-Original Message-
> Date: Wed, 2 Aug 2017 19:19:32 +
> From: "Eads, Gage"
> To: Jerin Jacob , "Rao, Nikhil"
>
> CC: "dev@dpdk.org" , "tho...@monjalon.net"
> , "Richardson, Bruce" ,
> "Van Haaren, Harry" , "hemant.agra...@nxp.com"
> , "nipun.gu...@nxp.com" ,
> "Vangati, Narender"
With below code, I am getting this warning.
warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
my_node_t *data_ptr = (my_node_t *) rte_malloc(NULL, sizeof(my_node_t), 0);
Thanks,
Rohit
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of R
Hi,
I have a use case in my application where I need to implement my own memory
manager, rather than doing malloc/free everytime to kernel.
Instead of writing my own memory manager, I am thinking to use dpdk rte_malloc
or rte_mempool. Please let me know if this is a good idea.
Also,
my_node_t *
This patchset has some trivial changes and make it easier
to read code.
Zhiyong Yang (2):
virtio: fix the wrong indent
virtio: use macro to replace magic number
drivers/net/virtio/virtio_pci.c | 2 +-
drivers/net/virtio/virtqueue.h | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
Signed-off-by: Zhiyong Yang
---
drivers/net/virtio/virtqueue.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio/virtqueue.h b/drivers/net/virtio/virtqueue.h
index 2e1208615..9c4f96d2b 100644
--- a/drivers/net/virtio/virtqueue.h
+++ b/drivers/net/virtio/vi
To use macro instead of magic number in order to enhance code
readability.
Signed-off-by: Zhiyong Yang
---
drivers/net/virtio/virtio_pci.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index e6da6802b..55b717
> On Aug 1, 2017, at 12:55 AM, Nelio Laranjeiro
> wrote:
>
> Changing the MTU is not related to changing the number of segments,
> activating or not the multi-segment support should be handled by the
> application.
>
> Fixes: 9964b965ad69 ("net/mlx5: re-add Rx scatter support")
> Cc: sta...@dp
> > >
> > > 5) specifying rte_event_eth_rx_adapter_conf.rx_event_port_id on
> > > rte_event_eth_rx_adapter_create() would waste one HW eventdev port
> > > if its happen to be used RX_ADAPTER_CAP_INBUILT_PORT on
> rte_event_eth_rx_adapter_queue_add().
> > > unlike SW eventdev port, HW eventdev po
Current mlx4 OFED version has bug which returns error to
ibv destroy functions when the device was plugged out, in
spite of the resources were destroyed correctly.
Hence, failsafe PMD was aborted, only in debug mode, when
it tries to remove the device in plug-out process.
The workaround added opt
> -Original Message-
> From: Aaron Conole [mailto:acon...@redhat.com]
> Sent: Wednesday, August 02, 2017 11:02 AM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; John Daley (johndale) ; Bruce
> Richardson
> Subject: [PATCH] enic: check for nb_free > 0
>
> Occasionally, the amount of packets t
Occasionally, the amount of packets to free from the work queue ends
perfectly on a boundary to have nb_free = 0 and pool = 0. This causes
a segfault as follows:
(gdb) bt
#0 rte_mempool_default_cache (mp=, mp=,
lcore_id=)
at
/usr/src/debug/openvswitch-2.6.1/dpdk-16.11/x86_64-nat
The original purpose of this deprecation is to make sure pci devices
are reset whenever DPDK apps crash. Since the
commit b58eedfc7dd5 ("igb_uio: issue FLR during open and release of device
file")
from Shijith can fix this problem without deprecating anything, now there
is no need to deprecate iom
The pinfo variable has wrong data. This has to have merged data of two
fields from Rx completion - pkt_info and hdr_type_etc.
Fixes: 6cb559d67b83 ("net/mlx5: add vectorized Rx/Tx burst for x86")
Signed-off-by: Yongseok Koh
Acked-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx_vec_sse.c | 5
If there's a Rx completion with error (e.g, MTU mismatch), it is handled
later out of main burst loop as a slow path for performance reason.
Statistics should be corrected by subtracting counters of errored packets.
Also, the last entry of mlx5_ptype_table[] must be RTE_PTYPE_ALL_MASK to
mark error
Do not fail upon detecting a set bus configuration when adding an
rte_devargs. i.e. adding a blacklisted PCI device while the bus is
configured in whitelist mode.
This failure condition has been introduced when the bus policies
were implemented. This implementation however was meant to strictly
fo
When the EAL parses the common options given to the application,
not all subsystems are available. Some device drivers are registered
afterward upon dynamic plugin loading.
Devices using those drivers are thus unable to be parsed by any drivers
and are rejected.
Store the device options first and
On Tue, Aug 01, 2017 at 02:09:26PM +0200, Nelio Laranjeiro wrote:
> This Series make some cleanup in mlx5 PMD by removing useless code or non
> working code.
>
> This series apply on top of:
> [PATCH 0/5] net/mlx5: enable/disable vPMD and some cleanups
> Specifically on patch: net/mlx5: cleanup R
Vector code is very young and can present some issues for users, to avoid
them to modify the selections function by commenting the code and recompile
the PMD, new devices parameters are added to deactivate the Tx and/or Rx
vector code.
By using such device parameters, the user will be able to fall
This series add more information and flexibility to the user to use or not the
vPMD. An information will be printed to inform the user which Tx/Rx function
have been automatically selected in addition, new device parameters are added
to disable Tx/Rx vector (see device documentation for more detai
On Wed, Aug 02, 2017 at 04:29:13PM +0200, Thomas Monjalon wrote:
> Hi Gaetan
>
> 02/08/2017 10:56, Gaetan Rivet:
> > When the EAL parses the common options given to the application,
> > not all subsystems are available. Some device drivers are registered
> > afterward upon dynamic plugin loading.
On Wed, Aug 02, 2017 at 04:32:38PM +0200, Thomas Monjalon wrote:
> 02/08/2017 10:55, Gaetan Rivet:
> > When adding rte_devargs to the list, a check is performed on the
> > intended bus that should use this rte_devargs. This bus configuration is
> > for the time being only set once when the first rt
02/08/2017 10:55, Gaetan Rivet:
> When adding rte_devargs to the list, a check is performed on the
> intended bus that should use this rte_devargs. This bus configuration is
> for the time being only set once when the first rte_devargs is added to
> it. If the bus configuration has previously been
Hi Gaetan
02/08/2017 10:56, Gaetan Rivet:
> When the EAL parses the common options given to the application,
> not all subsystems are available. Some device drivers are registered
> afterward upon dynamic plugin loading.
>
> Devices using those drivers are thus unable to be parsed by any drivers
This also bring back the RSS support on all flows created by default from
the control plane.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_flow.c | 437 ++-
1 file changed, 346 insertions(+), 91 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_
>From this commit the RSS support becomes un-available until it is replaced
by the generic flow implementation.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 2 -
drivers/net/mlx5/mlx5.h | 6 -
drivers/net/mlx5/mlx5_rxq.c | 470
RSS hash configuration is currently ignored by the PMD, this commits
removes the RSS feature.
This functionality will be added in a later commit.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 9 +-
drivers/net/mlx5/mlx5.h | 22 +--
drivers/net/mlx5/mlx5_defs.h
RSS hash configuration is currently ignored by the PMD, this commits
removes the RSS feature on promiscuous mode.
This functionality will be added in a later commit.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.h | 1 -
drivers/net/mlx5/mlx5_rxmode.c | 52 ++--
Since RSS configuration can also be used by flow API, there is no more
necessity to keep a list of RSS configurable for all protocols.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c| 24 +---
drivers/net/mlx5/mlx5.h| 6 +-
drivers/net/mlx5/mlx5_ethdev.c | 16
RSS hash configuration is currently ignored by the PMD, this commits
removes the RSS feature on promiscuous mode.
This functionality will be added in a later commit.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 1 +
drivers/net/mlx5/mlx5.h | 15 +++--
drivers
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_flow.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index f258567..90deb30 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -426,13 +42
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 3 +
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/mlx5/mlx5_flow.c | 137 +++-
drivers/net/mlx5/mlx5_rxq.c | 161 +++
drivers/net/mlx5/mlx5_rxtx.h |
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 16 +-
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/mlx5/mlx5_rxq.c | 492 +---
drivers/net/mlx5/mlx5_rxtx.h| 10 +
drivers/net/mlx5/mlx5_trigger.c | 45
5 files chang
In prevision of the huge rework on Rx hash queues and the fact this
feature will be totally different, the drop flow is removed from now and
will be re-introduced later.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.h | 1 -
drivers/net/mlx5/mlx5_flow.c | 228 +++--
Avoid to have for each Hash Rx queue it dedicated indirection table. On
verbs side, the indirection table only points to the Work Queue, two hash
Rx queues using the same set of WQ can use the same indirection table.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 3 +
dri
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 16 +-
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/mlx5/mlx5_mr.c | 37 ++--
drivers/net/mlx5/mlx5_rxtx.h| 13 +-
drivers/net/mlx5/mlx5_trigger.c | 56 ++
drivers/net/mlx5/mlx5_txq.c | 383
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 3 +
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/mlx5/mlx5_rxtx.h | 18 +-
drivers/net/mlx5/mlx5_txq.c | 437 ++-
4 files changed, 284 insertions(+), 175 deletions(-)
diff --git a
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 3 +
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_flow.c | 97 +++-
drivers/net/mlx5/mlx5_rxq.c | 564 ++-
drivers/net/mlx5/mlx5_rxtx.h | 26 +-
drivers/net/mlx5/mlx
Memory regions becomes shared by the same time, the control plane as it own
reference in addition of the Tx/Rx queues.
This should also avoid to un-register a memory region when the Tx queue
evicts it from its local cache.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.h | 8 ++
Debug tools to verify all flows are be un-register from the NIC.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 3 +++
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/mlx5/mlx5_flow.c | 22 ++
3 files changed, 26 insertions(+)
diff --git a/drivers/net/m
Prefix struct txq_ctrl with mlx5.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 8
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_mr.c | 20 +-
drivers/net/mlx5/mlx5_rxtx.c | 25 ---
This flag is already present in the Ethernet device.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.h | 1 -
drivers/net/mlx5/mlx5_flow.c| 6 +++---
drivers/net/mlx5/mlx5_rxq.c | 2 +-
drivers/net/mlx5/mlx5_trigger.c | 12
drivers/net/mlx5/mlx5_txq.c
Prefix struct rxq_ctrl with mlx5.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c | 8 ++---
drivers/net/mlx5/mlx5.h | 4 +--
drivers/net/mlx5/mlx5_flow.c | 12 +++
drivers/net/mlx5/mlx5_rxq.c | 61 +++-
Generic flow API should be use for flow steering as is provides a better
and easier way to configure flows.
Signed-off-by: Nelio Laranjeiro
---
doc/guides/nics/features/mlx5.ini |1 -
doc/guides/nics/mlx5.rst |2 -
drivers/net/mlx5/Makefile |1 -
drivers/net/mlx5/mlx
patch tested by dirk.lenz and works for one virtual device
On 08/01/2017 11:32 AM, Thomas Monjalon wrote:
01/08/2017 10:17, Gaëtan Rivet:
Hi,
On Mon, Jul 31, 2017 at 10:23:50PM +0200, Thomas Monjalon wrote:
31/07/2017 16:58, Dirk-Holger Lenz:
If dpdk is built with 'CONFIG_RTE_BUILD_SHARED_LI
mlx5_flow_create() and mlx5_flow_validate() are making common checks.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_flow.c | 151 +--
1 file changed, 72 insertions(+), 79 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx
This also bring back the RSS support on all flows created by default from
the control plane.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_flow.c | 437 ++-
1 file changed, 346 insertions(+), 91 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_
This series cleanups the control plane part and the way it uses the different
kind of objects (DPDK queues, Verbs Queues, ...). It has three goals:
1. Reduce the memory usage by sharing all possible objects.
2. Leave the configuration to the control plane and the creation/destruction
of qu
After fail to create a flow and if this is the first flow, the
mask_added flag should be reset, or it will prevent a new flow
which require different mask be created, since the mask config
remains impact.
Fixes: 72c135a89f8 ("net/ixgbe: create consistent filter")
Cc: sta...@dpdk.org
Signed-off-by
Hi Wenzhuo,
On Wed, Aug 02, 2017 at 01:00:47PM +, Lu, Wenzhuo wrote:
> Hi,
>
> > -Original Message-
> > From: Zhao1, Wei
> > Sent: Wednesday, August 2, 2017 3:28 AM
> > To: TETSURO NAKAMURA
> > Cc: TAKADA Naoki ; Satoshi NISHIYAMA
> > ; Lu, Wenzhuo ;
> > dev@dpdk.org
> > Subject: RE:
Hi Declan,
On 7/26/2017 7:16 PM, Declan Doherty wrote:
Hey Akhil, I like the proposal of allowing the rte_secruity API to be
supported on both NIC and crypto devices as I think it allows us to
cover all the protocol offload scenarios in a consist manner.
The main concern I have is in regards to
Hi,
> -Original Message-
> From: Zhao1, Wei
> Sent: Wednesday, August 2, 2017 3:28 AM
> To: TETSURO NAKAMURA
> Cc: TAKADA Naoki ; Satoshi NISHIYAMA
> ; Lu, Wenzhuo ;
> dev@dpdk.org
> Subject: RE: using rte_flow via tespmd with Intel X552
>
> Hi, NAKAMURA
>
> I have upload 2 documents ab
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Qi Zhang
>Sent: Wednesday, August 2, 2017 8:22 AM
>To: Wu, Jingjing
>Cc: dev@dpdk.org; jianfeng.t...@intel.com; Zhang, Qi Z ;
>sta...@dpdk.org; Tan, Jianfeng
>Subject: [dpdk-dev] [PATCH] net/i40e: fix data segment buffer length
>
>Buffer lengt
Hi, NAKAMURA
> -Original Message-
> From: TETSURO NAKAMURA [mailto:nakamura.tets...@lab.ntt.co.jp]
> Sent: Wednesday, August 2, 2017 5:02 PM
> To: Zhao1, Wei
> Cc: TAKADA Naoki ; Satoshi NISHIYAMA
> ; Lu, Wenzhuo ;
> dev@dpdk.org
> Subject: Re: using rte_flow via tespmd with Intel
Hi Zhao1, Wei - san,
I really appreciate your help !
I have no idea where to upload this,
but I'm sure it should be uploaded to somewhere public.
On 2017/08/02 16:28, Zhao1, Wei wrote:
Hi, NAKAMURA
I have upload 2 documents about command format when creating flow for NIC igb
and ixgbe in the
When the EAL parses the common options given to the application,
not all subsystems are available. Some device drivers are registered
afterward upon dynamic plugin loading.
Devices using those drivers are thus unable to be parsed by any drivers
and are rejected.
Store the device options first and
When adding rte_devargs to the list, a check is performed on the
intended bus that should use this rte_devargs. This bus configuration is
for the time being only set once when the first rte_devargs is added to
it. If the bus configuration has previously been set, then the rte_devargs
insertion fail
On Tue, Aug 01, 2017 at 05:35:30PM +, Legacy, Allain wrote:
> > -Original Message-
> > From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> > Sent: Tuesday, August 01, 2017 12:54 PM
> <...>
> > @@ -5946,12 +5949,11 @@ mlx4_arg_parse(const char *key, const char *val,
> > void *ou
Hi, NAKAMURA
I have upload 2 documents about command format when creating flow for NIC igb
and ixgbe in the email attachment.
I have decided to commit the context in the doc to doc\guides\nics\ixgbe.rst
before, BUT by now it seems the content of the document is too much
To store there. So, is th
58 matches
Mail list logo
