f913 ("pipeline: add IPsec")
> Coverity issue: 383139
>
> Signed-off-by: Cristian Dumitrescu
Applied, thanks.
> On 19 Feb 2023, at 20:35, Stephen Hemminger
> wrote:
>
> On Sun, 19 Feb 2023 03:30:59 +0300
> Levend Sayar mailto:levendsa...@gmail.com>> wrote:
>
>> rx no_mbufs stats counter update is added for another error case.
>>
>> Fixes: 4f6b1dd8240c ("net/gve: support basic statistics")
>> Cc: jun
Hi Ferruh,
Opps, I was not aware of that rejection.
Thanks for notification.
Let me supersede this patch.
And create a new one.
Best,
Levend
> On 19 Feb 2023, at 23:09, Ferruh Yigit wrote:
>
> On 2/19/2023 12:26 AM, Levend Sayar wrote:
>> Ferruh,
>>
>> Thanks for this detailed review.
>>
On 2/19/2023 12:31 AM, Levend Sayar wrote:
> Google Virtual NIC PMD is enriched with rx/tx queue statistics info.
>
> Signed-off-by: Levend Sayar
> ---
> drivers/net/gve/gve_ethdev.c | 8
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/net/gve/gve_ethdev.c b/drivers/net/gve/
On 2/19/2023 12:26 AM, Levend Sayar wrote:
> Ferruh,
>
> Thanks for this detailed review.
> I am setting superseded this patch and create a new one.
> You’re right at all points.
> For rx.no_mbufs counting, I probably overlooked while rebasing my patch
> and it is mixed with newly came patch.
>
>
On Fri, 3 Feb 2023 11:14:06 -0800
Stephen Hemminger wrote:
> This patchset keeps uncovering bad practices in the cmdline library
> around end of file and signal handling.
>
> Stephen Hemminger (3):
> cmdline: make rdline status not private
> cmdline: handle EOF in cmdline_poll
> testpmd:
The FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_fib.c | 22 +-
app/test/test_fib6.c | 24 ++--
app/test/test_fib6_perf.c | 15 +--
app/
Allow building on Windows, need to provide some comparability
wrappers for writev() and if_nametoindex.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
lib/pcapng/meson.build | 6 -
lib/pcapng/rte_pcapng.c | 59 +++--
2 files changed, 57 i
The RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_rib.c | 19 ---
app/test/test_rib6.c | 24 ++--
lib/rib/meson.build | 6 --
lib/rib/rte_rib.c| 1 +
li
This builds on Windows if sys/queue.h is included.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_reorder.c | 11 ---
lib/reorder/meson.build | 6 --
lib/reorder/rte_reorder.c | 1 +
3 files changed, 1 insertion(+), 17 deletions(-)
diff --git a/ap
This build works on Windows if sys/queue.h is included.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_ipfrag.c | 12
lib/ip_frag/ip_frag_common.h | 2 ++
lib/ip_frag/meson.build | 6 --
3 files changed, 2 insertions(+), 18 deletions(-)
This builds on Windows without changes.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
app/test/test_lpm.c | 15 +--
app/test/test_lpm6.c | 12
app/test/test_lpm6_data.h | 3 ++-
app/test/test_lpm6_perf.c | 14 +-
app/test/test_lpm
The command line arguments are stored in node_action_pair
and the name[] was sized to PATH_MAX which does not exist on Windows.
Since the name is either "CREATE" or "ATTACH" it is not
related to PATH_MAX (4096).
With this fix driver builds ok on windows, but need to modify the
test meson build to
Builds fine with current code, no changes needed.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
drivers/net/null/meson.build | 6 --
1 file changed, 6 deletions(-)
diff --git a/drivers/net/null/meson.build b/drivers/net/null/meson.build
index 4a483955a7a9..a51f8f5211b0 1006
While diagnosing some Windows cross build errors;
noticed that lots of important DPDK libraries are not
being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on
On Sun, 19 Feb 2023 03:30:59 +0300
Levend Sayar wrote:
> rx no_mbufs stats counter update is added for another error case.
>
> Fixes: 4f6b1dd8240c ("net/gve: support basic statistics")
> Cc: junfeng@intel.com
>
> Signed-off-by: Levend Sayar
> ---
> drivers/net/gve/gve_rx.c | 7 ++-
>
02/02/2023 14:47, Gregory Etelson:
> Quota action limits traffic according to pre-defined configuration.
> Quota reflects overall traffic usage regardless bandwidth.
> Quota flow action initialized with signed tokens number value.
> Quota flow action updates tokens number according to
> these rules
19/02/2023 09:30, Zhang, Qi Z:
> From: Mingjin Ye
> >
> > The scalar Tx path would send wrong mbuf that causes the kernel driver to
> > fire the MDD event.
> >
> > This patch adds mbuf detection in tx_prepare and enables it by defining
> > `RTE_ETHDEV_DEBUG_TX` macro to fix this issue. If valida
>-Original Message-
>From: Konstantin Ananyev
>Sent: Friday, February 17, 2023 11:15 AM
>To: Tomasz Duszynski ; Konstantin Ananyev
>;
>dev@dpdk.org
>Subject: RE: [EXT] Re: [PATCH v11 1/4] lib: add generic support for reading
>PMU events
>
>
>
>> >>
>> >> This is especially useful in ca
19/02/2023 12:08, Thomas Monjalon:
> 19/02/2023 11:04, Thomas Monjalon:
> > 29/01/2023 13:35, Zhang, Qi Z:
> > > From: Zhang, Qi Z
> > > >
> > > > PMD not announce pmd_supports_disable_iova_as_pa will not be build
> > > > when RTE_IOVA_AS_PA is not defined, but some AVX library for vector path
>
If IOVA as PA is disabled and the driver does not support IOVA as VA,
the build of the driver was disabled.
Unfortunately some drivers were building some sub-libraries
(with specific options for vector paths) which were not disabled.
The build parsing of those drivers need to be skipped earlier
to
19/02/2023 11:04, Thomas Monjalon:
> 29/01/2023 13:35, Zhang, Qi Z:
> > From: Zhang, Qi Z
> > >
> > > PMD not announce pmd_supports_disable_iova_as_pa will not be build
> > > when RTE_IOVA_AS_PA is not defined, but some AVX library for vector path
> > > is not skipped by the build system which ca
19/02/2023 11:01, Thomas Monjalon:
> 09/12/2022 10:33, fengchengwen:
> > The hns3 and idpf both use static_library to build advanced vector (e.g.
> > SVE for hns3 and AVX512 for idpf).
> > The static-library is still compiled even the PMD is not compiled, and this
> > lead to the compile fail p
17/02/2023 23:34, Thomas Monjalon:
> 13/12/2022 02:52, Zhang, Qi Z:
> > From: Xing, Beilei
> > > From: Wu, Jingjing
> > > > When enable_iova_as_pa option is disabled, idpf driver should avoid
> > > > the building in its build file.
> > > >
> > > > Fixes: 5bf87b45b2c8 (net/idpf: add AVX512 data pa
29/01/2023 13:35, Zhang, Qi Z:
> From: Zhang, Qi Z
> >
> > PMD not announce pmd_supports_disable_iova_as_pa will not be build
> > when RTE_IOVA_AS_PA is not defined, but some AVX library for vector path
> > is not skipped by the build system which cause compile error.
> >
> > The patch modify i4
09/12/2022 10:33, fengchengwen:
> The hns3 and idpf both use static_library to build advanced vector (e.g.
> SVE for hns3 and AVX512 for idpf).
> The static-library is still compiled even the PMD is not compiled, and this
> lead to the compile fail problem.
>
> Although the following could s
> -Original Message-
> From: Yan, Zhirun
> Sent: Tuesday, February 14, 2023 1:24 PM
> To: dev@dpdk.org; Zhang, Qi Z
> Cc: Yan, Zhirun
> Subject: [PATCH 2/2] net/ice: support mark action in switch
>
> This patch add mark action in switch filter.
>
> Signed-off-by: Zhirun Yan
Acked-
> -Original Message-
> From: Yan, Zhirun
> Sent: Tuesday, February 14, 2023 1:24 PM
> To: dev@dpdk.org; Zhang, Qi Z
> Cc: Yan, Zhirun
> Subject: [PATCH 1/2] net/ice/base: add ability to set markid via switch filter
>
> Support to add large action to set 32 bits markid via switch filt
> -Original Message-
> From: Mingjin Ye
> Sent: Monday, February 13, 2023 4:32 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; sta...@dpdk.org; Zhou, YidingX
> ; Ye, MingjinX ; Wu,
> Jingjing ; Xing, Beilei
> Subject: [PATCH v3] net/iavf: add debug checks for mbuf
>
> The scalar Tx path wo
> -Original Message-
> From: Xu, Ke1
> Sent: Friday, February 17, 2023 9:51 AM
> To: Zeng, ZhichaoX ; dev@dpdk.org
> Cc: Yang, Qiming ; Zhou, YidingX
> ; Zhang, Qi Z ; Wu, Jingjing
> ; Xing, Beilei ; Richardson,
> Bruce ; Konstantin Ananyev
>
> Subject: RE: [PATCH v3] net/iavf: enable
69 matches
Mail list logo
