Hi, Slava
> -邮件原件-
> 发件人: Slava Ovsiienko
> 发送时间: 2021年5月13日 18:49
> 收件人: Feifei Wang ; Matan Azrad
> ; Shahaf Shuler
> 抄送: dev@dpdk.org; nd ; sta...@dpdk.org; Ruifeng Wang
> ; nd ; nd
> 主题: RE: [PATCH v1 3/4] net/mlx5: fix rebuild bug for Memory Region cache
>
> Hi, Feifei
>
> .. sni
From: "Zou, Steven"
This patchset is for couple of new feature on DCF
1, enable multiple DCF instances
2, enhanced GRE in DCF support
3, enhanced VxLAN support
Signed-off-by: Zou, Steven
---
app/test-pmd/cmdline.c | 13 +-
drivers/common/iavf/virtchnl.h | 6 +-
d
On 2021/5/14 6:19, Honnappa Nagarahalli wrote:
>
>
>>
>> Currently, the SVE code is compiled only when -march supports SVE (e.g. '-
>> march=armv8.2a+sve'), there maybe some problem[1] with this approach.
>>
>> The solution:
>> a. If the minimum instruction set support SVE then compiles it.
>>
>
> On Thu, May 13, 2021 at 6:41 PM Chengwen Feng
> wrote:
> >
> > Currently, the soc_kunpeng930 declares '-march=armv8.2-a+crypto+sve',
> > but some compiler doesn't recognize the march because it doesn't
> > support sve.
> >
> > To solve this bug we use the following scheme:
> > 1. Define 'mar
>
> Currently, the SVE code is compiled only when -march supports SVE (e.g. '-
> march=armv8.2a+sve'), there maybe some problem[1] with this approach.
>
> The solution:
> a. If the minimum instruction set support SVE then compiles it.
> b. Else if the compiler support SVE then compiles it.
> c.
> >>> Currently, the SVE code is compiled only when -march supports SVE
> >>> (e.g. '- march=armv8.2a+sve'), there maybe some problem[1] with this
> >> approach.
> >>>
> >>> The solution:
> >>> a. If the minimum instruction set support SVE then compiles it.
> >>> b. Else if the compiler support S
The modify_field Flow API assumes that the META item is 32 bits wide.
But the C Register that is used for Meta item can be 16 or 32 bits
wide depending on kernel and firmware configurations.
Take this into consideration and use the appropriate META width.
Fixes: 641dbe4fb0 ("net/mlx5: support modi
On 4/7/2021 11:33 PM, Thomas Monjalon wrote:
> The script check-doc-vs-code.sh may be used to add
> some automatic checks of the doc.
>
> If run without any argument, a complete check is done.
> The optional argument is a git history reference point
> to check faster only what has changed since th
The vDPA PCI device unplug process should release all the private
device resources and also to unregister the device.
The device unregistration was missed what remained the device data
invalid in the rte_vhost library.
Unregister the device in unplug process via the remove operation.
Fixes: 9527
The vDPA PCI device unplug proccess should release all the private
device resources and also to unregister the device.
The device unregistration was missed what remained the device data
invalid in the rte_vhost library.
Unregister the device in unplug proccess via the remove operation.
Fixes: 95
On Thu, May 13, 2021 at 6:41 PM Chengwen Feng wrote:
>
> Currently, the soc_kunpeng930 declares '-march=armv8.2-a+crypto+sve',
> but some compiler doesn't recognize the march because it doesn't
> support sve.
>
> To solve this bug we use the following scheme:
> 1. Define 'march_base' tuple which d
On 5/13/2021 3:38 PM, Kevin Traynor wrote:
> On 13/05/2021 15:11, David Marchand wrote:
>> On Thu, May 13, 2021 at 2:38 PM Chenbo Xia wrote:
>>>
>>> This patch fixes an issue of application crash because of vhost iotlb
>>> not initialized when virtio has multiqueue enabled.
>>>
>>> iotlb messages
Hi Nipun,
> -Original Message-
> From: Nipun Gupta
> Sent: Thursday, May 13, 2021 4:02 AM
> To: Chautru, Nicolas ; Hemant Agrawal
> ; dev@dpdk.org; gak...@marvell.com
> Cc: david.march...@redhat.com
> Subject: RE: [PATCH v4 5/8] baseband/la12xx: add queue and modem config
> support
>
>
Hi.
This release notes update describes that patch:
http://patches.dpdk.org/project/dpdk/patch/20210428144142.85929-2-adamx.dybkow...@intel.com/
Adam
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, 13 May, 2021 14:36
> To: Dybkowski, AdamX
> Cc: dev@dpdk.org; Zhang, Roy
On 13/05/2021 15:11, David Marchand wrote:
> On Thu, May 13, 2021 at 2:38 PM Chenbo Xia wrote:
>>
>> This patch fixes an issue of application crash because of vhost iotlb
>> not initialized when virtio has multiqueue enabled.
>>
>> iotlb messages will be sent when some queues are not enabled. If w
On Thu, May 13, 2021 at 2:38 PM Chenbo Xia wrote:
>
> This patch fixes an issue of application crash because of vhost iotlb
> not initialized when virtio has multiqueue enabled.
>
> iotlb messages will be sent when some queues are not enabled. If we
> initialize iotlb in vhost_user_set_vring_num,
Akhil Goyal writes:
> Hi Ciara,
>
> I see that CI is reporting one error in autotest. Can you please check
> if it is because of your patchset?
>
> 9/96 DPDK:fast-tests / cycles_autotest FAIL 1.42s (exit status 255 or
> signal 127 SIGinvalid)
Just to note, I'm looking into that test case a bit m
On 13/05/2021 14:10, Ferruh Yigit wrote:
> On 5/13/2021 1:28 PM, Chenbo Xia wrote:
>> This patch fixes an issue of application crash because of vhost iotlb
>> not initialized when virtio has multiqueue enabled.
>>
>> iotlb messages will be sent when some queues are not enabled. If we
>> initialize
Currently, the soc_kunpeng930 declares '-march=armv8.2-a+crypto+sve',
but some compiler doesn't recognize the march because it doesn't
support sve.
To solve this bug we use the following scheme:
1. Define 'march_base' tuple which defines support march, it should
arrange from lower to higher.
e.g.
This patch set contains two bugfixes for Kunpeng SVE compile.
---
v4:
* fix compile option combine problem
v3:
* fix checkpatch error of 1/2 patch
* modify comments of 1/2 patch
v2:
* support select best -march parameter for kunpeng soc
Chengwen Feng (2):
config/arm: select most suitable -march
Currently, the SVE code is compiled only when -march supports SVE
(e.g. '-march=armv8.2a+sve'), there maybe some problem[1] with this
approach.
The solution:
a. If the minimum instruction set support SVE then compiles it.
b. Else if the compiler support SVE then compiles it.
c. Otherwise don't com
On 5/13/2021 1:28 PM, Chenbo Xia wrote:
> This patch fixes an issue of application crash because of vhost iotlb
> not initialized when virtio has multiqueue enabled.
>
> iotlb messages will be sent when some queues are not enabled. If we
> initialize iotlb in vhost_user_set_vring_num, it could hap
13/05/2021 10:35, Li Zhang:
> A meter may handle Rx queue reference in his sub-policies.
> In stop operation, all the Rx queues are released.
>
> Wrongly, the meter reference was not released before
> destroying the Rx queues what cause an error in stop.
>
> Release the Rx queues meter references
Hi,
I just noticed this, will try to fix it. Thanks.
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, May 13, 2021 8:58 PM
> To: Bing Zhao
> Cc: Slava Ovsiienko ; Matan Azrad
> ; sta...@dpdk.org; dev@dpdk.org; Ori Kam
> ; Raslan Darawsheh ; Asaf
> Penso
> Subject: Re: [dpd
13/05/2021 13:13, Bing Zhao:
> --- a/drivers/net/mlx5/mlx5_trigger.c
> +++ b/drivers/net/mlx5/mlx5_trigger.c
> @@ -19,6 +19,7 @@
> #include "mlx5_tx.h"
> #include "mlx5_utils.h"
> #include "rte_pmd_mlx5.h"
> +#include "mlx5_verbs.h"
You cannot include a Linux-only header here.
It fails on Windo
> > Currently, the counter offset support is discovered by creating the rule
> > with
> > invalid offset counter and drop action in root table. If the rule creation
> > fails
> > with EINVAL errno, that mean counter offset is not supported in root table.
> >
> > However, drop action may not be s
> Yunjian Wang (2):
> net/mlx4: fix mem leak when a device configured repeatedly
> net/mlx5: fix mem leak when a device configured repeatedly
Applied in next-net-mlx, thanks.
This patch fixes an issue of application crash because of vhost iotlb
not initialized when virtio has multiqueue enabled.
iotlb messages will be sent when some queues are not enabled. If we
initialize iotlb in vhost_user_set_vring_num, it could happen that
iotlb update comes when iotlb pool of dis
13/05/2021 05:38, wangyunjian:
> From: Yunjian Wang
>
> This series include two fixes patches for mlx4/mlx5.
>
> ---
> v2:
>* remove redundant empty line
Please mark your old patches as superseded in patchwork.
13/05/2021 14:05, Adam Dybkowski:
> This patch extends the DPDK 21.05 release notes with the QAT GEN3
> compression update.
Is it related to a recent change?
You should add a Fixes line here.
Next time, please update the release notes with the code.
> Signed-off-by: Adam Dybkowski
> ---
> +* **U
Release status meeting minutes {Date}
=
:Date: 13 May 2021
:toc:
.Agenda:
* Release Dates
* -rc2/3 status
* Subtrees
* LTS
* Defects
* Opens
.Participants:
* Arm
* Debian/Microsoft
* Intel
* Individual contributor(s)
* Marvell
* Nvidia
* Red Hat
Release Dates
> -Original Message-
> From: Dybkowski, AdamX
> Sent: Thursday, May 13, 2021 1:06 PM
> To: dev@dpdk.org; Zhang, Roy Fan ;
> gak...@marvell.com
> Cc: Dybkowski, AdamX
> Subject: [PATCH] doc: update QAT compression PMD in 21.05 rel notes
>
> This patch extends the DPDK 21.05 release notes
This patch extends the DPDK 21.05 release notes with the QAT GEN3
compression update.
Signed-off-by: Adam Dybkowski
---
doc/guides/rel_notes/release_21_05.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_05.rst
b/doc/guides/rel_notes/release_21_05.rst
Hi,
> -Original Message-
> From: Wan Junjie
> Sent: Thursday, May 13, 2021 6:52 PM
> To: Xia, Chenbo
> Cc: David Marchand ; Maxime Coquelin
> ; dev@dpdk.org; Peng He
> ; Zhihong Wang ;
> Thomas Monjalon
> Subject: Re: [External] RE: [dpdk-dev] [PATCH v2] vhost: avoid iotlb mempool
> all
Hi,
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, May 13, 2021 12:04 AM
> To: Bing Zhao ; Slava Ovsiienko
>
> Cc: Matan Azrad ; dev@dpdk.org; Ori Kam
> ; Raslan Darawsheh ;
> sta...@dpdk.org
> Subject: Re: [PATCH] net/mlx5: fix loopback for DV queue
>
> External email:
In the past, all the queues and other hardware objects were created
through Verbs interface. Currently, most objects creation are
migrated to Devx interface by default, including queues. Only when
the DV is disabled by device arg or eswitch is enabled, are all or
some of the objects created through
> > + .type = RTE_BBDEV_OP_LDPC_DEC,
> > + .cap.ldpc_dec = {
> > + .capability_flags =
> > + RTE_BBDEV_LDPC_CRC_TYPE_24A_CHECK |
> > + RTE_BBDEV_LDPC_LLR_COMPRESSION |
>
> Are you sure you support the 6b
> -Original Message-
> From: Wan Junjie
> Sent: Thursday, May 13, 2021 6:52 PM
> To: Xia, Chenbo
> Cc: David Marchand ; Maxime Coquelin
> ; dev@dpdk.org; Peng He
> ; Zhihong Wang ;
> Thomas Monjalon
> Subject: Re: [External] RE: [dpdk-dev] [PATCH v2] vhost: avoid iotlb mempool
> allocati
On Thu, May 13, 2021 at 6:24 PM Xia, Chenbo wrote:
>
> > -Original Message-
> > From: David Marchand
> > Sent: Thursday, May 13, 2021 4:39 PM
> > To: Xia, Chenbo ; Maxime Coquelin
> >
> > Cc: Wan Junjie ; dev@dpdk.org; Peng He
> > ; Zhihong Wang
> > Subject: Re: [dpdk-dev] [PATCH v2] vh
Hi, Feifei
.. snip..
>
> I can understand you worry that if there is no 'wmb at last', when agent_1
> leave the locked section, agent_2 still may observe unchanged global cache.
>
> However, when agent_2 take a lock and get(new MR) in time slot 7(Fig.1), it
> means
> agent_1 has finished updatin
Minutes of Technical Board Meeting, 2021-05-05
Members Attending
-
-Aaron
-Bruce
-Ferruh
-Hemant
-Honnappa
-Jerin
-Kevin (chair)
-Konstantin
-Maxime
-Olivier
-Stephen
-Thomas
NOTE: The technical board meetings every second Wednesday at
https://meet.jit.si/DPDK at 3 pm UTC.
Meeting
> -Original Message-
> From: David Marchand
> Sent: Thursday, May 13, 2021 4:39 PM
> To: Xia, Chenbo ; Maxime Coquelin
>
> Cc: Wan Junjie ; dev@dpdk.org; Peng He
> ; Zhihong Wang
> Subject: Re: [dpdk-dev] [PATCH v2] vhost: avoid iotlb mempool allocation while
> IOMMU disabled
>
> On Wed
On Wed, May 12, 2021 at 11:12:36PM +, Honnappa Nagarahalli wrote:
>
>
> >
> > Currently, the SVE code is compiled only when -march supports SVE (e.g. '-
> > march=armv8.2a+sve'), there maybe some problem[1] with this approach.
> >
> > The solution:
> > a. If the minimum instruction set supp
On Wed, May 12, 2021 at 09:43:57PM +0300, Medvedkin, Vladimir wrote:
> Hi Bruce,
>
> On 12/05/2021 19:32, Bruce Richardson wrote:
> > On Mon, May 10, 2021 at 06:53:19PM +0200, Heinrich Kuhn wrote:
> > > From: "Chaoyong.He"
> > >
> > > Set the Rx multi-queue mode to NONE when configuring a port t
From: Li Zhang
> A meter may handle Rx queue reference in his sub-policies.
> In stop operation, all the Rx queues are released.
>
> Wrongly, the meter reference was not released before destroying the Rx
> queues what cause an error in stop.
>
> Release the Rx queues meter references in stop o
From: Suanming Mou
> Currently, the counter offset support is discovered by creating the rule with
> invalid offset counter and drop action in root table. If the rule creation
> fails
> with EINVAL errno, that mean counter offset is not supported in root table.
>
> However, drop action may not
On 13/5/2021 10:09 AM, Pattan, Reshma wrote:
-Original Message-
From: Thomas Monjalon
Ping for review please.
21/04/2021 05:53, Min Hu (Connor):
From: HongBo Zheng
+ ep_hgh_edpi = EMPTY_POLL_HGH_THRESHOLD;
+ if (hgh_edpi > 0)
ep_hgh_edpi =
> -Original Message-
> From: Thierry Herbelot
> Sent: Friday, May 7, 2021 4:48 PM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; tho...@monjalon.net;
> Juraj Linkeš ; Honnappa Nagarahalli
> ; Ruifeng Wang
>
> Subject: [PATCH-V7] config/arm: add Qualcomm Centriq 2400 SoC config
>
> From the
> -Original Message-
> From: Thomas Monjalon
> Ping for review please.
>
> 21/04/2021 05:53, Min Hu (Connor):
> > From: HongBo Zheng
> >
> > + ep_hgh_edpi = EMPTY_POLL_HGH_THRESHOLD;
> >
> > + if (hgh_edpi > 0)
> > ep_hgh_edpi = hgh_edpi;
> >
Chang
On Wed, Apr 7, 2021 at 9:10 AM Xia, Chenbo wrote:
>
> > -Original Message-
> > From: Wan Junjie
> > Sent: Tuesday, February 2, 2021 4:15 PM
> > To: Xia, Chenbo ; Maxime Coquelin
> >
> > Cc: dev@dpdk.org; Wan Junjie ; Peng He
> > ; Zhihong Wang
> > Subject: [PATCH v2] vhost: avoid iotlb
A meter may handle Rx queue reference in his sub-policies.
In stop operation, all the Rx queues are released.
Wrongly, the meter reference was not released before
destroying the Rx queues what cause an error in stop.
Release the Rx queues meter references in stop operation.
Fixes: fc6ce56bba ("n
Hook check-symbol-maps.sh in the symbol check when in developer mode to
help developers catch issues before submitting their changes.
Signed-off-by: David Marchand
Reviewed-by: Ray Kinsella
---
Changes since v1:
- now that we have the developer mode, added this check in the build
process,
---
Add a check on versioned symbol duplicates in map files.
Signed-off-by: David Marchand
---
Changes since v1:
- rebased,
---
devtools/check-symbol-maps.sh | 15 +++
1 file changed, 15 insertions(+)
diff --git a/devtools/check-symbol-maps.sh b/devtools/check-symbol-maps.sh
index 0845
Avoid relying on GNU grep --exclude option.
Fixes: f8ad40dc998c ("devtools: check orphan symbols in map files")
Signed-off-by: David Marchand
---
devtools/check-symbol-maps.sh | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/devtools/check-symbol-maps.sh b/devtools/check-sy
On Thu, May 13, 2021 at 09:10:37AM +0800, Min Hu (Connor) wrote:
> Variable i is used as a denominator which may be zero, and
> this may result in segmentation fault.
>
> This patch fixed it.
>
> Fixes: 948bc3d6d095 ("test: add reciprocal based division")
> Cc: sta...@dpdk.org
>
> Signed-off-by:
Currently, the counter offset support is discovered by creating the
rule with invalid offset counter and drop action in root table. If
the rule creation fails with EINVAL errno, that mean counter offset
is not supported in root table.
However, drop action may not be supported in some rdma-core ver
A meter may handle Rx queue reference in his sub-policies.
In stop operation, all the Rx queues are released.
Wrongly, the meter reference was not released before
destroying the Rx queues what cause an error in stop.
Release the Rx queues meter references in stop operation.
Fixes: fc6ce56bba ("n
On Wed, May 12, 2021 at 10:38 PM Thomas Monjalon wrote:
>
> 12/05/2021 22:24, Thomas Monjalon:
> > > > Hook check-symbol-maps.sh in the symbol check when in developer mode to
> > > > help developers catch issues before submitting their changes.
> > > >
> > > > Signed-off-by: David Marchand
> > >
58 matches
Mail list logo
