The current machine='default' build name is not descriptive. The actual
default build is machine='native'. Add an alternative string which does
the same build and better describes what we're building:
machine='generic'. Leave machine='default' for backwards compatibility.
Signed-off-by: Juraj Link
mlx5 pmd supports various rx_burst functions.
Each function is enabled differently and supports different features.
Add more comprehensive information about each function.
Signed-off-by: Asaf Penso
Acked-by: Viacheslav Ovsiienko
---
v4: update the fallback for mprq burst functions
v3: checkpat
The documentation file contains some broken links to Mellanox's site.
Also now BlueField-2 platform is supported.
This patch provides new links and adds documentation for
BlueField-2 platform.
Signed-off-by: Asaf Penso
---
v2: styling updates and typo
---
doc/guides/platform/bluefield.rst | 16
Remove the operation of optical modules in the link function, because
when device binds from dpdk to kernel, it will be linked failed. This
function does not take effect in cable mode, therefore, optimization
is required.
Fixes: 54ac33869932 ("net/hinic: set link down and up")
Cc: sta...@dpdk.org
The maximum number of queues on the chip can be any value greater
than 0, it does not have to be the power of two, if it does not be
fixed, dpdk initialization may be failed with OVS mode firmware.
Fixes: 828d3e15a9dc ("net/hinic/base: support context and work queue")
Cc: sta...@dpdk.org
Signed-of
The maximum number of queues on the chip can be any value greater
than 0, it does not have to be the power of two, if it does not be
fixed, dpdk initialization may be failed with OVS mode firmware
and remove the operation of optical modules in the link function, because
when device binds from dpdk
Hi Hemminger,
Latest "dpdk-devbind.py --status-dev" just can support
{net,baseband,crypto,event,mempool,compress} devices,
But can not support show Misc devices, can you fix it or provide new method to
show?
Notes: Old dpdk-devbind.py can support show cbdma device by cmd "
./usertools/dpdk-d
The jumbo frame used the 'RTE_ETHER_MAX_LEN' as boundary condition, this
fix will change the boundary condition with 'RTE_ETHER_MTU'.
When the MTU(1500) set, the frame type of rx packet will be different
if used different overhead, it will cause the consistency issue, and the
normal packet will be
Acked-by: Jeff Guo
> -Original Message-
> From: Xing, Beilei
> Sent: Friday, November 20, 2020 4:50 PM
> To: dev@dpdk.org
> Cc: Guo, Jia ; Xing, Beilei ;
> sta...@dpdk.org
> Subject: [PATCH v2] net/i40e: fix global register recovery
>
> From: Beilei Xing
>
> PMD configures the global
hi, igor ryzhov and Thomas
Since this remain issue is report recently and we need to reproduce the issue
and evaluate the patch and guaranty no side affect for other case,
so I am not sure even I don't think it still have time window to hit 20.11. But
whatever we have begin to check your patch f
On Thursday, November 19, 2020 11:53 AM, Ferruh Yigit wrote:
> Removing unused functions, reported by cppcheck.
>
> Easy way to remove clutter, since the code is already in the git repo,
> they can be added back when needed.
>
> Signed-off-by: Ferruh Yigit
> ---
> drivers/net/txgbe/base/txgbe_e
在 2020/11/24 0:41, Ferruh Yigit 写道:
On 11/23/2020 2:32 PM, Lijun Ou wrote:
Since the hns3 NIC hardware features are not counted
and it is supported in fact. Besides, the flow director
is not supported and need to delete it.
Hi Lijun,
Normally these should be updated when the feature is ad
When doing rte_eth_dev_get_supported_ptypes on netvsc device
the values reported are incorrect if VF is not present.
If no VF is present the device uses rte_net_get_ptype()
which can return a wide range of values. Use the same table
as TAP device in that case.
Signed-off-by: Stephen Hemminger
--
This code is not called anywhere.
Signed-off-by: Stephen Hemminger
---
drivers/net/netvsc/hn_rndis.c | 31 ---
drivers/net/netvsc/hn_rndis.h | 1 -
2 files changed, 32 deletions(-)
diff --git a/drivers/net/netvsc/hn_rndis.c b/drivers/net/netvsc/hn_rndis.c
index 1ce2
A couple of patches motivated by recent cppcheck patches
from Ferruh. The handling of supported_ptypes is incorrect
in netvsc PMD for non VF case.
Stephen Hemminger (2):
net/netvsc: remove unused code
net/netvsc: fix ethdev get_supported_ptypes
drivers/net/netvsc/hn_ethdev.c | 48 +++
> -Original Message-
> From: Matan Azrad
> Sent: Monday, November 23, 2020 17:57
> To: Slava Ovsiienko
> Cc: dev@dpdk.org; NBU-Contact-Thomas Monjalon
> Subject: [PATCH] net/mlx5: reduce log level in hlist registration
>
> In mlx5 internal hash list tool, there is a log print when an en
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix shared RSS action update
>
> The shared RSS action update was not operational due to lack of kernel driver
> support of TIR object modification.
> This commit introduces the workaround to support shared RSS action modify
> using an indirect queue table u
The documentation file contains some broken links to Mellanox's site.
Also now BlueField-2 platform is supported.
This patch provides new links and adds documentation for
BlueField-2 platform.
Signed-off-by: Asaf Penso
---
doc/guides/platform/bluefield.rst | 16
1 file changed,
On Mon, 23 Nov 2020 18:29:32 +
Honnappa Nagarahalli wrote:
>
>
> >
> > The initialization me->locked=1 in lock() must happen before
> > next->locked=0 in unlock(), otherwise a thread may hang forever,
> > waiting me->locked become 0. On weak memory systems (such as ARMv8),
> >
On Sun, Nov 22, 2020 at 12:15 PM Thomas Monjalon wrote:
>
> 20/11/2020 18:30, Ajit Khaparde:
> > On Fri, Nov 6, 2020 at 12:59 PM Ajit Khaparde
> > wrote:
> > >
> > >
> > >
> > > On Fri, Nov 6, 2020 at 11:22 AM Kevin Traynor wrote:
> > >>
> > >> On 03/11/2020 17:26, Ajit Khaparde wrote:
> > >> >
>
> The initialization me->locked=1 in lock() must happen before
> next->locked=0 in unlock(), otherwise a thread may hang forever,
> waiting me->locked become 0. On weak memory systems (such as ARMv8),
> the current implementation allows me->locked=1 to be reordered with
> a
>
> Has anyone investigated later developments in concurrency?
> While researching MCS Lock discovered this quote:
> https://mfukar.github.io/2017/09/26/mcs.html
> Luckily, we don’t have to worry about this very much. MCS locks
> right now are mostly a teaching tool, and have mostly
> -Original Message-
> From: David Marchand
> Sent: Monday, November 23, 2020 4:10 PM
> To: Dumitrescu, Cristian
> Cc: dev ; Thomas Monjalon
> Subject: Re: [PATCH V2] doc: add the SWX pipeline type to the prog_guide
>
> On Sun, Nov 15, 2020 at 4:58 PM David Marchand
> wrote:
> >
> >
Document the pipeline example in the Sample Application User Guide.
Signed-off-by: Cristian Dumitrescu
---
doc/guides/sample_app_ug/index.rst| 1 +
doc/guides/sample_app_ug/pipeline.rst | 118 ++
2 files changed, 119 insertions(+)
create mode 100644 doc/guides/samp
Document the pipeline example in the Sample Application User Guide.
Signed-off-by: Cristian Dumitrescu
---
doc/guides/sample_app_ug/index.rst| 1 +
doc/guides/sample_app_ug/pipeline.rst | 118 ++
2 files changed, 119 insertions(+)
create mode 100644 doc/guides/samp
Document the pipeline example in the Sample Application User Guide.
Signed-off-by: Cristian Dumitrescu
---
doc/guides/sample_app_ug/index.rst| 1 +
doc/guides/sample_app_ug/pipeline.rst | 118 ++
2 files changed, 119 insertions(+)
create mode 100755 doc/guides/samp
Document the pipeline example in the Sample Application User Guide.
Signed-off-by: Cristian Dumitrescu
---
doc/guides/sample_app_ug/index.rst| 1 +
doc/guides/sample_app_ug/pipeline.rst | 118 ++
2 files changed, 119 insertions(+)
create mode 100755 doc/guides/samp
On 11/19/2020 11:32 AM, Anatoly Burakov wrote:
Currently, the Linux GSG mentions UIO drivers first. This is not ideal
as for the longest time, the recommended way to use DPDK with hardware
devices has been to use VFIO driver.
This commit simply moves UIO section after VFIO, with minor edits.
Si
On 11/20/2020 1:22 AM, Xu, Rosen wrote:
Hi,
-Original Message-
From: Walsh, Conor
Sent: Thursday, November 19, 2020 23:00
To: Xu, Rosen ; Zhang, Tianfei
; Pei, Andy
Cc: dev@dpdk.org; sta...@dpdk.org; Chen, LingliX ;
Walsh, Conor
Subject: [PATCH] raw/ifpga/base: fix uninitialized var
On 11/20/2020 4:54 PM, Anatoly Burakov wrote:
Currently, compilation instructions for igb_uio are missing. Add them,
as well as a top-level INSTALL file referring users to per-OS
subdirectories.
patch title should start with 'linux/igb_uio:'
Signed-off-by: Anatoly Burakov
Reviewed-by: Fer
There is a problem being discussed in the Linux memory management list
that maybe of interest to DPDK developers. The DPDK/ VFIO assumes that
it can do DMA from hugepaged memory because it is pinned. But in reality
the hugepages are still in ZONE_MOVABLE.
You can follow the thread here:
https://m
On 11/18/2020 11:56 PM, Qi Zhang wrote:
Add link for firmware/OOT kernel driver/DDP download
Add matching List.
Signed-off-by: Qi Zhang
Acked-by: Ferruh Yigit
On 11/23/2020 2:32 PM, Lijun Ou wrote:
Since the hns3 NIC hardware features are not counted
and it is supported in fact. Besides, the flow director
is not supported and need to delete it.
Hi Lijun,
Normally these should be updated when the feature is added but it seems we have
missed them.
On 11/19/20 4:52 AM, Ferruh Yigit wrote:
> diff --git a/lib/librte_vhost/fd_man.c b/lib/librte_vhost/fd_man.c
> index 55d4856f9e..942c5f145b 100644
> --- a/lib/librte_vhost/fd_man.c
> +++ b/lib/librte_vhost/fd_man.c
> @@ -100,21 +100,6 @@ fdset_add_fd(struct fdset *pfdset, int idx, int fd,
>
On Sun, Nov 15, 2020 at 4:58 PM David Marchand
wrote:
>
> On Fri, Nov 13, 2020 at 4:17 PM Dumitrescu, Cristian
> wrote:
> > > -Original Message-
> > > From: David Marchand
> > > Sent: Friday, November 13, 2020 1:04 PM
> > > To: Dumitrescu, Cristian
> > > Cc: dev
> > > Subject: Re: [PAT
On 11/23/2020 1:12 PM, Guoyang Zhou wrote:
The maximum number of queues on the chip can be any value greater
than 0, it does not have to be the power of two, and remove the
operation of optical modules in the link function.
--
v1:
- remove the limitation for max queue num
- remove the oper
On Mon, 23 Nov 2020 09:56:51 +
Ferruh Yigit wrote:
> On 11/23/2020 2:55 AM, Zhang, Tianfei wrote:
> >> drivers/raw/ifpga/base/opae_eth_group.c | 25 -
> >> drivers/raw/ifpga/base/opae_eth_group.h |1 -
> >> drivers/raw/ifpga/base/opae_hw_api.c | 212 --
> >> d
Regards,
Asaf Penso
>-Original Message-
>From: dev On Behalf Of Matan Azrad
>Sent: Monday, November 23, 2020 5:57 PM
>To: Slava Ovsiienko
>Cc: dev@dpdk.org; NBU-Contact-Thomas Monjalon
>Subject: [dpdk-dev] [PATCH] net/mlx5: reduce log level in hlist registration
>
>In mlx5 internal h
In mlx5 internal hash list tool, there is a log print when an entry
allocation is failed: Can't allocate hash list entry.
Some initialization checks triggers hash list registration in order to
check some capabilities. Here, the failure in registration doesn't
lead to failure in the initialization
On 11/23/2020 2:25 PM, Andrew Rybchenko wrote:
On 11/23/20 5:17 PM, Ferruh Yigit wrote:
On 11/23/2020 1:50 PM, Andrew Rybchenko wrote:
On 11/23/20 4:40 PM, Ferruh Yigit wrote:
Proposing to replace protocol header fields in the ``rte_flow_item_*``
structures with the protocol structs, like:
Cu
Hi Ibtisam,
> -Original Message-
> From: dev On Behalf Of Ibtisam Tariq
> Sent: Wednesday, July 22, 2020 1:50 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; Ibtisam Tariq
> Subject: [dpdk-dev] [PATCH 2/2] examples/l2fwd-crypto: Free resources on
> exit
>
> When exiting the application, r
On Mon, 23 Nov 2020 15:06:06 +
Honnappa Nagarahalli wrote:
>
>
> > > >
> > > > 07/10/2020 11:55, Diogo Behrens:
> > > > > Hi Thomas,
> > > > >
> > > > > we are still waiting for the comments from Honnappa. In our
> > > > > understanding, the missing barrier is a bug according to the
> > >
> -Original Message-
> From: dev On Behalf Of Ferruh Yigit
> Sent: Thursday, November 19, 2020 11:59 AM
> To: dev@dpdk.org; Doherty, Declan ; Kerlin,
> Marcin ; Mrozowicz, SlawomirX
> ; Kobylinski, Michal
> ; Piotr Azarewicz
>
> Cc: Yigit, Ferruh ; sta...@dpdk.org
> Subject: [dpdk-dev] [v
> > >
> > > 07/10/2020 11:55, Diogo Behrens:
> > > > Hi Thomas,
> > > >
> > > > we are still waiting for the comments from Honnappa. In our
> > > > understanding, the missing barrier is a bug according to the
> > > > model. We reproduced the scenario in herd7, which represents the
> > > > authori
On 11/23/2020 12:46 PM, Bing Zhao wrote:
Thanks a lot~
-Original Message-
From: Viacheslav Ovsiienko
Sent: Monday, November 23, 2020 8:30 PM
To: dev@dpdk.org
Cc: NBU-Contact-Thomas Monjalon ; Rani Sharoni
; ferruh.yi...@intel.com; Bing Zhao
; sta...@dpdk.org
Subject: [PATCH] ethdev: fi
Since the hns3 NIC hardware features are not counted
and it is supported in fact. Besides, the flow director
is not supported and need to delete it.
Signed-off-by: Lijun Ou
---
doc/guides/nics/features/hns3.ini | 8
doc/guides/nics/hns3.rst | 3 ++-
2 files changed, 10 insertio
On 11/23/20 5:17 PM, Ferruh Yigit wrote:
> On 11/23/2020 1:50 PM, Andrew Rybchenko wrote:
>> On 11/23/20 4:40 PM, Ferruh Yigit wrote:
>>> Proposing to replace protocol header fields in the ``rte_flow_item_*``
>>> structures with the protocol structs, like:
>>>
>>> Current ``struct rte_flow_item_eth
On 11/23/2020 11:00 AM, Jiawei Wang wrote:
When testpmd enabled the verbosity for the received packets, if two packets
were received at the same time, for example, sampling packet and normal
packet, the dump output of these packets may be overlapping due to multiple
core handling the multiple que
On 11/23/2020 1:50 PM, Andrew Rybchenko wrote:
On 11/23/20 4:40 PM, Ferruh Yigit wrote:
Proposing to replace protocol header fields in the ``rte_flow_item_*``
structures with the protocol structs, like:
Current ``struct rte_flow_item_eth``,
struct rte_flow_item_eth {
struct rte_ether_a
On 11/23/20 4:40 PM, Ferruh Yigit wrote:
> Proposing to replace protocol header fields in the ``rte_flow_item_*``
> structures with the protocol structs, like:
>
> Current ``struct rte_flow_item_eth``,
>
> struct rte_flow_item_eth {
> struct rte_ether_addr dst;
> struct rte_ether_addr
Proposing to replace protocol header fields in the ``rte_flow_item_*``
structures with the protocol structs, like:
Current ``struct rte_flow_item_eth``,
struct rte_flow_item_eth {
struct rte_ether_addr dst;
struct rte_ether_addr src;
rte_be16_t type;
uint32_t has_v
In the link up and link down functions, they do not need
to operate the optical module.
Fixes: 511b7371d32b ("net/hinic: fix hugepage memory leaks")
Cc: sta...@dpdk.org
Signed-off-by: Guoyang Zhou
---
drivers/net/hinic/base/hinic_pmd_niccfg.c | 38 ---
drivers/net/hin
The maximum number of queues on the chip can be any value
greater than 0, it does not have to be the power of two.
Fixes: 511b7371d32b ("net/hinic: fix hugepage memory leaks")
Cc: sta...@dpdk.org
Signed-off-by: Guoyang Zhou
---
drivers/net/hinic/base/hinic_pmd_nicio.c | 5 -
1 file changed,
The maximum number of queues on the chip can be any value greater
than 0, it does not have to be the power of two, and remove the
operation of optical modules in the link function.
--
v1:
- remove the limitation for max queue num
- remove the operation of optical modules in the link function
Thanks a lot~
> -Original Message-
> From: Viacheslav Ovsiienko
> Sent: Monday, November 23, 2020 8:30 PM
> To: dev@dpdk.org
> Cc: NBU-Contact-Thomas Monjalon ; Rani Sharoni
> ; ferruh.yi...@intel.com; Bing Zhao
> ; sta...@dpdk.org
> Subject: [PATCH] ethdev: fix eCPRI header generic data
There was a typo in eCPRI header definition.
Fixes: d164c609e70b ("ethdev: add eCPRI key fields to flow API")
Cc: sta...@dpdk.org
Reported-by: Rani Sharoni
Signed-off-by: Viacheslav Ovsiienko
---
lib/librte_net/rte_ecpri.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib
Hi, Wei
It was found this patch rejects the --txpkts command line settings.
set_tx_pkt_segments() is called before device started and
we have failure chain:
set_tx_pkt_segments()
nb_segs_is_invalid()
get_tx_ring_size ()
rte_eth_tx_queue_info_get()
It causes --txpkts testpmd command l
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Monday, November 23, 2020 11:00 AM
>
> 23/11/2020 10:30, Morten Brørup:
> > Bruce,
> >
> > Here's my input as a developer of hardware appliances. It is my
> opinion, and as such may contradict the trend towards making DPDK a
> library, r
Hi Ferruh,
> -Original Message-
> From: Ferruh Yigit
> Sent: Monday, November 23, 2020 6:23 PM
> To: Jiawei(Jonny) Wang ; wenzhuo...@intel.com;
> beilei.x...@intel.com; bernard.iremon...@intel.com; Ori Kam
> ; Slava Ovsiienko ; NBU-
> Contact-Thomas Monjalon ; Raslan Darawsheh
>
> Cc: de
When testpmd enabled the verbosity for the received packets, if two packets
were received at the same time, for example, sampling packet and normal
packet, the dump output of these packets may be overlapping due to multiple
core handling the multiple queues simultaneously.
The patch uses one strin
On 11/16/2020 7:55 AM, Xiaoyu Min wrote:
From: Xiaoyu Min
The rte_flow_item_vlan items are refined.
The structs do not exactly represent the packet bits captured on the
wire anymore so should only copy real header instead of the whole struct.
Replace the rte_flow_item_* with the existing corre
On 11/23/2020 6:13 AM, Jiawei(Jonny) Wang wrote:
Hi Ferruh,
-Original Message-
From: Ferruh Yigit
Sent: Saturday, November 21, 2020 1:50 AM
To: Jiawei(Jonny) Wang ; wenzhuo...@intel.com;
beilei.x...@intel.com; bernard.iremon...@intel.com; Ori Kam
; Slava Ovsiienko ; NBU-
Contact-Thomas
On 11/22/2020 2:15 PM, Thomas Monjalon wrote:
22/11/2020 14:28, Jack Min:
From: Ferruh Yigit
On 11/16/2020 7:55 AM, Xiaoyu Min wrote:
From: Xiaoyu Min
The rte_flow_item_vlan items are refined.
The structs do not exactly represent the packet bits captured on the
wire anymore so should only c
23/11/2020 10:30, Morten Brørup:
> Bruce,
>
> Here's my input as a developer of hardware appliances. It is my opinion, and
> as such may contradict the trend towards making DPDK a library, rather than a
> development kit.
>
> > DPDK build configuration - future enhancements
> >
On 11/23/2020 2:55 AM, Zhang, Tianfei wrote:
drivers/raw/ifpga/base/opae_eth_group.c | 25 -
drivers/raw/ifpga/base/opae_eth_group.h |1 -
drivers/raw/ifpga/base/opae_hw_api.c | 212 --
drivers/raw/ifpga/base/opae_hw_api.h | 36 -
drivers/raw/ifpga/bas
On 11/21/2020 4:29 AM, Stephen Hemminger wrote:
On Fri, 20 Nov 2020 23:33:40 +
Ferruh Yigit wrote:
On 11/20/2020 11:21 PM, Stephen Hemminger wrote:
On Fri, 20 Nov 2020 17:26:55 +
Ferruh Yigit wrote:
On 11/20/2020 11:50 AM, Min Hu (Connor) wrote:
From: Huisong Li
Currently, th
Bruce,
Here's my input as a developer of hardware appliances. It is my opinion, and as
such may contradict the trend towards making DPDK a library, rather than a
development kit.
> DPDK build configuration - future enhancements
>
> There are multiple req
Tested-by: Xie,WeiX < weix@intel.com>
Regards,
Xie Wei
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Murphy Yang
> Sent: Monday, November 23, 2020 4:35 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Zhang, Qi Z
> ; Yang, SteveX ; Rong, Leyi
> ; Lu, Wenzhuo
When received tunneled packets, the testpmd output log shows 'ol_flags'
value always is 'PKT_RX_OUTER_L4_CKSUM_UNKNOWN', but expected value is
'PKT_RX_OUTER_L4_CKSUM_GOOD' or 'PKT_RX_OUTER_L4_CKSUM_BAD'.
Add the 'PKT_RX_OUTER_L4_CKSUM_GOOD' and 'PKT_RX_OUTER_L4_CKSUM_BAD' to
'flags' for normal pat
Tested-by: Xie,WeiX < weix@intel.com>
Regards,
Xie Wei
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Murphy Yang
> Sent: Monday, November 23, 2020 3:05 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Zhang, Qi Z
> ; Yang, SteveX ; Yang,
> MurphyX
> Subject:
When testpmd enabled the verbosity for the received packets, if two packets
were received at the same time, for example, sampling packet and normal
packet, the dump output of these packets may be overlapping due to multiple
core handling the multiple queues simultaneously.
The patch uses one strin
> -Original Message-
> From: Asaf Penso
> Sent: Monday, November 23, 2020 10:05
> To: dev@dpdk.org
> Cc: NBU-Contact-Thomas Monjalon ; Slava Ovsiienko
> ; Matan Azrad ; Raslan
> Darawsheh
> Subject: [PATCH v3] doc: add information for Rx burst functions in mlx5
>
> mlx5 pmd supports vari
23/11/2020 08:54, Ori Kam:
> From: Matan Azrad
> > From: Thomas Monjalon
> > > 16/11/2020 08:55, Xiaoyu Min:
> > > > From: Xiaoyu Min
> > > >
> > > > The rte_flow_item_eth and rte_flow_item_vlan items are refined.
> > > > The structs do not exactly represent the packet bits captured on the
> > >
mlx5 pmd supports various rx_burst functions.
Each function is enabled differently and supports different features.
Add more comprehensive information about each function.
Signed-off-by: Asaf Penso
---
v3: checkpatch cleanup
v2: for mprq added note for rxqs_min_mprq
---
doc/guides/nics/mlx5.rst
74 matches
Mail list logo