On Mon, May 01, 2017 at 02:05:42PM -0700, Yongseok Koh wrote:
> If mlx5_dev_start() fails, it tries to rollback data structures related to
> rte_flow including drop queue. The destruction code doesn't assume the
> structures are created but priv_flow_delete_drop_queue() never does sanity
> check. T
-Original Message-
> Date: Mon, 1 May 2017 22:59:53 -0700
> From: Ashwin Sekhar T K
> To: byron.mar...@intel.com, pablo.de.lara.gua...@intel.com,
> jerin.ja...@caviumnetworks.com, jianbo@linaro.org
> Cc: dev@dpdk.org, Ashwin Sekhar T K
> Subject: [dpdk-dev] [PATCH v3] efd: support l
On Tue, Apr 25, 2017 at 12:28:46AM -0700, Rasesh Mody wrote:
> From: Rasesh Mody
>
> Changes included in this fix
> - limit CFLAGS to base files
> - fix to remove/mark unused members
> - add checks for debug config option
> - make qede_set_mtu() and qede_udp_dst_port_del() static and others
>
* Added file lib/librte_efd/rte_efd_arm64.h to hold arm64
specific definitions
* Verified the changes with efd_autotest unit test case
Signed-off-by: Ashwin Sekhar T K
---
v2:
* Slightly modified the content of the commit message body
* Added prefix [dpdk-dev] to the email subject line
v3:
* M
Since DPDK has only two timer sources,
Avoid &eal_timer_source memory read and followed
by the switch case statement when
RTE_LIBEAL_USE_HPET is not defined.
Signed-off-by: Jerin Jacob
---
v2:
- Rearrage the ifdef to avoiding the #else part by
"unifdef" the first occurence of rte_get_tsc_cycles(T
This problem is caused by a missing set of E1000_RXCSUM_CRCOFL
in eth_igb_rx_init(), it should be set to enable SCTP packet
L4 checksum.If it is not set, the printf message in cksum fwd
about L4 SCTP cksum flag is error.
Fixes: d15fcf76c8b7 ("net/e1000: move to drivers/net/")
Signed-off-by: Wei Z
Hi Thomas,
-Original Message-
From: Thomas Monjalon [mailto:tho...@monjalon.net]
Sent: Monday, May 1, 2017 9:56 PM
To: Pei, Yulong
Cc: dev@dpdk.org; Wu, Jingjing
Subject: Re: [dpdk-dev] [PATCH] app/testpmd: add bus info when display port info
26/04/2017 06:31, Yulong Pei:
> Display pc
On 28 April 2017 at 18:38, Sekhar, Ashwin wrote:
> On Friday 28 April 2017 03:36 PM, Jianbo Liu wrote:
>> On 27 April 2017 at 20:44, Ashwin Sekhar T K
>> wrote:
>>> * Added file lib/librte_efd/rte_efd_arm64.h to hold arm64
>>> specific definitions
>>> * Verified the changes with efd_autotest un
> > > >
> > > > int bnxt_link_update_op(struct rte_eth_dev *eth_dev, int
> > > > wait_to_complete) diff --git a/drivers/net/e1000/base/e1000_api.c
> > > > b/drivers/net/e1000/base/e1000_api.c
> > > > index f7cf83b..dcb53f7 100644
> > > > --- a/drivers/net/e1000/base/e1000_api.c
> > > > +++ b/drive
When counting max packet length from MTU, count
VLAN tag length twice for QinQ packets.
Signed-off-by: Wenzhuo Lu
Acked-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c| 3 +--
drivers/net/i40e/i40e_ethdev.h| 7 +++
drivers/net/i40e/i40e_ethdev_vf.c | 3 +--
3 files changed, 9 ins
Hi Ferruh,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Friday, April 28, 2017 8:38 PM
> To: Wu, Jingjing; Lu, Wenzhuo; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] net/i40e: consider QinQ when setting MTU
>
> On 4/28/2017 9:27 AM, Wu, Jingjing wrote:
> >
> >
> >> -Original M
Hi Wei,
> -Original Message-
> From: Dai, Wei
> Sent: Saturday, April 29, 2017 2:09 PM
> To: Lu, Wenzhuo; thomas.monja...@6wind.com; harish.pa...@cavium.com;
> rasesh.m...@cavium.com; stephen.h...@broadcom.com;
> ajit.khapa...@broadcom.com; Zhang, Helin; Ananyev, Konstantin; Wu,
> Jingjin
Hi,
> -Original Message-
> From: Jia Yu [mailto:j...@vmware.com]
> Sent: Friday, April 28, 2017 2:12 PM
> To: Lu, Wenzhuo; Ananyev, Konstantin
> Cc: dev@dpdk.org; Jia Yu
> Subject: [PATCH] ixgbe: check scatter_enable in set_mtu
>
> There exists case that software sets mtu (i.e jumbo fram
A new DPDK release candidate is ready for testing:
http://dpdk.org/browse/dpdk/tag/?id=v17.05-rc3
It is a quite big RC3 (125 patches) although thankfully smaller than RC2.
It was planned to release almost on time, i.e. this week.
Unfortunately it seems there are still some important bugs
These lists were unused and useless because they are maintained per bus:
struct rte_driver_list dev_driver_list
struct rte_device_list dev_device_list
Signed-off-by: Thomas Monjalon
---
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 2 --
lib/librte_eal/common/eal_common_dev.
22/02/2017 03:36, Yuanhan Liu:
> On Mon, Feb 20, 2017 at 10:04:47PM +0800, hanxue...@126.com wrote:
> > From: Huanle Han
> >
> > rte_eal_dev_detach() didn't remove dev from dev_device_list
> > after free the dev. So the following attached dev wrote to
> > the freed memory (tailq entry of previous
01/05/2017 08:29, gowrishankar muthukrishnan:
> On Sunday 30 April 2017 07:41 PM, Thomas Monjalon wrote:
> > These differences in Python 3 were causing errors:
> > - xrange is replaced by range
> > - dict values are a view (instead of list)
> > - has_key is removed
> Thanks Thomas.
>
> Please incl
01/05/2017 21:09, Mark Asselstine:
> The existing code used to search for module files via modinfo has
> several corner cases which can result in it failing where it should be
> successful.
>
> The call to lower() would cause results returned by 'modinfo' to be
> forced to lowercase, results which
27/04/2017 16:42, Michal Jastrzebski:
> Kuba Kozak (6):
> ethdev: revert patches extending xstats API in ethdev
> ethdev: retrieve xstats by ID
> ethdev: get xstats ID by name
> app/proc-info: support xstats by ID and by name
> net/e1000: support xstats by ID
> net/ixgbe: support xstats
01/05/2017 23:34, Thomas Monjalon:
> I am doing these small changes in your patches:
More explanations below
> --- a/drivers/net/e1000/igb_ethdev.c
> +++ b/drivers/net/e1000/igb_ethdev.c
> @@ -395,7 +395,7 @@ static const struct eth_dev_ops eth_igb_ops = {
> .link_update = eth_ig
27/04/2017 16:42, Michal Jastrzebski:
> From: Kuba Kozak
>
> This patchset fixes following patches:
>
> commit ea85e7d711b6 ("ethdev: retrieve xstats by ID")
> commit a954495245c4 ("ethdev: get xstats ID by name")
> commit 1223608adb9b ("app/proc-info: support xstats by ID")
> co
If mlx5_dev_start() fails, it tries to rollback data structures related to
rte_flow including drop queue. The destruction code doesn't assume the
structures are created but priv_flow_delete_drop_queue() never does sanity
check. This can cause a crash.
Fixes: 028761059aeb ("net/mlx5: use an RSS dro
> > Stop port before enabling QinQ.
> > Add commands to set inner and outer TPID's and start port.
> > Remove TPID's from flow validate and and flow create commands.
> >
> > Signed-off-by: Bernard Iremonger
>
> Acked-by: John McNamara
Applied, thanks
28/04/2017 09:43, Ferruh Yigit:
> Coverity issue: 1428042
> Fixes: 1131cbf0fb2b ("net/ark: stub PMD for Atomic Rules Arkville")
>
> Signed-off-by: Ferruh Yigit
Acked-by: John Miller
Applied, thanks
27/04/2017 15:51, Pascal Mazon:
> Rx queues configured with more than 1023 descriptors cause readv() calls to
> fail due to more iovec entries than permitted by the kernel. As a result,
> no packets can be received.
>
> Quietly limit internal Rx queue size to the maximum number of iovec entries
>
01/05/2017 20:41, Jerin Jacob:
> RBDR buffers are refilled when SW consumes the buffers from CQ.
> This creates deadlock case when CQ buffers exhausted due to lack
> of RBDR buffers. Fix is to refill the RBDR when rx_free_thresh
> meet, irrespective of the number of CQ buffers consumed.
>
> Fixes:
01/05/2017 22:07, Patil, Harish:
> Hi Remy,
> Have a small suggestion here.
> Since testpmd uses new libraries of librte_latencystats and
> librte_bitratestats it hurts packet processing performance.
> Many users who use testpmd to do the initial performance benchmarks may
> not be aware of such a
01/05/2017 22:10, Thomas Monjalon:
> From: Markos Chandras
>
> When RTE_DEVEL_BUILD is unset, -rpath is unset.
> So the ethtool app cannot link with ixgbe shared library
> which is required by ethtool lib:
>
> warning: librte_pmd_ixgbe.so.1, needed by
> examples/ethtool/lib/x86_64-native-linuxap
From: Markos Chandras
When RTE_DEVEL_BUILD is unset, -rpath is unset.
So the ethtool app cannot link with ixgbe shared library
which is required by ethtool lib:
warning: librte_pmd_ixgbe.so.1, needed by
examples/ethtool/lib/x86_64-native-linuxapp-gcc/lib/librte_ethtool.so,
not found (try using -
>From: Remy Horton
>
>Bit-rate collation should only be done by one core. This patch adds
>an option to select which core performs the bit-rate calculation,
>which is also disabled by default.
>
>Fixes: 7e4441c8efb9 ("app/testpmd: add bitrate statistics calculation")
>
>Signed-off-by: Remy Horto
01/05/2017 21:12, Jerin Jacob:
> Since DPDK has only two timer sources,
> Avoid &eal_timer_source memory read and followed
> by the switch case statement when
> RTE_LIBEAL_USE_HPET is not defined.
[...]
> static inline uint64_t
> rte_get_timer_cycles(void)
> {
> +#ifdef RTE_LIBEAL_USE_HPET
>
Since DPDK has only two timer sources,
Avoid &eal_timer_source memory read and followed
by the switch case statement when
RTE_LIBEAL_USE_HPET is not defined.
Signed-off-by: Jerin Jacob
---
lib/librte_eal/common/include/generic/rte_cycles.h | 12
1 file changed, 8 insertions(+), 4 de
The existing code used to search for module files via modinfo has
several corner cases which can result in it failing where it should be
successful.
The call to lower() would cause results returned by 'modinfo' to be
forced to lowercase, results which were subsequently passed to
exists() which is
RBDR buffers are refilled when SW consumes the buffers from CQ.
This creates deadlock case when CQ buffers exhausted due to lack
of RBDR buffers. Fix is to refill the RBDR when rx_free_thresh
meet, irrespective of the number of CQ buffers consumed.
Fixes: e2d7fc9f0a24 ("net/thunderx: add single an
On Monday, May 1, 2017 6:29:18 PM EDT Thomas Monjalon wrote:
> 01/05/2017 17:33, Mark Asselstine:
> > On Monday, May 1, 2017 11:24:13 AM EDT Mark Asselstine wrote:
> > > On Friday, April 28, 2017 11:38:17 AM EDT Thomas Monjalon wrote:
> > > > 25/11/2016 04:16, alloc:
> > > > > If the module path ha
On Monday, May 1, 2017 11:24:13 AM EDT Mark Asselstine wrote:
> On Friday, April 28, 2017 11:38:17 AM EDT Thomas Monjalon wrote:
> > 25/11/2016 04:16, alloc:
> > > If the module path has upper case chars, the dpdk-devbind.py script will
> > > crunch them to lower case. This will result in the scri
On Friday, April 28, 2017 11:38:17 AM EDT Thomas Monjalon wrote:
> 25/11/2016 04:16, alloc:
> > If the module path has upper case chars, the dpdk-devbind.py script will
> > crunch them to lower case. This will result in the script never
> > finding a module.
>
> I wonder why this "lower" was done
01/05/2017 17:33, Mark Asselstine:
> On Monday, May 1, 2017 11:24:13 AM EDT Mark Asselstine wrote:
> > On Friday, April 28, 2017 11:38:17 AM EDT Thomas Monjalon wrote:
> > > 25/11/2016 04:16, alloc:
> > > > If the module path has upper case chars, the dpdk-devbind.py script will
> > > > crunch them
27/04/2017 11:27, Andriy Berestovskyy:
> Port ID is not an index from 0 to n_nic_ports, but rather a value
> of nic_ports array.
>
> Signed-off-by: Andriy Berestovskyy
Fixes: af75078fece3 ("first public release")
Applied, thanks
28/04/2017 09:47, Yao, Lei A:
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> >
> > L3fwd power app monitors the RX queues to see if the polling frequency
> > should be adjusted (the busier the queue, the higher the frequency).
> > The app uses several thresholds in the ring
Coverity issue: 1428042
Fixes: 1131cbf0fb2b ("net/ark: stub PMD for Atomic Rules Arkville")
Signed-off-by: Ferruh Yigit
Acked-by: John Miller mailto:john.mil...@atomicrules.com>>
28/04/2017 12:47, Jingjing Wu:
> If the number of rx queues is zero, it is meaningless to enable
> rx interrupt. This patch fixes it.
>
> Fixes: aee3bc79cc34 ("examples/l3fwd-power: enable one-shot Rx interrupt and
> polling switch")
> Cc: sta...@dpdk.org
> Signed-off-by: Jingjing Wu
Applied, t
27/04/2017 14:37, Remy Horton:
> The l2fwd-keepalive example has infinite processing loops and as a
> result the only way to exit it is via SIGINT/SIGTERM (e.g. Control-C).
> The resulting shutdown is unclean, in particular leaving stale shared
> host memory handles. This is fixed by adding a signa
23/02/2017 03:00, Zhiyong Yang:
> To miss updating the variable prev_tsc in function l2fwd_main_loop()
> The patch fixes it.
>
> Fixes:e2366e74e029("examples: use buffered Tx")
> CC: sergio.gonzalez.mon...@intel.com
>
> Signed-off-by: Zhiyong Yang
Applied, thanks
28/04/2017 14:38, De Lara Guarch, Pablo:
> From: Bruce Richardson
> >
> > While later releases in the FreeBSD 10 series have a CPU_COUNT macro
> > defined, FreeBSD 10.0 and 10.1 do not have this macro. Therefore we
> > provide
> > a basic fallback implementation of the macro for platforms where it
31/03/2017 21:13, Allain Legacy:
> Enhancing test-pmd to be able to load CLI commands from file both at
> runtime and at startup. Usually the number of commands to run is small
> enough that a simple cut-n-paste does the trick, but while testing the flow
> API it was necessary to load several hund
26/04/2017 06:31, Yulong Pei:
> Display pci device bus info when show port info.
>
> Signed-off-by: Yulong Pei
> ---
> --- a/app/test-pmd/config.c
> +++ b/app/test-pmd/config.c
> @@ -469,6 +469,14 @@ port_infos_display(portid_t port_id)
> print_ethaddr("MAC address: ", &mac_addr);
> p
28/04/2017 13:00, Pablo de Lara:
> From: Remy Horton
>
> Bit-rate collation should only be done by one core. This patch adds
> an option to select which core performs the bit-rate calculation,
> which is also disabled by default.
>
> Fixes: 7e4441c8efb9 ("app/testpmd: add bitrate statistics calc
28/04/2017 04:02, Tan, Jianfeng:
> From: Hu, Jiayu
> >
> > When testpmd exists, it frees the acquired resources (e.g. stop ports).
> > However, when we terminate it by Ctrl-d, testpmd exists directly without
> > releasing the resources. In this patch, we fix this exit issue.
> >
> > Fixes:af75078
28/04/2017 10:30, Wu, Jingjing:
> From: Pei, Yulong
> >
> > Previous numa_support = 0 by default, it need to add --numa to testpmd
> > command line to enable numa, so port_numa and ring_numa were initialized at
> > function launch_args_parse(), now testpmd change numa_support = 1 as
> > default,
28/04/2017 10:51, Wu, Jingjing:
> From: Olivier Matz [mailto:olivier.m...@6wind.com]
> > Since
> > commit 999b2ee0fe45 ("app/testpmd: enable NUMA support by default"),
> > testpmd is started with numa enabled by default. This highlights a floating
> > point
> > exception when started with --total-
28/04/2017 17:51, Pablo de Lara:
> http://dpdk.org/git/next/dpdk-next-crypto
Pulled, thanks Pablo
Hi there,
We have a data collector software, written in Java, that processes several
data sources in streaming mode. One of most important sources is network
packets.
In a high volume customer we have hit limitations using tcpdump (only 1
thread) and we would like to leverage dpdk for a new progr
29/04/2017 02:38, Rasesh Mody:
> Inadvertently we failed to increment tx_pkts pointer while addressing
> coverity issue, this patch fixes that.
>
> Fixes: 41b52b732ec5 ("net/qede/base: fix code flow and remove unused code")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Rasesh Mody
Squashed, thanks
28/04/2017 14:53, Ferruh Yigit:
> http://dpdk.org/git/next/dpdk-next-net
Pulled, thanks Ferruh
55 matches
Mail list logo
