Hello Pawel,
See answers inline.
Regards
Sunil Kumar Kori
>-Original Message-
>From: David Marchand
>Sent: Wednesday, August 5, 2020 1:30 PM
>To: Pawel Wodkowski
>Cc: dev ; Jerin Jacob Kollanukkaran ;
>Sunil Kumar Kori
>Subject: [EXT] Re: [dpdk-dev] [PATCH] tracepoint: fix compilatio
Update release notes by adding new features for ice and iavf PMD.
Signed-off-by: Qi Zhang
Signed-off-by: Junfeng Guo
Signed-off-by: Jeff Guo
---
doc/guides/rel_notes/release_20_08.rst | 10 ++
1 file changed, 10 insertions(+)
diff --git a/doc/guides/rel_notes/release_20_08.rst
b/doc/
Hello Thomas,
Testing with dpdk 20.08-rc3 from Red Hat looks good.
We cover below 16 scenarios and and all get PASS on RHEL8 testing:
(1)Guest with device assignment(PF) throughput testing(1G hugepage size): PASS
(2)Guest with device assignment(PF) throughput testing(2M hugepage size) : PASS
(3)
Hi Thomas,
The commit message could be changed to "must".
In the HW offload, if the incoming packet match fails in the exact match
table, then the packet needs to use a default action. This default action
is being configured as part of this patch. The parif is a HW table and it
is being configured
Struct rte_eth_rxq_info will be modified to include a new field, indicating
the size of each buffer that could be used for hw to receive packets. Add
this field to rte_eth_rxq_info to expose relevant information to upper
layer users/application.
For more details:
https://mails.dpdk.org/archives/de
Struct rte_eth_rxq_info will be modified to include a new field, indicating
the size of each buffer that could be used for hw to receive packets. Add
this field to rte_eth_rxq_info to expose relevant information to upper
layer users/application.
For more details:
https://mails.dpdk.org/archives/de
vxlan_spec cannot be NULL since its already being accessed
before. Remove the redundant NULL check.
Signed-off-by: Gaurav Singh
---
drivers/net/bnxt/bnxt_flow.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 3
vxlan_spec cannot be NULL since its already being accessed
before. Remove the redundant NULL check.
Signed-off-by: Gaurav Singh
---
drivers/net/bnxt/bnxt_flow.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 3
vhost lib now does not have definition of reset status. This patch
adds the reset status definiton and changes related log.
Signed-off-by: Chenbo Xia
---
lib/librte_vhost/vhost.h | 1 +
lib/librte_vhost/vhost_user.c | 2 ++
2 files changed, 3 insertions(+)
diff --git a/lib/librte_vhost/vho
> Phil Yang writes:
>
>
>
> > > > > > > @@ -1790,9 +1792,9 @@ mlx5_rx_burst_mprq(void *dpdk_rxq,
> > > struct
> > > > > > > rte_mbuf **pkts, uint16_t pkts_n) void *buf_addr;
> > > > > > >
> > > > > > > /* Increment the refcnt of the whole chunk. */
> > > > > > > -rte_atomic16_add_return(&buf-
Update the release notes of mlx5 PMD for two new added devargs:
1. sys_mem_en
2. decap_en
Fixes: 5522da6b201b ("net/mlx5: add option to allocate memory from system")
Fixes: 50f95b23c9ba ("net/mlx5: add option to configure FCS or decapsulation")
Signed-off-by: Suanming Mou
---
doc/guides/rel_not
> -Original Message-
> From: Adrian Moreno
> Sent: Wednesday, August 5, 2020 10:45 PM
> To: dev@dpdk.org
> Cc: Wang, Zhihong ; Xia, Chenbo
> ; maxime.coque...@redhat.com; Adrian Moreno
>
> Subject: [PATCH v3 3/4] net/virtio: add GET_STATUS support to virtio-user
>
> This patch adds supp
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, August 6, 2020 4:18 AM
> To: Xia, Chenbo ; Wang, Yinan
> Cc: dev@dpdk.org; ajit.khapa...@broadcom.com; somnath.ko...@broadcom.com;
> kishore.padmana...@broadcom.com; venkatkumar.duvv...@broadcom.com;
> sta...@dpdk.
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v20.08-rc4
There are 27 new patches in this snapshot.
Release notes:
http://doc.dpdk.org/guides/rel_notes/release_20_08.html
More deprecation notices are pending for acceptance.
This is the last
On Wed, 5 Aug 2020 14:40:01 +0530
Jerin Jacob wrote:
> On Wed, Aug 5, 2020 at 2:16 PM Kinsella, Ray wrote:
> >
> >
> >
> > On 04/08/2020 17:20, Stephen Hemminger wrote:
> > > On Tue, 4 Aug 2020 11:41:53 +0100
> > > Bruce Richardson wrote:
> > >
> > >> On Mon, Aug 03, 2020 at 12:59:03PM +053
> -Original Message-
> From: Thomas Monjalon
> Sent: Wednesday, August 5, 2020 11:31 PM
> To: Suanming Mou
> Cc: John McNamara ; Marko Kovacevic
> ; dev@dpdk.org; Raslan Darawsheh
> ; Asaf Penso ; Gal Cohen
> (Product)
> Subject: Re: [dpdk-dev] [PATCH] doc: update release notes for ml
> > > A new API is planned to be introduced for sampling and mirroring
> > > with rte_flow. It should be more generic and allow more use cases.
> > >
> > > This deprecation is to show the direction, avoiding overlapping APIs.
> > >
> > > Signed-off-by: Thomas Monjalon
> Acked-by: Jerin Jacob
> >
> >>> The functions for L2 tunnel were missed when marking the legacy
> >>> filtering API as deprecated. That's why a separate notice is done
> >>> to make clear that it will be removed as well in DPDK 20.11.
> >>>
> >>> Signed-off-by: Thomas Monjalon
> >> Acked-by: Ajit Khaparde
> >> Acked-by: D
> >>> The flow director config, part of the legacy filtering API,
> >>> was marked as deprecated last year.
> >>> A separate notice is added to make clear that these specific structs
> >>> will be removed as well in DPDK 20.11, as the rest of the legacy
> >>> filtering API.
> >>>
> >>> Signed-off-b
> > > > Deprecation of rte_eth_dev_filter_ctrl() was announced in 2016,
> > > > and confirmed last year by avoiding include of rte_eth_ctrl.h.
> > > > After 4 years, it is time to complete the removal of the API
> > > > which is replaced with rte_flow.
> > > >
> > > > Signed-off-by: Thomas Monjalon
05/08/2020 12:33, Andrew Rybchenko:
> On 8/4/20 8:34 PM, Ajit Khaparde wrote:
> > On Tue, Aug 4, 2020 at 2:29 AM Hemant Agrawal
> > wrote:
> >
> >>> On Tue, Aug 4, 2020 at 2:50 PM Thomas Monjalon
> >>> wrote:
>
> Since DPDK 18.11, some drivers are migrating to a new behaviour,
> r
05/08/2020 11:28, Kinsella, Ray:
>
> On 31/07/2020 17:03, Thomas Monjalon wrote:
> > In order to prepare for adding more features requiring more space in mbuf,
> > some static fields must become dynamic.
> > Some small layout changes may have performance benefits as well.
> >
> > The deprecation
05/08/2020 11:36, Kinsella, Ray:
>
> On 02/07/2020 23:13, McDaniel, Timothy wrote:
> > From: "McDaniel, Timothy"
> >
> > Signed-off-by: McDaniel, Timothy
> > ---
> > doc/guides/rel_notes/deprecation.rst | 28
> > 1 file changed, 28 insertions(+)
> >
> > diff --g
> > > From: "McDaniel, Timothy"
> >
> > There is still "," in the name.
> >
> > >
> > > The ABI changes associated with this notification will better support
> > > devices that:
> > > 1. Have limits on the number or queues that may be linked to a port 2.
> > > Have ports that are limited to exac
25/06/2020 10:04, Maxime Coquelin:
> This patch makes rte_dev_probe() to return the
> rte_device pointer on success and NULL on error
> instead of returning 0 on success and negative
> value on error.
>
> The goal is to avoid that the calling application
> iterates the devices list afterwards to r
04/08/2020 13:20, Gaëtan Rivet:
> On 04/08/20 14:22 +0530, Jerin Jacob wrote:
> > On Tue, Aug 4, 2020 at 2:18 PM David Marchand
> > wrote:
> > >
> > > The PCI resources map API (pci_map_resource/pci_unmap_resource) was
> > > imposing use of Unix mmap flags while it does not make sense on Windows.
04/08/2020 16:44, Gaëtan Rivet:
> On 04/08/20 15:29 +0100, Bruce Richardson wrote:
> > On Tue, Aug 04, 2020 at 04:07:52PM +0200, David Marchand wrote:
> > > rte_kernel_driver is only used by the PCI subsystem and has polluted
> > > ethdev for no reason.
> > > Hide it in the PCI bus driver.
> > >
>
13/07/2020 14:31, Bruce Richardson:
> Put in deprecation notices in the 20.08 release for changes coming in 20.11.
>
> Bruce Richardson (2):
> doc: add deprecation notice for make build system removal
> doc: add deprecation notice for change of rawdev APIs
Applied, thanks
31/07/2020 19:23, Ajit Khaparde:
> From: Venkat Duvvuru
>
> While adding vfrep port to OVS bridge, vnic & svif information of
You should avoid justifying use cases with external applications.
> vfrep's endpoint(VF) would be needed to program default flow rules.
vfrep is not a well known word.
31/07/2020 19:22, Ajit Khaparde:
> From: Kishore Padmanabha
>
> The full offload egress flows for the VF rep interface should
> use loopback parif to offload missed flows.
>
> Fixes: fe82f3e02701 ("net/bnxt: support exact match templates")
"should" or "must"?
What was broken?
31/07/2020 19:23, Ajit Khaparde:
> From: Kishore Padmanabha
>
> The lookup default action record parif table is updated to catch
> the miss path for the entries in the exact match table.
>
> Fixes: fe82f3e02701 ("net/bnxt: support exact match templates")
It doesn't look like a fix.
What is "pa
31/07/2020 19:23, Ajit Khaparde:
> From: Somnath Kotur
>
> Set ulp_ctx explicitly to NULL in ulp_ctx_deinit() so that representor
> init is aborted if parent ulp context is not initialized.
> Also check for the same before creation of port default rules.
> Additional checks in VF rep dev ops for
04/08/2020 19:41, Chenbo Xia:
> Currenly the variable size of ethdev port id is 8 bits. This
> patch standarizes it to 16 bits.
>
> Fixes: 769de16872ab ("net/bnxt: fix port default rule create/destroy")
> Cc: sta...@dpdk.org
>
> Reported-by: Yinan Wang
> Signed-off-by: Chenbo Xia
> ---
> drive
31/07/2020 07:03, Ajit Khaparde:
> On Thu, Jul 30, 2020 at 7:35 PM Gaurav Singh wrote:
>
> > vxlan_mask cannot be NULL since its already being accessed
> > before. Remove the redundant NULL check.
Do you mean vxlan_spec?
> > Signed-off-by: Gaurav Singh
> >
> Acked-by: Ajit Khaparde
The title
03/08/2020 16:12, Somnath Kotur:
> OVS-DPDK seems to set the reset bit for every flow query. Honor
> the bit by resetting the SW counter values after assigning them.
> Also set the 'hit' bit only if the counter value retrieved by HW
> is non-zero.
> While querying flow stats, use max possible entri
03/08/2020 16:24, Ajit Khaparde:
> On Sun, Aug 2, 2020 at 11:51 PM Somnath Kotur
> wrote:
>
> > From: Shahaji Bhosle
> >
> > Adjusted resource allocations for the hardware resources
> > like TCAM entries, action records, stat counters, exact match records to
> > scale up offload flows.
> > Also
On Wed, 5 Aug 2020 17:46:21 +
"Trahe, Fiona" wrote:
> +1 for keeping at least the huge-page setup part - I use this all the time.
Huge page setup should be made its own script.
05/08/2020 18:57, Ferruh Yigit:
> On 8/5/2020 12:27 PM, Andrew Rybchenko wrote:
> > On 8/4/20 12:48 PM, Thomas Monjalon wrote:
> >> Some confusing comments were still present from old days,
> >> when most drivers were from Intel.
> >>
> >> Signed-off-by: Thomas Monjalon
> > Acked-by: Andrew Rybche
05/08/2020 04:41, Chen, Zhaoyan:
> From: ChenBo
>
> Update i40e PMD firmware/driver mapping table.
>
> Signed-off-by: ChenBo
Applied with this wording of name: Bo Chen
+1 for keeping at least the huge-page setup part - I use this all the time.
> -Original Message-
> From: dev On Behalf Of Thomas Monjalon
> Sent: Wednesday, August 5, 2020 5:27 PM
> To: Yigit, Ferruh
> Cc: dev@dpdk.org; david.march...@redhat.com; arybche...@solarflare.com;
> Richardson
04/08/2020 06:36, Zhang, Qi Z:
> From: Guo, Jia
> >
> > Different device has different hash capability, it should not be expected
> > that all
> > hash set would be successful to set into all devices by default. So remove
> > the
> > return checking when hash default set. And remove gtpu hash d
Hi, Tonghao
Thank you for the patch, and a lot of my apologizes - I mixed up your patch
with other
fix and thought we did not need it. Now I reviwed it again - we fixed
flow_dv_validate_item_meta()
(in other way - with explicit register check) but
flow_dv_validate_action_set_meta() still needs
05/08/2020 18:45, Bruce Richardson:
> On Wed, Aug 05, 2020 at 05:15:31PM +0200, Thomas Monjalon wrote:
> > 05/08/2020 17:07, Bruce Richardson:
> > > On Wed, Aug 05, 2020 at 04:57:42PM +0200, Thomas Monjalon wrote:
> > > > 05/08/2020 16:21, Bruce Richardson:
> > > > > The RTE_MACHINE_CPUFLAGS_* macr
On 8/5/2020 12:27 PM, Andrew Rybchenko wrote:
> On 8/4/20 12:48 PM, Thomas Monjalon wrote:
>> Some confusing comments were still present from old days,
>> when most drivers were from Intel.
>>
>> Signed-off-by: Thomas Monjalon
>
> Acked-by: Andrew Rybchenko
>
Acked-by: Ferruh Yigit
04/08/2020 12:53, Ferruh Yigit:
> On 8/1/2020 2:57 AM, Gaurav Singh wrote:
> > remove duplicate check
> >
> > Signed-off-by: Gaurav Singh
>
> Acked-by: Ferruh Yigit
Fixes: 62e94f7f66fb ("net/i40e: configure packet type mapping")
Cc: sta...@dpdk.org
Applied
On Wed, Aug 05, 2020 at 05:15:31PM +0200, Thomas Monjalon wrote:
> 05/08/2020 17:07, Bruce Richardson:
> > On Wed, Aug 05, 2020 at 04:57:42PM +0200, Thomas Monjalon wrote:
> > > 05/08/2020 16:21, Bruce Richardson:
> > > > The RTE_MACHINE_CPUFLAGS_* macros in DPDK build just duplicate info from
> >
> > The driver name was registered as "net_mlx5_regex".
> > It is renamed as "regex_mlx5".
> > The same name is used in mlx5_regex_driver.pci_driver.driver.name,
> > instead of "mlx5_regex", for consistency.
> >
> > The string used for log registration (pmd.regex.mlx5) could be derived
> > from th
On 7/29/2020 12:35 PM, wangyunjian wrote:
> From: Yunjian Wang
>
> When setup tx queues, we will create a mempool for the 'gso_ctx'.
> The mempool is not freed when closing tap device. If free the tap
> device and create it with different name, it will create a new
> mempool. This maybe cause an
> -Original Message-
> From: Thomas Monjalon
>
> The driver name was registered as "net_mlx5_regex".
> It is renamed as "regex_mlx5".
> The same name is used in mlx5_regex_driver.pci_driver.driver.name,
> instead of "mlx5_regex", for consistency.
>
> The string used for log registrati
05/08/2020 18:19, Ferruh Yigit:
> On 8/4/2020 11:26 AM, Thomas Monjalon wrote:
> > Environment configuration is the responsibility of distributions
> > or upper-level frameworks.
> > DPDK focus on documenting the requirements and some recommendations.
> >
> > Maintaining a good adaptative deployme
On 8/4/2020 11:26 AM, Thomas Monjalon wrote:
> Environment configuration is the responsibility of distributions
> or upper-level frameworks.
> DPDK focus on documenting the requirements and some recommendations.
>
> Maintaining a good adaptative deployment setup is a project by itself.
> Anyway th
02/08/2020 19:28, Stephen Hemminger:
> On Sat, 1 Aug 2020 23:04:28 -0700
> lon...@linuxonhyperv.com wrote:
>
> > From: Long Li
> >
> > The code should look into "slab" to figure out the index returned from
> > rte_bitmap_scan().
> >
> > Fixes: cc02518132 ("net/netvsc: split send buffers from T
Replace the arguments array by one argument.
All objects in the args array have the same values, so there is no need
to use an array, only one struct is enough.
The args object is a lot smaller, and the allocation can be replaced
with a stack variable.
The allocation of obj_table isn't needed eith
Remove the part that checks if there is enough room in the stack, it's
always true as long as size of stack >= MAX_BULK*rte_lcore_count().
This check used an atomic cmpset, and read / write to a shared size
variable. These operations result in some form of synchronization
that might get in the way
Use rte_eal_wait_lcore to wait and get the return value for all cores.
This is used to propagate any error to the main core.
Signed-off-by: Steven Lariau
Reviewed-by: Dharmik Thakkar
Reviewed-by: Phil Yang
Reviewed-by: Ruifeng Wang
---
app/test/test_stack.c | 10 --
1 file changed, 8
All the cores use the same argument object, so there is no need to use
a loop to launch the test on every core one by one.
Replace loop with one call to rte_eal_mp_remote_launch
Signed-off-by: Steven Lariau
Reviewed-by: Dharmik Thakkar
Reviewed-by: Phil Yang
Reviewed-by: Ruifeng Wang
---
app
The current multithread DPDK stack test is using atomics operations to
share information between threads.
The lockfree stack implementation also uses atomic operations.
This is an issue for testing. The atomics operations for the test may
add some extra synchronization to the stack implementation,
03/08/2020 04:23, Wei Hu (Xavier):
> Hi, Zhiguang He
>
> On 2020/8/2 20:27, 1913909...@qq.com wrote:
> > From: Zhiguang He <191390...@qq.com>
> >
> > when parament reta_size < RTE_RETA_GROUP_SIZE,
> > reta_count will be 0. Then this function will
> > be deadloop
> >
> > Fixes: 734ce47f71e0 ("bond
05/08/2020 17:26, Thomas Monjalon:
> From: Viacheslav Ovsiienko
>
> Add description about Tx scheduling timestamp upper limit.
> If timestamp exceed the value, it is marked by PMD as being
> into "too-distant-future" and not scheduled at all
> (is being sent without any wait).
>
> Signed-off-by:
19/07/2020 13:59, Suanming Mou:
> Update the release notes of mlx5 PMD for two new added devargs:
> 1. sys_mem_en
> 2. decap_en
That's a pity it was not sent as part of code implementation.
Please, at least, add the Fixes lines.
> Signed-off-by: Suanming Mou
> ---
> --- a/doc/guides/rel_notes/r
From: Viacheslav Ovsiienko
Add description about Tx scheduling timestamp upper limit.
If timestamp exceed the value, it is marked by PMD as being
into "too-distant-future" and not scheduled at all
(is being sent without any wait).
Signed-off-by: Viacheslav Ovsiienko
Signed-off-by: Thomas Monjal
On 05/08/2020 16:13, Nicolas Dichtel wrote:
Le 05/08/2020 à 16:53, Nick Connolly a écrit :
[snip]
+ if (check_numa()) {
+ ret = get_mempolicy(&cur_socket_id, NULL, 0, addr,
+ MPOL_F_NODE | MPOL_F_ADDR);
+ if (ret < 0) {
+ RTE_LOG(DEBUG, EAL, "%s(
This patch announces removal of RTE_CRYPTO_AUTH_AES_GMAC
from rte_crypto_auth_algorithm and addition of RTE_CRYPTO_AEAD_AES_GMAC
to rte_crypto_aead_algorithm.
AES-GMAC is variation of AES-GCM algorithm with the difference that
it does not perform encryption. As a matter of fact internally
there is
05/08/2020 17:07, Bruce Richardson:
> On Wed, Aug 05, 2020 at 04:57:42PM +0200, Thomas Monjalon wrote:
> > 05/08/2020 16:21, Bruce Richardson:
> > > The RTE_MACHINE_CPUFLAGS_* macros in DPDK build just duplicate info from
> > > the compiler macros, so we can remove them and just use the compiler
>
Le 05/08/2020 à 16:53, Nick Connolly a écrit :
[snip]
> + if (check_numa()) {
> + ret = get_mempolicy(&cur_socket_id, NULL, 0, addr,
> + MPOL_F_NODE | MPOL_F_ADDR);
> + if (ret < 0) {
> + RTE_LOG(DEBUG, EAL, "%s(): get_mempolicy: %s
On Wed, Aug 05, 2020 at 04:57:42PM +0200, Thomas Monjalon wrote:
> 05/08/2020 16:21, Bruce Richardson:
> > The RTE_MACHINE_CPUFLAGS_* macros in DPDK build just duplicate info from
> > the compiler macros, so we can remove them and just use the compiler
> > versions directly.
> >
> > Signed-off-by:
05/08/2020 16:21, Bruce Richardson:
> The RTE_MACHINE_CPUFLAGS_* macros in DPDK build just duplicate info from
> the compiler macros, so we can remove them and just use the compiler
> versions directly.
>
> Signed-off-by: Bruce Richardson
> ---
> --- a/doc/guides/rel_notes/deprecation.rst
> +++ b
I will send v2 soon then.
-Original Message-
From: Akhil Goyal
Sent: środa, 5 sierpnia 2020 16:40
To: Kusztal, ArkadiuszX ; dev@dpdk.org
Cc: ano...@marvell.com; Doherty, Declan ; Trahe,
Fiona ; asoma...@amd.com; rnagadhee...@marvell.com;
Hemant Agrawal ; De Lara Guarch, Pablo
; Zhang,
On 05/08/2020 15:36, Nicolas Dichtel wrote:
Le 05/08/2020 à 16:20, Nick Connolly a écrit :
[snip]
Fixes: 2a96c88be83e ("mem: ease init in a docker container")
I'm wondering if the bug existed before this commit.
Before this commit, it was:
move_pages(getpid(), 1, &addr, NULL, &cur_
The driver name was registered as "net_mlx5_regex".
It is renamed as "regex_mlx5".
The same name is used in mlx5_regex_driver.pci_driver.driver.name,
instead of "mlx5_regex", for consistency.
The string used for log registration (pmd.regex.mlx5) could be derived
from the driver name. A macro is cr
virtio 1.0 introduced a mechanism for the driver to verify that the
feature bits it sets are accepted by the device. This mechanism consists
in setting the VIRTIO_STATUS_FEATURE_OK status bit and re-reading it,
which gives a chance for the device to clear it if the the features
were not accepted.
For the sake of completeness, add the definition of the missing status
bit in accordance with the virtio spec
Reviewed-by: Chenbo Xia
Signed-off-by: Adrian Moreno
---
drivers/net/virtio/virtio_pci.h | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/net/vi
From: Maxime Coquelin
This patch adds support for VHOST_USER_SET_STATUS
request. It is used to make the backend aware of
Virtio devices status update.
It is useful for the backend to know when the Virtio
driver is done with the Virtio device configuration.
Reviewed-by: Chenbo Xia
Signed-off-by
Recently, two new messages have been added to the vhost-user protocol that
make the device initialization more robust.
VHOST_VIRTIO_SET_STATUS allows the driver to set the device status
VHOST_VIRTIO_GET_STATUS allows the driver to read the status back from
the device
This series implements these
This patch adds support for VHOST_USER_GET_STATUS request.
Only vhost-user backed is supported for now
Signed-off-by: Adrian Moreno
---
drivers/net/virtio/virtio_user/vhost_user.c | 2 +
.../net/virtio/virtio_user/virtio_user_dev.c | 44 +++
.../net/virtio/virtio_user/virtio
> Subject: RE: [PATCH] doc: announce move of aes gmac algorithm to aead
>
> > This patch announces removal of RTE_CRYPTO_AUTH_AES_GMAC
> > from rte_crypto_auth_algorithm and addition of
> > RTE_CRYPTO_AEAD_AES_GMAC
> > to rte_crypto_aead_algorithm.
> >
You should probably explain the reason of th
> This patch announces removal of RTE_CRYPTO_AUTH_AES_GMAC
> from rte_crypto_auth_algorithm and addition of
> RTE_CRYPTO_AEAD_AES_GMAC
> to rte_crypto_aead_algorithm.
>
> Signed-off-by: Arek Kusztal
> ---
Acked-by: Akhil Goyal
Le 05/08/2020 à 16:20, Nick Connolly a écrit :
[snip]
>>> Fixes: 2a96c88be83e ("mem: ease init in a docker container")
>> I'm wondering if the bug existed before this commit.
>>
>> Before this commit, it was:
>> move_pages(getpid(), 1, &addr, NULL, &cur_socket_id, 0);
>> if (cur_soc
05/08/2020 16:14, Thomas Monjalon:
> 03/08/2020 11:18, Dekel Peled:
> > Previous patch added definition of number of retries for UAR allocation.
> > This value is adequate for x86 systems with 4K pages.
> > On power9 system with 64K pages the required value is 32.
> > This patch updates the defined
> -Original Message-
> From: Andrew Rybchenko
> Sent: Wednesday, August 5, 2020 4:44 PM
> To: Dekel Peled ; dev@dpdk.org
> Cc: jerinjac...@gmail.com; step...@networkplumber.org;
> ajit.khapa...@broadcom.com; maxime.coque...@redhat.com;
> olivier.m...@6wind.com; david.march...@redhat.com;
This patch announces removal of RTE_CRYPTO_AUTH_AES_GMAC
from rte_crypto_auth_algorithm and addition of RTE_CRYPTO_AEAD_AES_GMAC
to rte_crypto_aead_algorithm.
Signed-off-by: Arek Kusztal
---
doc/guides/rel_notes/deprecation.rst | 4
1 file changed, 4 insertions(+)
diff --git a/doc/guides/r
On Wed, Aug 5, 2020 at 7:52 PM Bruce Richardson
wrote:
>
> The RTE_MACHINE_CPUFLAGS_* macros in DPDK build just duplicate info from
> the compiler macros, so we can remove them and just use the compiler
> versions directly.
>
> Signed-off-by: Bruce Richardson
Acked-by: Jerin Jacob
> ---
> d
The RTE_MACHINE_CPUFLAGS_* macros in DPDK build just duplicate info from
the compiler macros, so we can remove them and just use the compiler
versions directly.
Signed-off-by: Bruce Richardson
---
doc/guides/rel_notes/deprecation.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/g
I DPDK as part of our build on all architectures, we define a set of
RTE_MACHINE_CPUFLAG values which are then used in the code. However, as
best I can see, each of these macros just duplicates a shorter macro
provided by the compiler, e.g. RTE_MACHINE_CPUFLAG_SSE3 == __SSE3__,
RTE_MACHINE_CPUFLAG_
Hi Nicolas,
Thanks for the quick response.
On 05/08/2020 14:42, Nicolas Dichtel wrote:
Le 05/08/2020 à 14:26, Nick Connolly a écrit :
Running dpdk-helloworld on Linux with lib numa present,
but no kernel support for NUMA (CONFIG_NUMA=n) causes
ret_service_init() to fail with EAL: error allocat
03/08/2020 11:18, Dekel Peled:
> Previous patch added definition of number of retries for UAR allocation.
> This value is adequate for x86 systems with 4K pages.
> On power9 system with 64K pages the required value is 32.
> This patch updates the defined value from 2 to 32.
What about bigger pages
Hi Andrew,
> -Original Message-
> From: Andrew Rybchenko
>
> On 8/4/20 11:13 AM, Ori Kam wrote:
> > Using the rte_flow action RSS types field,
> > may result in undefined outcome.
> >
> > For example selecting both UDP and TCP,
> > selecting TCP RSS type but the pattern is targeting UDP
On 8/5/20 12:52 PM, Xia, Chenbo wrote:
> Hi Adrian,
>
>> -Original Message-
>> From: Adrian Moreno
>> Sent: Tuesday, August 4, 2020 3:24 PM
>> To: dev@dpdk.org
>> Cc: Wang, Zhihong ; Xia, Chenbo
>> ; maxime.coque...@redhat.com; Adrian Moreno
>>
>> Subject: [PATCH v2 1/4] net/virtio: a
On 8/5/20 12:55 PM, Xia, Chenbo wrote:
> Hi Adrian,
>
>> -Original Message-
>> From: Adrian Moreno
>> Sent: Tuesday, August 4, 2020 3:25 PM
>> To: dev@dpdk.org
>> Cc: Wang, Zhihong ; Xia, Chenbo
>> ; maxime.coque...@redhat.com; Adrian Moreno
>>
>> Subject: [PATCH v2 3/4] net/virtio: a
05/08/2020 15:34, Adrian Moreno:
> Hi Thomas,
>
> On 8/5/20 11:12 AM, Thomas Monjalon wrote:
> > 04/08/2020 09:38, Adrian Moreno:
> >> The PREFER_FALLTHROUGH check warns if a passthrough comment is found
> >> because, in the kernel, the special macro "fallthrough" is prefered.
> >>
> >> Since that
29/07/2020 13:35, wangyunjian:
> From: Yunjian Wang
>
> When setup tx queues, we will create a mempool for the 'gso_ctx'.
> The mempool is not freed when closing tap device. If free the tap
> device and create it with different name, it will create a new
> mempool. This maybe cause an OOM.
While
On 8/4/20 6:51 AM, Jerin Jacob wrote:
> On Mon, Aug 3, 2020 at 11:28 PM Ajit Khaparde
> wrote:
>>
>> On Mon, Aug 3, 2020 at 10:49 AM Thomas Monjalon wrote:
>>
>>> The flow director config, part of the legacy filtering API,
>>> was marked as deprecated last year.
>>> A separate notice is added to
On 8/5/20 4:31 PM, Dekel Peled wrote:
> Thanks, PSB.
>
>> -Original Message-
>> From: Andrew Rybchenko
>> Sent: Wednesday, August 5, 2020 1:49 PM
>> To: Dekel Peled ; dev@dpdk.org
>> Cc: jerinjac...@gmail.com; step...@networkplumber.org;
>> ajit.khapa...@broadcom.com; maxime.coque...@redh
Le 05/08/2020 à 14:26, Nick Connolly a écrit :
> Running dpdk-helloworld on Linux with lib numa present,
> but no kernel support for NUMA (CONFIG_NUMA=n) causes
> ret_service_init() to fail with EAL: error allocating
> rte services array.
>
> alloc_seg() calls get_mempolicy to verify that the allo
On 8/4/20 11:13 AM, Ori Kam wrote:
> Using the rte_flow action RSS types field,
> may result in undefined outcome.
>
> For example selecting both UDP and TCP,
> selecting TCP RSS type but the pattern is targeting UDP traffic.
> another option is that the PMD doesn't support all requested types.
>
Hi Thomas,
On 8/5/20 11:12 AM, Thomas Monjalon wrote:
> 04/08/2020 09:38, Adrian Moreno:
>> The PREFER_FALLTHROUGH check warns if a passthrough comment is found
>> because, in the kernel, the special macro "fallthrough" is prefered.
>>
>> Since that keyword is not defined in DPDK, ignore the warni
Thanks, PSB.
> -Original Message-
> From: Andrew Rybchenko
> Sent: Wednesday, August 5, 2020 1:49 PM
> To: Dekel Peled ; dev@dpdk.org
> Cc: jerinjac...@gmail.com; step...@networkplumber.org;
> ajit.khapa...@broadcom.com; maxime.coque...@redhat.com;
> olivier.m...@6wind.com; david.march...
> -Original Message-
> From: Burakov, Anatoly [mailto:anatoly.bura...@intel.com]
> Sent: Friday, July 31, 2020 7:55 PM
> To: wangyunjian ; dev@dpdk.org;
> david.march...@redhat.com
> Cc: Lilijun (Jerry) ; xudingke
> ; sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] eal/linux: do not cr
Hi All,
Can you please review and ack it? so it will be merged in 20.08 version.
Thanks,
Ori
> -Original Message-
> From: Ori Kam
>
> Using the rte_flow action RSS types field,
> may result in undefined outcome.
>
> For example selecting both UDP and TCP,
> selecting TCP RSS type but
Running dpdk-helloworld on Linux with lib numa present,
but no kernel support for NUMA (CONFIG_NUMA=n) causes
ret_service_init() to fail with EAL: error allocating
rte services array.
alloc_seg() calls get_mempolicy to verify that the allocation
has happened on the correct socket, but receives ENO
1 - 100 of 137 matches
Mail list logo
