Now we've setup both rx and tx queues.
Signed-off-by: Xiaolong Ye
---
examples/flow_filtering/main.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/examples/flow_filtering/main.c b/examples/flow_filtering/main.c
index 09dbbcea9..a0487be77 100644
--- a/examples/flow_filtering/main.c
+++ b/exa
Update feature for ICE driver
Signed-off-by: Beilei Xing
---
doc/guides/rel_notes/release_19_08.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/rel_notes/release_19_08.rst
b/doc/guides/rel_notes/release_19_08.rst
index 8da66fe..81ec5fa 100644
--- a/doc/guides/rel_notes/r
Enable Tx checksum offload for tunneling packets by
configuring tunneling parameters in Tx descriptors,
including outer L3/L4 checksum offload.
Signed-off-by: Beilei Xing
---
drivers/net/ice/ice_ethdev.c | 3 +-
drivers/net/ice/ice_rxtx.c | 78 +---
2 f
Enable Tx checksum offload for tunneling packets.
Update release notes.
v3 change:
- Add release notes.
v2 change:
- Parse udp checksum in tunneling parameter.
Beilei Xing (2):
net/ice: support Tx checksum offload for tunneling packets
doc: update for ICE driver
doc/guides/rel_notes/relea
Fixes: 3e0ceb9f17ff ("doc: add basic howto for flow API")
Cc: sta...@dpdk.org
Signed-off-by: Xiaolong Ye
---
doc/guides/howto/rte_flow.rst | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/doc/guides/howto/rte_flow.rst b/doc/guides/howto/rte_flow.rst
index e197376e2..27d4f2
On 6/18/19 8:24 AM, Noa Ezra wrote:
Hi,
Does anyone else want to review this patch?
I do, but it seems that the other 4 patches of the series aren't on
dev@dpdk.org, is that expected?
Thanks,
Noa.
-Original Message-
From: Matan Azrad
Sent: Sunday, May 26, 2019 1:23 PM
To: Noa Ezra
Hi Aaron,
>-Original Message-
>From: Aaron Conole
>Sent: Tuesday, June 18, 2019 2:55 AM
>To: Jerin Jacob Kollanukkaran
>Cc: dev@dpdk.org; Nithin Kumar Dabilpuram
>; Vamsi Krishna Attunuru
>; Pavan Nikhilesh Bhagavatula
>; Olivier Matz
>Subject: [EXT] Re: [dpdk-dev] [PATCH v3 25/27] memp
Tiwei Bie (4):
net/virtio: fix memory leak in in-order Rx
net/virtio: fix memory leak in mergeable Rx
net/virtio: fix memory leak in mergeable packed Rx
net/virtio: fix packets check in mergeable packed Rx
drivers/net/virtio/virtio_rxtx.c | 56 +++-
1 file chan
We should check the descriptor state instead of vq's internal
free count (i.e. the number of descriptors that we haven't made
available) for the remaining mergeable packets.
Fixes: a76290c8f1cf ("net/virtio: implement Rx path for packed queues")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
When there is no enough segments for a packet in mergeable
Rx path, we should free the whole mbuf chain instead of just
the last segment.
Fixes: bcac5aa207f8 ("net/virtio: improve batching in mergeable path")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_rxtx.c | 2
When there is no enough segments for a packet in mergeable
packed Rx path, we should free the whole mbuf chain instead
of just the last segment.
Fixes: a76290c8f1cf ("net/virtio: implement Rx path for packed queues")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_rxt
When there is no enough segments for a packet in in-order
mergeable Rx path, we should free the whole mbuf chain instead
of just the last segment.
Fixes: e5f456a98d3c ("net/virtio: support in-order Rx and Tx")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_rxtx.c | 2
On 6/10/2019 8:38 AM, Andrew Rybchenko wrote:
> checkpatches.sh generates warnings/errors because of a bit different
> coding style in base driver.
>
> Andrew Lee (1):
> net/sfc/base: fix signed/unsigned mismatch errors
>
> Andrew Rybchenko (2):
> net/sfc/base: do not rely on indirect header
On 6/18/2019 7:52 AM, Beilei Xing wrote:
> Add support for new i40e device.
>
> Signed-off-by: Beilei Xing
Can you please add the commit that introduces the device as Fixes line?
> ---
> v2 change:
> - Change indentation.
>
> doc/guides/rel_notes/release_19_08.rst | 4
> 1 file changed,
Multi-process support for memif PMD.
Primary process handles connection establishment.
Secondary process queries for memory regions.
Signed-off-by: Jakub Grajciar
---
drivers/net/memif/Makefile| 3 +
drivers/net/memif/memif_socket.c | 45 +--
drivers/net/memif/meson.build | 3 +
31/05/2019 15:15, Laatz, Kevin:
> On 29/05/2019 13:43, Reshma Pattan wrote:
> > Free the `values` pointer before returning
> > from rte_telemetry_command_ports_all_stat_values()
> > to avoid memory leak.
> >
> > Fixes: c12cefa379 ("telemetry: fix mapping of statistics")
> > CC: sta...@dpdk.org
> >
On 18-Jun-19 9:48 AM, Jakub Grajciar wrote:
Multi-process support for memif PMD.
Primary process handles connection establishment.
Secondary process queries for memory regions.
Signed-off-by: Jakub Grajciar
---
<...>
+/*
+ * Request regions
+ * Called by secondary process, when ports link s
17/05/2019 19:07, Reshma Pattan:
> telemetry has support for fetching port based stats
> from metrics library.
>
> Metrics library also has global stats which are
> not fetched by telemetry, so extend telemetry to
> fetch the global metrics.
>
> Signed-off-by: Reshma Pattan
> ---
I see some err
On Mon, Jun 17, 2019 at 11:21:28AM -0700, Nicolas Chautru wrote:
> Reorg within Intel teams causing transfer of maintainership
>
> Signed-off-by: Nicolas Chautru
> ---
Acked-by: Bruce Richardson
> MAINTAINERS | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/MAINTAINER
Hello Christian,
Could you have a look at this?
Is there someone working on it?
We will have to disable the table unit tests in CI otherwise.
Thanks.
--
David Marchand
On Mon, May 27, 2019 at 6:50 PM wrote:
> https://bugs.dpdk.org/show_bug.cgi?id=285
>
> Bug ID: 285
>
On Tue, Jun 18, 2019 at 12:59 PM Andrius Sirvys
wrote:
> crc_strip was removed from lib/librte_ethdev/rte_ethdev.c as the
> NICs carry out this operation themselves. However once removed,
> the references to it we're forgotten to be taken out in test-pmd.
>
> Fixes: 323e7b667f18 ("ethdev: make de
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
>
> I see some errors with GCC:
>
> rte_telemetry_parser.c:362:27: error: unused variable ‘p’
>
> rte_telemetry.c:550:11: error: unused variable ‘i’
>
> rte_telemetry.c:613:16: error: comparison of integer ex
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
On 6/6/2019 5:15 PM, Stephen Hemminger wrote:
> The VF spinlock was never initialized. It works because it is
> in zmalloc'd memory and an unlocked lock on x86 is 0.
> But for good practice, all spinlock's should be initialized.
>
> Fixes: dc7680e8597c ("net/netvsc: support integrated VF")
> Signe
On 6/11/2019 10:48 AM, Andy Pei wrote:
> This patch implemente statistics read and reset
> function for ipn3ke.
>
> Fixes: 70d6b7f550f4 ("net/ipn3ke: add representor")
> Cc: rosen...@intel.com
>
> Signed-off-by: Andy Pei
<...>
> +#define IPN3KE_RPST_TXQ_PRIO_XSTATS_CNT
> (sizeof(ipn3ke_rpst_t
Hello,
Can somebody please provide an equivalent patch for DPDK.
https://patchwork.ozlabs.org/patch/479888/
We are most likely hitting this issue in our setup, when we restart
our DPDK application multiple times on the i219LM NIC. If someone can
provide a patch, we can test it and see whether we
On 6/18/2019 12:59 PM, Ferruh Yigit wrote:
> On 6/11/2019 10:48 AM, Andy Pei wrote:
>> This patch implemente statistics read and reset
>> function for ipn3ke.
>>
>> Fixes: 70d6b7f550f4 ("net/ipn3ke: add representor")
>> Cc: rosen...@intel.com
>>
>> Signed-off-by: Andy Pei
>
> <...>
>
>> +#define
>
> The perf tests were removed in a previous commit.
>
> Fixes: 2ac67c32837a ('test/crypto: remove crypto perf tests')
> Signed-off-by: Thierry Herbelot
> ---
Acked-by: Akhil Goyal
Hi Arek,
> In case big number need to be freed, data it contains should be cleared
> before especially if it is critical data like private keys.
>
> Fixes: 3e9d6bd447fb ("crypto/openssl: add RSA and mod asym operations")
>
> Signed-off-by: Arek Kusztal
> ---
> config/common_base
> -Original Message-
> From: Xing, Beilei
> Sent: Monday, June 17, 2019 11:18 AM
> To: Zhang, Qi Z
> Cc: dev@dpdk.org
> Subject: [PATCH v4] net/i40e: allow VF to configure pctype mapping
>
> This patch allows VF to get/update/reset pctype mapping info.
>
> Signed-off-by: Beilei Xing
> -Original Message-
> From: Xing, Beilei
> Sent: Tuesday, June 18, 2019 3:05 PM
> To: Zhang, Qi Z ; Mcnamara, John
>
> Cc: dev@dpdk.org
> Subject: [PATCH v3 0/2] net/ice: support Tx checksum offload
>
> Enable Tx checksum offload for tunneling packets.
> Update release notes.
>
> v3
> -Original Message-
> From: Zhang, Qi Z
> Sent: Tuesday, June 18, 2019 9:14 PM
> To: Xing, Beilei
> Cc: dev@dpdk.org
> Subject: RE: [PATCH v4] net/i40e: allow VF to configure pctype mapping
>
>
>
> > -Original Message-
> > From: Xing, Beilei
> > Sent: Monday, June 17, 2019 1
On Wed, Jun 05, 2019 at 09:39:07PM +0100, Luca Boccassi wrote:
> On Wed, 2019-06-05 at 21:22 +0100, Bruce Richardson wrote:
> > After using meson to configure a build, we output a brief list of all
> > components that are enabled. However, of more use to most users is a
> > list of what is not bein
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Xu, Rosen
> Sent: Thursday, May 23, 2019 10:27 AM
> To: Pei, Andy ; dev@dpdk.org
> Cc: sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] net/ipn3ke: delete identical branch
>
> Hi,
>
> > -Original Message--
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Xu, Rosen
> Sent: Thursday, May 23, 2019 11:00 AM
> To: Pei, Andy ; dev@dpdk.org
> Cc: sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] net/ipn3ke: check input argument before
> other operation
>
> Hi,
>
>
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Xu, Rosen
> Sent: Thursday, May 23, 2019 1:20 PM
> To: Pei, Andy ; dev@dpdk.org
> Cc: sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] net/ipn3ke: fix null pointer dereference
>
>
>
> > -Original Message-
Hi Fan,
>
> This patch fixes the out-of-bounds coverity issue by adding
> missed algorithms to the array.
>
> Coverity issue: 337683
>
> Fixes: c68d7aa354f6 ("crypto/aesni_mb: use architecture independent macros")
>
> Signed-off-by: Fan Zhang
> ---
> drivers/crypto/aesni_mb/rte_aesni_mb_pmd_
telemetry has support for fetching port based stats
from metrics library.
Metrics library also has global stats which are
not fetched by telemetry, so extend telemetry to
fetch the global metrics.
Signed-off-by: Reshma Pattan
Acked-by: Kevin Laatz
---
v2: fix GCC compilation issues.
rebase with
> -Original Message-
> From: Xing, Beilei
> Sent: Wednesday, June 12, 2019 9:35 AM
> To: David Marchand ; dev@dpdk.org
> Cc: sta...@dpdk.org; Zhang, Qi Z
> Subject: RE: [dpdk-dev] [PATCH 1/2] net/i40e: fix dropped packets statistics
> name
>
>
>
> > -Original Message-
> > Fro
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Eelco Chaudron
> Sent: Monday, June 3, 2019 4:53 PM
> To: David Marchand
> Cc: dev@dpdk.org; sta...@dpdk.org; Yang, Qiming ;
> Lu, Wenzhuo
> Subject: Re: [dpdk-dev] [dpdk-stable] [PATCH 2/2] net/ice: fix dropped
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> Sent: Wednesday, May 29, 2019 5:12 PM
> To: Stephen Hemminger ; Xing, Beilei
> ; Zhang, Qi Z
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] net/i40e: remove queue_stats_mapping_set
>
>
>
> On 5/28
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Lavanya
> Govindarajan
> Sent: Monday, June 10, 2019 9:08 PM
> To: dev@dpdk.org
> Cc: Pattan, Reshma ; Wu, Jingjing
> ; Lu, Wenzhuo ;
> Parthasarathy, JananeeX M ;
> Govindarajan, LavanyaX
> Subject: [dpdk-dev]
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
> clang raise 'pointer-sign' warnings in __atomic_compare_exchange when
> passing 'uint64_t *' to parameter of type 'int64_t *' converts between
> pointers to integer types with different sign.
>
> Fixes: 7e6e609939a8 ("stack: add C11 atomic implementation")
>
> Suggested-by: Gage Eads
> Signed-
On Tue, 12 Mar 2019 17:22:32 +0800
Yangchao Zhou wrote:
> va2pa depends on the physical address and virtual address offset of
> current mbuf. It may get the wrong physical address of next mbuf which
> allocated in another hugepage segment.
>
> In rte_mempool_populate_default(), trying to allocat
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
>
> I see some errors with GCC:
On which GCC version the errors are shown?
>
> rte_telemetry_parser.c:362:27: error: unused variable ‘p’
>
> rte_telemetry.c:550:11: error: unused variable ‘i’
>
> rte_tele
On Thu, 6 Jun 2019 16:26:26 +0300
Igor Ryzhov wrote:
> As there is no ethtool support in KNI anymore,
> PCI related information is no longer needed.
>
> Signed-off-by: Igor Ryzhov
Several other fields are unused, and are removed by a patch I sent
a week ago.
On Thu, 6 Jun 2019 16:26:26 +0300
Igor Ryzhov wrote:
> As there is no ethtool support in KNI anymore,
> PCI related information is no longer needed.
>
> Signed-off-by: Igor Ryzhov
See http://patchwork.dpdk.org/patch/54627/
> +
> +static void
> +ntb_dev_info_get(struct rte_rawdev *dev, rte_rawdev_obj_t dev_info)
> +{
> + struct ntb_hw *hw = dev->dev_private;
> + struct ntb_attr *ntb_attrs = dev_info;
> +
> + strncpy(ntb_attrs[NTB_TOPO_ID].name, NTB_TOPO_NAME,
> NTB_ATTR_NAME_LEN);
> + switch (hw->topo)
One general comment:
Think about to use rte_read32() and rte_write32() when reading and writing
registers. Or you can define a Macro or inline function for NTB driver to touch
registers. Then you can omit lots of "(char *)hw->pci_dev->mem_resource[0].addr
" and "*((volatile uint32_t *)" like th
> -Original Message-
> From: Li, Xiaoyun
> Sent: Tuesday, June 18, 2019 10:11 AM
> To: Wu, Jingjing ; Wiles, Keith
> ; Liang,
> Cunming ; Maslekar, Omkar
> Cc: dev@dpdk.org; Li, Xiaoyun
> Subject: [PATCH v6 3/6] raw/ntb: add handshake process
>
> Add handshake process using doorbell
While testing KNI with netvsc, saw lots of places more code
could be safely removed from KNI kernel driver.
v3 - rebase to current master, add style fix patch
v2 - get rid of unnecessary padding, combine the unused field patches
Stephen Hemminger (8):
kni: don't need stubs for rx_mode or ioctl
The netdev subsystem already handles case where
network sevice does not support ioctl.
If device has no rx_mode hook it is not called.
Signed-off-by: Stephen Hemminger
---
kernel/linux/kni/kni_net.c | 19 ---
1 file changed, 19 deletions(-)
diff --git a/kernel/linux/kni/kni_net
Since kernel 2.6.28 the network subsystem has provided
dev->stats for devices to use statistics handling and is the
default if no ndo_get_stats is provided.
This allow allows for 64 bit (rather than just 32 bit)
statistics with KNI.
Signed-off-by: Stephen Hemminger
---
kernel/linux/kni/kni_dev.
netdev_alloc_skb is optimized to any alignment or setup
of skb->dev that is required. The kernel has chosen to not pad
packets on x86 (for many years), because it is faster.
Signed-off-by: Stephen Hemminger
---
kernel/linux/kni/kni_net.c | 17 +++--
1 file changed, 3 insertions(+), 1
The kni net structure only exists in driver no API/ABI.
Several fields were either totally unused or set and never used.
The fields in dev_info do need to stay in the ABI but
kernel can ignore them.
Signed-off-by: Stephen Hemminger
kni: drop unused status element
Yet another ethtool leftover.
Using void * instead of proper type is unsafe practice.
Signed-off-by: Stephen Hemminger
---
kernel/linux/kni/kni_dev.h | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/kernel/linux/kni/kni_dev.h b/kernel/linux/kni/kni_dev.h
index f3e6c3ca4efa..ceba5f73c1d9 100644
Make the KNI documentation reflect modern kernel networking.
Ifconfig has been superseded by iproute2 for 15 years.
Iproute2 is well maintained, supports current feature set.
Ethtool is no longer supported by KNI.
Tshark is a better replacement for tcpdump.
Signed-off-by: Stephen Hemminger
---
The correct thing to return if user gives a bad data
is to return -EFAULT. Logging is also discouraged because
it could be used as a DoS attack.
Signed-off-by: Stephen Hemminger
---
kernel/linux/kni/kni_misc.c | 16 +---
1 file changed, 5 insertions(+), 11 deletions(-)
diff --git a/
rte_kni does not follow standard style rules.
Noticed some extra \ line continuation etc.
Signed-off-by: Stephen Hemminger
---
lib/librte_kni/rte_kni.c | 30 +++---
1 file changed, 15 insertions(+), 15 deletions(-)
diff --git a/lib/librte_kni/rte_kni.c b/lib/librte_kni/r
Hi Fan,
> -Original Message-
> From: Zhang, Roy Fan
> Sent: Monday, June 17, 2019 3:31 PM
> To: dev@dpdk.org
> Cc: akhil.go...@nxp.com; De Lara Guarch, Pablo
> ; Zhang, Roy Fan
>
> Subject: [PATCH v2] aesni_mb: fix out-of-bounds access
>
> This patch fixes the out-of-bounds coverity issu
From: Stephen Hemminger
The mp_client crashes if run on Azure or any system where ethdev
ports are owned. This is because the flush loop is confusing the
index into the array of ports[] with the port id.
For example if the server has Ports 3 and 5. Then calling
rte_eth_tx_buffer_flush on any oth
From: Stephen Hemminger
The mp_client crashes if run on Azure or any system where ethdev
ports are owned. This is because the flush loop is confusing the
index into the array of ports[] with the port id.
For example if the server has Ports 3 and 5. Then calling
rte_eth_tx_buffer_flush on any oth
On Tue, Jun 18, 2019 at 6:41 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> From: Stephen Hemminger
>
> The mp_client crashes if run on Azure or any system where ethdev
> ports are owned. This is because the flush loop is confusing the
> index into the array of ports[] with the port
> -Original Message-
> From: Ola Liljedahl [mailto:ola.liljed...@arm.com]
> Sent: Saturday, June 15, 2019 4:00 PM
> To: Honnappa Nagarahalli ; Richardson,
> Bruce ; Eads, Gage ;
> dev@dpdk.org
> Cc: nd
> Subject: Re: [RFC,v2] lfring: lock-free ring buffer
>
> On Wed, 2019-06-05 at 18:21
> -Original Message-
> From: Andrew Rybchenko [mailto:arybche...@solarflare.com]
> Sent: Monday, May 13, 2019 7:22 AM
> To: Olivier Matz ; Eads, Gage
>
> Cc: dev@dpdk.org
> Subject: Re: ***Spam*** Re: Mempool handler ops index allocation issue
>
> Hi Olivier,
>
> On 5/13/19 3:14 PM, Ol
The mempool library assigns handler ops indexes based on the dynamic load
order of mempool handlers. Indexes are used so a mempool can be used by
multiple processes, but this only works if all processes agree on the
mapping from index to mempool handler.
When using the '-d' argument, it's possible
The rte_eth_dev_close() function now handles freeing resources for
devices (e.g., mac_addrs). To conform with the new close() behaviour we
are asserting the RTE_ETH_DEV_CLOSE_REMOVE flag so that
rte_eth_dev_close() releases all device level dynamic memory.
Second level memory allocated to each in
18/06/2019 18:55, Pattan, Reshma:
>
> > -Original Message-
> > From: Thomas Monjalon [mailto:tho...@monjalon.net]
>
> > I see some errors with GCC:
>
> On which GCC version the errors are shown?
GCC 8.3
Given the type of the errors I think it should be visible
with older versions.
> >
Clang complains about the following issues:
(1) unclear meaning of =- in assignment of pktgen.verbose
(2) unused function sct(), which is also defined in cli_cmds.c
(3) typo in include guard for rte_lua_vec.h
(4) duplicate inline due to expansion of __rte_always_inline
Tested the patch on both cla
> On Jun 18, 2019, at 3:58 PM, Adam Drescher wrote:
>
> Clang complains about the following issues:
> (1) unclear meaning of =- in assignment of pktgen.verbose
> (2) unused function sct(), which is also defined in cli_cmds.c
> (3) typo in include guard for rte_lua_vec.h
> (4) duplicate inline
Hi, xiaolong
> -Original Message-
> From: Ye, Xiaolong
> Sent: Tuesday, June 18, 2019 5:41 PM
> To: Yang, Qiming
> Cc: dev@dpdk.org; Zhao1, Wei
> Subject: Re: [dpdk-dev] [PATCH v2 1/3] net/ice: enable switch filter
>
> On 06/12, Qiming Yang wrote:
> >From: wei zhao
> >
> >The patch e
Hi, Ferruh
I will fix this in next version.
Thanks
-Original Message-
From: Yigit, Ferruh
Sent: Tuesday, June 18, 2019 8:40 PM
To: Pei, Andy ; dev@dpdk.org
Cc: Xu, Rosen
Subject: Re: [dpdk-dev] [PATCH v2 4/4] net/ipn3ke: implementation of statistics
On 6/18/2019 12:59 PM, Ferruh Yigit
mlx5 implements mlx5_flow_null_drv_ops to be used when a specific
flow typei/driver is not available or invalid.
This routines return error without modifying the rte_flow_error
parameter passed to them which causes testpmd, for example, to crash.
This commit addresses the issue by modifying the rte
Today, it is possible to offload an interface flow rules to the
hardware using DPDK flow commands.
With mlx5 it is also possible to offload a limited set of flow rules
to the mlxsw (or E-switch) using the same DPDK flow commands.
A 'transfer' attribute was added to the flow rule creation command in
When a tbl8 group is getting attached to a tbl24 entry, lookup
might fail even though the entry is configured in the table.
For ex: consider a LPM table configured with 10.10.10.1/24.
When a new entry 10.10.10.32/28 is being added, a new tbl8
group is allocated and tbl24 entry is changed to point
When a tbl8 group is getting attached to a tbl24 entry, lookup
might fail even though the entry is configured in the table.
For ex: consider a LPM table configured with 10.10.10.1/24.
When a new entry 10.10.10.32/28 is being added, a new tbl8
group is allocated and tbl24 entry is changed to point
Tests showed that the 'inline' keyword caused performance drop
on some x86 platforms after the memory ordering patches applied.
By removing the 'inline' keyword, the performance was recovered
as before on x86 and no impact to arm64 platforms.
Suggested-by: Medvedkin Vladimir
Signed-off-by: Ruifen
Add a missing include with the defines for vhost-user driver features.
Fixes: 5fbb3941da9f ("vhost: introduce driver features related APIs")
Cc: sta...@dpdk.org
Signed-off-by: Noa Ezra
Reviewed-by: Maxime Coquelin
Reviewed-by: Matan Azrad
---
lib/librte_vhost/rte_vhost.h | 1 +
1 file changed
Add support for disabling TSO and mrg-rxbuf in vhost.
Noa Ezra (2):
net/vhost: support TSO disabling
net/vhost: support mrg-rxbuf disabling
doc/guides/nics/vhost.rst | 10 +
drivers/net/vhost/rte_eth_vhost.c | 45 ---
2 files changed, 52 in
TSO (TCP Segmentation Offload) is enabled by default on vhost.
Add the ability to disable TSO on vhost.
The user should also disable the feature on the virtual machine's xml.
Signed-off-by: Noa Ezra
Reviewed-by: Matan Azrad
---
doc/guides/nics/vhost.rst | 5 +
drivers/net/vhost/rte
Rx mergeable buffers is a virtio feature that allows chaining of
multiple virtio descriptors to handle large packet size.
This behavior is supported and enabled by default, however in case
the user knows that rx mergeable buffers are not needed, he can disable
the feature.
The user should also set
Add an API that returns queue status for requested queue in the port.
The queue's status can be changed before the user has signed for the
queue state event interrupt. In this case the user can't know the
current queue's status. This API returns the current status.
Signed-off-by: Noa Ezra
Reviewe
In some situations, when a virtual machine is starting,
vring_state_changed can be called while there was no change in the
queue state. This fix makes sure that there was really a change in the
queue state before calling the callback for EVENT_QUEUE_STATE.
Fixes: ee584e9710b9 ("vhost: add driver o
92 matches
Mail list logo
