> Subject: Re: [dpdk-dev] [memnic PATCH 3/7] pmd: use helper macros
>
> 2014-09-11 07:47, Hiroshi Shimamoto:
> > Do not touch pktmbuf directly.
> >
> > Instead of direct access, use rte_pktmbuf_pkt_len() and
> > rte_pktmbuf_data_len()
> > to access the property.
>
> I guess this change is for co
> Subject: Re: [dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier
>
> 2014-09-11 07:48, Hiroshi Shimamoto:
> > x86 can keep store ordering with standard operations.
>
> Are we sure it's always the case (including old 32-bit CPU)?
> I would prefer to have a reference here. I know we already d
Hi Thomas,
Ouyang is right!
Yes, I had finished the patch validation, so I reply the email, thanks.
Regards,
Jingguo Fu
-Original Message-
From: Ouyang, Changchun
Sent: Wednesday, September 24, 2014 22:27
To: Thomas Monjalon; Fu, JingguoX
Cc: dev at dpdk.org; Ouyang, Changchun
Subject:
Hi Thomas, Keith,
> Subject: Re: [dpdk-dev] [memnic PATCH 7/7] pmd: split calling mbuf free
>
>
> On Sep 24, 2014, at 10:20 AM, Thomas Monjalon
> wrote:
>
> > 2014-09-11 07:52, Hiroshi Shimamoto:
> >> @@ -408,9 +408,9 @@ retry:
> >>
> >>rte_compiler_barrier();
> >>p->s
Tested-by: Haifeng Tang
This patch just includes one file, and has been tested by Intel.
Please see the detail information from the attachment.
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bruce Richardson
Sent: Thursday, September 18, 2014 6:56 PM
To: dev at
On Sep 24, 2014, at 8:12 PM, Hiroshi Shimamoto
wrote:
> Hi Thomas, Keith,
>
>> Subject: Re: [dpdk-dev] [memnic PATCH 7/7] pmd: split calling mbuf free
>>
>>
>> On Sep 24, 2014, at 10:20 AM, Thomas Monjalon
>> wrote:
>>
>>> 2014-09-11 07:52, Hiroshi Shimamoto:
@@ -408,9 +408,9 @@ retr
Tested-by: Liu Yong
This patch set has been tested by Intel.
Please see information as the following:
Host:
OS : Fedora 20 x86_64
Kernel : 3.11.10-301
GCC: 4.8.3
CPU: Intel Xeon CPU E5-2680 v2 @ 2.80GHz
NIC : Intel Nianti
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, September 25, 2014 12:22 AM
> To: Xie, Huawei
> Cc: dev at dpdk.org; Ouyang, Changchun
> Subject: Re: [dpdk-dev] [PATCH v4 0/5] lib/librte_vhost: user space vhost cuse
> driver library
>
Hi,
Multiqueue is supported by vhost user?
If not supported now , will it be supported in the future?
The patch set enables control packet filter on Fortville.
Control packet filter can assign packet to specific destination by filtering
with mac address and ethertype or only ethertype.
It mainly includes:
- Use new filter mechanism discussed at
http://dpdk.org/ml/archives/dev/2014-September/005
Define new APIs to support configure multi-kind filters using same APIs
- rte_eth_dev_filter_supported
- rte_eth_dev_filter_ctrl
As to the implemetation discussion, please refer to
http://dpdk.org/ml/archives/dev/2014-September/005179.html, and control packet
filter implementation is based on
define new filter type and its structure
- RTE_ETH_FILTER_CTRL_PKT
- struct rte_ctrl_pkt_filter
Signed-off-by: Jingjing Wu
---
lib/librte_ether/rte_eth_ctrl.h | 24
1 file changed, 24 insertions(+)
diff --git a/lib/librte_ether/rte_eth_ctrl.h b/lib/librte_ether/rte_et
Implement control packet filter, support add and delete operations.
It can assign packets to specific queue or vsi by filtering with mac address
and ethertype or only
ethertype on both rx and tx directions.
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev.c | 161
Add command to test control packet filter
- add/delete control packet filter
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 149 +
1 file changed, 149 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 225f6
2014-09-25 02:13, Tang, HaifengX:
> Tested-by: Haifeng Tang
>
> This patch just includes one file, and has been tested by Intel.
> Please see the detail information from the attachment.
Attachment is filtered out. Please do not try to attach some files for
the mailing list.
By the way, this pat
2014-09-25 00:47, Fu, JingguoX:
> Yes, I had finished the patch validation, so I reply the email, thanks.
I think there are 3 cases where a test report could be needed:
1) the patch is pending
2) an error is found during validation
3) the test is described with some useful information
Here the th
This is a patch to a problem that I have faced (described in the thread) and
this works for me.
1) Since the data_size_shl was getting its value from the key_size, the
table data entries were being corrupted when the calculation to shift the
number of bits was being made based on the key_
Tested-by: Zhaochen Zhan
This patch has been verified both on "native" and "default" target.
Please see environment information as the following:
CPU: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
OS: Linux fc20 3.11.10-301.fc20.x86_64
GCC: 4.8.3
NIC: Intel Niantic 82599
Reviewed-by: Helin Zhang
Great to have these patches!
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jingjing Wu
> Sent: Thursday, September 25, 2014 2:59 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 0/4] support control packet filter on Fortville
In DPDK1.7, PMDs are loaded by __constructor__ functions.
As there is no direct reference to each PMD driver structures, linker may
not link the PMD library.
You should use -Wl,--whole-archive and -Wl,--no-whole-archive for linking
DPDK applications.
See http://dpdk.org/browse/dpdk/tree/mk/rte.app.
>
> Hey guys
>Is it safe to add an entry to the rte_table_hash while the pipeline is
> being
> run - for instance if I were to try and add an entry on a port reader action
> when the packet enters the pipeline?
If the table is part of the pipeline, it is not safe. i.e. the table is not
thre
Tested-by: Liu Yong
This patch set has been tested by Intel.
Please see information as the following:
Host:
OS : Fedora 20 x86_64
Kernel : 3.11.10-301
GCC: 4.8.3
CPU: Intel Xeon CPU E5-2680 v2 @ 2.80GHz
NIC : 2*40G (8086
Fix several code style issues.
v2 changes:
* Put all code style fixes for testpmd in a single patch.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 28 +++-
app/test-pmd/config.c |
Fix several code style issues.
v2 changes:
* Put code style fixes of i40e into a single patch.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev_vf.c | 13 +
1 file changed, 9 insertions(+),
As e1000, ixgbe and i40e hardware use different sizes of redirection
table in PF or VF, ethdev and PMDs need to be reworked to support
multiple sizes of that table. In addition, commands in testpmd also
need to be reworked to support these changes.
v2 changes:
* Reorganize the patches.
* Add code
Add more annotation, to clearly tell the 'rte_eth_dev_info_get()'
users that the buffer should be cleared first.
v2 changes:
Add more annotation for 'rte_eth_dev_info_get()'.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by: Jingjing Wu
---
lib/librt
New function of 'eth_igbvf_infos_get' has been added for
VF ops of 'dev_infos_get', as more and more information
are different between PF and VF.
v2 changes:
* Added new function for ops of 'dev_infos_get' specifically
for igb VF.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by:
New function of 'ixgbevf_dev_info_get' has been added for
VF ops of 'dev_infos_get', as more and more information
are different between PF and VF.
v2 changes:
* Added new function for ops of 'dev_infos_get' specifically
for ixgbe VF.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewe
Add support of setting hash lookup table size according to the
hardawre capability.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev.c | 14 +-
lib/librte_pmd_i40e/i40e_ethdev.h | 1 +
2 fi
Add support of getting redirection table size in both PF
and VF.
v2 changes:
* Put getting reta size of both i40e PF and VF into a single
patch.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev.c
Add support of updating/querying redirection table in VF.
v2 changes:
* Add support of updating/querying i40e reta of VF.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev_vf.c | 89
As multiple sizes of redirection table have been supported,
the commands of updating/querying reta need to be reworked
as well. In addition, the reta size can be queried by the
existing command of 'show port info <>'.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
As ethdev has been changed to support multiple sizes of
reta, updating/querying reta should be reworked to support
that change.
v2 changes:
* Put rework of updating/querying igb reta to a single patch.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by:
As ethdev has been changed to support multiple sizes of reta,
updating/querying reta should be reworked to support that
change.
v2 changes:
* Put rework of updating/querying ixgbe reta to a single patch.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by
To support possible different sizes of redirection table,
structures and functions need to be redefined. In detail,
* 'struct rte_eth_rss_reta' has been redefined.
* 'uint16_t reta_size' has been added into
'struct rte_eth_dev_info'.
* Updating/querying reta have been reimplemented with one
mor
As ethdev has been changed to support multiple sizes of reta,
updating/querying reta should be reworked to support that change.
v2 changes:
* Put rework of updating/querying i40e reta to a single patch.
Signed-off-by: Helin Zhang
Reviewed-by: Jijiang Liu
Reviewed-by: Cunming Liang
Reviewed-by:
> > > master |<- put packets ->| |<- get packets ->|
> > > slave | |<- rx packets ->|<- tx packets ->| |
> > > |<- set ->|
> > >
> > > Measuring how many sets in the certain period, that represents
> > > the MEMNIC PMD performance. The master workl
Hi Ouyang,
2014-07-23 12:48, Ouyang Changchun:
> Update comments for the field start_rx_per_q for better readability.
> Rename the field name to rx_enable_queue for better readability too.
> Accordingly Update its reference in sample vhost.
> - uint8_t start_rx_per_q; /**< start rx per queue.
Since the introduction of vector PMD, a bug in ixgbe_rxq_rearm could
cause a crash. As long as the memory pool allocated to the RX queue
has mbufs available, there is no problem. After allocation of _all_
mbufs from the memory pool, previously returned mbufs by
rte_eth_rx_burst could be accessed by
Hi Balazs,
I didn't read your patch because of this disclaimer:
2014-09-25 09:36, Balazs Nemeth:
> This e-mail and any attachments may contain confidential material for the
> sole use of the intended recipient(s). Any review or distribution by others
> is strictly prohibited. If you are not the
On Wed, Sep 24, 2014 at 06:26:13PM +0100, Alan Carew wrote:
> Moved the current librte_power implementation to rte_power_acpi_cpufreq, with
> renaming of functions only.
> Added rte_power_kvm_vm implmentation to support Power Management from a VM.
>
> librte_power now hides the implementation base
Hi,
Is there any BSD users available to test this patch?
I'd like to add this comment in the log:
"The initial implementation simply raised the IOPL of the current thread
when open(2) was called on the device. This behaviour is retained in the
current implementation as legacy support for bot
On Thu, Sep 25, 2014 at 07:46:16AM +, Saha, Avik (AWS) wrote:
> This is a patch to a problem that I have faced (described in the thread) and
> this works for me.
>
> 1) Since the data_size_shl was getting its value from the key_size, the
> table data entries were being corrupted when t
Change alarm cancel function to thread-safe.
It eliminates a race between threads using rte_alarm_cancel and
rte_alarm_set.
Signed-off-by: Pawel Wodkowski
Reviewed-by: Michal Jastrzebski
---
lib/librte_eal/common/include/rte_alarm.h |3 +-
lib/librte_eal/linuxapp/eal/eal_alarm.
Hi Thomas,
I will work with team to see if we can improve test report.
Because intel validation team will continue to upgrade test cases to verify
feature,
I think that it's still worth to verify patch or features even it has already
integrated branch.
Thanks
Waterman
>-Original Messa
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Michal Jastrzebski
> Sent: Thursday, September 25, 2014 1:56 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2] Change alarm cancel function to thread-safe:
>
> Change alarm cancel function to thread-safe.
> It eliminates a r
This patch cleans up some IXGBE base codes, such as remove unnecessary return
statement,
and reduce goto statement etc.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82598.c | 2 --
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 7 +++
lib/librte_pmd_ixgbe/ixgbe/i
This patch implements a function to check command complete for flow director in
IXGBE base code, and replaces related code snippet with this function.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 63 +++-
1 file changed, 38 insertions
This patch series update IXGBE base code (a.k.a. share code) from
package 2014.03.13 to package 2014.09.04, all in all, it includes the
following changes:
1. Change comments and fix typo in IXGBE base code.
2. Clean up IXGBE base code.
3. Implement a function to check command complete for flo
This patch supports cloud mode in IXGBE base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 70
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 10 +
2 files changed, 80 insertions(+)
diff --git a/lib/librte_pmd_ixgbe/ixgbe/i
This patch refines function to let eeprom checksum calculation return
either a negative error code on error, or the 16-bit checksum in IXGBE
base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_common.c | 96 ++--
lib/librte_pmd_ixgbe/ixgbe/ixgbe
This patch introduces a new argument to let caller determine if it need read
and
return data or not after executing host interface command in IXGBE base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_common.c | 66 ++-
lib/librte_pmd_ixgb
This patch extend mask from 16 bits to 32 bits for releasing or
acquiring SWFW semaphore in IXGBE base code. It is used in reading and
writing I2C byte.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_api.c| 4 +-
lib/librte_pmd_ixgbe/ixgbe/ixgbe_api.h| 4 +-
l
This patch support new device id 82599_QSFP and 82599_LS in IXGBE
base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 79 +++--
lib/librte_pmd_ixgbe/ixgbe/ixgbe_api.c| 2 +
lib/librte_pmd_ixgbe/ixgbe/ixgbe_common.c | 8 +-
lib/librte_pmd_ixgbe/ixg
It need wait for 5 ms for polling EEC register in IXGBE X540 share codes.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_x540.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_x540.c
b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_x5
This patch defines new error type in IXGBE share codes; they are
used to report different kinds of error.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_osdep.h | 10 ++
1 file changed, 10 insertions(+)
diff --git a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_osdep.h
b/lib/
This patch stores lan_id and physical semaphore mask into hardware physical
information,
and use them to control read and write physical registers in IXGBE base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_phy.c | 23 +++
1 file changed, 11 ins
This patch updates comments and fixes some comments typo, such as 'tx' is
changed into 'Tx',
'cloude' is changed into 'cloud' etc.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 36 +++
lib/librte_pmd_ixgbe/ixgbe/ixgbe_common.c | 2 +
This patch removes unnecessary delay when setting up physical link
and negotiating in IXGBE share code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_phy.c | 53 --
1 file changed, 6 insertions(+), 47 deletions(-)
diff --git a/lib/librte_p
This patch implements a function to reset VF register to initial
values in IXGBE base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_vf.c | 46 +++
1 file changed, 46 insertions(+)
diff --git a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_vf.c
b
This patch uses hardware MAC type to determine I2C control, clock
in/out, and data in/out in IXGBE base code.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_phy.c | 58 -
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 12 ---
2 files changed
This patch updates device id and PF driver in IXGBE PMD to support X550.
Signed-off-by: Changchun Ouyang
---
lib/librte_eal/common/include/rte_pci_dev_ids.h | 14 ++
lib/librte_ether/rte_ethdev.h | 2 +-
lib/librte_pmd_ixgbe/ixgbe_bypass_api.h | 9 +++
lib
This patch adds new file to support controller X550, therefore update the
Makefile and README file.
It also updates the API functions, DCB related functions, mailbox related
functions etc to support X550.
In addition, some new MACROs used by X550 are added.
Signed-off-by: Changchun Ouyang
---
This patch implement functions to do I2C byte read and write in
IXGBE base code; it also relocates function of ixgbe_mng_enabled.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 136 +-
lib/librte_pmd_ixgbe/ixgbe/ixgbe_common.c | 27 +
This patch supports these functionalities in IXGBE base code:
Thermal sensor, DMA coalescing, EEE support, Source address pruning,
Anti-spoofing, Iosf buffer reading and writing, Malicious driver detection.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_82599.c | 4 +
Hi Thomas,
Thanks very much for your comments!
I will rework this patch and also update for i40e.
Thanks and best regards,
Changchun
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, September 25, 2014 5:31 PM
> To: Ouyang, Changchun
>
I am preparing v5 patch for vhost lib and vhost example, merging the merge-able
feature. The weird thing is merge-able feature ever works smoothly in my old
code base which is based on
1.7.1, while after I merge with latest repo, mergeable could cause mbuf
allocation failure.
I did code comparis
It is not supported yet. Do you want to send patches to add this support?
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Zhangkun (K)
Sent: Thursday, September 25, 2014 11:28 AM
To: dev at dpdk.org
Subject: [dpdk-dev] multiqueue is supported by vhost user?
Hi,
2014-09-25 13:07, Cao, Waterman:
> I will work with team to see if we can improve test report.
> Because intel validation team will continue to upgrade test cases to verify
> feature,
> I think that it's still worth to verify patch or features even it has
> already integrated branch.
Of cours
On Thu, Sep 25, 2014 at 01:56:08PM +0100, Michal Jastrzebski wrote:
> Change alarm cancel function to thread-safe.
> It eliminates a race between threads using rte_alarm_cancel and
> rte_alarm_set.
>
> Signed-off-by: Pawel Wodkowski
> Reviewed-by: Michal Jastrzebski
>
> ---
> lib/l
Hi,
There were a few DPDK demos at IDF, and from one of them I gathered that you
can use DDIO to enhance performance by using certain lookup APIs in DPDK. Can
someone shed light on this? Is DDIO enabled by default? It is available only on
the v3 chip and needs DPDK 1.7 or how can get the perfor
This set of patches fixes bugs in the packet framework. Some of the
bugs occur in corner cases (i.e. when a lookup is performed on a few
packets or when buckets are in extended states) while others can cause
memory to be accessed beyond what is reserved during initialization
time.
Balazs Nemeth (4
When an entry is deleted from an extensible rte_table_hash, the bucket
that stored the entry can become empty. If this is the case, the
bucket needs to be removed from the chain of buckets.
During removal of the bucket, the chain should be updated first. If
the bucket that will be removed is clear
If a key is not found in a bucket and the bucket has been extended,
the extended buckets also have to checked for potentially matching
keys. The extended buckets are checked at the end of the lookup. In
most cases, this logic is skipped as it is uncommon to have buckets in
an extended state.
In ca
During initialization of rte_hash_table_ext and rte_hash_table_lru,
t->data_size_shl is calculated. This member contains the number of
bits to shift left during calculation of the location of entries in
the hash table. To determine the number of bits to shift left, the
size of the entry (as provi
During initialization of rte_table_hash_ext and rte_table_hash_lru, a
contiguous region of memory is allocated to store meta data, buckets,
extended buckets, keys, stack of keys, stack of extended buckets and
data entries. The size of each region depends on the hash table
configuration.
The addres
On Thu, Sep 25, 2014 at 03:24:35PM +, Balazs Nemeth wrote:
> When an entry is deleted from an extensible rte_table_hash, the bucket
> that stored the entry can become empty. If this is the case, the
> bucket needs to be removed from the chain of buckets.
>
> During removal of the bucket, the c
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> Sent: Thursday, September 25, 2014 4:08 PM
> To: Jastrzebski, MichalX K
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] Change alarm cancel function to
> thread-safe:
>
> On Thu, Sep
On Mon, Sep 08, 2014 at 07:18:10AM +, Wodkowski, PawelX wrote:
> > -Original Message-
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Sent: Friday, August 29, 2014 17:41
> > To: dev at dpdk.org
> > Cc: Wodkowski, PawelX
> > Subject: Re: [dpdk-dev] [PATCH] ixgbe_rx_sc
On Thu, Sep 25, 2014 at 03:19:31PM +, Anjali Kulkarni wrote:
> Hi,
>
> There were a few DPDK demos at IDF, and from one of them I gathered that you
> can use DDIO to enhance performance by using certain lookup APIs in DPDK. Can
> someone shed light on this? Is DDIO enabled by default? It is
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Thursday, September 25, 2014 11:10 AM
> To: Carew, Alan
> Cc: dev at dpdk.org
> Subject: Re: [PATCH v2 07/10] librte_power common interface for Guest and
> Host
>
> On Wed, Sep 24, 2014 at 06:26:13PM +0100,
On Thu, Sep 25, 2014 at 04:03:48PM +, Ananyev, Konstantin wrote:
>
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> > Sent: Thursday, September 25, 2014 4:08 PM
> > To: Jastrzebski, MichalX K
> > Cc: dev at dpdk.org
> > Subject: Re: [d
On Thu, Sep 25, 2014 at 05:06:11PM +, Carew, Alan wrote:
> > -Original Message-
> > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > Sent: Thursday, September 25, 2014 11:10 AM
> > To: Carew, Alan
> > Cc: dev at dpdk.org
> > Subject: Re: [PATCH v2 07/10] librte_power common inter
I turned off CONFIG_RTE_IXGBE_INC_VEC, everything works fine.
> -Original Message-
> From: Xie, Huawei
> Sent: Thursday, September 25, 2014 10:27 PM
> To: Xie, Huawei; Thomas Monjalon
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v4 0/5] lib/librte_vhost: user space vhost cuse
>
Add initial pass header files to support symbol versioning.
---
Change notes
v2)
* Fixed ifdef in rte_compat.h to test for RTE_BUILD_SHARED_LIB instead of the
non-existant RTE_SYMBOL_VERSIONING
* Fixed VERSION_SYMBOL macro to add the needed extra @ to make versioning work
properly
* Improved/Cla
On Thu, Sep 25, 2014 at 09:11:24AM -0700, Jeff Shaw wrote:
> Intel(R) Data Direct I/O Technology (Intel(R) DDIO) is a feature introduced
> with the Intel(R) Xeon(R) processor E5 family.
>
> It has been around for several years and is available at least on all Xeon
> E5 processors. DDIO is part o
Actually, in the demo that I saw they had probably used as many of the
accelerations as possible to get the kind of rates they described. Even if
we could see (a documentation of) what all things they used in this
particular application, it would help.
>From my discussions, it seemed as if there we
On Thu, Sep 25, 2014 at 07:27:21PM +, Anjali Kulkarni wrote:
> Actually, in the demo that I saw they had probably used as many of the
> accelerations as possible to get the kind of rates they described. Even if
> we could see (a documentation of) what all things they used in this
> particular a
> > > > When Flow Director was used together with bulk alloc, id and hash was
> > > > swapped when packet matches flow director filter due to improper fdir
> > > > field initialization.
> > > >
> > > > Signed-off-by: Pawel Wodkowski
> >
> > Reviewed-by: Helin Zhang
> > Reviewed-by: Jingjing Wu
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Thursday, September 25, 2014 6:24 PM
> To: Ananyev, Konstantin
> Cc: Jastrzebski, MichalX K; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] Change alarm cancel function to
> thread-safe:
>
> On Thu, Sep 25, 2014 at 04:03:48PM +00
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Thursday, September 25, 2014 4:05 PM
> To: Cao, Waterman
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] patches validation
>
> 2014-09-25 13:07, Cao, Waterman:
> > I will work with team to see if we can improve te
91 matches
Mail list logo