From: Dapeng Yu
In original implementation, if DCF is created on PF1, the PF ID is
still 0, but not 1. Without the right PF ID, the ACL will not work.
This patch makes VF to get its parent's physical function ID.
Fixes: 0b02c9519432 ("net/ice: handle PF initialization by DCF")
CC: sta...@dpdk.o
>
>> In crypto adapter metadata, reserved bytes in request info structure is
>> a space holder for response info. It enforces an order of operation if
>> the structures are updated using memcpy to avoid overwriting response
>> info. It is logical to move the reserved space out of request info. It
>
From: Qiming Chen
Sent: Tuesday, August 31, 2021 14:41
To: Wang, Haiyue
Cc: dev@dpdk.org
Subject: Re: [PATCH] net/ixgbe: slove mbuf leak
Do you mean let me sync to mailto:sta...@dpdk.org?
Yes. ;-)
Qiming Chen
mailto:chenqiming_hua...@163.com
签名由 https://mail.163.com/dashi/dlpro.html?from=mail
Do you mean let me sync to sta...@dpdk.org?
| |
Qiming Chen
|
|
chenqiming_hua...@163.com
|
签名由网易邮箱大师定制
On 8/31/2021 14:04，Wang, Haiyue wrote：
-Original Message-
From: Qiming Chen
Sent: Tuesday, August 31, 2021 11:28
To: dev@dpdk.org
Cc: Wang, Haiyue ; Qiming Chen
Subject: [PATCH] net/
> In crypto adapter metadata, reserved bytes in request info structure is
> a space holder for response info. It enforces an order of operation if
> the structures are updated using memcpy to avoid overwriting response
> info. It is logical to move the reserved space out of request info. It
> also
Hi Fan,
> Hi Akhil,
>
> > +__rte_internal
> > +static inline void *
> > +_rte_cryptodev_dequeue_prolog(uint8_t dev_id, uint8_t qp_id)
> > +{
> > + struct rte_cryptodev *dev = &rte_cryptodevs[dev_id];
> > +
> > + return dev->data->queue_pairs[qp_id];
> > +}
>
> [Fan: the function name looks un
> -Original Message-
> From: Qiming Chen
> Sent: Tuesday, August 31, 2021 11:28
> To: dev@dpdk.org
> Cc: Wang, Haiyue ; Qiming Chen
>
> Subject: [PATCH] net/ixgbe: slove mbuf leak
>
> A local test found that repeated port start and stop operations during
> the continuous SSE vector buff
From: Long Li
The driver code had wrong assumption that all the addresses to ring buffers
in the secondary process are the same as those in the primary process. This
is not always correct as the channels could be mapped to different
addresses in the secondary process.
Fix this by keeping track o
Hi Wenwu,
> -Original Message-
> From: Ma, WenwuX
> Sent: Friday, August 27, 2021 10:01 PM
> To: dev@dpdk.org
> Cc: maxime.coque...@redhat.com; Xia, Chenbo ; Jiang,
> Cheng1 ; Hu, Jiayu ; Ma, WenwuX
> ; sta...@dpdk.org
> Subject: [PATCH v2] examples/vhost: fix memory leak on forwarding pa
Hi Gaoxiang,
> -Original Message-
> From: Gaoxiang Liu
> Sent: Friday, August 27, 2021 10:19 PM
> To: maxime.coque...@redhat.com; Xia, Chenbo
> Cc: dev@dpdk.org; liugaoxi...@huawei.com; Gaoxiang Liu
> Subject: [PATCH v6] vhost: fix crash on port deletion
>
> The rte_vhost_driver_unregi
From: Satheesh Paul
This patch adds an ROC API to merge base steering rule with rules
added by VF.
Signed-off-by: Satheesh Paul
Reviewed-by: Kiran Kumar Kokkilagadda
---
drivers/common/cnxk/roc_npc.c | 27 +++
drivers/common/cnxk/roc_npc.h | 5 ++---
drivers/commo
ok，Has been resubmitted to LTS
https://inbox.dpdk.org/stable/20210831035207.5590-1-chenqiming_hua...@163.com/T/#u
| |
Qiming Chen
|
|
chenqiming_hua...@163.com
|
签名由网易邮箱大师定制
On 8/30/2021 13:23，Xing, Beilei wrote：
-Original Message-
From: dev On Behalf Of Qiming Chen
Sent: Monday, Aug
>
> In crypto adapter metadata, reserved bytes in request info structure is a
> space
> holder for response info. It enforces an order of operation if the structures
> are
> updated using memcpy to avoid overwriting response info. It is logical to move
> the reserved space out of request info. I
A local test found that repeated port start and stop operations during
the continuous SSE vector bufflist receiving process will cause the mbuf
resource to run out. The final positioning is when the port is stopped,
the mbuf of the pkt_first_seg pointer is not released. Resources leak.
The patch sc
From: Zhihong Peng
Malloc cl in the cmdline_stdin_new function, so release in the
cmdline_stdin_exit function is logical, so that cl will not be
released alone.
Fixes: af75078fece3 (first public release)
Cc: sta...@dpdk.org
Signed-off-by: Zhihong Peng
---
lib/cmdline/cmdline_socket.c | 1 +
1
Hi Eugenio,
> -Original Message-
> From: Eugenio Perez Martin
> Sent: Tuesday, August 31, 2021 2:10 AM
> To: Xia, Chenbo
> Cc: Maxime Coquelin ; dev@dpdk.org; Pei Zhang
> ; Jason Wang
> Subject: Re: [PATCH] vhost: Clean iotlb cache on vring stop
>
> On Mon, Aug 30, 2021 at 1:58 PM Xia,
>
> On Tue, 24 Aug 2021 20:03:03 +
> Honnappa Nagarahalli wrote:
>
> > > One difference between this implementation and the previous one is
> > > this busy loop. rte_pause() relaxes the cpu, but will not make the
> > > calling thread to sleep and wait for the sync event. So here we can
> >
>
> On Mon, 30 Aug 2021 11:34:12 -0500
> Honnappa Nagarahalli wrote:
>
> > + /* Control thread status.
> > +*
> > +* _RTE_CTRL_THREAD_LAUNCHING - Yet to call pthread_create
> function
> > +* _RTE_CTRL_THREAD_RUNNING - Control thread is running
> successfully
> > +* _RTE_CTRL_
> -Original Message-
> From: Akhil Goyal
> Sent: Tuesday, August 10, 2021 8:50 PM
> To: dev@dpdk.org
> Cc: ano...@marvell.com; Nicolau, Radu ; Doherty,
> Declan ; hemant.agra...@nxp.com;
> ma...@nvidia.com; Ananyev, Konstantin ;
> tho...@monjalon.net; Zhang, Roy Fan ;
> asoma...@amd.com; r
> -Original Message-
> From: Akhil Goyal
> Sent: Wednesday, August 11, 2021 9:20 AM
> To: Matan Azrad ; dev@dpdk.org
> Cc: Anoob Joseph ; Nicolau, Radu
> ; Doherty, Declan ;
> hemant.agra...@nxp.com; Ananyev, Konstantin
> ; NBU-Contact-Thomas Monjalon
> ; Zhang, Roy Fan ;
> asoma...@amd.co
> -Original Message-
> From: Akhil Goyal
> Sent: Thursday, August 5, 2021 9:03 PM
> To: dev@dpdk.org
> Cc: ano...@marvell.com; Nicolau, Radu ; Doherty,
> Declan ; hemant.agra...@nxp.com;
> ma...@nvidia.com; Ananyev, Konstantin ;
> tho...@monjalon.net; Zhang, Roy Fan ;
> asoma...@amd.com; r
Removed freeing of unallocated mempool in event timer adapter create
unit test.
Fixes: d1f3385d0076 ("test: add event timer adapter auto-test")
Signed-off-by: Shijith Thotton
---
app/test/test_event_timer_adapter.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/app/test/test_event_timer_a
> -Original Message-
> From: Akhil Goyal
> Sent: Thursday, August 5, 2021 9:03 PM
> To: dev@dpdk.org
> Cc: ano...@marvell.com; Nicolau, Radu ; Doherty,
> Declan ; hemant.agra...@nxp.com;
> ma...@nvidia.com; Ananyev, Konstantin ;
> tho...@monjalon.net; Zhang, Roy Fan ;
> asoma...@amd.com; r
Hi Akhil,
> -Original Message-
> From: Akhil Goyal
> Sent: Sunday, August 29, 2021 1:52 PM
> To: dev@dpdk.org
> Cc: ano...@marvell.com; Nicolau, Radu ; Doherty,
> Declan ; hemant.agra...@nxp.com;
> ma...@nvidia.com; Ananyev, Konstantin ;
> tho...@monjalon.net; Zhang, Roy Fan ;
> asoma...@
Replace the inline functions in the eventdev user application API with
regular non-inline API calls. This allows for a cleaner and more
simple API/ABI, but might well also cause performance regressions.
The purpose of this RFC patch is to allow for performance testing.
The rte_eventdev struct dec
In crypto adapter metadata, reserved bytes in request info structure is
a space holder for response info. It enforces an order of operation if
the structures are updated using memcpy to avoid overwriting response
info. It is logical to move the reserved space out of request info. It
also solves the
In crypto adapter metadata, reserved bytes in request info structure is
a space holder for response info. It enforces an order of operation if
the structures are updated using memcpy to avoid overwriting response
info. It is logical to move the reserved space out of request info. It
also solves the
Added changes to receive packets as event vector. By default this is
disabled and can be enabled using the option --enable-vector. Vector
size and timeout to form the vector can be configured using options
--vector-size and --vector-tmo-ns.
Example:
dpdk-l3fwd -l 0-3 -n 4 -- -p 0x03 --mode
Hi Ferruh,
Sorry, I forgot to mention, Yes, I was compiling with -Denable_kmods=true.
I just partially found a root cause. The target system Photon OS (It
VMware distro)
It has 4 kernels variation regular, RT, ESX, security. ( I was using ESX
variation that optimized for
ESXi but it looks lik
On Mon, Aug 30, 2021 at 1:58 PM Xia, Chenbo wrote:
>
> Hi Eugenio,
>
Hi Chenbo,
> > -Original Message-
> > From: Eugenio Pérez
> > Sent: Saturday, August 28, 2021 12:13 AM
> > To: Maxime Coquelin ; Xia, Chenbo
> >
> > Cc: dev@dpdk.org; Pei Zhang ; Jason Wang
> >
> > Subject: [PATCH] v
On Thu, Aug 26, 2021 at 9:06 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Thu, 26 Aug 2021 20:46:47 +0300
> Igor Ryzhov wrote:
>
> > Could you please clarify where exactly do I need to use rtnl lock?
> > From what I understand, netif_carrier_on/off can be called without the
> lo
On 8/30/2021 5:47 PM, spyroot wrote:
> Hi Folks,
>
> I troubleshooting already for a couple of hours but still not clear why
> exactly the build process
> failing on the KNI module. I've tried to disable KNI but it same issue.
> Any ideas what might be a root cause. It x86_64
>
>
> [1/2] Gener
The lock sh->txpp.mutex was not correctly released if the function
returned in these two branches, which may led to deadlock if the
function was acquired again.
Fixes: d133f4cdb7 ("net/mlx5: create clock queue for packet pacing")
Cc: sta...@dpdk.org
Signed-off-by: Chengfeng Ye
---
drivers/net/m
The lock sh->txpp.mutex was not correctly released if the function
returned in these two branches, which may led to deadlock if the
function was acquired again.
Fixes: d133f4cdb7 ("net/mlx5: create clock queue for packet pacing")
Cc: sta...@dpdk.org
Signed-off-by: Chengfeng Ye
---
drivers/net/ml
https://bugs.dpdk.org/show_bug.cgi?id=800
Bug ID: 800
Summary: rte_malloc fails to allocate memory block, but
succeeds allocating bigger block
Product: DPDK
Version: 18.11
Hardware: All
OS: All
S
On 8/30/2021 10:41 AM, David Marchand wrote:
> Hi Ferruh,
>
> On Fri, Aug 27, 2021 at 3:26 AM Ferruh Yigit wrote:
>>
>> Add 'RTE_ETH' namespace to all enums & macros in a backward compatible
>
> For completeness, there are some struct renames too in this patch.
>
ack, will add to the commit lo
Hi Folks,
I troubleshooting already for a couple of hours but still not clear why
exactly the build process
failing on the KNI module. I've tried to disable KNI but it same issue.
Any ideas what might be a root cause. It x86_64
[1/2] Generating rte_kni with a custom command
FAILED: kernel/linu
On Mon, 30 Aug 2021 11:34:12 -0500
Honnappa Nagarahalli wrote:
> + /* Control thread status.
> + *
> + * _RTE_CTRL_THREAD_LAUNCHING - Yet to call pthread_create function
> + * _RTE_CTRL_THREAD_RUNNING - Control thread is running successfully
> + * _RTE_CTRL_THREAD_ERROR -
>
> On Mon, 30 Aug 2021 11:34:13 -0500
> Honnappa Nagarahalli wrote:
>
> > +static int
> > +test_crtl_thread(void)
>
> Is this a typo? Did you mean test_ctrl_thread?
Yes, thanks
On Mon, 30 Aug 2021 11:34:13 -0500
Honnappa Nagarahalli wrote:
> +static int
> +test_crtl_thread(void)
Is this a typo? Did you mean test_ctrl_thread?
Add a testcase to test launching of control threads.
Signed-off-by: Honnappa Nagarahalli
---
app/test/test_lcores.c | 41 +
1 file changed, 41 insertions(+)
diff --git a/app/test/test_lcores.c b/app/test/test_lcores.c
index 19a7ab9fce..738148f339 100644
-
Remove the usage of pthread barrier and replace it with
synchronization using atomic variable. This also removes
the use of reference count required to synchronize freeing
the memory.
Signed-off-by: Honnappa Nagarahalli
---
v3:
1) RFC to Patch
2) Changed commit log and API description [Olivier]
3
Avoid using stashing option of stype in NPA in cn10k-a0 stepping.
This is a workaround for a HW Errata due to which NPA stashing operations
will never result in writing the data into L2 cache. But instead, it will
be written into LLC.
Signed-off-by: Ashwin Sekhar T K
---
drivers/common/cnxk/roc
Update roc models.
Signed-off-by: Ashwin Sekhar T K
---
drivers/common/cnxk/roc_model.c | 51 +++
drivers/common/cnxk/roc_model.h | 53 +
2 files changed, 67 insertions(+), 37 deletions(-)
diff --git a/drivers/common/cnxk/roc_model.c b
Reduced max chunk pool cache size from RTE_MEMPOOL_CACHE_MAX_SIZE(512)
to 128.
If chunk pool cache is empty, it gets filled during arm. Filling 512
entries at a time will fail arm if timeout is shorter.
Fixes: 0e792433d051 ("event/cnxk: create and free timer adapter")
Signed-off-by: Shijith Thot
Type of kvargs value and handler function argument should match to avoid
spilling memory.
Fixes: 7ffa7379965e ("event/cnxk: add option to configure getwork mode")
Signed-off-by: Shijith Thotton
---
v2:
* Rebased.
drivers/event/cnxk/cnxk_eventdev.c | 6 +++---
drivers/event/cnxk/cnxk_tim_evdev
> -Original Message-
> From: dev On Behalf Of Akhil Goyal
> Sent: Saturday, July 31, 2021 8:13 PM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; david.march...@redhat.com;
> hemant.agra...@nxp.com; ano...@marvell.com; De Lara Guarch, Pablo
> ; Trahe, Fiona ;
> Doherty, Declan ; ma...@nvi
There is no reason to re-register a interrupt handler for LSC if this
feature was not requested in the first place.
A simple usecase is when asking for Rx interrupts without LSC interrupt.
Fixes: 26b683b4f7d0 ("net/virtio: setup Rx queue interrupts")
Cc: sta...@dpdk.org
Signed-off-by: David March
> -Original Message-
> From: pbhagavat...@marvell.com
> Sent: Tuesday, August 24, 2021 1:10 AM
> To: jer...@marvell.com; Gujjar, Abhinandan S ;
> Jayatheerthan, Jay ;
> Carrillo, Erik G
> Cc: Ananyev, Konstantin ; dev@dpdk.org; Pavan
> Nikhilesh
> Subject: [dpdk-dev] [RFC 14/15] eventd
Hello Pavan,
On Mon, Aug 23, 2021 at 9:41 PM wrote:
>
> From: Pavan Nikhilesh
>
> Use new driver interface for the fastpath enqueue/dequeue inline
> functions.
>
> Signed-off-by: Pavan Nikhilesh
> ---
> lib/eventdev/rte_event_crypto_adapter.h | 13 +-
> lib/eventdev/rte_event_eth_tx_adapte
> -Original Message-
> From: pbhagavat...@marvell.com
> Sent: Tuesday, August 24, 2021 1:10 AM
> To: jer...@marvell.com; Jayatheerthan, Jay ; Ray
> Kinsella
> Cc: Ananyev, Konstantin ; dev@dpdk.org; Pavan
> Nikhilesh
> Subject: [dpdk-dev] [RFC 13/15] eventdev: promote event vector API
> -Original Message-
> From: pbhagavat...@marvell.com
> Sent: Tuesday, August 24, 2021 1:10 AM
> To: jer...@marvell.com; Gujjar, Abhinandan S ;
> Jayatheerthan, Jay
> Cc: Ananyev, Konstantin ; dev@dpdk.org; Pavan
> Nikhilesh
> Subject: [dpdk-dev] [RFC 10/15] eventdev: remove rte prefix
> -Original Message-
> From: pbhagavat...@marvell.com
> Sent: Tuesday, August 24, 2021 1:10 AM
> To: jer...@marvell.com; Gujjar, Abhinandan S ;
> Jayatheerthan, Jay
> Cc: Ananyev, Konstantin ; dev@dpdk.org; Pavan
> Nikhilesh
> Subject: [dpdk-dev] [RFC 06/15] eventdev: use new API for i
According to flow action MARK definition, PMDs must set both
PKT_RX_FDIR and PKT_RX_FDIR_ID if the packet contains a mark.
Fixes: 1aacc3d388d3 ("net/sfc: support user mark and flag Rx for EF100")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Malov
Reviewed-by: Andrew Rybchenko
---
drivers/net/sfc/sf
Hi Sunil,
On 8/23/21 11:53 AM, Sunil Pai G wrote:
> Note to the reader:
> ---
> The intent of this patch is to explore possible different approaches
> of async implementations.
> Please consider this patch for discussions only and not for
> merge/upstream.
Thanks for sharing, that
If the child process returns -1, WEXITSTATUS converts to 255 because it
takes only 8 least significant bits. As a result, the test is not
considered failed when it should.
Signed-off-by: Igor Ryzhov
---
app/test/test_kni.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/t
It allows inspecting link parameters using ethtool.
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/kni_net.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/kernel/linux/kni/kni_net.c b/kernel/linux/kni/kni_net.c
index 611719b5ee27..931b9daf104f 100644
--- a/kerne
Current implementation doesn't allow us to update KNI carrier if the
interface is not yet UP in kernel. It means that we can't use it in the
same thread which is processing rte_kni_ops.config_network_if, which is
very convenient, because it allows us to have correct carrier status
of the interface
Tx prepare method calls rte_net_intel_cksum_prepare(), which
handles tunnel packets correctly, but Tx burst path does not
take tunnel presence into account when computing the offsets.
Signed-off-by: Ivan Malov
Reviewed-by: Andrew Rybchenko
---
drivers/net/virtio/virtqueue.h | 9 ++---
1 fil
> -Original Message-
> From: Jerin Jacob
> Sent: Monday, August 30, 2021 5:31 PM
> To: Xueming(Steven) Li
> Cc: dpdk-dev ; Ferruh Yigit ;
> NBU-Contact-Thomas Monjalon ;
> Andrew Rybchenko
> Subject: Re: [PATCH v2 01/15] ethdev: introduce shared Rx queue
>
> On Sat, Aug 28, 2021 at 7
Avoid using stype STSTP in batch alloc.
This is a workaround for errata IPBUNPA-37480.
Signed-off-by: Ashwin Sekhar T K
---
drivers/common/cnxk/roc_nix_queue.c | 5 +
drivers/common/cnxk/roc_nix_tm_ops.c | 8
drivers/common/cnxk/roc_npa.h| 2 +-
3 files changed, 14 inserti
On Mon, Aug 30, 2021 at 12:12 AM Michał Krawczyk wrote:
[...]
> Hi William,
>
> It's great to hear that you're working on ENA support for Windows!
>
> ENA PMD uses admin interrupt for processing all the commands like
> creating the IO queues, setting up the MTU, etc., and also for the
> AENQ event
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Implements vDPA ops set_vring_state to configure vring state.
>
> Signed-off-by: Vijay Kumar Srivastava
> ---
> drivers/vdpa/sfc/sfc_vdpa_ops.c | 54
> ++---
> 1 file changed,
NPA stack should be aligned to ROC cache line size.
Fixes: f765f5611240 ("common/cnxk: add NPA pool HW operations")
Signed-off-by: Ashwin Sekhar T K
---
drivers/common/cnxk/roc_npa.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/common/cnxk/roc_npa.c b/drivers/comm
Hi,
I'm testing DPDK on virtio device using a Linux VM (ubuntu 21.04) from
google cloud. But I couldn't get it to work. I wonder
1. Is "no modern virtio pci device found." expected?
2. Should I use vfio-pci, instead uio_pci_generic?
3. maybe on the public cloud, promiscuous mode is not allowed?
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Add support for unicast and broadcast MAC filter configuration.
I wonder whether we should not add an API to the vDPA framework for this
MAC management, as this is something generic.
What do you think?
> Signed-of
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Implement the vDPA ops get_notify_area to get the notify
> area info of the queue.
>
> Signed-off-by: Vijay Kumar Srivastava
> ---
> drivers/vdpa/sfc/sfc_vdpa_ops.c | 166
> ++-
v2:
* Fixed typo in commit message
* changed subject line
v1:
RX adapter uses memove() to move unprocessed events to the beginning of
the packet enqueue buffer. The use memmove() was found to consume good
amount of CPU cycles (about 20%).
This patch removes the use of memove() while implementing
> -Original Message-
> From: Dmitry Kozlyuk
> Sent: August 9, 2021, 12:45
> [...]
> Hi Anatoly, could you review please?
Hi Anatoly, did you have time to look?
Patch 1/3 may be useful on its own, because it makes adding and testing patches
like [1] easier.
[1]:
http://patchwork.dpdk.or
On Mon, Aug 30, 2021 at 1:59 PM Thomas Monjalon wrote:
>
> 30/08/2021 13:26, Ferruh Yigit:
> > On 8/30/2021 8:52 AM, Kamaraj P wrote:
> > > Hello All,
> > > Some more info.
> > > We are using DPDK 19.11.
> > >
> > > When we bringup application with --security-opt seccomp=unconfined
> > > option..
30/08/2021 13:45, Kamaraj P:
> Thanks Ferruh.
> One more option which we have figured out to resolve this error by adding
> SYS_NICE capability ( --cap-add=NET_ADMIN --cap-add=SYS_RESOURCE
> --cap-add=SYS_NICE).
> Not sure whether it will work for all the scenarios. However the above
> options will
Hi, Chengfeng
Thank you for the fix, we can adopt this (ie Upstream ML is not "wrong place" 😊)
I have some minor comment about:
- the problem should be described in past tense (should b e"sh->txpp.mutex WAS
not correctly")
- headline should be more specific, like "fix mutex unlock in txpp cleanup
30/08/2021 13:26, Ferruh Yigit:
> On 8/30/2021 8:52 AM, Kamaraj P wrote:
> > Hello All,
> > Some more info.
> > We are using DPDK 19.11.
> >
> > When we bringup application with --security-opt seccomp=unconfined
> > option..
> > THe application bringup successfully and didnt observe any error when
Hi Eugenio,
> -Original Message-
> From: Eugenio Pérez
> Sent: Saturday, August 28, 2021 12:13 AM
> To: Maxime Coquelin ; Xia, Chenbo
>
> Cc: dev@dpdk.org; Pei Zhang ; Jason Wang
>
> Subject: [PATCH] vhost: Clean iotlb cache on vring stop
Clean -> clean
>
> Old IOVA cache entries are
Thanks Ferruh.
One more option which we have figured out to resolve this error by adding
SYS_NICE capability ( --cap-add=NET_ADMIN --cap-add=SYS_RESOURCE
--cap-add=SYS_NICE).
Not sure whether it will work for all the scenarios. However the above
options will ensure DPDK application can run as non-p
On Mon, Aug 30, 2021 at 5:25 AM Robin Zhang wrote:
>
> In DPDK 21.05, iavf already became the default VF for i40e devices.
> So remove i40evf due to it's no need to maintain now.
>
> Robin Zhang (3):
> net/i40e: remove i40evf
> net/iavf: remove i40evf devargs option
> doc: remove old depreca
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Implement vDPA ops dev_conf and dev_close for DMA mapping,
> interrupt and virtqueue configurations.
>
> Signed-off-by: Vijay Kumar Srivastava
> ---
> drivers/vdpa/sfc/sfc_vdpa.c | 6 +
> drivers/vdpa/sfc/sf
On 8/30/2021 8:52 AM, Kamaraj P wrote:
> Hello All,
> Some more info.
> We are using DPDK 19.11.
>
> When we bringup application with --security-opt seccomp=unconfined
> option..
> THe application bringup successfully and didnt observe any error when DPDK
> init logs.
> Will that --security-opt se
Set crypto adapter enqueue and dequeue operations for CN10K.
Signed-off-by: Shijith Thotton
---
doc/guides/rel_notes/release_21_11.rst | 3 ++
drivers/event/cnxk/cn10k_eventdev.c | 45 +++-
drivers/event/cnxk/cn10k_worker.c| 11
drivers/event/cnxk/cn10k_worker.h
Added crypto adapter enqueue and dequeue operations for CN10K.
Signed-off-by: Shijith Thotton
---
drivers/crypto/cnxk/cn10k_cryptodev_ops.c | 95 +++
drivers/crypto/cnxk/cn10k_cryptodev_ops.h | 6 ++
drivers/crypto/cnxk/version.map | 2 +
3 files changed, 103 inse
Set crypto adapter enqueue and dequeue operations for CN9K.
Signed-off-by: Shijith Thotton
---
drivers/event/cnxk/cn9k_eventdev.c | 94 +++-
drivers/event/cnxk/cn9k_worker.c | 22 +
drivers/event/cnxk/cn9k_worker.h | 41 -
drivers/eve
Added crypto adapter enqueue and dequeue operations for CN9K.
Signed-off-by: Shijith Thotton
---
drivers/crypto/cnxk/cn9k_cryptodev_ops.c | 235 ---
drivers/crypto/cnxk/cn9k_cryptodev_ops.h | 6 +
drivers/crypto/cnxk/cnxk_cryptodev_ops.h | 28 +++
drivers/crypto/cnxk/meson
Added eventdev ops required to initialize crypto adapter.
Signed-off-by: Shijith Thotton
---
drivers/crypto/cnxk/cnxk_cryptodev_ops.h | 9 +++
drivers/event/cnxk/cn10k_eventdev.c | 46
drivers/event/cnxk/cn9k_eventdev.c | 45
drivers/event/cnxk/cnxk_eventdev
Added flow control based check to determine CPT IQ is full.
Signed-off-by: Shijith Thotton
---
drivers/common/cnxk/roc_cpt.c | 6 --
drivers/common/cnxk/roc_cpt.h | 11 +++
drivers/common/cnxk/roc_cpt_priv.h | 6 --
3 files changed, 15 insertions(+), 8 deletions(-)
d
Added a common macro to set eventdev enqueue and dequeue operations to
reduce code.
Signed-off-by: Shijith Thotton
Signed-off-by: Nithin Dabilpuram
---
drivers/event/cnxk/cn10k_eventdev.c | 133 +-
drivers/event/cnxk/cn9k_eventdev.c | 267 +++-
2 files chang
CPT can be told to submit events to SSO upon completion. Crypto adapter
uses this feature and the new flag can be used to optimize receive path
in those cases.
Signed-off-by: Shijith Thotton
---
drivers/net/cnxk/cn10k_rx.h | 5 +++--
drivers/net/cnxk/cn9k_rx.h | 3 ++-
2 files changed, 5 insert
First three patches are to support crypto adapter implementation and the
remaining patches add event crypto adapter OP_FORWARD mode support on
Marvell CN9K and CN10K SOC.
Shijith Thotton (8):
net/cnxk: add flag to show CPT can enqueue events
event/cnxk: add macro to set eventdev ops
common/c
Hi Vijay,
> -Original Message-
> From: Vijay Srivastava
> Sent: Wednesday, July 7, 2021 12:44 AM
> To: dev@dpdk.org
> Cc: maxime.coque...@redhat.com; Xia, Chenbo ;
> andrew.rybche...@oktetlabs.ru; Vijay Kumar Srivastava
> Subject: [PATCH 02/10] vdpa/sfc: add support for device initializa
A lot of flags are parts of a group but are documented alone.
The Doxygen syntax @{ and @} for grouping is used
to make flags appear together and have a common description.
Some Rx/Tx offload flags and RSS definitions are not grouped
because they need to be all properly documented first.
Signed-o
On 2021-08-24 09:43, Mattias Rönnblom wrote:
> On 2021-08-23 21:40, pbhagavat...@marvell.com wrote:
>> From: Pavan Nikhilesh
>>
>> Mark all the driver specific functions as internal, remove
>> `rte` prefix from `struct rte_eventdev_ops`.
>> Remove experimental tag from internal functions.
>> Remov
> -Original Message-
> From: Jerin Jacob
> Sent: 30 August 2021 15:11
> To: Kundapura, Ganapati
> Cc: Jayatheerthan, Jay ; dpdk-dev
>
> Subject: Re: [PATCH v1] eventdev: change packet enqueue buffer in RX
> adapter to circular buffer
>
> On Fri, Aug 27, 2021 at 3:12 PM Ganapati Kundap
On 8/27/2021 7:28 AM, Xueming(Steven) Li wrote:
>
>
>> -Original Message-
>> From: Kevin Traynor
>> Sent: Thursday, August 26, 2021 6:46 PM
>> To: Ferruh Yigit ; Ben Magistro
>> ; dev@dpdk.org; Beilei Xing ;
>> Luca Boccassi ; Christian Ehrhardt
>> ; Xueming(Steven) Li
>>
>> Cc: ben.m
Hi Ferruh,
On Fri, Aug 27, 2021 at 3:26 AM Ferruh Yigit wrote:
>
> Add 'RTE_ETH' namespace to all enums & macros in a backward compatible
For completeness, there are some struct renames too in this patch.
> way. The macros for backward compatibility can be removed in next LTS.
Olivier had pro
On Fri, Aug 27, 2021 at 3:12 PM Ganapati Kundapura
wrote:
>
> RX adapter user memove() to move unprocessed events to the beginning of
> the packet enqueue buffer. The use memmove() was found to consume good
> amount of CPU cycles (about 20%).
>
> This patch removes the use of memove() while implem
Hi Jerin,
Thanks for your comments and suggestions.
I will try to update this in V2.
On Mon, Aug 23, 2021 at 8:51 PM Jerin Jacob wrote:
> On Mon, Aug 23, 2021 at 2:14 PM Aman Kumar wrote:
> >
> > This patch provides rte_memcpy* calls optimized for
> > AMD EPYC Gen2 platforms. This option is dis
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Implement vDPA ops get_vfio_device_fd to get the VFIO device fd.
>
> Signed-off-by: Vijay Kumar Srivastava
> ---
> drivers/vdpa/sfc/sfc_vdpa_ops.c | 24
> 1 file changed, 24 insertions(+)
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Implement vDPA ops get_queue_num to get the maximum number
> of queues supported by the device.
>
> Signed-off-by: Vijay Kumar Srivastava
> ---
> drivers/vdpa/sfc/sfc_vdpa_ops.c | 16 +---
> 1 file cha
On 7/6/21 6:44 PM, Vijay Srivastava wrote:
> From: Vijay Kumar Srivastava
>
> Implement vDPA ops get_feature and get_protocol_features.
> This patch retrieves device supported features and enables
> protocol features.
>
> Signed-off-by: Vijay Kumar Srivastava
> ---
> doc/guides/vdpadevs/fea
Hi,
I'm new to posting patches to DPDK repository. I've submitted a patch recently
and one of the tests performed on it failed. I went through the failed test log
and it looks like it does not have anything to do with my patch. You can check
details below. What should be the next steps I should
On Sat, Aug 28, 2021 at 7:46 PM Xueming(Steven) Li wrote:
>
>
>
> > -Original Message-
> > From: Jerin Jacob
> > Sent: Thursday, August 26, 2021 7:58 PM
> > To: Xueming(Steven) Li
> > Cc: dpdk-dev ; Ferruh Yigit ;
> > NBU-Contact-Thomas Monjalon ;
> > Andrew Rybchenko
> > Subject: Re:
1 - 100 of 119 matches
Mail list logo
