Fix the vector mapping with queue by changing the recircle when
exceeds RX_VEC_START + nb_msix;
Signed-off-by: Jingjing Wu
---
drivers/net/iavf/iavf_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/iavf/iavf_ethdev.c b/drivers/net/iavf/iavf_ethdev.c
index
On Thu, Jan 7, 2021 at 2:42 AM Ashish Sadanandan
wrote:
>
> Hi Olivier,
>
> On Wed, Jan 6, 2021 at 6:21 AM Olivier Matz wrote:
> >
> > Hi Ashish,
> >
> > Yes, it should reference the patch that introduced the issue. In this case,
> > it should be:
> >
> > Fixes: 4958ca3a443a ("mbuf: support dyn
From: Alvin Zhang
The command here does not create a queue region, but only sets the
lookup table, so the descriptions in the doc is not exact.
Signed-off-by: Alvin Zhang
Fixes: feaae285b342 ("net/i40e: support hash configuration in RSS flow")
Cc: sta...@dpdk.org
---
V2: Divide the patch into
From: Alvin Zhang
The api should return the system error status, but it returned the
hardware error status, this is confused for the caller.
This patch adds check on hardware execution status and returns -EIO
in case of hardware execution failure.
Signed-off-by: Alvin Zhang
Fixes: 1d4b2b4966bb
From: Alvin Zhang
1. Delete original code.
2. Add 2 tables(One maps flow pattern and RSS type to PCTYPE,
another maps RSS type to input set).
3. Parse RSS pattern and RSS type to get PCTYPE.
4. Parse RSS action to get queues, RSS function and hash field.
5. Create and destroy RSS filters.
6. C
Currently, when use 'flow' command to create a rule that combine with
several RSS types, even the RSS type combination is invalid, it also
be created successfully.
Here list some invalid RSS combinations:
- ETH_RSS_IPV4 | ETH_RSS_NONFRAG_IPV4_TCP
- ETH_RSS_IPV6 | ETH_RSS_NONFRAG_IPV6_TCP
For 'E
This patchset aims to support AVX512 vPMD on i40e.
And the changes are only target to AVX512 vector path.
---
v2:
- Add return value check on rte_mempool_default_cache().
Leyi Rong (3):
net/i40e: remove devarg use-latest-supported-vec
net/i40e: add AVX512 vector path
net/i40e: optimize Tx b
Add AVX512 support for i40e PMD. This patch adds i40e_rxtx_vec_avx512.c
to support i40e AVX512 vPMD.
This patch aims to enable AVX512 on i40e vPMD. Main changes are focus
on Rx path compared with AVX2 vPMD.
Signed-off-by: Leyi Rong
Signed-off-by: Bruce Richardson
---
drivers/net/i40e/i40e_rxtx
Optimize Tx path by using AVX512 instructions and vectorize the
tx free bufs process.
Signed-off-by: Leyi Rong
Signed-off-by: Bruce Richardson
---
drivers/net/i40e/i40e_rxtx.c| 19 +++
drivers/net/i40e/i40e_rxtx.h| 4 +
drivers/net/i40e/i40e_rxtx_vec_avx512.c | 152 ++
As eal parameter --force-max-simd-bitwidth is already introduced,
to make it more clear when setting rx/tx function, remove
devarg use-latest-supported-vec support.
Signed-off-by: Leyi Rong
---
doc/guides/nics/i40e.rst | 9 ---
drivers/net/i40e/i40e_ethdev.c | 63 +--
dr
On Thu, Dec 17, 2020 at 11:19 PM Stephen Hemminger
wrote:
>
> On Thu, 17 Dec 2020 16:16:16 +0500
> Sarosh Arif wrote:
>
> > +if get_hugepages(path) != pages:
> > +print("Unable to reserve required pages. The pages reserved are:")
> > +global SHOW_HUGEPAGES
> > +SHOW_HU
The Geneve tunneling protocol is designed to allow the user to specify
some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to
present the user data.
In order to support GENEVE TLV Option the new rte_flow item
"rte_flow_item_geneve_opt" is introduce
The Geneve tunneling protocol is designed to allow the
user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option
is the mean intended to present the user data.
In order to support GENEVE TLV Option the new rte_flow
item "rte_flow_item_geneve_opt" is added.
The n
From: Viacheslav Ovsiienko
The patch adds the GENEVE option rte flow item support to
command line interpreter. The flow command with GENEVE
option items looks like:
flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is
This is preparation step to support match on GENEVE TLV option.
In this Patch we add the HCA attributes that will allow
supporting GENEVE TLV option matching.
Signed-off-by: Shiri Kuzin
Acked-by: Viacheslav Ovsiienko
---
drivers/common/mlx5/mlx5_devx_cmds.c | 7 +++
drivers/common/mlx5/ml
TLV object is a special firmware maintained entity used
to support match on GENEVE header extension option.
The TLV object is created with DevX API and accepts
the option class, type and lehgth fields.
The class type and length fields are set using MLX5_SET
and the Devx object is created using ml
Currently firmware supports the only TLV object per device
to match on the GENEVE header option.
This patch adds the simple TLV object management to the mlx5 PMD.
Signed-off-by: Shiri Kuzin
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.c | 2 +
drivers/net/mlx5/mlx5.h
This patch adds validation routine for the GENEVE
header TLV option.
The GENEVE TLV option match must include all fields
with full masks due to NIC does not support masking
on option class, type and length.
The option data length must be non zero and provided
data pattern should be zero neither d
The GENEVE TLV option matching flows must be created
using a translation function.
This function checks whether we already created a Devx
object for the matching and either creates the objects
or updates the reference counter.
Signed-off-by: Shiri Kuzin
Acked-by: Viacheslav Ovsiienko
---
drive
GENEVE TLV option support added to mlx5 PMD.
The limitations and support were updated in
documentation.
Signed-off-by: Shiri Kuzin
Acked-by: Viacheslav Ovsiienko
---
doc/guides/nics/mlx5.rst | 23 ++-
doc/guides/rel_notes/release_21_02.rst | 8
2 fil
This series introduces a new net virtual device called iavf_client which
is based on vfio-user client interface. Through vfio-user client interface,
PCI liked device could be used similar as Intel® Ethernet Adaptive Virtual
Function specification.
The code to enable iavf_client mainly contains two
This patch implements Emulated pci interfaces on vfio-user client
which is located in common/iavf/vfio_user/.
Four main functions provided to upper layer (driver) to set
up or talk to device:
- client_vfio_user_setup
- client_vfio_user_release
- client_vfio_user_get_bar_addr
- client_vfio_user_
This patch add a new net driver based on vdev abstraction,
i.e. iavf_client_ethdev.c. It is using common iavf functions
to talk with Emulated pci interfaces based on vfio-user.
--
| -- |
| | iavf driver | |> (iavf_client_ethdev.c)
New devarg "intr": if intr=1, use interrupt mode on control queue
Signed-off-by: Jingjing Wu
---
drivers/net/iavf/iavf.h | 19
drivers/net/iavf/iavf_client_ethdev.c | 131 ++
drivers/net/iavf/iavf_ethdev.c| 18 +---
3 files changed, 134 insert
Signed-off-by: Jingjing Wu
---
drivers/net/iavf/iavf_client_ethdev.c | 29 +--
drivers/net/iavf/iavf_ethdev.c| 9 +++--
2 files changed, 26 insertions(+), 12 deletions(-)
diff --git a/drivers/net/iavf/iavf_client_ethdev.c
b/drivers/net/iavf/iavf_client_ethde
Fix the vector mapping with queue by changing the recircle when
exceeds RX_VEC_START + nb_msix;
Fixes: d6bde6b5eae9 ("net/avf: enable Rx interrupt")
Signed-off-by: Jingjing Wu
---
drivers/net/iavf/iavf_ethdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/
Hi Beilei,
> -Original Message-
> From: Xing, Beilei
> Sent: Thursday, January 7, 2021 3:19 PM
> To: Xia, Chenbo ; dev@dpdk.org; tho...@monjalon.net;
> david.march...@redhat.com
> Cc: step...@networkplumber.org; Liang, Cunming ; Lu,
> Xiuchun ; Li, Miao ; Wu, Jingjing
>
> Subject: RE: [d
Currently, when use 'flow' command to create a rule with following invalid
RSS type combination, it can be created successfully.
Invalid RSS combinations list:
- ETH_RSS_IPV4 | ETH_RSS_NONFRAG_IPV4_TCP
- ETH_RSS_IPV6 | ETH_RSS_NONFRAG_IPV6_TCP
This patch adds these combinations in 'invalid_rss_
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, January 7, 2021 6:12 AM
> To: Guo, Jia
> Cc: Zhang, Qi Z ; Wu, Jingjing
> ; Yang, Qiming ; Wang,
> Haiyue ; dev@dpdk.org; Yigit, Ferruh
> ; andrew.rybche...@oktetlabs.ru
> Subject: Re: [dpdk-dev] [dpdk-dev v2 1/2] ethdev: ad
Acked-by: Jeff Guo
> -Original Message-
> From: Murphy Yang
> Sent: Thursday, January 7, 2021 5:17 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Guo, Jia ;
> Zhang, Qi Z ; Yang, SteveX ;
> Wu, Jingjing ; Xing, Beilei ;
> Yang, MurphyX
> Subject: [PATCH v8] net/iavf: fix invalid RSS combina
On 1/7/21 12:32 PM, Guo, Jia wrote:
-Original Message-
From: Thomas Monjalon
Sent: Thursday, January 7, 2021 6:12 AM
To: Guo, Jia
Cc: Zhang, Qi Z ; Wu, Jingjing
; Yang, Qiming ; Wang,
Haiyue ; dev@dpdk.org; Yigit, Ferruh
; andrew.rybche...@oktetlabs.ru
Subject: Re: [dpdk-dev] [dpdk-de
07/01/2021 10:32, Guo, Jia:
> From: Thomas Monjalon
> > 24/12/2020 07:59, Jeff Guo:
> > > Add type of RTE_TUNNEL_TYPE_ECPRI into the enum of ethdev tunnel
> > type.
> > >
> > > Signed-off-by: Jeff Guo
> > > Reviewed-by: Qi Zhang
> > [...]
> > > --- a/lib/librte_ethdev/rte_ethdev.h
> > > +++ b/li
On Wed, Jan 6, 2021 at 9:48 PM Xueming Li wrote:
>
> To support extended representor syntax, this patch refactor represntor
Typo in 'representor'
> infrastructure:
> 1. introduces representor type enum
> 2. devargs representor port range extraction from partial value
>
> Signed-off-by: Xueming Li
On 12/3/20 12:36 AM, Xueming Li wrote:
> For better performance and latency, this patch sets default event
> handling mode to polling mode which uses dedicate thread per device to
> poll and process event.
>
> Signed-off-by: Xueming Li
> Acked-by: Matan Azrad
> ---
> doc/guides/vdpadevs/mlx5
Cc'ing sta...@dpdk.org.
On 12/11/20 6:11 PM, Jiawei Zhu wrote:
> From: Jiawei Zhu
>
> When i < VIRTIO_MAX_VIRTQUEUES and j == i,
> dev->callfds[i] and dev->kickfds[i] are default 0.
> So it will close(0), close the standard input (stdin).
> And when the code fails in kickfd creation,
> it will l
On 1/5/21 3:27 PM, Maxime Coquelin wrote:
>
>
> On 1/5/21 3:16 PM, Maxime Coquelin wrote:
>>
>>
>> On 11/17/20 11:06 AM, Joyce Kong wrote:
>>> Optimize packed ring Rx batch path with NEON instructions.
>>>
>>> Signed-off-by: Joyce Kong
>>> Reviewed-by: Ruifeng Wang
>>> ---
>>> drivers/net/v
On Thu, Jan 7, 2021 at 12:08 PM Xueming(Steven) Li wrote:
>
>
>
> >-Original Message-
> >From: Somnath Kotur
> >Sent: Thursday, January 7, 2021 2:32 PM
> >To: Xueming(Steven) Li
> >Cc: NBU-Contact-Thomas Monjalon ; Ferruh Yigit
> >; Andrew Rybchenko
> >; Olivier Matz ;
> >Slava Ovsiienko
-Original Message-
From: Sebastian, Selwin
Sent: Wednesday, January 6, 2021 1:31 PM
To: dev@dpdk.org
Cc: Somalapuram, Amaranath ;
ferruh.yi...@intel.com
Subject: [PATCH v3] net/axgbe: add support for reading FW version
From: Selwin Sebastian
Added support for fw_version_get API
Si
On 12/25/20 9:28 AM, Jiayu Hu wrote:
> Async enqueue offloads large copies to DMA devices, and small copies
> are still performed by the CPU. However, it requires users to get
> enqueue completed packets by rte_vhost_poll_enqueue_completed(), even
> if they are completed by the CPU when rte_vhos
From: Akhil Goyal
the descriptor sharing needed to be changed for ZUC+ZUC as we
were getting invalid CHA combination error due to sharing
being done on DECOs simultaneously.
Signed-off-by: Akhil Goyal
---
drivers/common/dpaax/caamflib/desc/pdcp.h | 4 ++--
1 file changed, 2 insertions(+), 2 de
This patch adds test case for AES-XCBC hash only for
Digest and Digest-verify
Signed-off-by: Hemant Agrawal
---
app/test/test_cryptodev_hash_test_vectors.h | 41 +
1 file changed, 41 insertions(+)
diff --git a/app/test/test_cryptodev_hash_test_vectors.h
b/app/test/test_cryp
From: Akhil Goyal
This patch add support for AES-XCBC-MAC for following cases
- AES-XCBC-MAC auth only
- AES-CBC/CTR + AES-XCBC-MAC (non-proto)
- AES-CBC/CTR + AES-XCBC-MAC (protocol offload)
- DES-CBC + AES-XCBC-MAC (non-proto)
- 3DES-CBC + AES-XCBC-MAC (non-proto)
Signed-off-by: Barry Cao
Sig
This patch adds support for AES_CMAC integrity in non-security mode.
This patch modifies the camm flib to handles the AES CMAC
without conflicting the proto ALG operations. i.e. by creating
another ALG operation routine.
Signed-off-by: Hemant Agrawal
---
doc/guides/cryptodevs/dpaa2_sec.rst
if for some reason the queue is not close properly,
specially in test cases.
The QUEUE retire prints are flooding the screen.
They are not really required as WARNING.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa_sec/dpaa_sec.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
On 1/7/2021 5:39 AM, George Prekas wrote:
On 1/6/2021 12:02 PM, Ferruh Yigit wrote:
On 12/5/2020 5:42 AM, George Prekas wrote:
Strict-aliasing rules are violated by cast to uint16_t* in flowgen.c
and the calculated IP checksum is wrong on GCC 9 and GCC 10.
Signed-off-by: George Prekas
---
v
> Subject: Re: [dpdk-dev] [PATCH] mlx5: fix __mlx5_bit_off macro warning for
> Windows
>
> External email: Use caution opening links or attachments
>
>
> On Wed, 6 Jan 2021 15:42:21 +0200, Tal Shnaiderman wrote:
> > While compiling with clang 11 the callers of the __mlx5_bit_off macro
> > warns
While compiling with clang 11 the callers of the
__mlx5_bit_off macro warns on the cast of pointers to
unsigned long which is a smaller int type in Windows.
warning: cast to smaller integer type 'unsigned long'
from 'u8 (*)[16]' [-Wpointer-to-int-cast]
To resolve it the type is changed to uintptr
Add support for secondary processes in ioat devices. The update
allocates a memzone for a primary process or returns it in a
secondary process.
Signed-off-by: Kumar Amber
---
v5
* add error check for memzone lookup
v6
* fix compilation
v7
* include dev ops for secondary
---
drivers/raw/ioat/ioa
On Thu, Jan 07, 2021 at 05:53:12PM +0530, Kumar Amber wrote:
> Add support for secondary processes in ioat devices. The update
> allocates a memzone for a primary process or returns it in a
> secondary process.
>
> Signed-off-by: Kumar Amber
>
> ---
> v5
> * add error check for memzone lookup
>
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, January 7, 2021 6:12 PM
> To: Guo, Jia
> Cc: Zhang, Qi Z ; Wu, Jingjing ;
> Yang, Qiming ; Wang, Haiyue
> ; dev@dpdk.org; Yigit, Ferruh
> ; andrew.rybche...@oktetlabs.ru; or...@nvidia.com;
> getel...@nvidia.com
> Subject: R
From: Tal Shnaiderman
> While compiling with clang 11 the callers of the __mlx5_bit_off macro warns
> on the cast of pointers to unsigned long which is a smaller int type in
> Windows.
>
> warning: cast to smaller integer type 'unsigned long'
> from 'u8 (*)[16]' [-Wpointer-to-int-cast]
>
> To
On 1/7/2021 11:32 AM, Ferruh Yigit wrote:
On 1/7/2021 5:39 AM, George Prekas wrote:
On 1/6/2021 12:02 PM, Ferruh Yigit wrote:
On 12/5/2020 5:42 AM, George Prekas wrote:
Strict-aliasing rules are violated by cast to uint16_t* in flowgen.c
and the calculated IP checksum is wrong on GCC 9 and G
multi-threaded flows feature uses pthread function pthread_key_create
but for Windows the destruction option in the function is unimplemented.
to resolve it Windows will implement destruction mechanism to cleanup
mlx5_flow_workspace object for each terminated thread.
Linux flow will keep the curr
07/01/2021 13:47, Zhang, Qi Z:
>
> > -Original Message-
> > From: Thomas Monjalon
> > Sent: Thursday, January 7, 2021 6:12 PM
> > To: Guo, Jia
> > Cc: Zhang, Qi Z ; Wu, Jingjing
> > ;
> > Yang, Qiming ; Wang, Haiyue
> > ; dev@dpdk.org; Yigit, Ferruh
> > ; andrew.rybche...@oktetlabs.ru;
On Thu, Jan 7, 2021 at 2:33 PM Thomas Monjalon wrote:
> > Yes that may break the ABI but fortunately the checking-abi-compatibility
> > tool shows negative :) , thanks Ferruh' s guide.
> > https://github.com/ferruhy/dpdk/actions/runs/468859673
>
> That's very strange. An enum value is changed.
>
Latest VIC adapters support 64B CQ (completion queue) entries as well
as 16B entries available on all VIC models. 64B entries can greatly
reduce cache contention (CPU stall cycles) between DMA writes (Rx
packet descriptors) and polling CPU. The effect is very noticeable on
Intel platforms with DDIO
On 12/21/20 3:23 PM, Joyce Kong wrote:
> As desc_is_used has a load-acquire or rte_io_rmb inside
> and wait for used desc in virtqueue, it is ok to remove
> virtio_rmb behind it.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> drivers/net/virtio/virtio_ethdev.c | 6 +++---
>
> Tuesday, January 5, 2021 17:17, Thomas Monjalon
> > RTE Flows API lacks the ability to save an arbitrary header field in
> > order to use it later for advanced packet manipulations. Examples
> > include the usage of VxLAN ID after the packet is decapsulated or
> > storing this ID inside the pack
David Marchand writes:
> On Thu, Jan 7, 2021 at 2:33 PM Thomas Monjalon wrote:
>> > Yes that may break the ABI but fortunately the checking-abi-compatibility
>> > tool shows negative :) , thanks Ferruh' s guide.
>> > https://github.com/ferruhy/dpdk/actions/runs/468859673
>>
>> That's very stran
On 12/21/20 3:23 PM, Joyce Kong wrote:
> Replace rte_smp_wmb/rmb with rte_io_wmb/rmb as they are the same on x86
> and ppc platforms. Then, for function virtqueue_fetch_flags_packed/
> virtqueue_store_flags_packed, the if and else branch are still identical
> for the platforms except Arm.
>
> S
On Wed, 6 Jan 2021 22:35:53 +0200, Tal Shnaiderman wrote:
[...]
> +int
> +rte_thread_tls_key_create(rte_tls_key *key, void (*destructor)(void *))
> +{
> + int err;
> +
> + *key = malloc(sizeof(**key));
> + if ((*key) == NULL) {
> + RTE_LOG(DEBUG, EAL, "Cannot allocate TLS k
On 12/21/20 3:23 PM, Joyce Kong wrote:
> Relax the full write barriers to one-way barriers for virtio
> control path for Arm platform
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> drivers/net/virtio/virtio_ethdev.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(
26/11/2020 12:15, Wisam Jaddo:
> The clock() function is not good practice to use for multiple
> cores/threads, since it measures the CPU time used by the process
> and not the wall clock time, while when running through multiple
> cores/threads simultaneously, we can burn through CPU time much
> f
26/11/2020 12:15, Wisam Jaddo:
> After this series the application will start supporting testing
> multiple threaded insertion and deletion rates.
>
> Also it will provide the latency & throughput rates of all threads.
>
>
> Wisam Jaddo (4):
> app/flow-perf: refactor flows handler
> app/flow
07/01/2021 15:17, Alexander Kozyrev:
> > Tuesday, January 5, 2021 17:17, Thomas Monjalon
> > > RTE Flows API lacks the ability to save an arbitrary header field in
> > > order to use it later for advanced packet manipulations. Examples
> > > include the usage of VxLAN ID after the packet is decaps
> > > Thursday, January 7, 2021 10:07, Thomas Monjalon
> > > > RTE Flows API lacks the ability to save an arbitrary header field in
> > > > order to use it later for advanced packet manipulations. Examples
> > > > include the usage of VxLAN ID after the packet is decapsulated or
> > > > storing th
Hi
I would very much like a review for this patch.
Thank you,
Ori
> -Original Message-
> From: Ori Kam
> Sent: Thursday, December 17, 2020 12:38 PM
> Subject: [PATCH 1/2] regexdev: add resource limit reached rsp flag
>
> When scanning a buffer it is possible that the scan will abort
>
07/01/2021 16:10, Alexander Kozyrev:
> > > > Thursday, January 7, 2021 10:07, Thomas Monjalon
> > > > > RTE Flows API lacks the ability to save an arbitrary header field in
> > > > > order to use it later for advanced packet manipulations. Examples
> > > > > include the usage of VxLAN ID after the
> 07/01/2021 16:10, Alexander Kozyrev:
> > > > > Thursday, January 7, 2021 10:18, Thomas Monjalon
>
> > > > > > RTE Flows API lacks the ability to save an arbitrary header field in
> > > > > > order to use it later for advanced packet manipulations. Examples
> > > > > > include the usage of VxLAN
On 1/7/2021 2:20 PM, George Prekas wrote:
On 1/7/2021 5:32 AM, Ferruh Yigit wrote:
On 1/7/2021 5:39 AM, George Prekas wrote:
On 1/6/2021 12:02 PM, Ferruh Yigit wrote:
On 12/5/2020 5:42 AM, George Prekas wrote:
Strict-aliasing rules are violated by cast to uint16_t* in flowgen.c
and the calcul
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, January 7, 2021 9:34 PM
> To: Guo, Jia ; Zhang, Qi Z
> Cc: Wu, Jingjing ; Yang, Qiming
> ; Wang, Haiyue ;
> dev@dpdk.org; Yigit, Ferruh ;
> andrew.rybche...@oktetlabs.ru; or...@nvidia.com; getel...@nvidia.com;
> Dodji Seket
On 1/6/2021 8:54 PM, Lance Richardson wrote:
On Wed, Jan 6, 2021 at 4:27 AM Ferruh Yigit wrote:
On 12/14/2020 6:53 PM, Lance Richardson wrote:
Limit number of representor packets transferred per poll
to requested burst size.
Hi Lance,
Can you please describe the impact of the change? Sinc
On Thu, 7 Jan 2021 13:06:35 +0500
Sarosh Arif wrote:
> On Thu, Dec 17, 2020 at 11:19 PM Stephen Hemminger
> wrote:
> >
> > On Thu, 17 Dec 2020 16:16:16 +0500
> > Sarosh Arif wrote:
> >
> > > +if get_hugepages(path) != pages:
> > > +print("Unable to reserve required pages. The page
On Wed, 6 Jan 2021 23:39:39 -0600
George Prekas wrote:
> On 1/6/2021 12:02 PM, Ferruh Yigit wrote:
> > On 12/5/2020 5:42 AM, George Prekas wrote:
> >> Strict-aliasing rules are violated by cast to uint16_t* in flowgen.c
> >> and the calculated IP checksum is wrong on GCC 9 and GCC 10.
> >>
> >>
On 1/7/2021 3:50 PM, Stephen Hemminger wrote:
On Wed, 6 Jan 2021 23:39:39 -0600
George Prekas wrote:
On 1/6/2021 12:02 PM, Ferruh Yigit wrote:
On 12/5/2020 5:42 AM, George Prekas wrote:
Strict-aliasing rules are violated by cast to uint16_t* in flowgen.c
and the calculated IP checksum is wro
On 12/21/20 3:23 PM, Joyce Kong wrote:
> Replace the smp barriers with atomic thread fence for synchronization
> between different threads, if there are no load/store operations.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> drivers/net/virtio/virtqueue.h | 8
>
On Thu, 7 Jan 2021 15:59:59 +
Ferruh Yigit wrote:
> >>
> >> int main(void)
> >> {
> >>pkt_burst_flow_gen();
> >>return 0;
> >> }
> >
> > If I change your code like this to use union, Gcc 10 is still broken.
>
> This worked fine for me: https://godbolt.org/z/vdsxh9
I was looking
On 12/21/20 4:50 PM, Joyce Kong wrote:
> Use C11 atomic APIs with one-way barriers to replace two-way
> barriers when operating enqueue/dequeue. Used->idx and avail->idx
> are the synchronization points for split vring.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> exampl
On 12/21/20 4:50 PM, Joyce Kong wrote:
> Simply replace the rte_smp_mb barriers with SEQ_CST atomic thread fence,
> if there is no load/store operations.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> examples/vhost_blk/vhost_blk.c | 8
> 1 file changed, 4 inserti
On 12/21/20 4:50 PM, Joyce Kong wrote:
> As function desc_is_avail performs a load-acquire barrier to
> enforce the ordering between desc flags and desc content, it is
> unnecessary to add a rte_smp_rmb barrier around the trace which
> follows desc_is_avail.
>
> Signed-off-by: Joyce Kong
> Rev
On 12/21/20 4:50 PM, Joyce Kong wrote:
> The ordering between avail index and desc reads has been enforced
> by load-acquire for split vring, so smp_rmb barrier is not needed
> behind it.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> lib/librte_vhost/virtio_net.c | 7 ++--
On 12/21/20 4:50 PM, Joyce Kong wrote:
> Relax the full read barrier to one-way barrier for desc flags in
> packed vring.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> lib/librte_vhost/virtio_net.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
Reviewed-
On 12/21/20 4:50 PM, Joyce Kong wrote:
> Used idx can be synchronized by one-way barrier instead of full
> write barrier for split vring.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> lib/librte_vhost/vdpa.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
Rev
07/01/2021 16:22, Alexander Kozyrev:
> > 07/01/2021 16:10, Alexander Kozyrev:
> > > > > > Thursday, January 7, 2021 10:18, Thomas Monjalon
> >
> > > > > > > RTE Flows API lacks the ability to save an arbitrary header field
> > > > > > > in
> > > > > > > order to use it later for advanced packet m
This series adds a few fixes and improvements to the Nvidia RegEx PMD.
V2:
* Fix small issue in combined rule patch.
* Add new patch for number of queues.
Ori Kam (5):
regex/mlx5: fix memory rule alignment
regex/mlx5: add support for combined rule file
regex/mlx5: fix support for group id
The rof file holds programming instructions for
a given HW version.
In order to support future generation of HW it
was decided that the rof file will hold number
of rule configurations, and the driver will use
the one that matches the HW version.
In current code we force sync after each write bloc
In order to know which groups in the RegEx engine
should be used there is a need to check the req_flags.
This commit adds the missing check.
Cc: sta...@dpdk.org
Fixes: 4d4e245ad637 ("regex/mlx5: support enqueue")
Signed-off-by: Ori Kam
---
drivers/regex/mlx5/mlx5_regex_fastpath.c | 21 +++
Due to Kernel requirement the memory allocated must be aligned to 2M.
Fixes: b34d816363b5 ("regex/mlx5: support rules import")
Cc: sta...@dpdk.org
Signed-off-by: Ori Kam
---
drivers/regex/mlx5/mlx5_rxp.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/regex/mlx5/mlx
The high priority match request flags means that the
RegEx engine should stop on the first match.
This commit add this flag check to the RegEx engine.
Signed-off-by: Ori Kam
---
drivers/regex/mlx5/mlx5_regex_fastpath.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drive
The RegEx engine as no limitation on number of queues.
This commits modifies the max supported queues reported to the application.
Fixes: fbc8c7003b93 ("regex/mlx5: add completion queue creation")
Cc: sta...@dpdk.org
Signed-off-by: Ori Kam
---
drivers/regex/mlx5/mlx5_rxp.c | 3 +--
1 file chan
> 07/01/2021 16:22, Alexander Kozyrev:
> > > 07/01/2021 16:10, Alexander Kozyrev:
> > > > > > > Thursday, January 7, 2021 10:18, Thomas Monjalon
> > >
> > > > > > > > RTE Flows API lacks the ability to save an arbitrary header
> > > > > > > > field in
> > > > > > > > order to use it later for adv
07/01/2021 16:24, Zhang, Qi Z:
> From: Thomas Monjalon
> > 07/01/2021 13:47, Zhang, Qi Z:
> > > From: Thomas Monjalon
> > > > 07/01/2021 10:32, Guo, Jia:
> > > > > From: Thomas Monjalon
> > > > > > 24/12/2020 07:59, Jeff Guo:
> > > > > > > --- a/lib/librte_ethdev/rte_ethdev.h
> > > > > > > +++ b
07/01/2021 17:57, Alexander Kozyrev:
> > 07/01/2021 16:22, Alexander Kozyrev:
> > > > 07/01/2021 16:10, Alexander Kozyrev:
> > > > > > > > Thursday, January 7, 2021 10:18, Thomas Monjalon
> > > >
> > > > > > > > > RTE Flows API lacks the ability to save an arbitrary header
> > > > > > > > > field
On 12/21/20 4:50 PM, Joyce Kong wrote:
> Simply relace smp barriers with atomic thread fence for
> virtio packed vring.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> lib/librte_vhost/virtio_net.c | 9 +
> 1 file changed, 5 insertions(+), 4 deletions(-)
>
Reviewe
The header was missing the extern "C" directive which causes name
mangling of functions by C++ compilers, leading to linker errors
complaining of undefined references to these functions.
Fixes: 4958ca3a443a ("mbuf: support dynamic fields and flags")
Cc: olivier.m...@6wind.com
Cc: sta...@dpdk.org
On 12/21/20 4:50 PM, Joyce Kong wrote:
> Simply replace the smp barriers with atomic thread fence for vhost control
> path, if there are no synchronization points.
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> lib/librte_vhost/vhost.c | 18 +-
> lib/l
On 1/6/21 4:06 AM, Xueming Li wrote:
> This patch introduces new parameters for VirtQ CQ moderation, used for
> performance tuning.
>
> Signed-off-by: Xueming Li
> ---
> drivers/common/mlx5/mlx5_devx_cmds.c | 3 +++
> drivers/common/mlx5/mlx5_devx_cmds.h | 3 +++
> drivers/common/mlx5/mlx5_pr
Hi David,
On Thu, Jan 7, 2021 at 1:02 AM David Marchand
wrote:
> On Thu, Jan 7, 2021 at 2:42 AM Ashish Sadanandan
> wrote:
> >
> > Hi Olivier,
> >
> > On Wed, Jan 6, 2021 at 6:21 AM Olivier Matz
> wrote:
> > >
> > > Hi Ashish,
> > >
> > > Yes, it should reference the patch that introduced the
On 1/6/21 4:06 AM, Xueming Li wrote:
> The next parameters control the HW queue moderation feature.
> This feature helps to control the traffic performance and latency
> tradeoff.
>
> Each packet completion report from HW to SW requires CQ processing by SW
> and triggers interrupt for the guest
Meeting minutes of 7 January 2021
-
Agenda:
* Release Dates
* Subtrees
* LTS
* Opens
Participants:
* Arm
* Broadcom
* Debian/Microsoft
* Intel
* Marvell
* Nvidia
* NXP
* Red Hat
Release Dates
-
* Happy new year!
* v21.02 dates
* Proposal/V1 passe
1 - 100 of 148 matches
Mail list logo
