> -Original Message-
> From: Xu, HailinX
> Sent: Thursday, June 29, 2023 2:50 PM
> To: Thomas Monjalon ; annou...@dpdk.org;
> dev@dpdk.org
> Cc: Kovacevic, Marko ; Mcnamara, John
> ; Richardson, Bruce
> ; Ferruh Yigit
> Subject: RE: release candidate 23.07-rc2
>
> > -Original Message-
> -Original Message-
> From: Zeng, ZhichaoX
> Sent: Thursday, July 6, 2023 2:08 PM
> To: Xu, Ke1 ; dev@dpdk.org
> Cc: Lu, Wenzhuo ; sta...@dpdk.org
> Subject: RE: [PATCH v2] net/iavf: fix Tunnel TSO path selecting.
>
> > -Original Message-
> > From: Xu, Ke1
> > Sent: Thursday,
> -Original Message-
> From: Xu, Ke1
> Sent: Thursday, July 6, 2023 10:45 AM
> To: dev@dpdk.org
> Cc: Xu, Ke1 ; Zeng, ZhichaoX
> ; Lu, Wenzhuo ;
> sta...@dpdk.org
> Subject: [PATCH v2] net/iavf: fix Tunnel TSO path selecting.
>
> IAVF currently supports TSO and Tunnel TSO. Both these two
> -Original Message-
> From: Xu, Ke1
> Sent: Thursday, July 6, 2023 11:52 AM
> To: Zeng, ZhichaoX ; dev@dpdk.org
> Cc: Zhang, Qi Z ; Wu, Jingjing ;
> Xing, Beilei ; Lu, Wenzhuo
> Subject: RE: [PATCH] net/iavf: fix AVX2 path selection
>
>
> > -Original Message-
> > From: Zeng,
> -Original Message-
> From: Luca Boccassi
> Sent: Thursday, July 6, 2023 3:44 AM
> To: Zhou, YidingX ; dev@dpdk.org
> Subject: Re: [PATCH] ice: fix build error on 32bit configure
>
> On Wed, 5 Jul 2023 at 08:21, Yiding Zhou wrote:
> >
> > Replace 'rte_memcpy' with 'memcpy' like other
On Wed, 2023-07-05 at 19:57 -0700, Stephen Hemminger wrote:
> On Tue, 29 Nov 2022 14:04:45 -0800
> Tyler Retzlaff wrote:
>
> > > Markus and I ultimately use the function in the patch to call
> > > rte_thread_setname() (which takes the pthread id as an argument) to
> > > rename our
> > > lcore wor
Hi Chengwen,
Thanks for reviewing.
The memory leak occurs similar as what mentioned in
https://patches.dpdk.org/project/dpdk/patch/20230705092511.362484-2-wenjun1...@intel.com/.
I will perfect the commit description in V2 version.
Thanks,
Wenjun
> -Original Message-
> From: fengchengwen
On Wed, Jul 5, 2023 at 3:26 PM Hemant Agrawal wrote:
>
> This patch update the dpdk sample app policy to support
> vendor PMD specific test examples in the respective drivers
> sub-directory in examples.
>
> Signed-off-by: Hemant Agrawal
> Acked-by: Jerin Jacob
> ---
> doc/guides/sample_app_ug/
> -Original Message-
> From: Zeng, ZhichaoX
> Sent: Thursday, July 6, 2023 11:20 AM
> To: dev@dpdk.org
> Cc: Zhang, Qi Z ; Xu, Ke1 ;
> Zeng, ZhichaoX ; Wu, Jingjing
> ; Xing, Beilei ; Lu,
> Wenzhuo
> Subject: [PATCH] net/iavf: fix AVX2 path selection
>
> The AVX2 path does not support
The AVX2 path does not support outer checksum offload, when AVX2 is
forcibly selected and outer checksum offload is configured, the basic
Tx path will be selected to ensure proper functionality.
Fixes: 5712bf9d6e14 ("net/iavf: add Tx AVX2 offload path")
Signed-off-by: Zhichao Zeng
---
drivers/ne
The AVX2 path does not support outer checksum offload, when AVX2 is
forcibly selected and outer checksum offload is configured, the basic
Tx path will be selected to ensure proper functionality.
Fixes: 5712bf9d6e14 ("net/iavf: add Tx AVX2 offload path")
Signed-off-by: Zhichao Zeng
---
drivers/ne
On Tue, 29 Nov 2022 14:04:45 -0800
Tyler Retzlaff wrote:
> > Markus and I ultimately use the function in the patch to call
> > rte_thread_setname() (which takes the pthread id as an argument) to rename
> > our
> > lcore workers from "lcore-worker-X" to something more meaningful in the
> > scope
On Thu, 31 Mar 2022 21:54:35 -0700
Mahdi Rakhshandehroo wrote:
> Currently, building DPDK with clang + LTO is unsupported, which is a
> showstopper for cross-language interprocedural optimizations. The root
> cause is that pmdinfogen expects to scan through the section headers of
> generated ELF
在 2023/7/5 19:41, Ferruh Yigit 写道:
On 7/5/2023 10:40 AM, lihuisong (C) wrote:
在 2023/7/5 16:02, Ferruh Yigit 写道:
On 7/5/2023 4:16 AM, lihuisong (C) wrote:
在 2023/7/4 18:59, Ferruh Yigit 写道:
On 7/4/2023 9:45 AM, Jie Hai wrote:
Function update_queue_state updates queue state of all queues
of
IAVF currently supports TSO and Tunnel TSO. Both these two features
are implemented in scalar path. As there are missed flags for Tunnel
TSO, it selects vector paths wrongly when only Tunnel TSO is enabled
after introducing the patch "net/iavf: add offload path for Tx AVX512".
This patch added
On Wed, 5 Jul 2023 19:32:48 -0700
Stephen Hemminger wrote:
> The mailmap file is supposed to be in sorted order,
> but several entries are in the wrong place.
>
> Signed-off-by: Stephen Hemminger
> ---
> v2 - use en_us.UTF-8 sort order
>
> .mailmap | 6 +++---
> 1 file changed, 3 insertions(
On Sun, 21 Aug 2022 15:12:13 -0700
Shinae Woo wrote:
> - explicitly set default value for intr_mode
> - give read permission on wc_activate
> - applying patch gives below visibility
>
> ```
> $ more /sys/module/igb_uio/parameters/* | cat
> ::
> /sys/module/igb_uio/parameters/intr_mod
IAVF currently supports TSO and Tunnel TSO. Both these two features
are implemented in scalar path. As there are missed flags for Tunnel
TSO, it selects vector paths wrongly when only Tunnel TSO is enabled
after introducing the patch "net/iavf: add offload path for Tx AVX512".
This patch added
The mailmap file is supposed to be in sorted order,
but several entries are in the wrong place.
Signed-off-by: Stephen Hemminger
---
v2 - use en_us.UTF-8 sort order
.mailmap | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/.mailmap b/.mailmap
index d200f363394d..bfd3528
Hi, kaisen
Your patch missed prefix app/test.
> -Original Message-
> From: You, KaisenX
> Sent: Wednesday, July 5, 2023 5:35 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Zhou, YidingX
> ; You, KaisenX ;
> sta...@dpdk.org
> Subject: [PATCH] child process synchronization NIC startup parameter
On Thu, 6 Jul 2023 01:58:03 +0100
Ferruh Yigit wrote:
> On 7/6/2023 12:45 AM, Stephen Hemminger wrote:
> > The mailmap file is supposed to be in sorted order,
> > but several entries are in the wrong place.
> >
> > Signed-off-by: Stephen Hemminger
> >
>
> Hi Stephen,
>
> What are you using
> -Original Message-
> From: Lu, Wenzhuo
> Sent: Thursday, July 6, 2023 9:27 AM
> To: Zhang, Qi Z ; Jiale, SongX ;
> dev@dpdk.org
> Subject: RE: [PATCH] net/iavf: fix SCTP tunnel packet forwarding issue
>
> Hi Qi,
>
> > -Original Message-
> > From: Zhang, Qi Z
> > Sent: Thurs
Hi Qi,
> -Original Message-
> From: Zhang, Qi Z
> Sent: Thursday, June 29, 2023 4:58 PM
> To: Jiale, SongX ; Lu, Wenzhuo
> ; dev@dpdk.org
> Cc: Lu, Wenzhuo
> Subject: RE: [PATCH] net/iavf: fix SCTP tunnel packet forwarding issue
>
>
>
> > -Original Message-
> > From: Jiale, So
On 2023/7/5 17:25, Wenjun Wu wrote:
> The device should be detached before quit, otherwise it will
> cause memory leak.
Which memory will leak?
For mp, if secondary process quit, it only needs to properly handle the memory
shared with other process.
>
> Fixes: 05f1d6842fc3 ("examples/multi_pro
On 7/6/2023 12:45 AM, Stephen Hemminger wrote:
> The mailmap file is supposed to be in sorted order,
> but several entries are in the wrong place.
>
> Signed-off-by: Stephen Hemminger
>
Hi Stephen,
What are you using for sorting?
When I use 'sort' binary [1] I get only a few lines of diff,
but
On 2023/7/5 17:25, Wenjun Wu wrote:
> In hotplug usecase, devargs will be allocated in secondary process
> in the function alloc_devargs. Since it will not be insert into the
> devarg_list, it will have no chance to be freed.
>
> This patch adds additional memory free for device structure member d
The mailmap file is supposed to be in sorted order,
but several entries are in the wrong place.
Signed-off-by: Stephen Hemminger
---
.mailmap | 334 +++
1 file changed, 167 insertions(+), 167 deletions(-)
diff --git a/.mailmap b/.mailmap
index
On Mon, 24 Aug 2020 13:54:12 +0530
rohit@nxp.com wrote:
> From: Rohit Raj
>
> As per the current code we have API for bus probe, but the
> bus close API is missing. This breaks the multi process
> scenarios as objects are not cleaned while terminating the
> secondary processes.
>
> This pat
On Thu, 8 Oct 2020 21:00:44 +0530
rohit@nxp.com wrote:
> +static void
> +signal_handler(int signal)
> +{
> + if (signal == SIGINT)
> + rte_eal_cleanup();
NAK
Call rte_eal_cleanup in signal handler is not safe.
Need to set a flag and handle it in main code.
Good evening DPDK Community,
September 12-13, 2023 we will have the DPDK Summit at the Gibson Hotel in
Point Square in Dublin. Registration and attendance for this event are
*free!*
Our Call for Speakers is just wrapping up - the deadline is this Friday,
Jul. 7th at 11:59 PDT. If y
Please put you
On Wed, 6 Apr 2022 02:14:46 +0300
Dmitry Kozlyuk wrote:
> >
> > Don't merge this patch as is please; it would cause a lot of pain
> > in a cloud environment.
> >
> > In our environment core dumps are collected (via systemd) and uploaded
> > to a central server. With this kind of change the proc
On Fri, 31 Mar 2023 12:37:40 +0200
Maxime Coquelin wrote:
> Indeed!
>
> On 3/31/23 12:34, Thomas Monjalon wrote:
> > We missed this patch, there was no comment.
> > Please review.
> >
> > 10/03/2022 18:59, luca.bocca...@gmail.com:
> >> From: Luca Boccassi
> >>
> >> Allow more flexibility wit
Mail to marayam.tah...@intel.com is bouncing.
Presume no longer at Intel.
Signed-off-by: Stephen Hemminger
---
MAINTAINERS | 1 -
1 file changed, 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index b5adba69d82d..d688b047b402 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1797,7 +1797,6 @
When debugging driver or startup issues, it is useful to have
a timestamp on each message printed. The messages in syslog
already have a timestamp, but often syslog is not available
during testing. The timestamp format is chosen to look
like the default Linux dmesg timestamp.
The first few lines a
It is useful for application to be able to set the default log
stream before call rte_eal_init(). This makes all messages go
to the new default.
For example, to skip using syslog; just doing
rte_openlog_stream(stderr);
There is no reason for helper command line applications to clutter
sys
FreeBSD logging code was not using syslog and did not have
the same options as Linux. Use a common set of functions for that.
Pre-parsing for log level is common to all OS's.
Now the pre-scan can return an error.
A bad argument give to --log-level option was given the
code would keep going.
Use
The --syslog flag takes facility argument.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_options.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/eal/common/eal_common_options.c
b/lib/eal/common/eal_common_options.c
index 03059336987d..005da4d12001 1006
This aligns getopt, getopt_long, etc to have the same const
attributes as Linux and FreeBSD. The changes are derived from
the FreeBSD version of getopt_long.
Signed-off-by: Stephen Hemminger
---
lib/eal/windows/getopt.c | 23 ---
lib/eal/windows/include/getopt.h | 8
Patchset that includes:
- unified code for more of log argument handling
- fix for duplicate option errors
- timestamp option for logging
v7
- consolidate patches
- fix windows to have same getopt args as Linux and FreeBSD
Stephen Hemminger (5):
windows: make getopt functions have con
Tested-by: Jeremy Spewock
On Wed, Jun 21, 2023 at 2:33 PM Jeremy Spewock wrote:
> Acked-by: Jeremy Spewock
>
> On Fri, Jun 9, 2023 at 5:46 AM Juraj Linkeš
> wrote:
>
>> Pexpect is not a dedicated SSH connection library while Fabric is. With
>> Fabric, all SSH-related logic is provided and we
On Wed, 5 Jul 2023 at 08:21, Yiding Zhou wrote:
>
> Replace 'rte_memcpy' with 'memcpy' like other PMD code to avoid errors when
> compiling with GCC-12 on 32-bit configure.
>
> Compiler reports the follow error:
>
> error: array subscript 8 is outside array bounds of "struct rte_mbuf *[32]"
> [-We
> > Subject: [PATCH] doc: support IPsec Multi-buffer lib v1.4
> >
> > Updated AESNI MB and AESNI GCM, KASUMI, ZUC, SNOW3G and
> > CHACHA20_POLY1305 PMD documentation guides with information about
> > the latest Intel IPsec Multi-buffer library supported.
> >
> > Signed-off-by: Ciara Power
> Acked-
> 05/07/2023 11:36, Ciara Power пишет:
> > The socket ID field for a cryptodev device data was unsigned int.
> > Due to recent changes to the default device socket ID,
> > this caused an issue when the socket ID was unknown and set to -1.
> > The device socket ID wrapped around to 255,
> > and cau
> > Subject: [PATCH] crypto/qat: fix legacy sm4 ecb capability
> >
> > Following the deprecation of insecure algorithms in QAT, SM4-ECB should be
> > included as legacy, to be disabled by default.
> >
> > Fixes: cffb726b7797 ("crypto/qat: enable insecure algorithms")
> >
> > Signed-off-by: Ciara Po
> Due to recent changes to the default device socket ID,
> before being used as an index for session mempool list,
> the socket ID should be set to 0 if unknown (-1).
>
> Fixes: 7dcd73e37965 ("drivers/bus: set device NUMA node to unknown by
> default")
> Cc: olivier.m...@6wind.com
> Cc: sta...@dpd
Integrated zero-copy put API in mempool cache in i40e PMD.
On Ampere Altra server, l3fwd single core's performance improves by 5%
with the new API
Signed-off-by: Kamalakshitha Aligeri
Reviewed-by: Ruifeng Wang
Reviewed-by: Feifei Wang
---
drivers/net/i40e/i40e_rxtx_vec_common.h | 27 ++
From: Morten Brørup
Zero-copy access to mempool caches is beneficial for PMD performance.
Furthermore, having a zero-copy mempool API is considered a precondition
for fixing a certain category of bugs, present in some PMDs: For
performance reasons, some PMDs had bypassed the mempool API in order
Due to recent changes to the default device socket ID, before
being used as an index for session mempool list,
set socket ID to 0 if unknown (-1).
Fixes: 7dcd73e37965 ("drivers/bus: set device NUMA node to unknown by default")
Cc: olivier.m...@6wind.com
Cc: sta...@dpdk.org
Signed-off-by: Kai Ji
Integrated zero-copy put API in mempool cache in i40e PMD.
On Ampere Altra server, l3fwd single core's performance improves by 5%
with the new API
Signed-off-by: Kamalakshitha Aligeri
Reviewed-by: Ruifeng Wang
Reviewed-by: Feifei Wang
---
.mailmap| 1 +
drivers
From: Morten Brørup
Zero-copy access to mempool caches is beneficial for PMD performance.
Furthermore, having a zero-copy mempool API is considered a precondition
for fixing a certain category of bugs, present in some PMDs: For
performance reasons, some PMDs had bypassed the mempool API in order
On Wed, Jul 5, 2023 at 7:02 PM Maxime Coquelin
wrote:
>
> On 7/5/23 15:36, David Marchand wrote:
> > On Wed, Jul 5, 2023 at 3:22 PM Maxime Coquelin
> > wrote:
> >> @@ -950,9 +954,14 @@ rte_vhost_driver_register(const char *path, uint64_t
> >> flags)
> >> * two values.
> >> */
Hi David,
On 7/5/23 15:36, David Marchand wrote:
On Wed, Jul 5, 2023 at 3:22 PM Maxime Coquelin
wrote:
@@ -950,9 +954,14 @@ rte_vhost_driver_register(const char *path, uint64_t flags)
* two values.
*/
vsocket->use_builtin_virtio_net = true;
- vsocket->support
On Mon, 5 Apr 2021 21:39:51 +0200
Thomas Monjalon wrote:
> After argument parsing done by rte_eal_init(),
> the remaining arguments are to be parsed by the application
> by progressing in the argv array.
> In this context, the first string represented by argv[0] is still
> the same program name
On 7/5/2023 3:32 PM, Ferruh Yigit wrote:
> For some drivers [1], testpmd forwarding is broken with commit [2].
>
> This is because with [2] testpmd gets queue state from ethdev and
> forwarding is done only on queues in started state, but some drivers
> don't update queue status properly, and this
Hi Rakesh,
On Tue, May 23, 2023 at 03:39:53PM +, Rakesh Kudurumalla wrote:
> Ping
>
> Regards,
> Rakesh
>
> > -Original Message-
> > From: Rakesh Kudurumalla
> > Sent: Wednesday, April 26, 2023 2:58 PM
> > To: Olivier Matz
> > Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran ; Nithin
>
Hi Ciara,
> -Original Message-
> From: Power, Ciara
> Sent: Wednesday 5 July 2023 14:23
> To: dev@dpdk.org
> Cc: Ji, Kai ; O'Donovan, Saoirse
> ; Power, Ciara
> Subject: [PATCH] crypto/qat: fix legacy sm4 ecb capability
>
> Following the deprecation of insecure algorithms in QAT, SM4-EC
The mlx5 provides the send scheduling on specific moment of time,
and for the related kind of applications it would be extremely useful
to have extra debug information - when and how packets were scheduled
and when the actual sending was completed by the NIC hardware (it helps
application to track
The Python script is intended to analyze mlx5 PMD
datapath traces and report:
- tx_burst routine timings
- how packets are pushed to WQEs
- how packet sending is completed with timings
Signed-off-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/tools/mlx5_trace.py | 307 ++
The mlx5 provides the send scheduling on specific moment of time,
and for the related kind of applications it would be extremely useful
to have extra debug information - when and how packets were scheduled
and when the actual sending was completed by the NIC hardware (it helps
application to track
There is the demand to trace the send completions of
every WQE if time scheduling is enabled.
The patch extends the size of completion queue and
requests completion on every issued WQE in the
send queue. As the result hardware provides CQE on
each completed WQE and driver is able to fetch
completi
There is an intention to engage DPDK tracing capabilities
for mlx5 PMDs monitoring and profiling in various modes.
The patch introduces tracepoints for the Tx datapath in
the ethernet device driver.
To engage this tracing capability the following steps
should be taken:
- meson option -Denable_tra
Hey Ciara,
> -Original Message-
> From: Ciara Power
> Sent: Wednesday, July 5, 2023 3:34 PM
> To: dev@dpdk.org
> Cc: Ji, Kai ; De Lara Guarch, Pablo
> ; Power, Ciara
> Subject: [PATCH] doc: support IPsec Multi-buffer lib v1.4
>
> Updated AESNI MB and AESNI GCM, KASUMI, ZUC, SNOW3G and
>
On Wed, 5 Jul 2023 12:37:41 +0100
Ferruh Yigit wrote:
> On 7/4/2023 10:02 PM, Stephen Hemminger wrote:
> > Support rte_eth_dev_set_mtu for pcap driver when the
> > pcap device is convigured to point to a network interface.
> >
> > This is rebased an consolidated from earlier version.
> > Added s
Updated AESNI MB and AESNI GCM, KASUMI, ZUC, SNOW3G
and CHACHA20_POLY1305 PMD documentation guides
with information about the latest Intel IPsec Multi-buffer
library supported.
Signed-off-by: Ciara Power
---
doc/guides/cryptodevs/aesni_gcm.rst | 6 +++---
doc/guides/cryptodevs/aesni_mb.r
For some drivers [1], testpmd forwarding is broken with commit [2].
This is because with [2] testpmd gets queue state from ethdev and
forwarding is done only on queues in started state, but some drivers
don't update queue status properly, and this breaks forwarding for those
drivers.
Drivers shou
On Tue, Jul 4, 2023 at 10:45 AM Nithin Dabilpuram
wrote:
>
> Announce addition of new security IPsec SA option to enable
> out of place processing in Ingress Inline inbound SA's.
>
> Signed-off-by: Nithin Dabilpuram
> Acked-by: Akhil Goyal
Acked-by: Jerin Jacob
> ---
>
> v2:
> - Modified depr
On Wed, Jul 5, 2023 at 5:01 PM Konstantin Ananyev
wrote:
>
> 04/07/2023 20:44, Akhil Goyal пишет:
> > Structure rte_security_ops and rte_security_ctx are meant to
> > be used by rte_security library and the PMDs associated.
> > These will be moved to an internal header in DPDK 23.11 release.
> >
>
Acked-by: Kai Ji mailto:kai...@intel.com>>
From: Power, Ciara
Sent: 05 July 2023 14:23
To: dev@dpdk.org
Cc: Ji, Kai ; O'Donovan, Saoirse
; Power, Ciara
Subject: [PATCH] crypto/qat: fix legacy sm4 ecb capability
Following the deprecation of insecure algorithms i
> > Subject: [PATCH v2] crypto/ipsec_mb: fix jobs array used for burst
> >
> > The jobs variable was global, which meant it was not thread safe.
> > This caused a segmentation fault when running the crypto performance app,
> > using more than one lcore for crypto processing.
> >
> > Moving this to
Hi Konstantin,
Can you review this patch?
> UDP header length is included in sa->hdr_len. Take care of that in
> L3 header and pakcet length calculation.
>
> Fixes: 01eef5907fc3 ("ipsec: support NAT-T")
>
> Signed-off-by: Xiao Liang
> ---
> lib/ipsec/esp_outb.c | 2 +-
> lib/ipsec/sa.c |
VFIO module provides configurable dma_entry_limit
parameter to store the DMA entries. By default this
is 64K and if we are using --no-huge, we shall need
to increase the value of dma_entry_limit. Add
commands in linux_gsg document to change the
dma_entry_limit.
Signed-off-by: Nipun Gupta
---
Cha
On Wed, Jul 5, 2023 at 3:22 PM Maxime Coquelin
wrote:
> @@ -950,9 +954,14 @@ rte_vhost_driver_register(const char *path, uint64_t
> flags)
> * two values.
> */
> vsocket->use_builtin_virtio_net = true;
> - vsocket->supported_features = VIRTIO_NET_SUPPORTED_FEATURES
> > Subject: [PATCH] drivers/ipsec_mb: fix aesni_mb set session ID
> >
> > In the case of multiprocess, when the same session is being used for both
> > primary and secondary processes, the session ID will be the same.
> > However the pointers are not available to the secondary process, so in this
Following the deprecation of insecure algorithms in QAT,
SM4-ECB should be included as legacy, to be disabled by default.
Fixes: cffb726b7797 ("crypto/qat: enable insecure algorithms")
Signed-off-by: Ciara Power
---
drivers/crypto/qat/dev/qat_crypto_pmd_gen3.c | 6 +++---
1 file changed, 3 inse
This features was mistakenly removed, add it back.
Fixes: 0adb8eccc6a6 ("vhost: add VDUSE device creation and destruction")
Signed-off-by: Maxime Coquelin
---
lib/vhost/vduse.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/vhost/vduse.h b/lib/vhost/vduse.h
index cd55bfd858..46753fec73
This small series brings a couple of VDUSE fixes
for v23.07, discovered during testing with OVS-DPDK.
Maxime Coquelin (2):
vhost: fix vduse features negotiation
vduse: fix missing event index features
lib/vhost/socket.c | 19 +--
lib/vhost/vduse.c | 28 +++---
The series introducing VDUSE support missed the
application capability to disable supported features.
This results in TSO being negotiated while not supported by
the application.
Fixes: 0adb8eccc6a6 ("vhost: add VDUSE device creation and destruction")
Signed-off-by: Maxime Coquelin
---
lib/vho
Acked-by: Kai Ji mailto:kai...@intel.com>>
From: Ciara Power
Sent: 05 July 2023 11:30
To: dev@dpdk.org
Cc: gak...@marvell.com ; Power, Ciara
; Matz, Olivier ;
sta...@dpdk.org
Subject: [PATCH] app/crypto-perf: fix socket ID default value
Due to recent changes t
> On Wed, Jul 5, 2023 at 4:48 PM Mattias Rönnblom
> wrote:
> >
> > Announce the removal of the single-event enqueue and dequeue
> > operations from the eventdev ABI.
> >
> > Signed-off-by: Mattias Rönnblom
>
> Acked-by: Jerin Jacob
Acked-by: Pavan Nikhilesh
>
>
> >
> > ---
> > PATCH v2: Fix
On Wed, Jul 5, 2023 at 4:48 PM Mattias Rönnblom
wrote:
>
> Announce the removal of the single-event enqueue and dequeue
> operations from the eventdev ABI.
>
> Signed-off-by: Mattias Rönnblom
Acked-by: Jerin Jacob
>
> ---
> PATCH v2: Fix commit subject prefix.
> ---
> doc/guides/rel_notes/de
05/07/2023 12:53, Nipun Gupta:
> VFIO module provides configurable dma_entry_limit
> parameter to store the DMA entries. By default this
> is 64K and if we are using --no-huge, we shall need
> to increase the value of dma_entry_limit. Add
> commands in linux_gsg document to change the
> dma_entry_l
> -Original Message-
> From: Power, Ciara
> Sent: Wednesday, July 5, 2023 11:26 AM
> To: dev@dpdk.org
> Cc: gak...@marvell.com; De Lara Guarch, Pablo
> ; Power, Ciara ;
> Cornu, Marcel D ; Ji, Kai
> Subject: [PATCH v2] crypto/ipsec_mb: fix jobs array used for burst
>
> The jobs variab
On 7/5/2023 10:40 AM, lihuisong (C) wrote:
>
> 在 2023/7/5 16:02, Ferruh Yigit 写道:
>> On 7/5/2023 4:16 AM, lihuisong (C) wrote:
>>> 在 2023/7/4 18:59, Ferruh Yigit 写道:
On 7/4/2023 9:45 AM, Jie Hai wrote:
> Function update_queue_state updates queue state of all queues
> of all ports, usi
On 7/4/2023 10:02 PM, Stephen Hemminger wrote:
> Support rte_eth_dev_set_mtu for pcap driver when the
> pcap device is convigured to point to a network interface.
>
> This is rebased an consolidated from earlier version.
> Added support for FreeBSD.
>
As far as I understand motivation is to make
04/07/2023 09:10, Feifei Wang пишет:
To support mbufs recycle mode, announce the coming ABI changes
from DPDK 23.11.
Signed-off-by: Feifei Wang
Reviewed-by: Ruifeng Wang
---
doc/guides/rel_notes/deprecation.rst | 4
1 file changed, 4 insertions(+)
diff --git a/doc/guides/rel_notes/dep
18/04/2023 09:25, Sivaprasad Tummala пишет:
A new flag RTE_CPUFLAG_MONITORX is added to rte_cpu_flag_t in
DPDK 23.07 release to support monitorx instruction on EPYC processors.
This results in ABI breakage for legacy apps.
Signed-off-by: Sivaprasad Tummala
---
doc/guides/rel_notes/deprecation
04/07/2023 20:44, Akhil Goyal пишет:
Structure rte_security_ops and rte_security_ctx are meant to
be used by rte_security library and the PMDs associated.
These will be moved to an internal header in DPDK 23.11 release.
Signed-off-by: Akhil Goyal
---
doc/guides/rel_notes/deprecation.rst | 3 +
05/07/2023 11:36, Ciara Power пишет:
The socket ID field for a cryptodev device data was unsigned int.
Due to recent changes to the default device socket ID,
this caused an issue when the socket ID was unknown and set to -1.
The device socket ID wrapped around to 255,
and caused errors when alloc
Announce the removal of the single-event enqueue and dequeue
operations from the eventdev ABI.
Signed-off-by: Mattias Rönnblom
---
PATCH v2: Fix commit subject prefix.
---
doc/guides/rel_notes/deprecation.rst | 8
1 file changed, 8 insertions(+)
diff --git a/doc/guides/rel_notes/depre
> -Original Message-
> From: Zeng, ZhichaoX
> Sent: Wednesday, July 5, 2023 2:03 PM
> To: dev@dpdk.org
> Cc: Zhang, Qi Z ; Zeng, ZhichaoX
> ; Wu, Jingjing ; Xing, Beilei
>
> Subject: [PATCH] doc: update iavf feature list
>
> The iavf supports timestamp offload on vector path.
>
> Sig
The mlx5 provides the send scheduling on specific moment of time,
and for the related kind of applications it would be extremely useful
to have extra debug information - when and how packets were scheduled
and when the actual sending was completed by the NIC hardware (it helps
application to track
The Python script is intended to analyze mlx5 PMD
datapath traces and report:
- tx_burst routine timings
- how packets are pushed to WQEs
- how packet sending is completed with timings
Signed-off-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/tools/mlx5_trace.py | 307 ++
There is an intention to engage DPDK tracing capabilities
for mlx5 PMDs monitoring and profiling in various modes.
The patch introduces tracepoints for the Tx datapath in
the ethernet device driver.
To engage this tracing capability the following steps
should be taken:
- meson option -Denable_tra
There is the demand to trace the send completions of
every WQE if time scheduling is enabled.
The patch extends the size of completion queue and
requests completion on every issued WQE in the
send queue. As the result hardware provides CQE on
each completed WQE and driver is able to fetch
completi
The mlx5 provides the send scheduling on specific moment of time,
and for the related kind of applications it would be extremely useful
to have extra debug information - when and how packets were scheduled
and when the actual sending was completed by the NIC hardware (it helps
application to track
Acked-by: Hemant Agrawal
On 05-Jul-23 1:14 AM, Akhil Goyal wrote:
Caution: This is an external email. Please take care when clicking links or
opening attachments. When in doubt, report the message using the 'Report this
email' button
Structure rte_security_ops and rte_security_ctx are meant
VFIO module provides configurable dma_entry_limit
parameter to store the DMA entries. By default this
is 64K and if we are using --no-huge, we shall need
to increase the value of dma_entry_limit. Add
commands in linux_gsg document to change the
dma_entry_limit.
Signed-off-by: Nipun Gupta
---
doc
The socket ID field for a cryptodev device data was unsigned int.
Due to recent changes to the default device socket ID,
this caused an issue when the socket ID was unknown and set to -1.
The device socket ID wrapped around to 255,
and caused errors when allocating memory.
Changing this field type
IAVF curerently supports TSO and Tunnel TSO. Both these two features
are implemented in scalar path. As there are missed flags for Tunnel
TSO, it selects vector paths wrongly when only Tunnel TSO is enabled.
This patch added the missed flags to fix the Tunnel TSO path selecting.
Signed-off-by:
Due to recent changes to the default device socket ID,
before being used as an index for session mempool list,
the socket ID should be set to 0 if unknown (-1).
Fixes: 7dcd73e37965 ("drivers/bus: set device NUMA node to unknown by default")
Cc: olivier.m...@6wind.com
Cc: sta...@dpdk.org
Signed-of
1 - 100 of 120 matches
Mail list logo
