On Wed, May 11, 2016 at 08:22:59AM -0700, Stephen Hemminger wrote:
> On Wed, 11 May 2016 19:17:58 +0530
> Hemant Agrawal wrote:
>
> > IGB_UIO not supported for arm64 arch in kernel so disable.
> >
> > Signed-off-by: Hemant Agrawal
> > Reviewed-by: Santosh Shukla
>
> Really, I have use
On Wed, May 11, 2016 at 08:22:59AM -0700, Stephen Hemminger wrote:
> On Wed, 11 May 2016 19:17:58 +0530
> Hemant Agrawal wrote:
>
> > IGB_UIO not supported for arm64 arch in kernel so disable.
> >
> > Signed-off-by: Hemant Agrawal
> > Reviewed-by: Santosh Shukla
>
> Really, I have use
This patch introduces dpaa2 machine target to address difference
in cpu parameter, number of core to 8 and no numa support
w.r.t default armv8-a machine
Signed-off-by: Hemant Agrawal
---
config/defconfig_arm64-dpaa2-linuxapp-gcc | 42 ++
mk/machine/dpaa2/rte.vars.mk
IGB_UIO not supported for arm64 arch in kernel so disable.
Signed-off-by: Hemant Agrawal
Reviewed-by: Santosh Shukla
---
config/defconfig_arm64-armv8a-linuxapp-gcc | 2 ++
1 file changed, 2 insertions(+)
diff --git a/config/defconfig_arm64-armv8a-linuxapp-gcc
On Fri, Apr 29, 2016 at 01:18:34AM +, Jianfeng Tan wrote:
> +static void
> +vdev_read_dev_config(struct virtio_hw *hw, uint64_t offset,
> + void *dst, int length)
> +{
> + int i;
> + struct virtio_user_hw *uhw = (struct virtio_user_hw *)hw->vdev_private;
Unnecessary
On 5/11/2016 8:35 AM, Alejandro Lucero wrote:
> On Tue, May 10, 2016 at 4:59 PM, Stephen Hemminger <
> stephen at networkplumber.org> wrote:
>
>> On Tue, 10 May 2016 19:21:41 +0800
>> Zhe Tao wrote:
>>
>>> Problem:
>>> The following operations will cause the igb_uio based DPDK
>>> operation
On 5/10/2016 9:02 PM, Harish Patil wrote:
> Hi Ferruh,
>
>> fix errors:
>> icc: command line warning #10006: ignoring unknown option
>> '-Wno-unused-value'
>> icc: command line warning #10006: ignoring unknown option
>> '-Wno-format-nonliteral'
>> icc: command line warning #10006: ignoring
On Fri, Apr 29, 2016 at 01:18:33AM +, Jianfeng Tan wrote:
> +static int
> +kick_one_vq(struct virtio_user_hw *hw, struct virtqueue *vq,
> + unsigned queue_sel)
Firstly, "dev" is a more common word than "hw" here. So, name the struct
to "virtio_user_device", and name the var to "dev".
Fix issue reported by Coverity.
Coverity ID 97920
munmap structure of hugepage
leaked_storage: Variable hugepage going out of scope leaks the storage
it points to.
The system resource will not be reclaimed and reused, reducing the future
availability of the resource.
In rte_eal_hugepage_init:
> -Original Message-
> From: Mrozowicz, SlawomirX
> Sent: Wednesday, May 11, 2016 10:15 AM
> To: Dumitrescu, Cristian
> Cc: dev at dpdk.org; Singh, Jasvinder
> Subject: RE: [PATCH v3] examples/qos_meter: fix unchecked return value
>
>
>
> >-Original Message-
> >From:
On 5/9/2016 5:13 PM, Olivier Matz wrote:
> From: Maxime Leroy
>
> The functions rte_log_cur_msg_loglevel() and rte_log_cur_msg_logtype()
> return the current log level/type for the message being processed. They
> are used when implementing a user-defined logging stream.
>
> The current log
This patch updates the parsing routines related to packet queues
(pktq_in/out fields in the PIPELINE section) and message queues
(msgq_in/out fields of in the MSGQ Section) specified in ip_pipeline
configuration file.
In the updated routines, function "strtok_r()" is used for parsing the
string
This patch enables rss (receive side scaling) per network interface
through the configuration file. The user can specify following
parameters in LINK section for enabling the rss feature - rss_qs,
rss_proto_ipv4, rss_proto_ipv6 and ip_proto_l2.
The "rss_qs" is mandatory parameter which indicates
Signed-off-by: Rasesh Mody
Signed-off-by: Harish Patil
---
drivers/net/bnx2x/bnx2x.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
index f7ee77e..f1bc2dd 100644
--- a/drivers/net/bnx2x/bnx2x.c
+++
- Process Tx completions based on configured Tx free threshold and
determine how much TX BDs are required before invoking bnx2x_tx_encap()
- Change bnx2x_tx_encap() to void function
Signed-off-by: Rasesh Mody
Signed-off-by: Harish Patil
---
drivers/net/bnx2x/bnx2x.c | 14
Fix stats_get() routine to display drop counters under imissed counter.
Fixes: 540a211084a7 ("bnx2x: driver core")
Signed-off-by: Rasesh Mody
Signed-off-by: Harish Patil
---
drivers/net/bnx2x/bnx2x_ethdev.c | 16
drivers/net/bnx2x/bnx2x_rxtx.c |2 ++
2 files changed,
Many drivers provide their own implementation of rte_mbuf_raw_alloc(),
duplicating the code. Introduce a new public function in rte_mbuf to
allocate a raw mbuf (uninitialized).
Signed-off-by: Olivier Matz
---
app/test-pmd/flowgen.c| 13 +
app/test-pmd/txonly.c
The cycles test was wrongly in the group of mempool perf test.
It is moved in parallel test group 1 with timer test.
Signed-off-by: Thomas Monjalon
---
app/test/autotest_data.py | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/app/test/autotest_data.py
Signed-off-by: Thomas Monjalon
---
app/test/test_lpm6.c| 2 --
app/test/test_memcpy.c | 15 ---
app/test/test_mempool.c | 2 --
app/test/test_ring.c| 2 --
4 files changed, 21 deletions(-)
diff --git a/app/test/test_lpm6.c b/app/test/test_lpm6.c
index 9545982..458a10b
Adjust memory parameter (--socket-mem) of the autotests
to be able to make fast_test in a constrained environment (e.g. a VM).
Signed-off-by: Thomas Monjalon
---
app/test/autotest_data.py | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git
The autotests are too long to be run often.
This patch reduces the needed time of some tests in fast_test.
Signed-off-by: Thomas Monjalon
---
app/test/autotest_test_funcs.py | 14 ++--
app/test/test_alarm.c | 48 -
app/test/test_hash.c
In order to make autotests easy to run often,
the time and memory constraints are reduced.
These patches depend on the LPM autotest split.
The current status in a small VM is below:
> make fast_test
/root/dpdk/build/app/test -c f -n 4
Test name Test result
On Fri, Apr 29, 2016 at 01:18:29AM +, Jianfeng Tan wrote:
> This patch is to move phys addr check from virtio_dev_queue_setup
> to pci ops. To makt that happen, make sure virtio_ops.setup_queue
> return the result if we pass through the check.
>
> Signed-off-by: Huawei Xie
> Signed-off-by:
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen Hemminger
> Sent: Wednesday, May 11, 2016 4:21 PM
> To: Yang, Ziye
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] pci: Add the class_id support in pci probe
>
> On Wed, 11 May 2016 14:08:15
When submitting an update to a single patch in a large patchset, what's the
preferred method? Do I send a single "[PATCH v2 01/40]" email with the
modified patch or do I send the entire series of 40 patches again?
--
Stephen Hurd
Principal Engineer - Software Development
Broadcom Corporation
The LPM performance tests should not be mixed with
functional tests: they are long and show some statistics.
LPM6 functional tests are added to parallel tests.
Signed-off-by: Thomas Monjalon
---
v2: move LPM6 autotest in parallel tests
app/test/Makefile | 2 +
This patch is used to add the class_id (class_code,
subclass_code, programming_interface) support for
pci_device probe. With this patch, it will be
flexible for users to probe a class of devices
by class_id.
Signed-off-by: Ziye Yang
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 4
> -Original Message-
> From: Yang, Ziye
> Sent: Wednesday, May 11, 2016 2:39 PM
> To: Richardson, Bruce
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] librte_eal: fix wrong args operation in
> eal_parse_args
>
> Hi Bruce,
>
> Could you tell me the documentation file name?
Hi Bruce,
Could you tell me the documentation file name? Then I can conduct the following
documentation relate d patch.
-Original Message-
From: Richardson, Bruce
Sent: Wednesday, May 11, 2016 8:21 PM
To: Yang, Ziye
Cc: dev at dpdk.org
Subject: RE: [dpdk-dev] [PATCH] librte_eal: fix
This patch is used to fix wrong operation on user
input args. eal_parse_args function should not operate
the args passed by the user. If the element in argv
is generated by malloc function, changing it will cause
memory issues when free the args.
Signed-off-by: Ziye Yang
---
From: Zyta Szpak
rte_eth_dev_get_reg_length and rte_eth_dev_get_reg callbacks
do not provide register size to the app in any way. Example assuming
they are 32-bit wide always allocates not enough memory if the
registers are 64-bit wide. It results in memory corruption.
On Wed, May 11, 2016 at 01:28:21PM +0800, Ziye Yang wrote:
> This patch is used to fix wrong operation on user
> input args. eal_parse_args function should not operate
> the args passed by the user. If the element in argv
> is generated by malloc function, changing it will cause
> memory issues
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jerin Jacob
> Sent: Saturday, May 7, 2016 4:16 PM
> To: dev at dpdk.org
> Cc: thomas.monjalon at 6wind.com; Richardson, Bruce
> ; Jerin Jacob
> ; Maciej Czekaj
> ; Kamil Rytarowski
> ; Zyta Szpak
> ; Slawomir
> -Original Message-
> From: Yang, Ziye
> Sent: Wednesday, May 11, 2016 12:51 PM
> To: Richardson, Bruce
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] librte_eal: fix wrong args operation in
> eal_parse_args
>
> Hi Bruce,
>
> Could it be fixed later? If not, it should be
Hi All,
I am using DPDK 16.04 and pktgen 3.0.00
Issue is, Packet reaches to wrong queue at DPDK, if source IP of a packet
changed at pktgen-dpdk.
On my DPDK machine, i have configured RSS rule just for Destination IP
(172.10.10.2).
[image: rss]
Now that the mempool objects are chained into a list, we can use it to
browse them. This implies a rework of rte_mempool_obj_iter() API, that
does not need to take as many arguments as before. The previous function
is kept as a private function, and renamed in this commit. It will be
removed in a
Hi Lazaros,
Sorry for the late review. Please find some comments,
in addition to what Konstantin already said.
On 04/04/2016 05:43 PM, Lazaros Koromilas wrote:
> --- a/app/test/test_mempool.c
> +++ b/app/test/test_mempool.c
> @@ -79,6 +79,7 @@
>
> static struct rte_mempool *mp;
> static
Hi Bruce,
Could it be fixed later? If not, it should be documented. I faced this issued
today, and found that dpdk changed my last arg. In my mind, dpdk should not
change the argv[last], which will confuse the users.
Best Regards,
Ziye Yang
-Original Message-
From: Richardson, Bruce
The perf test lists were not consistent.
Add timer perf test to the list.
Add LPM6 test to fast tests.
And fix the phony target.
Signed-off-by: Thomas Monjalon
---
mk/rte.sdktest.mk | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/mk/rte.sdktest.mk
We can define any custom whitelist of tests to run.
The predefined lists for ring and mempool tests are not quite useful.
Signed-off-by: Thomas Monjalon
---
v2: completely remove targets
mk/rte.sdktest.mk | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/mk/rte.sdktest.mk
Ok.
So if you build DPDK from the sources, you would have an 'app' directory
in you build directory.
You can read the Getting Started Guide for more info:
http://dpdk.readthedocs.io/en/v16.04/linux_gsg/index.html
Once you ran the 'test' DPDK application (located inside the 'app'
directory) you
We can define any custom whitelist of tests to run.
The predefined lists for ring and mempool tests are not quite useful.
Signed-off-by: Thomas Monjalon
---
mk/rte.sdktest.mk | 2 --
1 file changed, 2 deletions(-)
diff --git a/mk/rte.sdktest.mk b/mk/rte.sdktest.mk
index 5b30ec8..c448303 100644
On Wed, 11 May 2016 22:32:16 +0530
Jerin Jacob wrote:
> On Wed, May 11, 2016 at 08:22:59AM -0700, Stephen Hemminger wrote:
> > On Wed, 11 May 2016 19:17:58 +0530
> > Hemant Agrawal wrote:
> >
> > > IGB_UIO not supported for arm64 arch in kernel so disable.
> > >
> > > Signed-off-by: Hemant
Fix issue reported by Coverity.
Coverity ID 30690: Bad bit shift operation
large_shift: In expression 1ULL << i, left shifting by more than 63 bits
has undefined behavior. The shift amount, i, is as much as 127.
Fixes: de3cfa2c9823 ("sched: initial import")
Signed-off-by: Slawomir Mrozowicz
On 5/10/2016 6:18 PM, Dumitrescu, Cristian wrote:
>
>
>> -Original Message-
>> From: Mrzyglod, DanielX T
>> Sent: Tuesday, May 10, 2016 11:11 AM
>> To: Dumitrescu, Cristian
>> Cc: dev at dpdk.org; Mrzyglod, DanielX T
>> Subject: [PATCH] sched: fix useless call
>>
>> Fix issue reported
On 10 May 2016 at 22:49, Bruce Richardson wrote:
> On Fri, May 06, 2016 at 11:55:46AM +0530, Jianbo Liu wrote:
>> use ARM NEON intrinsic to implement ixgbe vPMD
>>
>> Signed-off-by: Jianbo Liu
>> ---
>> drivers/net/ixgbe/Makefile | 4 +
>> drivers/net/ixgbe/ixgbe_rxtx_vec_neon.c
> -Original Message-
> From: Mrozowicz, SlawomirX
> Sent: Wednesday, May 11, 2016 9:48 AM
> To: Dumitrescu, Cristian
> Cc: dev at dpdk.org; Singh, Jasvinder ;
> Mrozowicz, SlawomirX
> Subject: [PATCH v4] examples/qos_sched: fix bad bit shift operation
>
> Fix issue reported by
CC-ing Tomasz, who is the original author of RED implementation and its
autotest. Tomasz, what do you think?
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, May 11, 2016 8:45 AM
> To: Dumitrescu, Cristian
> Cc: dev at dpdk.org
>
The autotest for librte_sched red takes more than a minute.
Would it be possible to reduce it to a second please?
If it is really impossible, it must be removed from fast_test.
To be consistent with the naming for ARM NEON implementation,
ixgbe_rxtx_vec.c is renamed to ixgbe_rxtx_vec_sse.c.
Signed-off-by: Jianbo Liu
---
drivers/net/ixgbe/Makefile | 2 +-
drivers/net/ixgbe/{ixgbe_rxtx_vec.c => ixgbe_rxtx_vec_sse.c} | 0
2 files
>-Original Message-
>From: Dumitrescu, Cristian
>Sent: Tuesday, May 10, 2016 7:42 PM
>To: Mrozowicz, SlawomirX
>Cc: dev at dpdk.org; Singh, Jasvinder
>Subject: RE: [PATCH v3] examples/qos_meter: fix unchecked return value
>
>
>
>> -Original Message-
>> From: Mrozowicz,
On 05/07/2016 06:16 PM, Jerin Jacob wrote:
> diff --git a/drivers/net/thunderx/Makefile b/drivers/net/thunderx/Makefile
> new file mode 100644
> index 000..69bb750
> --- /dev/null
> +++ b/drivers/net/thunderx/Makefile
> @@ -0,0 +1,64 @@
> +# BSD LICENSE
> +#
> +# Copyright(c) 2016 Cavium
On Tue, May 10, 2016 at 4:59 PM, Stephen Hemminger <
stephen at networkplumber.org> wrote:
> On Tue, 10 May 2016 19:21:41 +0800
> Zhe Tao wrote:
>
> > Problem:
> > The following operations will cause the igb_uio based DPDK
> > operation failed.
> > --Any device assignment through the
On Wed, 11 May 2016 19:17:58 +0530
Hemant Agrawal wrote:
> IGB_UIO not supported for arm64 arch in kernel so disable.
>
> Signed-off-by: Hemant Agrawal
> Reviewed-by: Santosh Shukla
Really, I have use IGB_UIO on ARM64
On Wed, 11 May 2016 14:08:15 +0800
Ziye Yang wrote:
> This patch is used to add the class_id (class_code,
> subclass_code, programming_interface) support for
> pci_device probe. With this patch, it will be
> flexible for users to probe a class of devices
> by class_id.
>
> Signed-off-by: Ziye
On 05/06/2016 10:25 PM, Stephen Hurd wrote:
> Initial skeleton simply fails init.
> Add nic guide and tie into build system.
>
> Signed-off-by: Stephen Hurd
> ---
[...]
> diff --git a/drivers/net/bnxt/Makefile b/drivers/net/bnxt/Makefile
> new file mode 100644
> index 000..89b4a27
> ---
>On Tue, May 03, 2016 at 10:38:20PM -0700, Rasesh Mody wrote:
>> Change the Tx routine logic to ring the doorbell once per burst and not
>> on every Tx packet. This driver-level optimization is necessary to
>>achieve
>> line rates for larger frame sizes (1k or more).
>>
>> Fixes: 540a211084a7
> From: Remy Horton [mailto:remy.horton at intel.com]
> Sent: Tuesday, May 10, 2016 3:27 AM
>
>
> On 07/05/2016 01:14, Rasesh Mody wrote:
> [..]
> >>> Enhance the stats_get() routine to display drop counters under
> >>> imissed counter.
> >>> Added extended stats get support to provide
58 matches
Mail list logo