Thank You Andriy, You are right
Below I put tested KVM configuration that put packets into vhost dpdk plane
space:
export TLBFS_DIR=/mnt/huge
export UVH_PREFIX=/home/ubuntu/esi_ee/dpdk/examples/vhost/usvhost
export VM1_MAC=00:01:04:00:01:00
kvm -cpu host -smp 2 -enable-kvm \
-drive if=virtio,
ICMP echo replies with invalid checksums may be dropped by network nodes or
ignored by the ping utility.
Signed-off-by: Adrien Mazarguil
Acked-by: Ivan Boule
---
v2:
- Compute correct checksum value by taking overflow into account.
app/test-pmd/icmpecho.c | 11 +--
1 file changed, 9 i
Hi Maciej,
I guess it's nice to make sure guest memory is actually allocated by
hugepages.
So please check like below.
$ cat /sys/kernel/mm/hugepage/x/free_hugepages
$ ./start_qemu.sh
$ cat /sys/kernel/mm/hugepage/x/free_hugepages
If qemu guest allocates memory from hugepages, 2nd cat co
Update MAINTAINERS file to claim responsibility for the cxgbe PMD.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
MAINTAINERS |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 0477e7d..37b7472 100644
--- a/MAINTAINERS
Changes to:
1. config/common_linuxapp to add macros for cxgbe pmd.
2. lib/Makefile to add cxgbe pmd to the compile list.
3. mk/rte.app.mk to add cxgbe pmd to link.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
config/common_linuxapp | 10 ++
lib/Makefile |
Adds cxgbe poll mode driver documentation under the usual doc/guides/nics/
directory with the rest of the drivers. The documentation covers cxgbe
implementation details, features and limitations, prerequisites, configuration,
and a sample application usage.
Signed-off-by: Rahul Lakkireddy
Signed
Adds cxgbe poll mode driver for DPDK under lib/librte_pmd_cxgbe directory. This
patch:
1. Adds the Makefile to compile cxgbe pmd.
2. Registers and initializes the cxgbe pmd driver.
3. Implements the necessary eth_dev_ops.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
lib/lib
Adds hardware specific api for all the Chelsio T5 adapters under
lib/librte_pmd_cxgbe/cxgbe directory.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
lib/librte_pmd_cxgbe/cxgbe/adapter.h| 560 ++
lib/librte_pmd_cxgbe/cxgbe/common.h | 401
lib/librte_
This series of patches add the CXGBE Poll Mode Driver support for Chelsio
Terminator 5 series of 10G/40G adapters. The CXGBE PMD is split into two
patches. The first patch adds the hardware specific api for all supported
Chelsio T5 adapters and the second patch adds the actual DPDK CXGBE PMD.
Mo
As now subtree_id is not used acl_merge_trie() any more,
there is no point to calculate and maintain that information.
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl.h | 7 ---
lib/librte_acl/acl_bld.c | 119 +--
2 files changed, 12 ins
Signed-off-by: Konstantin Ananyev
---
app/test/test_acl.c | 147 ++--
1 file changed, 143 insertions(+), 4 deletions(-)
diff --git a/app/test/test_acl.c b/app/test/test_acl.c
index 7119ad3..6a032f9 100644
--- a/app/test/test_acl.c
+++ b/app/test/te
Reported by Zi Hu :
"...
cat test_data/rule1
@192.168.0.0/24 192.168.0.0/24 400 : 500 0 : 52 6/0xff
@192.168.0.0/24 192.168.0.0/24 400 : 500 54 : 65280 6/0xff
@192.168.0.0/24 192.168.0.0/24 400 : 500 0 : 65535 6/0xff
cat test_data/trace1
0xc0a80005 0xc0a80009 450 53 0x06
I run the test by:
sudo
ACL: fix a problem in acl_merge_trie
ACL: add new test case for ranges build
ACL: remove subtree_id calculations at build stage
app/test/test_acl.c | 147 +--
lib/librte_acl/acl.h | 7 ---
lib/librte_acl/acl_bld.c | 121 --
Hi,
2015-01-06 14:53, Zhigang Lu:
> Add defconfig and rte.vars.mk files for Tile architecture.
>
> Signed-off-by: Zhigang Lu
> Signed-off-by: Cyril Chemparathy
This series is pending without having any comments.
So it should be accepted and merged.
Do you accept to be the maintainer of Tile a
On Fri, May 22, 2015 at 04:09:45PM +0200, Thomas Monjalon wrote:
> > Series Acked-by: John McNamara
>
> Acked-by: Thomas Monjalon
>
> Applied with the changes discussed in this thread.
> Some patches were split/merge to make each driver move more atomic.
>
Thanks Thomas.
As unified packet types are used instead, those old bit masks and
the relevant macros for packet type indication need to be removed.
Signed-off-by: Helin Zhang
---
lib/librte_mbuf/rte_mbuf.c | 6 --
lib/librte_mbuf/rte_mbuf.h | 6 --
2 files changed, 12 deletions(-)
v2 changes:
* Used r
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
examples/l3fwd/main.c | 71 +--
1 file changed, 40 insertions(+), 31 deletions(-)
v2 changes:
* Used
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
examples/l3fwd-power/main.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
v2 changes:
* Used redefined packet types and enlarged packet_
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
examples/l3fwd-acl/main.c | 19 ++-
1 file changed, 6 insertions(+), 13 deletions(-)
v2 changes:
* Used redefined packet types and en
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
examples/ip_reassembly/main.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
v2 changes:
* Used redefined packet types and enlarged pac
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
examples/ip_fragmentation/main.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
v2 changes:
* Used redefined packet types and enlarged
Severl useless code lines are added accidently, which blocks packet
type unification. They should be deleted at all.
Signed-off-by: Helin Zhang
---
app/test/packet_burst_generator.c | 10 --
1 file changed, 10 deletions(-)
v4 changes:
* Removed several useless code lines which block pac
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
Signed-off-by: Jijiang Liu
---
app/test-pmd/csumonly.c | 10 +--
app/test-pmd/rxonly.c | 178 ++--
2 files
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
app/test-pipeline/pipeline_hash.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
v2 changes:
* Used redefined packet types and enlarged
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_fm10k/fm10k_rxtx.c | 30 +-
1 file changed, 21 insertions(+), 9 deletions(-)
v4 changes:
* Supported unifi
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_vmxnet3/vmxnet3_rxtx.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
v2 changes:
* Used redefined packet types and enlarge
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_enic/enic_main.c | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
v2 changes:
* Used redefined packet types and en
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_i40e/i40e_rxtx.c | 786 ++--
1 file changed, 512 insertions(+), 274 deletions(-)
v2 changes:
* Use
To unify packet type among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Note that around 2.5% performance drop (64B) was observed of doing
4 ports (1 port per 82599 card) IO forwarding on the same SNB core.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_i
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_e1000/igb_rxtx.c | 98 ++---
1 file changed, 83 insertions(+), 15 deletions(-)
v2 changes:
* Used
As there are only 6 bit flags in ol_flags for indicating packet
types, which is not enough to describe all the possible packet
types hardware can recognize. For example, i40e hardware can
recognize more than 150 packet types. Unified packet type is
composed of L2 type, L3 type, L4 type, tunnel type
To unify the packet type, bit masks of packet type for ol_flags are
replaced. In addition, more packet types (UDP, TCP and SCTP) are
supported in vectorized ixgbe PMD.
Note that around 2% performance drop (64B) was observed of doing 4
ports (1 port per 82599 card) IO forwarding on the same SNB core
In order to unify the packet type, the field of 'packet_type' in
'struct rte_mbuf' needs to be extended from 16 to 32 bits.
Accordingly, some fields in 'struct rte_mbuf' are re-organized to
support this change for Vector PMD. As 'struct rte_kni_mbuf' for
KNI should be right mapped to 'struct rte_mb
Currently only 6 bits which are stored in ol_flags are used to indicate
the packet types. This is not enough, as some NIC hardware can recognize
quite a lot of packet types, e.g i40e hardware can recognize more than 150
packet types. Hiding those packet types hides hardware offload capabilities
whi
Fix warning messages "cast to pointer from integer of different size" when
compiling DPDK in 32 bit with Mellanox PMD.
SGE addresses are 64 bit integers, converting them to pointers must be done
through uintptr_t to avoid compilation warnings when those have a different
size.
Signed-off-by: Nelio
> Series Acked-by: John McNamara
Acked-by: Thomas Monjalon
Applied with the changes discussed in this thread.
Some patches were split/merge to make each driver move more atomic.
On Fri, May 22, 2015 at 02:42:08PM +0200, Olivier MATZ wrote:
> Hi Adrien,
Hi Olivier,
> On 05/22/2015 01:07 PM, Adrien Mazarguil wrote:
> > ICMP echo replies with invalid checksums may be dropped by network nodes or
> > ignored by the ping utility.
> >
> > Signed-off-by: Adrien Mazarguil
> > A
On Thu, May 21, 2015 at 05:51:34PM +, Shyam Sundar Govindaraj wrote:
> Hi
>
> I am interested in DPDK LPM implementation. I understand the high
> performance of this LPM table but, when it comes to multithreaded
> application, I read the following safety issue in DPDK documentation. I
> u
Hi Adrien,
On 05/22/2015 01:07 PM, Adrien Mazarguil wrote:
> ICMP echo replies with invalid checksums may be dropped by network nodes or
> ignored by the ping utility.
>
> Signed-off-by: Adrien Mazarguil
> Acked-by: Ivan Boule
> ---
> app/test-pmd/icmpecho.c | 5 +++--
> 1 file changed, 3 inse
On Thu, 21 May 2015 16:55:57 +0800
Cunming Liang wrote:
> This patch does below:
> - Create VFIO eventfds for each interrupt vector (move to next)
> - Assign per interrupt vector's eventfd to VFIO by ioctl
>
> Signed-off-by: Danny Zhou
> Signed-off-by: Cunming Liang
One non-trivial performa
ICMP echo replies with invalid checksums may be dropped by network nodes or
ignored by the ping utility.
Signed-off-by: Adrien Mazarguil
Acked-by: Ivan Boule
---
app/test-pmd/icmpecho.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/app/test-pmd/icmpecho.c b/app/test-p
In the icmpecho forwarding mode, ARP replies from testpmd contain
invalid zero-filled MAC addresses. This is broken since the commit below.
Fixes: 31db4d38de72 ("net: change arp header struct declaration")
Signed-off-by: Adrien Mazarguil
Acked-by: Ivan Boule
---
app/test-pmd/icmpecho.c | 2 +-
Hi guys,
I guess you just miss the qemu flag to map the memory as shared, i.e.:
-object memory-backend-file,id=mem,size=1024M,mem-path=/mnt/huge,share=on
(the keyword is share=on)
Here is an example script:
https://github.com/srajag/contrail-vrouter/blob/dpdk-devel/examples/vms/VROUTER1/80.start-v
On Fri, 22 May 2015 18:03:04 +0200
Thomas Monjalon wrote:
> Hi,
>
> 2015-01-06 14:53, Zhigang Lu:
> > Add defconfig and rte.vars.mk files for Tile architecture.
> >
> > Signed-off-by: Zhigang Lu
> > Signed-off-by: Cyril Chemparathy
>
> This series is pending without having any comments.
> So
I checked this, results below
#before script:
root@# cat /sys/kernel/mm/hugepages/hugepages-2048kB/free_hugepages
494
#after 1 qemu script
root@# cat /sys/kernel/mm/hugepages/hugepages-2048kB/free_hugepages
366
So qemu consume 262144k~262MB that is correct with script
On Fri, May 22, 2015 at 11:
"Do you use some command I suggest before,
In case of you miss the previous mail, just copy it again:"
-Yes but it didn't help me ;/
I will describe step by step to esure that configuration is made by right
way
I started vhost:
./build/app/vhost-switch -c f -n 4 --huge-dir /mnt/huge --socket-
On 22 May 2015 at 10:05, Maciej Grochowski
wrote:
> What I'm going to do today is to compile newest kernel for vhost and guest
> and debug where packet flow stuck, I will report the result
>
Compiling the guest virtio-net driver with debug printouts enabled can be
really helpful in these situati
Added new test that verifies that rte_jhash_1words,
rte_jhash_2words and rte_jhash_3words return the same
values as rte_jhash.
Note that this patch has been added after the update
of the jhash function because these 3 functions did not
return the same values as rte_jhash before
Signed-off-by: Pab
Changed name to something more meaningful,
and mark rte_jhash2 as deprecated.
Signed-off-by: Pablo de Lara
---
app/test/test_func_reentrancy.c |2 +-
app/test/test_hash.c|4 ++--
app/test/test_hash_functions.c |6 +++---
lib/librte_hash/rte_jhash.h | 17 +++
rte_jhash is basically like _rte_jhash_2hashes but it returns only 1 hash,
instead of 2.
In order to remove duplicated code, rte_jhash calls _rte_jhash_2hashes,
passing 0 as the second seed and returning just the first hash value.
(performance penalty is negligible)
The same is done with rte_jhas
From: De Lara Guarch, Pablo
With the jhash update, two new functions were introduced:
- rte_jhash_2hashes: Same as rte_jhash, but takes two seeds
and return two hashes (uint32_ts)
- rte_jhash2_2hashes: Same as rte_jhash2, but takes two seeds
and return
Jenkins hash function was developed originally in 1996,
and was integrated in first versions of DPDK.
The function has been improved in 2006,
achieving up to 60% better performance, compared to the original one.
This patch integrates that code into the rte_jhash library.
It also updates the precal
In order to make sure that the hash functions are returning
the correct values, new tests have been added:
- First test compares precalculated hash values with values calculated
from the existing hash functions.
- Second test compares values returned from rte_jhash2 and rte_jhash,
expecting same r
In order to see more clearly the performance difference
between different hash functions, order of the loops
have been changed, so it iterates first through initial values,
then key sizes and then the hash functions.
Signed-off-by: Pablo de Lara
---
app/test/test_hash_functions.c | 20
Previous key sizes used for testing did not have much purpose.
This patch substitutes them with some more meaninful
(standard multiple of 2 key sizes, plus IPv4/v6 tuple and others)
Also an arbitrary initial value has been added
to increase the test coverage.
Signed-off-by: Pablo de Lara
---
ap
From: De Lara Guarch, Pablo
Cycles per hash calculation were measured per single operation.
It is much more accurate to run several iterations between measurements
and divide by number of iterations.
Signed-off-by: Pablo de Lara
---
app/test/test_hash_functions.c | 18 +-
1 f
From: De Lara Guarch, Pablo
This patch moves hash function performance tests to a separate file,
so user can check performance of the existing hash functions quicker,
without having to run all the other hash operation performance tests,
which takes some time.
Signed-off-by: Pablo de Lara
---
a
Jenkins hash function was developed originally in 1996,
and was integrated in first versions of DPDK.
The function has been improved in 2006,
achieving up to 60% better performance, compared to the original one.
This patchset updates the current jhash in DPDK,
including two new functions that gene
On 5/22/2015 2:20 AM, Stephen Hemminger wrote:
> On Thu, 21 May 2015 16:56:00 +0800
> Cunming Liang wrote:
>
>> +int
>> +rte_eth_dev_rx_intr_ctl_q(uint8_t port_id, uint16_t queue_id,
>> + int epfd, int op, void *data)
>> +{
>> +uint32_t vec;
>> +struct rte_eth_dev *d
On 5/22/2015 2:17 AM, Stephen Hemminger wrote:
> On Thu, 21 May 2015 16:55:54 +0800
> Cunming Liang wrote:
>
>> +static int
>> +eal_epoll_process_event(struct epoll_event *evs, int n,
>> +struct rte_epoll_event *events)
>> +{
>> +int i;
>> +int count = 0;
>> +stru
Unfortunately not, I have the same issue in rte_vhost_dequeue_burst
function.
What kernel version are You using on host/guest? In my case on host I
had 3.13.0 and on guests old 3.2 debian.
I just looked deeper into virtio back-end (vhost) but at first glace it
seems like nothing coming from virt
On Fri, 22 May 2015 00:05:36 +
Neil Horman wrote:
> On Thu, May 21, 2015 at 11:14:00AM -0700, Stephen Hemminger wrote:
> > On Thu, 21 May 2015 13:58:46 -0400
> > Neil Horman wrote:
> >
> > > On Thu, May 21, 2015 at 10:43:00AM -0700, Stephen Hemminger wrote:
> > > > On Thu, 21 May 2015 06:32
Hi,
pcap.h is available by development version of libpcap third-party software.
Please install it. e.g. apt-get install libpcap-dev
2015-05-22 7:35 GMT+09:00 Navneet Rao :
> Getting the following message when using
>
> dpdk-2.0.0 version
>
> to building the libraries and test application using
>
On Fri, 22 May 2015 18:54:20 +0530
Rahul Lakkireddy wrote:
> +static int cxgbe_dev_link_update(struct rte_eth_dev *eth_dev,
> + __rte_unused int wait_to_complete)
> +{
> + struct port_info *pi = (struct port_info *)(eth_dev->data->dev_private);
> + struct adap
On Fri, 22 May 2015 18:54:20 +0530
Rahul Lakkireddy wrote:
> +/**
> + * fls - find last (most-significant) bit set
> + * @x: the word to search
> + *
> + * This is defined the same way as ffs.
> + * Note fls(0) = 0, fls(1) = 1, fls(0x8000) = 32.
> + */
> +static inline int fls(int x)
Isn't t
On Fri, 22 May 2015 18:54:20 +0530
Rahul Lakkireddy wrote:
> +#define pr_err(y, args...) dev_err(0, y, ##args)
> +#define pr_warn(y, args...) dev_warn(0, y, ##args)
> +#define pr_info(y, args...) dev_info(0, y, ##args)
> +#define BUG() pr_err("BUG at %s:%d", __func__, __LINE__)
> +
> +#define ASS
On 21/05/2015 17:59, Harish Patil wrote:
> Hello dpdk-dev,
>
> I understand that the reserved memzones cannot be freed, as mentioned in
> the DPDK specs. But I would like to know why? Is there any limitations?
There should be a few threads in the mailing list related to this topic.
Last that comes
Hi
What kernel version are You using on host/guest?
>>I use ubuntu 12.04(3.11.0-15-generic) in host. In vm I use ubuntu 12.04 and
>>ubuntu14.04 both, but the result is same.
Do you use some command I suggest before, In case of you miss the previous
mail, just copy it again:
>> I try it but the r
On 5/22/15 2:05 AM, Ouyang, Changchun wrote:
> Hi Thomas,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas F Herbert
>> Sent: Friday, May 22, 2015 9:39 AM
>> To: dpdk >> dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH 4/6] vhost: Add new command
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Maciej Grochowski
> Sent: Friday, May 22, 2015 4:06 PM
> To: Gaohaifeng (A); dev at dpdk.org
> Subject: Re: [dpdk-dev] FW: Vhost user no connection vm2vm
>
> Unfortunately not, I have the same issue in rte_
On 5/21/15 6:35 PM, Navneet Rao wrote:
> Getting the following message when using
>
> dpdk-2.0.0 version
This isn't a problem with dpdk. You need libpcap which includes the
headers, http://www.tcpdump.org/release/libpcap-1.7.3.tar.gz or
libpcap-devel package.
>
> to building the libraries and
On Fri, May 22, 2015 at 04:44:07PM +0800, Helin Zhang wrote:
> In order to unify the packet type, the field of 'packet_type' in
> 'struct rte_mbuf' needs to be extended from 16 to 32 bits.
> Accordingly, some fields in 'struct rte_mbuf' are re-organized to
> support this change for Vector PMD. As '
Hi Thomas,
> -Original Message-
> From: Thomas F Herbert [mailto:therbert at redhat.com]
> Sent: Friday, May 22, 2015 9:13 AM
> To: Ouyang, Changchun; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 0/6] Support multiple queues in vhost
>
>
>
> On 5/21/15 3:49 AM, Ouyang Changchun wrot
Hi Thomas,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas F Herbert
> Sent: Friday, May 22, 2015 9:39 AM
> To: dpdk >> dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 4/6] vhost: Add new command line option:
> rxq
>
>
>
> On 5/21/15 3:49 AM, Ouyan
74 matches
Mail list logo
