2015-06-05 16:07, Andrew Harvey:
> On 6/5/15, 3:46 AM, "Thomas Monjalon" wrote:
> >Stephen and me say the same thing about using the ethdev API.
>
> And your would have a point would be valid if dpdk were available to every
> interface we support (it is not) and on every processor architecture
2015-06-05 17:01, Bruce Richardson:
> The macro to turn on additional debug output when the app was compiled
> with "-DDEBUG" was missing a ";".
It shows that such dead code is almost never tested.
It would be saner if this command would return no result:
git grep 'ifdef.*DEBUG' examples
Hi Avinash,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Yeddula, Avinash
> Sent: Friday, June 5, 2015 6:06 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] 4 Traffic classes per Pipe limitation
>
> Hi,
> This is related to the QOS scheduler functionality
Extend dump_cfg to also display statistcs information for given DPDK
ports and rename the application to proc_info as it's now a utility
doing a little more than just dumping the memory information for DPDK.
Signed-off-by: Maryam Tahhan
---
app/Makefile | 2 +-
app/dump_cfg/Makefile
Extend testpmd to show additional aggregate extended stats.
Signed-off-by: Maryam Tahhan
---
app/test-pmd/config.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index f788ed5..b42d83f 100644
--- a/app/test-pmd/config.c
+++
Extend rte_eth_xstats_get to retrieve additional stats from the device
driver as well the top level extended stats. Add additional drop
counters to the extended stats.
Signed-off-by: Maryam Tahhan
---
lib/librte_ether/rte_ethdev.c | 12
lib/librte_ether/rte_ethdev.h | 4
2
Implement xstats_get() and xstats_reset() in dev_ops for ixgbe to expose
detailed error statistics to DPDK applications.
Signed-off-by: Maryam Tahhan
---
drivers/net/ixgbe/ixgbe_ethdev.c | 160 +--
1 file changed, 138 insertions(+), 22 deletions(-)
diff
This patch implements xstats_get() and xstats_reset() in dev_ops for ixgbe to
expose detailed error statistics to DPDK applications.
The dump_cfg application is extended to demonstrate the usage of retrieving
statistics for DPDK interfaces and renamed to proc_info in order reflect this
new
There is same problem by using ovs-dpdk
Maybe that is because recv of VM is poor,so no free tx?descriptors,ovs-dpdk
drop the
pktsRTFSC*--Zhou,
Tianlin ?2015?6?5?(???) 17:23dev at
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Sugumaran, Varthamanan
> Sent: Friday, June 05, 2015 6:22 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] ACL-Dynamic Adding or Deleting rules
>
> Hi,
> Is there is a way to add/delete a single ACL rule in
On 6/5/15, 5:47 AM, "Bruce Richardson" wrote:
>On Fri, Jun 05, 2015 at 11:25:09AM +, Wang, Liang-min wrote:
>>
>>
>> > -Original Message-
>> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
>> > Sent: Friday, June 05, 2015 6:47 AM
>> > To: Andrew Harvey (agh)
>> > Cc:
On 05/06/15 17:06, Jay Rolette wrote:
> The past few days I've been trying to chase down why operations over KNI
> are so bloody slow. To give you an idea how bad it is, we did a simple test
> over an NFS mount:
>
> # Mount over a non-KNI interface (eth0 on vanilla Ubuntu 14.04 LTS)
> $ time
Hello Stephen,
Wednesday, June 3, 2015, 8:40:14 PM, you wrote:
> On Tue, 2 Jun 2015 15:11:30 +0200
> Roman Dementiev wrote:
>>
>> This series of patches adds methods that use hardware memory transactions
>> (HTM)
>> on fast-path for DPDK locks (a.k.a. lock elision). Here the methods are
>>
From: "Chen Jing D(Mark)"
Both PF and VF shared code in function fm10k_stats_get().
The function works well with PF, but has problem with VF since
VF has less queues than PF.
Signed-off-by: Chen Jing D(Mark)
---
drivers/net/fm10k/fm10k_ethdev.c |2 +-
1 files
From: "Chen Jing D(Mark)"
Original implementation required mbuf size should be greater than
ETHER_MAX_VLAN_FRAME_LEN, which is not necessary. If it's less
than that value, scatter function will be selected and incoming
packets greater than mbuf size will be filled into
From: "Chen Jing D(Mark)"
Add functions to support promiscuous/allmulticast enable and
disable.
Signed-off-by: Chen Jing D(Mark)
---
drivers/net/fm10k/fm10k_ethdev.c | 118 +-
1 files changed, 117 insertions(+), 1 deletions(-)
diff
From: "Chen Jing D(Mark)"
The patch set add promiscuous mode configuration and 2 bug fixes.
Chen Jing D(Mark) (3):
fm10k: Add promiscuous mode support
fm10k: remove mbuf size sanity check
fm10k: Fix improper max queue number for VF
drivers/net/fm10k/fm10k_ethdev.c
The macro to turn on additional debug output when the app was compiled
with "-DDEBUG" was missing a ";".
Fixes: 07db4a975094 ("examples/distributor: new sample app")
Signed-off-by: Anbarasan Murugesan
Signed-off-by: Bruce Richardson
---
examples/distributor/main.c | 2 +-
1 file changed, 1
Fix RTE_MBUF_METADATA macros to allow for unaligned accesses to
meta-data fields.
Forcing aligned accesses is not really required, so this is removing an
unneeded constraint.
This issue was met during testing of the new version of the ip_pipeline
application. There is no performance impact.
This
RTE_EAL_RX_INTR will be removed from v2.2. It's only used to avoid
ABI(unannounced) broken in v2.1.
The users should make sure understand the impact before turning on the feature.
There are two abi changes required in this interrupt patch set.
They're 1) struct rte_intr_handle; 2) struct
Demonstrate how to handle per rx queue interrupt in a NAPI-like
implementation in usersapce. PDK polling thread mainly works in
polling mode and switch to interrupt mode only if there is no
any packet received in recent polls.
User space interrupt notification generally takes a lot more cycles
The patch does below for igb PF:
- Setup NIC to generate MSI-X interrupts
- Set the IVAR register to map interrupt causes to vectors
- Implement interrupt enable/disable functions
Signed-off-by: Danny Zhou
Signed-off-by: Cunming Liang
---
v9 changes
- move queue-vec mapping init from
The patch does below things for ixgbe PF and VF:
- Setup NIC to generate MSI-X interrupts
- Set the IVAR register to map interrupt causes to vectors
- Implement interrupt enable/disable functions
Signed-off-by: Danny Zhou
Signed-off-by: Yong Liu
Signed-off-by: Cunming Liang
---
v10 changes
-
The patch adds two dev_ops functions to enable and disable rx queue interrupts.
In addtion, it adds rte_eth_dev_rx_intr_ctl/rx_intr_q to support per port or
per queue rx intr event set.
Signed-off-by: Danny Zhou
Signed-off-by: Cunming Liang
fix by http://www.dpdk.org/dev/patchwork/patch/4784/
To make bsd compiling happy with new intr changes.
Signed-off-by: Cunming Liang
---
v8 changes
- add stub for new function
v7 changes
- remove stub 'linux only' function from source file
lib/librte_eal/bsdapp/eal/eal_interrupts.c | 19 ++
The new UIO generic handle type was introduced by patch.
http://dpdk.org/ml/archives/dev/2015-April/017008.html
When using uio_pci_generic and turning on lsc interrupt, it complains fd read
error.
The root cause is the 'count' size of read is not correct.
Reported-by: Yong Liu
Signed-off-by:
The patch exposes intr event fd create and release for PMD.
The device driver can assign the number of event associated with interrupt
vector.
It also provides misc functions to check 1) allows other slowpath intr(e.g.
lsc);
2) intr event on fastpath is enabled or not.
Signed-off-by: Cunming
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
---
v8 changes
- move eventfd creation out of the setup_interrupts to a standalone function
Signed-off-by: Cunming Liang
---
lib/librte_eal/linuxapp/eal/eal_interrupts.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_interrupts.c
b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
index 08dc2ab..4499055 100644
---
The patch adds 'rte_intr_rx_ctl' to add or delete interrupt vector events
monitor on specified epoll instance.
Signed-off-by: Cunming Liang
---
v10 changes:
- add RTE_INTR_HANDLE_UIO_INTX for uio_pci_generic
v8 changes
- fix EWOULDBLOCK and EINTR processing
- add event status check
v7
The patch adds 'rte_epoll_wait' and 'rte_epoll_ctl' for async event wakeup.
It defines 'struct rte_epoll_event' as the event param.
The 'op' uses the same enum as epoll_wait/ctl does.
The epoll event support to carry a raw user data and to register a callback
which is executed during wakeup.
The patch adds interrupt vectors support in rte_intr_handle.
'vec_en' is set when interrupt vectors are detected and associated event fds
are set.
Those event fds are stored in efds[].
'intr_vec' is reserved for device driver to initialize the vector mapping table.
When the event fds add to a
v11 changes
- typo cleanup and check kernel style
v10 changes
- code rework to return actual error code
- bug fix for lsc when using uio_pci_generic
v9 changes
- code rework to fix open comment
- bug fix for igb lsc when both lsc and rxq are enabled in vfio-msix
- new patch to turn off
Refactor the ieee1588_fwd mode in testpmd to use the new ethdev
APIs to enable and read IEEE1588 PTP timestamps.
Signed-off-by: John McNamara
---
app/test-pmd/ieee1588fwd.c | 443 ++---
1 file changed, 13 insertions(+), 430 deletions(-)
diff --git
Add ixgbe support for new ethdev APIs to enable and read IEEE1588
PTP timestamps.
Signed-off-by: John McNamara
---
drivers/net/ixgbe/ixgbe_ethdev.c | 118 +++
1 file changed, 118 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c
Add e1000/igb support for new ethdev APIs to enable and read
IEEE1588 PTP timestamps.
Signed-off-by: John McNamara
---
drivers/net/e1000/igb_ethdev.c | 118 +
1 file changed, 118 insertions(+)
diff --git a/drivers/net/e1000/igb_ethdev.c
Add ethdev API to enable and read IEEE1588 PTP timestamps from
nics that support it. The following functions are added:
rte_eth_ieee1588_enable()
rte_eth_ieee1588_disable()
rte_eth_ieee1588_read_rx_timestamp()
rte_eth_ieee1588_read_tx_timestamp()
Signed-off-by: John McNamara
---
This patchset adds ethdev API to enable and read IEEE1588 PTP timestamps from
devices that support it. The following functions are added:
rte_eth_ieee1588_enable()
rte_eth_ieee1588_disable()
rte_eth_ieee1588_read_rx_timestamp()
rte_eth_ieee1588_read_tx_timestamp()
The "ieee1588"
Signed-off-by: Jingjing Wu
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 761172e..2cd3461 100644
---
enable mirror functionality in i40e driver
.mirror_rule_set
.mirror_rule_reset
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 334 +
drivers/net/i40e/i40e_ethdev.h | 23 +++
2 files changed, 357 insertions(+)
diff --git
This path renames the mirror type in rte_eth_mirror_conf and macros,
and rework the mirror set in ixgbe dirvers by using new definition.
It also fixes some coding style.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 42 +--
rename rte_eth_vmdq_mirror_conf to rte_eth_mirror_conf and move
the maximum rule id check from ethdev level to driver
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 22 +++---
drivers/net/ixgbe/ixgbe_ethdev.c | 11 +++
drivers/net/ixgbe/ixgbe_ethdev.h
This patch set enables mirror functionality in i40e driver, and redefines
structure and macros used to configure mirror.
v2 changes:
- correct comments style
- add doc change
v3 changes:
- change the mirror rule type to support bit mask and avoid ABI broken
- fix code style
Jingjing Wu
On 6/5/15, 3:46 AM, "Thomas Monjalon" wrote:
>2015-06-04 22:10, Andrew Harvey:
>> On 6/4/15, 7:58 AM, "Stephen Hemminger"
>>wrote:
>> >"Andrew Harvey (agh)" wrote:
>> >> I believe that their is value in this interface for software stacks
>>not
>> >> based on Linux being moved toward DPDK that
Keith, your mail is very long but it's maybe on purpose to show that
there are too many logs ;)
2015-06-05 12:32, Wiles, Keith:
> On 6/5/15, 5:00 AM, "Thomas Monjalon" wrote:
> >2015-05-27 15:10, Wiles, Keith:
> >> I would like to have the log-level default changed to not log
> >>everything,
>
2015-06-05 11:25, Wang, Liang-min:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Stephen and me say the same thing about using the ethdev API.
> > We don't understand why using a fake ethtool lib would be easier.
> > Though you are saying it "facilitated [your] adoption to
It announces the planned ABI changes for interrupt mode on v2.2.
The feature will turn off by default so as to avoid v2.1 ABI broken.
Signed-off-by: Cunming Liang
---
doc/guides/rel_notes/abi.rst | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc/guides/rel_notes/abi.rst
Signed-off-by: Pablo de Lara
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 9362c19..189c41c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -339,6 +339,7 @@ F: doc/guides/sample_app_ug/l3_forward_access_ctrl.rst
Hashes
M: Bruce Richardson
Usually hash tables not only store keys, but also data associated
to them. In order to maintain the existing API,
the key index will still be returned when
looking up/deleting an entry, but user will be able
to store/look up data associated to a key.
Signed-off-by: Pablo de Lara
---
Added rehash function to be able to keep adding more entries,
when a key cannot be added, as a result of a loop
evicting entries infinitely.
Also added reset function to be able to empty the table,
without having to destroy and create it again.
Signed-off-by: Pablo de Lara
---
Previous implementation was lacking a function
to look up a burst of entries, given precalculated hash values.
This patch implements such function, quite useful for
looking up keys from packets that have precalculated hash values
from a 5-tuple key.
Added the function in the hash unit test as
This patch replaces the existing hash library with another approach,
using the Cuckoo Hash method to resolve collisions (open addressing),
which pushes items from a full bucket when a new entry tries
to be added in it, storing the evicted entry in an alternative location,
using a secondary hash
rte_prefetchX functions included volatile void *p as parameter,
but the function does not modify it, so it should include the const keyword.
Signed-off-by: Pablo de Lara
---
lib/librte_eal/common/include/arch/ppc_64/rte_prefetch.h | 6 +++---
This patchset is to replace the existing hash library with
a more efficient and functional approach, using the Cuckoo hash
method to deal with collisions. This method is based on using
two different hash functions to have two possible locations
in the hash table where an entry can be.
So, if a
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Daniel Mrzyglod
> Sent: Friday, June 5, 2015 3:55 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] lib: fix RTE_MBUF_METADATA macros
>
> Fix RTE_MBUF_METADATA macros to allow for unaligned accesses to
>
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, June 05, 2015 9:41 AM
> To: Wang, Liang-min
> Cc: Andrew Harvey (agh); Stephen Hemminger; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/2] ethtool: add new library to provide
>
On 6/5/15, 8:58 AM, "Thomas Monjalon" wrote:
>Keith, your mail is very long but it's maybe on purpose to show that
>there are too many logs ;)
Yes it was kind of the point, but only because I wanted to show the
complete output.
>
>2015-06-05 12:32, Wiles, Keith:
>> On 6/5/15, 5:00 AM, "Thomas
On Fri, Jun 05, 2015 at 11:25:09AM +, Wang, Liang-min wrote:
>
>
> > -Original Message-
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Sent: Friday, June 05, 2015 6:47 AM
> > To: Andrew Harvey (agh)
> > Cc: Stephen Hemminger; Wang, Liang-min; dev at dpdk.org
> >
This patch sets the setup_EEE function pointer to NULL for the
interfaces which do not support EEE. Currently only the KR backplane
interface (0x15AB) supports EEE. Setting this pointer to NULL prevents
EEE registers from being incorrectly modified and gives base drivers a
flag to check for EEE
This patch adds x550em PHY reset function ixgbe_reset_phy_t_X550em.
ixgbe_reset_phy_t_X550em calls the reset PHY generic, and then enables
the x550em PHY LASI(Link Alarm Status Interrupt) interrupts.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 21 ++---
Set the lan_id before the first I2C access. The existing call was
clearly being done after a previous I2C access in the same function
and that can't be right, so call the set_lan_id method earlier. At
this point it probably doesn't matter for this QSFP function, but
it makes sense to do it
This patch adds ixgbe_check_link_t_X550em for checking x550em
PHY link. We check that both the MAC and external x550em PHY have link.
This is to avoid a false link up between the internal and external PHY
when the external PHY doesn't have link.
Signed-off-by: Wenzhuo Lu
---
This patch adds x550em external PHY interrupt and forced 1G/10G
support. Support includes enabling and handling Link Status
Change and Thermal Sensor interrupt. ixgbe_handle_lasi has been added
to the API for handling the interrupts received from x550em PHY.
ixgbe_enable_lasi_ext_t_x550em and
This patch adds x550em Auto neg Flow Control support to
ixgbe_device_supports_autoneg_fc and sets the x550em setup_fc function
pointer to ixgbe_setup_fc_generic. ixgbe_setup_fc_generic is used for
x550em because flow control is setup on the external PHY via MDIO, when
ixgbe_setup_fc_X550em sets up
This patch cleans up the ixgbe_setup_internal_phy_ x550em() function as follows:
- Renames it to ixgbe_setup_internal_phy_t_x550em to clarify that it is
specific to copper
- Returns an error if called for non-copper devices
- Corrects the comments
- Removed the LASI(Link Alarm Status
This patch changes the return value for ixgbe_setup_internal_phy_t_x550em
when link is down to IXGBE_SUCCESS.
The driver will call ixgbe_setup_internal_phy_t_x550em when a link status
change is reported. The links status change can occur on link up or link
down, and if the link status change is
The following patch moves the handling of the I2C MUX (which is only
used for x550em SFP+ devices) out of the ixgbe_x540.c file and
into the ixgbe_x550.c file where it belongs.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x540.c | 30 +-
The init flow is simplified. We no longer wait for the PHY FW init
complete bit to be set as this bit is only set once by the PHY at power
on and then cleared on the first read. So only the first instance of
running SW (or possibly MAC FW) needs to initialize the PHY.
The PHY initialization has
For the KR backplane which is different from other backplane,
in that we can't use auto-negotiation to determine the
mode. Instead, use whatever the user configured.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_api.c| 12 +
drivers/net/ixgbe/base/ixgbe_api.h| 1 +
This patch adds SW Low Power Link Up (LPLU) support for x550em PHY.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_api.c | 13 +++
drivers/net/ixgbe/base/ixgbe_api.h | 1 +
drivers/net/ixgbe/base/ixgbe_type.h | 17 +++-
drivers/net/ixgbe/base/ixgbe_x550.c | 173
This patch adds SFP+ dual-speed support.
82599 fiber link code was moved from ixgbe_82599.c to ixgbe_commom.c
for use by X550em, and the API was updated to support the common code
usage. SFP MAC link code is added to x550em.
Signed-off-by: Changchun Ouyang
Signed-off-by: Wenzhuo Lu
---
The introduction of ixgbe_get_bus_info_X550em failed to call the
set_lan_id method to set the func and lan_id and deal with port-
swapped configurations. Add the call to resolve the problem.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 2 ++
1 file changed, 2
The FEC feature can improve BER(Bit Error Rate) but uses more power
to do so. It also cannot be used with EEE(Energy Efficient Ethernet).
EEE is an important feature, and we have no known BER issues, so FEC
is not needed.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 4
The I2C mux control relies on the SDP setting in the ESDP register
so it is necessary to restore the value after a MAC reset. So,
put the code in a function so it can be used in more than one place.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 41
On some hardware platforms, the CS4227 does not initialize properly.
Detect those cases and reset it appropriately.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_phy.h | 12 +++
drivers/net/ixgbe/base/ixgbe_x550.c | 184
2 files changed, 196
The driver now needs to issue a firmware command to inform the
firmware that a driver is coming up. This prevents the possibility
of the firmware and the driver configuring the PHY at the same
time. Upon completion of the command, the firmware will no longer
be configuring the PHY.
Signed-off-by:
A retry count of 10 is likely to run into problems on X550 devices
that have to detect and reset unresponsive CS4227 devices. So,
reduce the I2C retry count to 3 for X550 and above. This should
avoid any possible regressions in existing devices.
Signed-off-by: Wenzhuo Lu
---
Most I2C accesses take and release semaphores for each access. It's
also necessary to perform multiple I2C operations under the same
holding of the semaphore, so provide unlocked I2C methods for that
purpose.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_api.c | 68
Remove the function which is not called by the drivers.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_common.c | 30 --
drivers/net/ixgbe/base/ixgbe_common.h | 1 -
2 files changed, 31 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_common.c
All bits in FDIRTCPM and FDIRUDPM are set to 1 when
ixgbe_fdir_set_input_mask_82599 is called. Not settings these bits will cause
TCP and UDP packets to be filtered out when NVGRE or VXLAN mode is enabled.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_82599.c | 7 ++-
1 file
This patch moves the check of the return value from
ixgbe_start_hw_generic after the function is called.
Previously we had the code to disable relaxed ordering in
between, which seems a bit out of place.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_82598.c | 5 +++--
1 file
Hi,
Is there is a way to add/delete a single ACL rule in librte_acl?
I had looked at the library and found no method to add/delete a rule
dynamically to the existing ACL context.
Please let me know if there are any alternate ways of doing it.
Thanks
Vartha
There's a typo in the code comment for FC end
of Frame Exception (FCEOFe/IPE), so fixed the typo.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
Remove the redundant "from".
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_common.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_common.c
b/drivers/net/ixgbe/base/ixgbe_common.c
index 0174ecb..8179354 100644
---
Update copyright in every file.
Update README file.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/README| 4 ++--
drivers/net/ixgbe/base/ixgbe_82598.c | 2 +-
drivers/net/ixgbe/base/ixgbe_82598.h | 2 +-
drivers/net/ixgbe/base/ixgbe_82599.c | 2 +-
Short summary:
*update copyright and readme
*fix code comment, double from
*fix typo error in code comment
*check return value after calling
*allow tunneled UDP and TCP frames to reach their destination
*erase ixgbe_get_hi_status
*provide unlocked I2C methods
*reduce I2C retry count on X550
Hi,
This is related to the QOS scheduler functionality provided by dpdk.
I see a limit on the number of traffic classes to be 4. I'm exploring the
available options to increase that limit to 8.
This is what I found when I researched on this topic.
The limitation on number's of TC (and pipes)
2015-06-04 22:10, Andrew Harvey:
> On 6/4/15, 7:58 AM, "Stephen Hemminger" wrote:
> >"Andrew Harvey (agh)" wrote:
> >> I believe that their is value in this interface for software stacks not
> >> based on Linux being moved toward DPDK that need simple operations like
> >> getting the mac
On 6/5/15, 5:00 AM, "Thomas Monjalon" wrote:
>2015-05-27 15:10, Wiles, Keith:
>> I would like to have the log-level default changed to not log
>>everything,
>> but the user needs to enable the log messages if he needs to see more
>> information. Normally applications or systems are not so
2015-05-28 19:26, Dumitrescu, Cristian:
> I think we have the following options identified so far for stats collection
> configuration:
>
> 1. Stats configuration through the RTE_LOG_LEVEL
> 2. Single configuration flag global for all DPDK libraries
> 3. Single configuration flag per DPDK
On Fedora 22, the "ar" binary operates by default in deterministic mode,
making the "u" parameter irrelevant, and leading to warning messages
getting printed in the build output like below.
INSTALL-LIB librte_kvargs.a
ar: `u' modifier ignored since `D' is the default (see `U')
There are two
2015-05-27 15:10, Wiles, Keith:
> I would like to have the log-level default changed to not log everything,
> but the user needs to enable the log messages if he needs to see more
> information. Normally applications or systems are not so verbose, but if
> needed the user enables the verbose or
This is a modification of qos_sched example to use
librte_cfgfile for parsing configuration file.
Signed-off-by: Michal Jastrzebski
---
examples/qos_sched/cfg_file.c | 157 ++---
examples/qos_sched/cfg_file.h | 35 ++---
examples/qos_sched/init.c |
rte_vhost_driver_unregister will remove the listenfd from event list, and then
close it.
Signed-off-by: Huawei Xie
Signed-off-by: Peng Sun
---
lib/librte_vhost/rte_virtio_net.h| 3 ++
lib/librte_vhost/vhost_cuse/vhost-net-cdev.c | 9
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, June 05, 2015 6:47 AM
> To: Andrew Harvey (agh)
> Cc: Stephen Hemminger; Wang, Liang-min; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/2] ethtool: add new library to provide
>
MAC address with fixed VLAN 0 was removed. VF MAC/VLAN filter was enabled
for the default value. Removed all VLAN and MAC address table entries when
the system(e.g. testpmd) was closed.
Signed-off-by: Shaopeng He
---
drivers/net/fm10k/fm10k_ethdev.c | 38 +++---
On Thu, Jun 04, 2015 at 02:01:19PM -0700, Navneet Rao wrote:
> Running ---
>
>
>
> ./testpmd -c7 -n3 --vdev=eth_pcap0,iface=vEth0 --vdev=eth_pcap1,iface=vEth1
> -- -i --nb-cores=2 --nb-ports=2 --total-num-mbufs=1024
>
>
>
> results in a
>
>
>
> EAL: Error - exiting with code: 1
>
>
When we get the address of vring descriptor table in VHOST_SET_VRING_ADDR
message,
will try to reallocate virtio_net device and virtqueue to the same numa node.
Signed-off-by: Huawei Xie
---
config/common_linuxapp| 1 +
lib/librte_vhost/Makefile | 4 ++
use rte_malloc/free for virtio_net and virt queue allocation/free
Signed-off-by: Huawei Xie
---
lib/librte_vhost/virtio-net.c | 19 ++-
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/lib/librte_vhost/virtio-net.c b/lib/librte_vhost/virtio-net.c
index
The virtio_net device and vhost virt queue should be allocated on the same numa
node as vring descriptors.
When we firstly allocate the virtio_net device and vhost virt queue, we don't
know the numa node of vring descriptors.
When we receive the VHOST_SET_VRING_ADDR message, we get the numa node
1 - 100 of 124 matches
Mail list logo