> -Original Message-
> From: Zhang, Qi Z
> Sent: Friday, September 28, 2018 11:20 AM
> To: Xing, Beilei
> Cc: dev@dpdk.org; Yigit, Ferruh ; Wu, Jingjing
> ; Rawat, Anand ; Menon,
> Ranjit ; Zhang, Qi Z ;
> sta...@dpdk.org
> Subject: [PATCH] net/i40e: fix send admin queue command before
> -Original Message-
> From: Gagandeep Singh
> Sent: Friday, September 28, 2018 10:47 AM
> To: dev@dpdk.org; ferruh.yi...@intel.com
> Cc: Pankaj Chauhan ; Gagandeep Singh
>
> Subject: [PATCH v3 0/3] introduces the enetc PMD driver
>
> *ENETC* PMD driver which integrates with the existi
Add enetc usage document to compile and run the
DPDK application on enetc supported platform.
This document introduces the enetc driver, supported
platforms and supported features.
Signed-off-by: Gagandeep Singh
---
MAINTAINERS| 7 ++
doc/guides/nics/enetc.rst
Signed-off-by: Gagandeep Singh
---
drivers/net/enetc/enetc_ethdev.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/drivers/net/enetc/enetc_ethdev.c b/drivers/net/enetc/enetc_ethdev.c
index 2d90d8fd5..3e1c934cb 100644
--- a/drivers/net/enetc/enetc_ethdev.c
+++ b/drivers/n
*ENETC* PMD driver which integrates with the existing PCI bus.
Document is also part of the set
v2->v3 Change-log:
* Added a release note
* commom part of documentation is removed
* cflag Werror added in Makefile
* private list of RX/TX queues removed
* added the dev_start and dev_stop APIs
* adde
Add RX and TX queue setup, datapath functions
Signed-off-by: Gagandeep Singh
---
drivers/net/enetc/Makefile| 3 +-
drivers/net/enetc/base/enetc_hw.h | 19 +-
drivers/net/enetc/enetc.h | 25 ++-
drivers/net/enetc/enetc_ethdev.c | 326 +-
drivers/ne
The vdpa sample application creates vhost-user sockets by using the
vDPA backend. vDPA stands for vhost Data Path Acceleration which utilizes
virtio ring compatible devices to serve virtio driver directly to enable
datapath acceleration. As vDPA driver can help to set up vhost datapath,
this applic
It's used to get number of available registered vDPA devices.
Reviewed-by: Maxime coquelin
Acked-by: Xiao Wang
Signed-off-by: Xiaolong Ye
---
lib/librte_vhost/rte_vdpa.h| 3 +++
lib/librte_vhost/rte_vhost_version.map | 1 +
lib/librte_vhost/vdpa.c| 6 ++
3 files
Hi,
This patchset introduces vdpa sample to demonstrate the vDPA use case.
v7 changes:
* make vports static
* avoid unnecessary static variable initialization
v6 changes:
* improve the document according to Xiao's comments
* fix a typo, PRIu64 -> PRIx64
v5 changes:
* improve print format and co
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_dev_release_port_second
This patch cover the multi-process hotplug case when a device
attach/detach request be issued from a secondary process
device attach on secondary:
a) secondary send sync request to the primary.
b) primary receive the request and attach the new device if
failed goto i).
c) primary forward attach
The sample code demonstrates device (ethdev only) management
at a multi-process environment. The user can attach/detach a
device on primary process and see it is synced on secondary
process automatically.
How to start?
./hotplug_mp --proc-type=auto
Command Line Example:
>help
>list
/* attach a
With the enabling for hotplug on multi-process,
rte_eth_dev_pci_generic_remove can be used to detach the device from
a secondary process also. But we need to take care of the uninit callback
parameter to make sure it handles the secondary case correctly.
Signed-off-by: Qi Zhang
---
drivers/net/b
We are going to introduce the solution to handle hotplug in
multi-process, it includes the below scenario:
1. Attach a device from the primary
2. Detach a device from the primary
3. Attach a device from a secondary
4. Detach a device from a secondary
In the primary-secondary process model, we ass
Add driver API rte_eth_release_port_secondary to support the
case when an ethdev need to be detached on a secondary process.
Local state is set to unused and shared data will not be reset
so the primary process can still use it.
Signed-off-by: Qi Zhang
---
lib/librte_ethdev/rte_ethdev.c
v16:
- rebase to patch "simplify parameters of hotplug functions"
http://patchwork.dpdk.org/patch/45463/ include:
* keep rte_eal_hotplug_add/rte_eal_hotplug_move unchanged.
* the IPC sync logic is moved to rte_dev_probe/rte_dev_remove.
* simplify the IPC message by removing busname and dev
Some users want to use their own epoll instances to control both
DPDK rxq interrupt fds and their own other fds. So added a function
to get rxq interrupt fd based on port id and queue id.
Signed-off-by: Xiaoyun Li
---
lib/librte_ethdev/rte_ethdev.c | 37 ++
lib/li
We should not issue Admin Queue command before Admin Queue is initialized.
But this happened in i40e_hw_init and i40e_filter_input_set_init.
The patch fixes the issue by proper reordering.
Fixes: b6a0ec418274 ("i40e: use AQ for Rx control register read/write")
Cc: sta...@dpdk.org
Reported-by: Ra
Add counter action support
Signed-off-by: John Daley
---
doc/guides/nics/enic.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/nics/enic.rst b/doc/guides/nics/enic.rst
index 438a83d5f..86941fdb2 100644
--- a/doc/guides/nics/enic.rst
+++ b/doc/guides/nics/enic.rst
@@ -260,
Support counter action for 1400 series adapters.
The adapter API for allocating and freeing counters is independent of
the adapter match/action API. If the filter action is requested, a
counter is first allocated and then assigned to the filter, and when
the filter is deleted, the counter must als
rte_flow structures were not being freed when destroyed or flushed.
Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters
enabled")
Cc: sta...@dpdk.org
Signed-off-by: Hyong Youb Kim
Signed-off-by: John Daley
---
v2: fix signoff
drivers/net/enic/enic_flow.c | 2 ++
1 file ch
From: Hyong Youb Kim
Move a number of Rx functions to the header file so that the avx2
based Rx handler can use them.
Signed-off-by: Hyong Youb Kim
Reviewed-by: John Daley
---
drivers/net/enic/enic_rxtx.c| 263 +-
drivers/net/enic/enic_rxtx_common.h | 2
From: Hyong Youb Kim
Add the vectorized version of the no-scatter Rx handler. It aims to
process 8 descriptors per loop using AVX2 SIMD instructions. This
handler is in its own file enic_rxtx_vec_avx2.c, and makefile and
meson.build are modified to compile it when the compiler supports
AVX2. Unde
Add counter action support
Signed-off-by: John Daley
---
doc/guides/nics/enic.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/nics/enic.rst b/doc/guides/nics/enic.rst
index 438a83d5f..86941fdb2 100644
--- a/doc/guides/nics/enic.rst
+++ b/doc/guides/nics/enic.rst
@@ -260,
Support counter action for 1400 series adapters.
The adapter API for allocating and freeing counters is independent of
the adapter match/action API. If the filter action is requested, a
counter is first allocated and then assigned to the filter, and when
the filter is deleted, the counter must als
rte_flow structures were not being freed when destroyed or flushed.
Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters
enabled")
Cc: sta...@dpdk.org
Signed-off-by: Hyong Youb Kim
Reviewed-by: John Daley
---
drivers/net/enic/enic_flow.c | 2 ++
1 file changed, 2 insertions
Reply inlined:
>-Original Message-
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Honnappa Nagarahalli
>Sent: Thursday, September 6, 2018 10:12 AM
>To: Richardson, Bruce ; De Lara Guarch, Pablo
>
>Cc: dev@dpdk.org; honnappa.nagaraha...@dpdk.org; gavin...@arm.com;
>steve.cap...@arm
Reply inlined:
>-Original Message-
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Honnappa Nagarahalli
>Sent: Thursday, September 6, 2018 10:12 AM
>To: Richardson, Bruce ; De Lara Guarch, Pablo
>
>Cc: dev@dpdk.org; honnappa.nagaraha...@dpdk.org; gavin...@arm.com;
>steve.cap...@arm
On 09/27, Maxime Coquelin wrote:
[snip]
>> +devcnt = 0;
>This isn't needed as it is static.
>
>> +interactive = 0;
>Ditto
>
>> +memset(iface, 0, sizeof(iface));
>Ditto
>
>> +memset(vports, 0, sizeof(vports));
>And this one too when it will be made static.
>
>
>Appart these small com
Some general comments for the various __atomic_store/load added,
1. Although it passes the compiler check, but I just want to confirm that if we
should use GCC/clang builtins, or if
There are higher level APIs in DPDK to do atomic operations?
2. We believe compiler will translate the atomic_st
> On Sep 19, 2018, at 1:27 AM, Xueming Li wrote:
>
> Current code uses PF links status for representor port, not the
> representor
> interface itself. This caused wrong representor port link status when
> toggling linterface up or down.
>
> Fixes: 2b7302638898 ("net/mlx5: probe all port repres
> On Sep 19, 2018, at 12:55 AM, Xueming Li wrote:
>
> This patch fixes the issue that representor port shows xstats of PF.
>
> Fixes: 2b7302638898 ("net/mlx5: probe all port representors")
> Cc: adrien.mazarg...@6wind.com
>
> Signed-off-by: Xueming Li
> ---
Acked-by: Yongseok Koh
Thanks
>-Original Message-
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Honnappa Nagarahalli
>Sent: Thursday, September 6, 2018 10:12 AM
>To: Richardson, Bruce ; De Lara Guarch, Pablo
>
>Cc: dev@dpdk.org; honnappa.nagaraha...@dpdk.org; gavin...@arm.com;
>steve.cap...@arm.com; ola.liljed
On Thu, Sep 27, 2018 at 8:44 PM, Ferruh Yigit wrote:
>> Well, yes the link_status (link up, link down) _is_ applied to the KNI
>> interface. When that occurs, most people want to know what the link
>> speed is that the link came up at.
>
> +1 to this, people would like to know link speed of the i
Hi Honnappa,
Reply inlined:
>-Original Message-
>
>Currently, reader-writer concurrency problems in rte_hash are
>addressed using reader-writer locks. Use of reader-writer locks
>results in following issues:
>
>1) In many of the use cases for the hash table, writer threads
On 9/28/2018 12:05 AM, Dan Gora wrote:
> On Thu, Sep 27, 2018 at 6:49 PM, Ferruh Yigit wrote:
It would be useful if it writes the values of virtual interface, but this
API
prints user input.
>>> I'm sorry, Ferruh, I really don't understand what you are referring to
>>> here.
> On Aug 20, 2018, at 6:34 AM, viacheslavo wrote:
>
> Mellanox mlx5 PMD supports Flow Counters via Verbs library.
> The current implementation is based on the Mellanox proprietary
> Verbs library included in MLNX OFED packages. The Flow Counter
> support is recently added into linux-rdma release
On Thu, Sep 27, 2018 at 6:49 PM, Ferruh Yigit wrote:
>>> It would be useful if it writes the values of virtual interface, but this
>>> API
>>> prints user input.
>>>
>> I'm sorry, Ferruh, I really don't understand what you are referring to
>> here. What virtual interface are you talking about?
>
On Thu, Sep 27, 2018 at 7:04 PM, Ferruh Yigit wrote:
>>
>> With no traffic it's worse because the worker threads perform these
>> increments even more often!
>
> Dear Dan,
>
> Your implementation doesn't prevent the race when there is traffic, it can be
Correct. To do that would require per-cpu
On 9/27/2018 4:53 PM, Dan Gora wrote:
> On Thu, Sep 27, 2018 at 8:40 AM Ferruh Yigit wrote:
Not sure about introducing an extra check to datapath for possible error on
stats zero. I am for dropping this patch, what do you think?
>>>
>>> This is literally adding one instruction to the dat
On 9/27/2018 4:40 PM, Dan Gora wrote:
> On Thu, Sep 27, 2018 at 8:35 AM Ferruh Yigit wrote:
>>
>> On 9/26/2018 7:56 PM, Dan Gora wrote:
>>> On Wed, Sep 26, 2018 at 1:42 PM, Ferruh Yigit
>>> wrote:
> There is nothing to "reflect" to the kernel interface, nor to apply to
> the kernel inter
On 25/09/18 17:46, Akhil Goyal wrote:
>
>
> On 9/16/2018 8:48 AM, Dmitry Eremin-Solenikov wrote:
>> IPsec requires truncated HMAC operations support. Extend OpenSSL crypto
>> PMD to support truncated HMAC operations necessary for IPsec.
>>
>> Signed-off-by: Dmitry Eremin-Solenikov
>>
>> ---
>> C
Hi,
27/09/2018 10:24, Ophir Munk:
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > The function rte_devargs_remove(), which is intended to be internal, can
> > take a devargs structure as argument.
> > The matching is still using string comparison of bus name and device name.
> > It is sim
> > >
> > > Signed-off-by: Yipeng Wang
> > > ---
> > > lib/librte_hash/rte_cuckoo_hash.c | 326
> > > +-
> > > lib/librte_hash/rte_cuckoo_hash.h | 5 +
> > > lib/librte_hash/rte_hash.h| 3 +
> > > 3 files changed, 292 insertions(+), 42 deletions(-)
On 9/26/2018 5:21 PM, Igor Ryzhov wrote:
> Long time ago preallocation of memory for KNI was introduced in commit
> 0c6bc8e. It was done because of lack of ability to free previously
> allocated memzones, which led to memzone exhaustion. Currently memzones
> can be freed and this patch uses this ab
Hi Keith and Raslan,
Thanks for your review Keith.
It reminds me that you had some comments in my v3 which were not addressed.
Raslan, it seems you missed the v3 I had sent (you sent another v3).
Please check the comments done by Keith and Ferruh during the summer.
Thank you all
On 09/27/2018 08:05 PM, Kevin Traynor wrote:
On 09/27/2018 01:37 PM, Maxime Coquelin wrote:
On 09/27/2018 02:24 PM, Maxime Coquelin wrote:
On 09/17/2018 05:54 AM, Jiayu Hu wrote:
When perform enqueue operations on the split and packed ring,
s/perform/performing/
s/ring/rings/
if the r
On 09/27/2018 01:37 PM, Maxime Coquelin wrote:
>
>
> On 09/27/2018 02:24 PM, Maxime Coquelin wrote:
>>
>>
>> On 09/17/2018 05:54 AM, Jiayu Hu wrote:
>>> When perform enqueue operations on the split and packed ring,
>> s/perform/performing/
>> s/ring/rings/
>>> if the reserved buffer length from t
> -Original Message-
> From: Singh, Jasvinder
> Sent: Thursday, September 27, 2018 6:11 PM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian
> Subject: [PATCH] librte_sched: allocate memory on the given socket id
>
> Replace rte_zmalloc() with rte_zmalloc_socket() to allocate
> memory on t
Replace rte_zmalloc() with rte_zmalloc_socket() to allocate
memory on the socket id provided by the application.
Signed-off-by: Jasvinder Singh
---
lib/librte_sched/rte_sched.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/librte_sched/rte_sched.c b/lib/librte_sched/r
On 09/26/2018 11:06 AM, Xiaolong Ye wrote:
The vdpa sample application creates vhost-user sockets by using the
vDPA backend. vDPA stands for vhost Data Path Acceleration which utilizes
virtio ring compatible devices to serve virtio driver directly to enable
datapath acceleration. As vDPA drive
> -Original Message-
> From: Burakov, Anatoly
> Sent: Thursday, September 27, 2018 11:41 AM
> To: dev@dpdk.org
> Cc: Iremonger, Bernard ;
> laszlo.madara...@ericsson.com; laszlo.vadke...@ericsson.com;
> andras.kov...@ericsson.com; winnie.t...@ericsson.com;
> daniel.andr...@ericsson.com; jan
On Thu, Sep 27, 2018 at 8:40 AM Ferruh Yigit wrote:
> >> Not sure about introducing an extra check to datapath for possible error on
> >> stats zero. I am for dropping this patch, what do you think?
> >
> > This is literally adding one instruction to the datapath. Not even an
> > atomic instructi
Hi Qiang,
On 09/07/2018 02:53 AM, Qiang Zhou wrote:
when rte_vhost_driver_unregister delete the connection fd,
the fd lock will prevent the vsocket to be freed. But when
vhost_user_msg_handler return error, it will delete vsocket
conn_list. And then the fd lock will become invalid. So the
vsocke
On Thu, Sep 27, 2018 at 8:35 AM Ferruh Yigit wrote:
>
> On 9/26/2018 7:56 PM, Dan Gora wrote:
> > On Wed, Sep 26, 2018 at 1:42 PM, Ferruh Yigit
> > wrote:
> >>> There is nothing to "reflect" to the kernel interface, nor to apply to
> >>> the kernel interface. This is exactly how every other ker
Meeting notes for the DPDK technical board meeting
held on 2018-09-26
Attendees:
- Bruce Richardson
- Ferruh Yigit
- Hemant Agrawal
- Jerin Jacob
- Konstantin Ananyev
- Maxime Coquelin
- Olivier Matz
- Stephen Hemminger
- Thom
On 8/17/2018 11:48 AM, Jeff Guo wrote:
> When a device is hotplugged out, the PCI resource is released in the
> kernel, the UIO file descriptor will disappear and the irq will be
> released. After this, a kernel crash will be caused if the igb uio driver
> tries to access or release these resources
This patch adds glue functions for operations:
- Create packet reformat (encap/decap) flow action.
- Destroy flow action.
Signed-off-by: Dekel Peled
---
drivers/net/mlx5/mlx5_glue.c | 33 +
drivers/net/mlx5/mlx5_glue.h | 7 +++
2 files changed, 40 insertions(
This patch add the required definitions for DV encap/decap actions.
It also adds usage of the new actions definition in validation
function of existing drop operation.
Signed-off-by: Dekel Peled
---
drivers/net/mlx5/mlx5_flow.c | 10 ++
drivers/net/mlx5/mlx5_flow.h | 7 +++
2 files
This patch adds support for Direct Verbs decap operations, L2 and L3.
Signed-off-by: Dekel Peled
---
drivers/net/mlx5/mlx5_flow_dv.c | 258 ++--
1 file changed, 248 insertions(+), 10 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5
This patch adds support for Direct Verbs encap operations, L2 and L3.
Signed-off-by: Dekel Peled
---
drivers/net/mlx5/mlx5_flow_dv.c | 249 +++-
1 file changed, 244 insertions(+), 5 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/
This series adds support of encap and decap actions in DV format.
It is using the generic encapsulation framework from [1].
Note that encap action is relevant for egress only.
Hence this series should be merged on top of [2].
[1] "ethdev: add generic L2/L3 tunnel encapsulation actions"
http://
On Thu, Sep 27, 2018 at 02:24:07PM +0100, Agalya Babu RadhaKrishnan wrote:
> FreeBSD compilation was failing through meson build.
> RTE_EAL_VFIO is not supported in FreeBSD.
> But RTE_EAL_VFIO was enabled for both linux and freebsd.
> So RTE_EAL_VFIO is removed from config/rte_config.h and
> based
On Thu, Sep 27, 2018 at 02:24:07PM +0100, Agalya Babu RadhaKrishnan wrote:
> FreeBSD compilation was failing through meson build.
> RTE_EAL_VFIO is not supported in FreeBSD.
> But RTE_EAL_VFIO was enabled for both linux and freebsd.
> So RTE_EAL_VFIO is removed from config/rte_config.h and
> based
On Thu, Sep 27, 2018 at 02:48:38PM +0200, Thomas Monjalon wrote:
> 27/09/2018 14:30, Bruce Richardson:
> > On Thu, Sep 27, 2018 at 01:36:06PM +0200, Thomas Monjalon wrote:
> > > 27/09/2018 13:26, Raslan Darawsheh:
> > > > v2:
> > > > - fix includes order
> > >
> > > I'm afraid you will nee
This patch complements [1], adding to MLX5 PMD the option to set
flow rule for egress traffic.
[1] "net/mlx5: support metadata as flow rule criteria"
http://mails.dpdk.org/archives/dev/2018-September/113275.html
Signed-off-by: Dekel Peled
---
drivers/net/mlx5/mlx5_flow.c | 54
As described in series starting at [1], it adds option to set
metadata value as match pattern when creating a new flow rule.
This patch adds metadata support in mlx5 driver, in two parts:
- Add the setting of metadata value in matcher when creating
a new flow rule.
- Add the passing of metadata
On Fri, Aug 31, 2018 at 5:44 AM Chao Zhu wrote:
> Thomas,
>
> Sorry for the late reply. DPDK will definitely be supported by IBM. I think
> Pradeep and the team is working on this. Personally, I don't have enough
> bandwidth to do this, we'll have a discussion internally to decide who will
> take
On 27-Sep-18 2:42 PM, Alejandro Lucero wrote:
On Thu, Sep 27, 2018 at 2:22 PM Burakov, Anatoly
mailto:anatoly.bura...@intel.com>> wrote:
On 27-Sep-18 2:14 PM, Alejandro Lucero wrote:
> On Thu, Sep 27, 2018 at 11:41 AM Anatoly Burakov
mailto:anatoly.bura...@intel.com>>
> wro
As described in [1],[2] this series adds option to set metadata value as
match pattern when creating a new flow rule.
This patch introduces code for debug porpuse only.
The new debug command takes a 32 bit value and stores it per port.
testpmd will add to any Tx packet sent from this port the meta
As described in [1], this series adds option to set metadata value as
match pattern when creating a new flow rule.
This patch introduces additional options in testpmd commands.
New item type "meta" "data", new offload flag "match_metadata".
[1] "ethdev: support metadata as flow rule criteria"
Si
As described in [1], a new rte_flow item is added to support metadata
to use as flow rule match pattern.
The metadata is an opaque item, fully controlled by the application.
The use of metadata is relevant for egress rules only.
It can be set in the flow rule using the RTE_FLOW_ITEM_META.
In orde
This series implements the match-metadata feature described in [1].
[1] "[RFC v2] ethdev: support metadata as flow rule criteria"
http://mails.dpdk.org/archives/dev/2018-August/110194.html
V3:
* Add link to RFC email.
* Add cover letter subject line.
V2:
* Fix some checkpatch coding style is
> -Original Message-
> From: Babu Radhakrishnan, AgalyaX
> Sent: Thursday, September 27, 2018 2:24 PM
> To: dev@dpdk.org
> Cc: alejandro.luc...@netronome.com; allain.leg...@windriver.com; Singh,
> Jasvinder ; Wiles, Keith ;
> ma...@mellanox.com; Richardson, Bruce ;
> Pattan, Reshma ; Bab
> On Sep 27, 2018, at 8:24 AM, Babu Radhakrishnan, AgalyaX
> wrote:
>
> Disabled tap build in FreeBSD because it is not supported
> Added changes to enable tap build if it is Linux OS and
> disable in FreeBSD.
Good, the old build system restricted tap to Linux only.
Acked-by: Keith Wiles
>
On 27.09.2018 12:40, Hideyuki Yamashita wrote:
[...]
>>> Is it correct understanding that race condition
>>> includes
>>> - read information before port is available
>>> - other device may be plugged (or unplugged)
>>> and so using "eth_dev_last_created_port" is
>>> NOT reliable?
>>
>> I am think
On Thu, Sep 27, 2018 at 2:22 PM Burakov, Anatoly
wrote:
> On 27-Sep-18 2:14 PM, Alejandro Lucero wrote:
> > On Thu, Sep 27, 2018 at 11:41 AM Anatoly Burakov <
> anatoly.bura...@intel.com>
> > wrote:
> >
> >> We will be assigning "invalid" socket ID's to external heap, and
> >> malloc will now be
On Thu, 2018-09-27 at 14:24 +0100, Agalya Babu RadhaKrishnan wrote:
> Disabled vdev_netvsc build in FreeBSD because it is not supported
> Added changes to enable vdev_netvsc build if it is Linux OS and
> disable in FreeBSD.
>
> Fixes: 6e8a721044 ("vfio: export functions even when disabled")
>
> S
On 9/21/2018 5:25 PM, Paul Luse wrote:
> From: paul luse
>
> This patch fixes an issue caught with ASAN where a vdev_scan()
> to a secondary bus was failing to free some memory.
>
> Signed-off-by: paul luse
(carry from Anatoly's ack from other thread)
Acked-by: Anatoly Burakov
Acked-by: Ferr
FreeBSD compilation was failing through meson build.
RTE_EAL_VFIO is not supported in FreeBSD.
But RTE_EAL_VFIO was enabled for both linux and freebsd.
So RTE_EAL_VFIO is removed from config/rte_config.h and
based on the platform RTE_EAL_VFIO flag is enabled/disabled appropriately.
Fixes: 6e8a7210
Disabled tap build in FreeBSD because it is not supported
Added changes to enable tap build if it is Linux OS and
disable in FreeBSD.
Fixes: 6e8a721044 ("vfio: export functions even when disabled")
Signed-off-by: Agalya Babu RadhaKrishnan
---
drivers/net/tap/meson.build | 3 +++
1 file changed,
Compilation issues were observed in FreeBSD when built via meson.
VFIO flags are enabled/disabled based on linux/freebsd.
Changes are done in meson.build files to ensure
the build of drivers are enabled/disabled based on platform.
1/6: set/unset of vfio flags based on platforms
2/6: updated net/nf
On Thu, Sep 27, 2018 at 2:18 PM Burakov, Anatoly
wrote:
> On 27-Sep-18 2:01 PM, Alejandro Lucero wrote:
> > On Thu, Sep 27, 2018 at 11:47 AM Anatoly Burakov <
> anatoly.bura...@intel.com>
> > wrote:
> >
> >> Switch over all parts of EAL to use heap ID instead of NUMA node
> >> ID to identify heap
On 27-Sep-18 2:14 PM, Alejandro Lucero wrote:
On Thu, Sep 27, 2018 at 11:41 AM Anatoly Burakov
wrote:
We will be assigning "invalid" socket ID's to external heap, and
malloc will now be able to verify if a supplied socket ID is in
fact a valid one, rendering parameter checks for sockets
obsole
Disabled avp build in FreeBSD because it is not supported.
Added changes to enable avp build if it is Linux OS and
disable in FreeBSD.
Fixes: 6e8a721044 ("vfio: export functions even when disabled")
Signed-off-by: Agalya Babu RadhaKrishnan
---
drivers/net/avp/meson.build | 3 +++
1 file changed
Disabled vdev_netvsc build in FreeBSD because it is not supported
Added changes to enable vdev_netvsc build if it is Linux OS and
disable in FreeBSD.
Fixes: 6e8a721044 ("vfio: export functions even when disabled")
Signed-off-by: Agalya Babu RadhaKrishnan
---
drivers/net/vdev_netvsc/meson.build
Disabled nfp build in FreeBSD because it is not supported
Added changes to enable NFP build if it is Linux OS and
disable in FreeBSD.
Fixes: 6e8a721044 ("vfio: export functions even when disabled")
Signed-off-by: Agalya Babu RadhaKrishnan
---
drivers/net/nfp/meson.build | 3 +++
1 file changed,
Disabled softnic build in FreeBSD because it is not supported
Added changes to enable softnic build if it is Linux OS and
disable in FreeBSD.
Fixes: 6e8a721044 ("vfio: export functions even when disabled")
Signed-off-by: Agalya Babu RadhaKrishnan
---
drivers/net/softnic/meson.build | 3 +++
1 f
> On Sep 27, 2018, at 6:19 AM, Raslan Darawsheh wrote:
>
> change the fds for the queues to be pointers and add new process private
> structure and make the queue fds point to it.
>
> Signed-off-by: Raslan Darawsheh
> ---
> drivers/net/tap/rte_eth_tap.c | 63 -
On 27-Sep-18 2:01 PM, Alejandro Lucero wrote:
On Thu, Sep 27, 2018 at 11:47 AM Anatoly Burakov
wrote:
Switch over all parts of EAL to use heap ID instead of NUMA node
ID to identify heaps. Heap ID for DPDK-internal heaps is NUMA
node's index within the detected NUMA node list. Heap ID for
exte
On Thu, Sep 27, 2018 at 11:41 AM Anatoly Burakov
wrote:
> We will be assigning "invalid" socket ID's to external heap, and
> malloc will now be able to verify if a supplied socket ID is in
> fact a valid one, rendering parameter checks for sockets
> obsolete.
>
> This changes the semantics of wha
> On Sep 27, 2018, at 6:19 AM, Raslan Darawsheh wrote:
>
> In the case the device is created by the primary process,
> the secondary must request some file descriptors to attach the queues.
> The file descriptors are shared via IPC Unix socket.
>
> Thanks to the IPC synchronization, the secon
On Thu, Sep 27, 2018 at 11:47 AM Anatoly Burakov
wrote:
> Switch over all parts of EAL to use heap ID instead of NUMA node
> ID to identify heaps. Heap ID for DPDK-internal heaps is NUMA
> node's index within the detected NUMA node list. Heap ID for
> external heaps will be order of their creatio
27/09/2018 14:30, Bruce Richardson:
> On Thu, Sep 27, 2018 at 01:36:06PM +0200, Thomas Monjalon wrote:
> > 27/09/2018 13:26, Raslan Darawsheh:
> > > v2:
> > > - fix includes order
> >
> > I'm afraid you will need a v3 to fix spacing :)
> >
> > > --- a/examples/multi_process/client_server_mp/mp_
> > The rte_afu_driver is assigned to rte_afu_device.driver during probing.
> > There is no need of accessing the rte_afu_driver via rte_device.driver and
> > type casting to its container.
> >
> > Signed-off-by: Thomas Monjalon
>
> Acked-by: Rosen Xu
Applied
26/09/2018 10:12, Shreyansh Jain:
> On Wednesday 26 September 2018 12:52 AM, Thomas Monjalon wrote:
> > The probing functions of NXP buses were missing to set
> > the driver used for successfully probing a device.
> >
> > The NXP driver and the generic rte_driver are now set
> > in the device stru
On 09/27/2018 02:24 PM, Maxime Coquelin wrote:
On 09/17/2018 05:54 AM, Jiayu Hu wrote:
When perform enqueue operations on the split and packed ring,
s/perform/performing/
s/ring/rings/
if the reserved buffer length from the descriptor table execeeds
65535, the returned length by fill_vec_
> -Original Message-
> From: Richardson, Bruce
> Sent: Thursday, September 27, 2018 1:28 PM
> To: Ananyev, Konstantin
> Cc: Honnappa Nagarahalli ; Wang, Yipeng1
> ; dev@dpdk.org;
> mic...@digirati.com.br
> Subject: Re: [dpdk-dev] [PATCH v2 5/7] hash: add extendable bucket feature
>
>
Invoking rte_pci_read/write_config functions requires device with
a intr_handle type for using VFIO or UIO driver related functions.
Secondary processes rely on primary processes for device initialization
so they do not usually require using these functions. However, some PMDs,
like NFP PMD, requi
On Thu, Sep 27, 2018 at 01:36:06PM +0200, Thomas Monjalon wrote:
> 27/09/2018 13:26, Raslan Darawsheh:
> > v2:
> > - fix includes order
>
> I'm afraid you will need a v3 to fix spacing :)
>
> > --- a/examples/multi_process/client_server_mp/mp_server/main.c
> > +++ b/examples/multi_process/cli
1 - 100 of 184 matches
Mail list logo
