2015-11-17 09:14, Simon Kagstrom:
> Otherwise mbufs will leak when the port is destroyed. The
> rte_sched_port_qbase() and rte_sched_port_qsize() functions are used
> in free now, so move them up.
>
> Signed-off-by: Simon Kagstrom
Applied, thanks
2015-10-25 18:51, Thomas Monjalon:
> 2015-09-11 08:31, Mcnamara, John:
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > > 2015-09-10 20:33, Ferruh Yigit:
> > > > Correct, those macros available after kernel version >= 2.6.34.
> > > > ...
> > >
> > > Why should we
2015-09-10 17:01, Ferruh Yigit:
> To get pci_dev and vf number from dev, benefit from
> existing macros in pci.h
>
> Signed-off-by: Ferruh Yigit
Applied, thanks
2015-11-11 13:59, Polehn, Mike A:
> Adds ethdev driver prefetch of variable structure to CPU cache 0
> while calling into tx or rx device driver operation.
>
> RFC 2544 test of NIC task test measurement points show improvement
> of lower latency and/or better packet throughput indicating clock
>
On 24 November 2015 at 13:53, Panu Matilainen wrote:
> On 11/24/2015 05:30 PM, Martinx - ? wrote:
>>
>> On 24 November 2015 at 13:22, Panu Matilainen wrote:
>>>
>>> On 11/24/2015 04:46 PM, Sergio Gonzalez Monroy wrote:
On 24/11/2015 13:57, Panu Matilainen wrote:
>
>
Any review, please?
2015-11-13 09:30, Tom Kiely:
> If all rx descriptors are processed while transient
> mbuf exhaustion is present, the rx ring ends up with
> no available descriptors. Thus no packets are received
> on that ring. Since descriptor refill is performed post
> rx descriptor
2015-11-17 15:19, Zhigang Lu:
> Without it, compiling error occurs when CONFIG_RTE_BUILD_SHARED_LIB
> is enabled.
>
> Reported-by: Guo Xin
> Signed-off-by: Zhigang Lu
Applied, thanks
2015-11-22 18:28, Stephen Hemminger:
> On Sun, 22 Nov 2015 14:13:35 -0500
> Zhihong Wang wrote:
>
> > The kernel fills new allocated (huge) pages with zeros.
> > DPDK just has to populate page tables to trigger the allocation.
> >
> > Signed-off-by: Zhihong Wang
>
> Nice, especially on slow
2015-11-24 16:59, Pablo de Lara:
> Fixes following error when Ubuntu 12.04 uses kernel 3.13.0-30-generic,
> since skb_set_hash() is implemented in the kernel from 3.13.0-30,
> which is declared as UBUNTU_KERNEL_VERSION(3,13,0,30,0) and not
> UBUNTU_KERNEL_VERSION(3,13,0,30,54)
>
> In file
2015-11-24 09:43, Victor Kaplansky:
> Yuanhan, looks fine. Thanks for handling this.
> -- Victor
>
> On Tue, Nov 24, 2015 at 03:25:35PM +0800, Yuanhan Liu wrote:
> > From: Victor Kaplansky
> >
> > The VHOST_USER_SET_VRING_ENABLE request was sent for each queue-pair.
> > However, it's changed to
2015-11-24 15:15, Yuanhan Liu:
> On Tue, Nov 24, 2015 at 03:45:35PM +0900, Tetsuya Mukawa wrote:
> > The patch fixes reset_owner message handling not to clear callfd,
> > because callfd will be valid while connection is establihed.
> >
> > Signed-off-by: Tetsuya Mukawa
>
> Acked-by: Yuanhan Liu
2015-11-12 14:06, Jianfeng Tan:
> This patch fixes a bug under lower version linux kernel, mmap()
> fails when length is not aligned with hugepage size. mmap()
> without flag of MAP_ANONYMOUS, should be called with length
> argument aligned with hugepagesz at older longterm version
> Linux, like
2015-11-24 17:49, Mcnamara, John:
> Very useful. We should add these scripts to the contributing guide at some
> point.
Yes
> One minor nit. "q" is used for "quick" in this script and "quiet" in the
> checkpatch script. Maybe for consistency use "f" for fast or "n" for no-doc
> or something
> > Problem:if I firstly insert my kmod_test.ko, then insert eventfd_link.ko,
> > error will happen with hint " Device or resource busy". This is because
> > the default minor device number, 0, has been occupied by my kmod_test.ko .
> >
> > root at distro:~/test$ lsmod
> > Module
2015-11-19 17:44, Ferruh Yigit:
> examples/tep_termination/main.c| 2 +-
> examples/vhost/main.c | 2 +-
> examples/vhost_xen/main.c | 2 +-
> lib/librte_eal/common/eal_thread.h | 6 ++
It cannot compile
On Tue, Nov 24, 2015 at 06:54:15PM +0100, Thomas Monjalon wrote:
> 2015-11-19 17:44, Ferruh Yigit:
> > examples/tep_termination/main.c| 2 +-
> > examples/vhost/main.c | 2 +-
> > examples/vhost_xen/main.c | 2 +-
> >
Thanks for the idea
2015-11-06 13:48, Ferruh Yigit:
> post-merge: Same script as "post-commit", but triggered after git pull
A symbolic link would be sufficient.
As it must be installed in the git hooks (probably making a symbolic link),
duplicating the file is not needed at all.
> pre-commit:
On Tue, Nov 24, 2015 at 06:44:36PM +0100, Thomas Monjalon wrote:
> Thanks for the idea
>
> 2015-11-06 13:48, Ferruh Yigit:
> > post-merge: Same script as "post-commit", but triggered after git pull
>
> A symbolic link would be sufficient.
> As it must be installed in the git hooks (probably
This helper function is used to convert port id to virtio device
pointer. To use this function, a port should be managed by vhost PMD.
After getting virtio device pointer, it can be used for calling vhost
library APIs. But some library APIs should not be called with vhost PMD.
Here is.
-
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost. It means librte_vhost is also needed to compile the PMD.
The vhost messages will be handled only when a port is started. So start
a port first, then invoke QEMU.
The PMD has 2 parameters.
- iface: The
The vhost PMD will be a wrapper of vhost library, but some of vhost
library APIs cannot be mapped to ethdev library APIs.
Becasue of this, in some cases, we still need to use vhost library APIs
for a port created by the vhost PMD.
Currently, when virtio device is created and destroyed, vhost
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost.
PATCH v5 changes:
- Rebase on latest master.
- Fix RX/TX routine to count RX/TX bytes.
- Fix RX/TX routine not to count as error packets if enqueue/dequeue
cannot send all packets.
- Fix if-condition
Hi Daniel,
So for my own curiosity: what went wrong here?
Did compiler avoid to generate a code for while {} loop?
Or something else?
BTW, it is an internal function, so instead of introducing new typedef,
we can just change the type of buf?
Let say to uint8_t *?
>From gcc manual page: " A
On 11/24/2015 05:30 PM, Martinx - ? wrote:
> On 24 November 2015 at 13:22, Panu Matilainen wrote:
>> On 11/24/2015 04:46 PM, Sergio Gonzalez Monroy wrote:
>>>
>>> On 24/11/2015 13:57, Panu Matilainen wrote:
On 11/23/2015 08:37 PM, Martinx - ? wrote:
>
> Hello!
>
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Monday, November 23, 2015 1:40 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2 2/2] scripts: add build tests
>
> ...
>
> +while getopts hj:q ARG ; do
> + case $ARG in
> +
The functions rte_eth_rx_queue_count and rte_eth_descriptor_done are
supported by very few PMDs. Therefore, it is best to check for support
for the functions in the ethdev library, so as to avoid run-time crashes
at run-time if the application goes to use those APIs. Similarly, the
port parameter
The functions for rx/tx burst, for rx_queue_count and descriptor_done in
the ethdev library all had two copies of the code. One copy in
rte_ethdev.h was inlined for performance, while a second was in
rte_ethdev.c for debugging purposes only. We can eliminate the second
copy of the functions by
This patchset performs two cleanups:
1. Four functions in ethdev.c which were enabled for debug only have been
merged into their inlined header-file counterparts. This change required that
a number of macros be renamed and moved to the header file too. The macro
changes
are in patches 1 &
revise pcap_ring.rst file to match latest ring PMD code
Signed-off-by: Bernard Iremonger
Acked-by: Bruce Richardson
---
doc/guides/nics/pcap_ring.rst | 28
1 file changed, 16 insertions(+), 12 deletions(-)
diff --git a/doc/guides/nics/pcap_ring.rst
Use command line option --vdev=eth_ring0 to create port 0.
Create two rings and five ethdevs in test_pmd_ring for ports 1 to 5.
Improve test output by adding the port number to printf statements,
and adding a printf describing each test.
Signed-off-by: Bernard Iremonger
Acked-by: Bruce
Changes in v2:
move doc changes to a seperate patch.
Bernard Iremonger (2):
app/test: create ring and ethdevs in pmd_ring_autotest
doc: revise ring-based PMD to match latest ring PMD code
app/test/test_pmd_ring.c | 360 +-
> Adrien Mazarguil (6):
> mlx4: fix possible crash when clearing device statistics
> mlx4: fix memory registration for indirect mbuf data
> mlx4: fix TX for scattered mbufs with too many segments
> mlx5: fix possible crash when clearing device statistics
> mlx5: fix memory registration
This fix is for IPv6 checksum offload error on RHEL65.
Any optimalisation above -O0 provide error in IPv6 checksum
flag "-fstrict-aliasing" is default for optimalisation above -O0.
Fixes: 2b039d5f20a3 ("net: fix build with gcc 4.4.7 and strict aliasing")
Signed-off-by: Daniel Mrzyglod
---
This fix is for IPv6 checksum offload error on RHEL65.
Any optimalisation above -O0 provide error in IPv6 checksum
flag "-fstrict-aliasing" is default for optimalisation above -O0.
The solution is to add typedef with __attribute__((__may_alias__) for uint16_t.
Step 1 : start testpmd
./testpmd -c
On 11/24/2015 04:46 PM, Sergio Gonzalez Monroy wrote:
> On 24/11/2015 13:57, Panu Matilainen wrote:
>> On 11/23/2015 08:37 PM, Martinx - ? wrote:
>>> Hello!
>>>
>>> My name is Thiago, I'm trying to compile DPDK 2.0, 2.1 and/or 2.2-rc1,
>>> on Ubuntu with Xen support but, it does not build...
Fixes following error when Ubuntu 12.04 uses kernel 3.13.0-30-generic,
since skb_set_hash() is implemented in the kernel from 3.13.0-30,
which is declared as UBUNTU_KERNEL_VERSION(3,13,0,30,0) and not
UBUNTU_KERNEL_VERSION(3,13,0,30,54)
In file included
from
On Fri, Nov 06, 2015 at 02:04:54PM +0100, Thomas Monjalon wrote:
> 2015-11-06 12:57, Bruce Richardson:
> > So, any thoughts or comments on this? There has been lots of discussion in
> > this
> > general area but nothing yet going into the release to try and improve the
> > situation.
> >
> >
On 11/20/2015 09:46 PM, Eric Kinzie wrote:
> On Tue Nov 03 11:48:57 + 2015, Ferruh Yigit wrote:
>> On Tue, Nov 03, 2015 at 01:31:45PM +0200, Panu Matilainen wrote:
>>> On 11/03/2015 01:02 PM, Ferruh Yigit wrote:
On Tue, Nov 03, 2015 at 08:48:16AM +0200, Panu Matilainen wrote:
> On
2015-11-24 15:45, Bruce Richardson:
> On Tue, Nov 24, 2015 at 04:29:12PM +0100, Thomas Monjalon wrote:
> > 2015-11-24 14:56, Bruce Richardson:
> > > On Tue, Nov 17, 2015 at 07:53:09AM -0800, Stephen Hemminger wrote:
> > > > On Tue, 17 Nov 2015 12:21:07 +
> > > > Bruce Richardson wrote:
> > >
On Mon, Nov 23, 2015 at 02:40:07AM +0100, Thomas Monjalon wrote:
> This script can be used to call checkpatch.pl from Linux with some
> custom DPDK options.
>
> The path to the original Linux script must be set in an environment
> variable. A script is added to load any configuration variables
>
The physically linked-together combined library has been an increasing
source of problems, as was predicted when library and symbol versioning
was introduced. Replace the complex and fragile construction with a
simple linker script which achieves the same without all the problems,
remove the
> -Original Message-
> From: Iremonger, Bernard
> Sent: Tuesday, November 24, 2015 4:29 PM
> To: Richardson, Bruce
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH 1/1] app/test: create ring and ethdevs in
> pmd_ring_autotest
>
> Hi Bruce,
>
> > -Original Message-
> >
2015-11-24 14:56, Bruce Richardson:
> On Tue, Nov 17, 2015 at 07:53:09AM -0800, Stephen Hemminger wrote:
> > On Tue, 17 Nov 2015 12:21:07 +
> > Bruce Richardson wrote:
> > > -static inline uint32_t
> > > +static inline int
Are we talking about this change only?
Or the move in the first patch
Hi Bruce,
> -Original Message-
> From: Richardson, Bruce
> Sent: Tuesday, November 24, 2015 4:14 PM
> To: Iremonger, Bernard
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/1] app/test: create ring and ethdevs in
> pmd_ring_autotest
>
> On Mon, Nov 23, 2015 at 03:20:46PM +,
Neil Horman writes:
> On Tue, Nov 24, 2015 at 04:31:17PM +0200, Panu Matilainen wrote:
>> The physically linked-together combined library has been an increasing
>> source of problems, as was predicted when library and symbol versioning
>> was introduced. Replace the complex and fragile
2015-11-13 18:58, Declan Doherty:
> +/**
> + * A macro that returns the physical address that points to the start of the
> + * data in the mbuf
> + *
> + * @param m
> + * The packet mbuf.
> + * @param o
> + * The offset into the data to calculate address from.
> + */
> +#define
On Mon, Nov 23, 2015 at 03:20:46PM +, Bernard Iremonger wrote:
> Use command line option --vdev=eth_ring0 to create port 0.
> Create two rings and five ethdevs in test_pmd_ring for ports 1 to 5.
> Improve test output by adding the port number to printf statements,
> and adding a printf
This is fix for GCC 4.4.7.
flag "-fstrict-aliasing" is default for optimalisation above -O0.
Fixes: 2b039d5f20a3 ("net: fix build with gcc 4.4.7 and strict aliasing")
Signed-off-by: Daniel Mrzyglod
---
lib/librte_net/rte_ip.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git
On 11/23/2015 08:37 PM, Martinx - ? wrote:
> Hello!
>
> My name is Thiago, I'm trying to compile DPDK 2.0, 2.1 and/or 2.2-rc1,
> on Ubuntu with Xen support but, it does not build...
>
> Also, initially, I'm using DPDK sources from Ubuntu APT repository
> but, it is also reproducible using
On Tue, Nov 24, 2015 at 03:47:28PM +, Mrzyglod, DanielX T wrote:
> This error fix this situation for IPv6 checksum offload error on RHEL65
> Any optimalisation above -O0 provide error in IPv6 checksum
>
>
> Step 1 : start testpmd
> ./x86_64-native-linuxapp-gcc/app/testpmd -c 0x6 -n 4 -- -i
This error fix this situation for IPv6 checksum offload error on RHEL65
Any optimalisation above -O0 provide error in IPv6 checksum
Step 1 : start testpmd
./x86_64-native-linuxapp-gcc/app/testpmd -c 0x6 -n 4 -- -i --portmask=0x3
--disable-hw-vlan --enable-rx-cksum --crc-strip --txqflags=0
On Tue, Nov 24, 2015 at 04:29:12PM +0100, Thomas Monjalon wrote:
> 2015-11-24 14:56, Bruce Richardson:
> > On Tue, Nov 17, 2015 at 07:53:09AM -0800, Stephen Hemminger wrote:
> > > On Tue, 17 Nov 2015 12:21:07 +
> > > Bruce Richardson wrote:
> > > > -static inline uint32_t
> > > > +static
The patch fixes reset_owner message handling not to clear callfd,
because callfd will be valid while connection is establihed.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_vhost/virtio-net.c | 39 +--
1 file changed, 29 insertions(+), 10 deletions(-)
diff
From: Victor Kaplansky
The VHOST_USER_SET_VRING_ENABLE request was sent for each queue-pair.
However, it's changed to be sent per queue in the queue-pair by QEMU
commit dc3db6ad ("vhost-user: start/stop all rings"). The change
is reasonable, as we send all other request per
We still have a compilation error when zlib.h is missing.
Please keep it disabled until we have a solution (probably a "configure"
script).
2015-11-23 23:44, Rasesh Mody:
> +F: doc/guides/nics/bnx2x.rst
Strictly speaking, this must be updated in the patch adding the doc.
2015-11-23 23:44, Rasesh Mody:
> Signed-off-by: Harish Patil
> Signed-off-by: Rasesh Mody
> ---
> doc/guides/nics/bnx2x.rst | 314
> +
> 1 file changed, 314 insertions(+)
> create mode 100644 doc/guides/nics/bnx2x.rst
Please add this new doc in
On Tue, Nov 24, 2015 at 03:45:35PM +0900, Tetsuya Mukawa wrote:
> The patch fixes reset_owner message handling not to clear callfd,
> because callfd will be valid while connection is establihed.
>
> Signed-off-by: Tetsuya Mukawa
Acked-by: Yuanhan Liu
Thanks.
2015-11-23 23:44, Rasesh Mody:
> --- a/drivers/net/bnx2x/bnx2x.h
> +++ b/drivers/net/bnx2x/bnx2x.h
> @@ -16,6 +16,12 @@
> #ifndef __BNX2X_H__
> #define __BNX2X_H__
>
> +#ifdef __FreeBSD__
> +#define __LITTLE_ENDIAN _LITTLE_ENDIAN
> +#define __BIG_ENDIAN_BIG_ENDIAN
> +#define __BYTE_ORDER
On Tue, Nov 17, 2015 at 07:53:09AM -0800, Stephen Hemminger wrote:
> On Tue, 17 Nov 2015 12:21:07 +
> Bruce Richardson wrote:
>
> > The functions rte_eth_rx_queue_count and rte_eth_descriptor_done are
> > supported by very few PMDs. Therefore, it is best to check for support
> > for the
2015-11-20 18:43, Rahul Lakkireddy:
> This series of patches fix some issues and recent regressions in cxgbe pmd.
>
> Patch 1 fixes a regression where queue setup for cxgbe pmd fails with
> -EINVAL because cxgbe pmd requires a min of 1024 descriptors, but most
> examples and apps initialize with
On Tue, 24 Nov 2015 16:31:17 +0200
Panu Matilainen wrote:
> The physically linked-together combined library has been an increasing
> source of problems, as was predicted when library and symbol versioning
> was introduced. Replace the complex and fragile construction with a
> simple linker
On 24/11/2015 13:57, Panu Matilainen wrote:
> On 11/23/2015 08:37 PM, Martinx - ? wrote:
>> Hello!
>>
>> My name is Thiago, I'm trying to compile DPDK 2.0, 2.1 and/or 2.2-rc1,
>> on Ubuntu with Xen support but, it does not build...
>>
>> Also, initially, I'm using DPDK sources from Ubuntu APT
On Tue, 24 Nov 2015 22:13:28 +0100
Thomas Monjalon wrote:
> 2015-11-22 18:28, Stephen Hemminger:
> > On Sun, 22 Nov 2015 14:13:35 -0500
> > Zhihong Wang wrote:
> >
> > > The kernel fills new allocated (huge) pages with zeros.
> > > DPDK just has to populate page tables to trigger the
On Tue, 24 Nov 2015 17:31:17 +0100
Daniel Mrzyglod wrote:
> This fix is for IPv6 checksum offload error on RHEL65.
> Any optimalisation above -O0 provide error in IPv6 checksum
> flag "-fstrict-aliasing" is default for optimalisation above -O0.
> The solution is to add typedef with
Anyone to review please?
2015-11-03 12:26, Chas Williams:
> Signed-off-by: Chas Williams <3chas3 at gmail.com>
> ---
> drivers/net/bnx2x/bnx2x.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
> index
> > The vector number may change during 'dev_start'. Before enabling a new
> > vector mapping, it's necessary to disable/unmap the previous setting.
> >
> > Fixes: fe685de2b1b6 ("igb: fix VF start with PF stopped")
> >
> > Reported-by: Yong Liu
> > Signed-off-by: Cunming Liang
> Acked-by:
On Fri, Nov 20, 2015 at 12:21:05PM +, Bruce Richardson wrote:
> On Thu, Nov 19, 2015 at 05:44:26PM +, Ferruh Yigit wrote:
> > Fixes: 67b6d3039e9e ("eal: set name to threads")
> >
> > pthread_setname_np() function added in glibc 2.12, using this function
> > in older glibc versions cause
The tbl8 and tbl24 structures were essentially identical except for
slightly different names for one or two fields. Merge these two
structures into a single structure definition.
Two fields have been renamed as part of this change: the
"ext_entry" field in the tbl24 has been renamed to
Following on from the discussion in Release 2.2 about changes needed to LPM,
this
is a small cleanup patch for 2.3 to get things started.
This clean-up is inspired by a similar change made by Vladimir in his patchset
proposal for 2.2 for LPM, and I would hope it would make further changes that
From: "Chen Jing D(Mark)"
fm10k driver will meet compile error on non-x86 platforms due to
SSE instructions. Original implementation didn't have switch to
turn off vPMD.
The improvement introduces a macro to turn on/off vPMD functions,
it's on by default. On non-x86
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Yu Nemo Wenbin
> Sent: Thursday, November 19, 2015 1:15 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] hash: add rte_hash_set_cmp_func() function.
>
> Give user a chance to costomize the hash key
From: "Chen Jing D(Mark)"
After introducing vPMD feature, fm10k driver will select best
Rx/Tx in running time. Original implementation selects Rx/Tx
silently without notifications.
This patch adds debug info to notify user what actual Rx/Tx
func are used.
Signed-off-by:
On 2015/11/20 20:21, Yuanhan Liu wrote:
> On Thu, Nov 19, 2015 at 06:07:00PM +0900, Tetsuya Mukawa wrote:
>> The patch fixes reset_owner message handling not to clear callfd,
>> because callfd will be valid while connection is establihed.
>>
>> Signed-off-by: Tetsuya Mukawa
>
>>
>> /*
>>
On 2015/11/21 9:15, Rich Lane wrote:
> On Thu, Nov 12, 2015 at 9:20 PM, Tetsuya Mukawa wrote:
>
>> +static uint16_t
>> +eth_vhost_tx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
>> +{
>>
> ...
>
>> +
>> + /* Enqueue packets to guest RX queue */
>> + nb_tx =
On 24 November 2015 at 13:22, Panu Matilainen wrote:
> On 11/24/2015 04:46 PM, Sergio Gonzalez Monroy wrote:
>>
>> On 24/11/2015 13:57, Panu Matilainen wrote:
>>>
>>> On 11/23/2015 08:37 PM, Martinx - ? wrote:
Hello!
My name is Thiago, I'm trying to compile DPDK 2.0, 2.1
Add note to the documentation that only x86_64 versions of the external
libraries are currently provided.
Signed-off-by: Matej Vido
---
doc/guides/nics/szedata2.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/nics/szedata2.rst b/doc/guides/nics/szedata2.rst
index
On 2015/11/24 12:40, Yuanhan Liu wrote:
> On Tue, Nov 24, 2015 at 11:48:04AM +0900, Tetsuya Mukawa wrote:
>> On 2015/11/20 20:43, Yuanhan Liu wrote:
>>> On Fri, Nov 13, 2015 at 02:20:31PM +0900, Tetsuya Mukawa wrote:
>>>
+static pthread_mutex_t internal_list_lock =
2015-11-24 14:15, Chen Jing D:
> fm10k driver will meet compile error on non-x86 platforms due to
> SSE instructions. Original implementation didn't have switch to
> turn off vPMD.
> The improvement introduces a macro to turn on/off vPMD functions,
> it's on by default. On non-x86 platforms, it
On 2015/11/20 20:43, Yuanhan Liu wrote:
> On Fri, Nov 13, 2015 at 02:20:31PM +0900, Tetsuya Mukawa wrote:
>
>> +static pthread_mutex_t internal_list_lock = PTHREAD_MUTEX_INITIALIZER;
>> +
>> +static rte_atomic16_t nb_started_ports;
>> +pthread_t session_th;
> static?
Hi Yuanhan,
I
On Tue, Nov 24, 2015 at 11:48:04AM +0900, Tetsuya Mukawa wrote:
> On 2015/11/20 20:43, Yuanhan Liu wrote:
> > On Fri, Nov 13, 2015 at 02:20:31PM +0900, Tetsuya Mukawa wrote:
> >
> >> +static pthread_mutex_t internal_list_lock = PTHREAD_MUTEX_INITIALIZER;
> >> +
> >> +static rte_atomic16_t
2015-11-24 11:10, Chen Jing D:
> When the fm10k port is closed, both func tx_queue_clean() and
> fm10k_tx_queue_release_mbufs_vec() will try to release buffer in
> SW ring. The latter func won't do sanity check on those pointers
> and cause crash.
>
> The fix removed Vector TX buffer release func
2015-11-24 14:00, Chen Jing D:
> This patch adds debug info to notify user what actual Rx/Tx
> func are used.
[...]
> + if (rx_using_sse)
> + PMD_INIT_LOG(ERR, "Use vector Rx func");
> + else
> + PMD_INIT_LOG(ERR, "Use regular Rx func");
debug info != LOG(ERR
2015-11-24 02:20, Chen, Jing D:
> From: Thomas Monjalon
> > 2015-11-12 12:57, Chen Jing D:
> > > + if (rx_using_sse)
> > > + PMD_INIT_LOG(ERR, "Use vector Rx func");
> > > + else
> > > + PMD_INIT_LOG(ERR, "Use regular Rx func");
> >
> > Why using en error log level?
>
> Because
> > vlan_tci should be copied from RX descriptor to mbuf,
> > and flag PKT_RX_VLAN_PKT should be set for every RX packet,
> > this patch fixes this issue.
> > Packets in fm10k device always carry at least one VLAN tag.
> > For those packets coming in without VLAN tag,
> > the port default VLAN tag
From: "Chen Jing D(Mark)"
When the fm10k port is closed, both func tx_queue_clean() and
fm10k_tx_queue_release_mbufs_vec() will try to release buffer in
SW ring. The latter func won't do sanity check on those pointers
and cause crash.
The fix removed Vector TX buffer
On Tue, Nov 24, 2015 at 04:31:17PM +0200, Panu Matilainen wrote:
> The physically linked-together combined library has been an increasing
> source of problems, as was predicted when library and symbol versioning
> was introduced. Replace the complex and fragile construction with a
> simple linker
Yuanhan, looks fine. Thanks for handling this.
-- Victor
On Tue, Nov 24, 2015 at 03:25:35PM +0800, Yuanhan Liu wrote:
> From: Victor Kaplansky
>
> The VHOST_USER_SET_VRING_ENABLE request was sent for each queue-pair.
> However, it's changed to be sent per queue in the queue-pair by QEMU
>
We have encountered a CPU where the AES-NI instruction set is disabled
due to export restrictions. Since the build machine and target machine
is different, using -native configs doesn't work, and on this CPU, the
application refuses to run due to the AES CPU flags being amiss.
The patch passes
On 23/11/2015 15:33, Nissim Nisimov wrote:
> Hi all,
>
> We are working on a system which requires allocating a big number of mem
> zones.
>
> We are now reaching the max limit of MEMZONEs allowed (RTE_MAX_MEMZONE).
>
> I see today dpdk limit the above number to 2560.
> Is there any specific
On Tue, 24 Nov 2015 11:32:46 +0100
Thomas Monjalon wrote:
> 2015-11-24 14:00, Chen Jing D:
> > This patch adds debug info to notify user what actual Rx/Tx
> > func are used.
> [...]
> > + if (rx_using_sse)
> > + PMD_INIT_LOG(ERR, "Use vector Rx func");
> > + else
> > +
> -Original Message-
> From: Zhuangyanying [mailto:ann.zhuangyanying at huawei.com]
> Sent: Tuesday, November 24, 2015 11:53 AM
> To: Tan, Jianfeng; dev at dpdk.org
> Cc: mst at redhat.com; mukawa at igel.co.jp; nakajima.yoshihiro at
> lab.ntt.co.jp;
> Qiu, Michael; Guohongzhen;
> -Original Message-
> From: Jianfeng Tan [mailto:jianfeng.tan at intel.com]
> Sent: Friday, November 06, 2015 2:31 AM
> To: dev at dpdk.org
> Cc: mst at redhat.com; mukawa at igel.co.jp; nakajima.yoshihiro at
> lab.ntt.co.jp;
> michael.qiu at intel.com; Guohongzhen; Zhoujingbin;
Tested-by: Yong Liu
- Tested Commit: 4522eca1352d8ed7c0823e328e2a68c2ba6459ab
- OS: Fedora20 3.11.10-301.fc20.x86_64
- GCC: gcc version 4.8.3 20140911
- CPU: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
- NIC: Intel Corporation Device 82599ES [8086:10fb]
- Default x86_64-native-linuxapp-gcc
PATCH v1: configure the vector PMD option as no for FVL
PATCH v2: Update the "know issues" for 2.2 release note
Issue: l3fwd app need the ptype in the mbuf to forwarding the packets properly.
but now some drivers like virtio driver and FVL vPMD will not set the ptype in
mbuf, so l3fwd cannot work
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Tuesday, November 24, 2015 6:55 AM
> To: Chen, Jing D
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] fm10k: fix a crash bug when quit from
> testpmd
>
> 2015-11-12 12:57, Chen Jing D:
> >
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Tuesday, November 24, 2015 6:17 AM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] i40e: skip any phy config as a workaround
>
> 2015-11-17 15:09, Helin Zhang:
> > As
created the new xgene1 machine target to address the difference
in optional armv8-a CRC extension availability compared to
default armv8-a machine target(enabled CRC extension by default)
Signed-off-by: Jerin Jacob
---
config/defconfig_arm64-xgene1-linuxapp-gcc | 56
select hash function for cuckoo, fbk as rte_hash_crc_4byte
if arm64-CRC extension available
Signed-off-by: Jerin Jacob
---
lib/librte_hash/rte_cuckoo_hash.c | 2 +-
lib/librte_hash/rte_fbk_hash.h| 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git
armv8-a has optional CRC32 extension, march=armv8-a+crc enables code
generation for the ARMv8-A architecture together with
the optional CRC32 extensions.
added RTE_MACHINE_CPUFLAG_CRC32 to detect the availability of
CRC32 extension in compile time. At run-time, The RTE_CPUFLAG_CRC32
can be used
1 - 100 of 102 matches
Mail list logo