2015-06-24 14:34, Neil Horman:
> +Some ABI changes may be too significant to reasonably maintain multiple
> +versions. In those cases ABI's may be updated without backward compatibility
> +being provided. The requirements for doing so are:
> +
> +#. At least 3 acknowledgments of the need to do so m
2015-06-24 16:15, Neil Horman:
> On Wed, Jun 24, 2015 at 09:41:08PM +0200, Thomas Monjalon wrote:
> > 2015-06-24 14:34, Neil Horman:
> > > @@ -103,7 +103,7 @@
> > > #define VERSION_SYMBOL(b, e, v)
> >
> > Should be (b, e, n)
> >
> > > #define __vsym
> > > #define BASE_SYMBOL(b, n)
> >
> > Sho
2015-06-24 14:34, Neil Horman:
> @@ -103,7 +103,7 @@
> #define VERSION_SYMBOL(b, e, v)
Should be (b, e, n)
> #define __vsym
> #define BASE_SYMBOL(b, n)
Should be (b, e)
> -#define BIND_DEFAULT_SYMBOL(b, v)
> +#define BIND_DEFAULT_SYMBOL(b, e, n)
2015-06-24 18:15, Daniel Mrzyglod:
> The problem occure when we have config files on NFS filesystem.
You forgot to describe the problem.
> Solution with minimal change to library is to add buffer for I/O operations.
> Without buffering there is 3-10% of faulty Reading from NFS.
>
> The problem
The problem occure when we have config files on NFS filesystem.
Solution with minimal change to library is to add buffer for I/O operations.
Without buffering there is 3-10% of faulty Reading from NFS.
The problem seems to be only NFS filesystem.
Signed-off-by: Daniel Mrzyglod
---
lib/librte_c
Acked-by: David Harton
> -Original Message-
> From: Liang-Min Larry Wang [mailto:liang-min.wang at intel.com]
> Sent: Wednesday, June 17, 2015 6:22 PM
> To: dev at dpdk.org
> Cc: Andrew Harvey (agh); David Harton (dharton); Roger Melton (rmelton);
> Liang-Min Larry Wang
> Subject: [PATCH
On Wed, Jun 24, 2015 at 09:41:08PM +0200, Thomas Monjalon wrote:
> 2015-06-24 14:34, Neil Horman:
> > @@ -103,7 +103,7 @@
> > #define VERSION_SYMBOL(b, e, v)
>
> Should be (b, e, n)
>
> > #define __vsym
> > #define BASE_SYMBOL(b, n)
>
> Should be (b, e)
>
> > -#define BIND_DEFAULT_SYMBOL(b,
This patch depends on the Port Hotplug Framework.
It implements the eth_dev_uninit functions for rte_ixgbe_pmd and
rte_ixgbevf_pmd.
Changes in V5:
Set nb_rx_queues and nb_tx_queues to 0 in uninit functions.
Rebased to latest ixgbe code.
Changes in V4:
Release rx and tx queues in dev_uninit() func
When close one port twice, testpmd will give out wrong messagse.
testpmd> port stop 0
Stopping ports...
Checking link statuses...
Port 0 Link Up - speed 0 Mbps - full-duplex
Port 1 Link Up - speed 0 Mbps - full-duplex
Done
testpmd> port close 0
Closing ports...
Done
testpmd> port close 0
Closing
This patch adds management of PKT_RX_FDIR and PKT_RX_RSS_HASH ol_flags in vPMD.
Signed-off-by: Tomasz Kulasek
---
drivers/net/ixgbe/ixgbe_rxtx_vec.c | 41 +++-
1 file changed, 31 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_rxtx_vec.c
b/
2015-06-24 07:22, Chui Hui Chiu:
> How do I get the latest code for DPDK 2 with all fixes and patches since
> DPDK 2.0.0? Are they already integrated into the master version?
It is a work in progress. The master branch is the most up-to-date:
http://dpdk.org/browse/dpdk/log/
A new release
People have been asking for ways to use the ABI macros, heres some docs to
clarify their use. Included is:
* An overview of what ABI is
* Details of the ABI deprecation process
* Details of the versioning macros
* Examples of their use
* Details of how to use the ABI validator
Thanks to John Mcn
Clean up some macro definition typos and comments
Signed-off-by: Neil Horman
CC: thomas.monjalon at 6wind.com
---
lib/librte_compat/rte_compat.h | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/lib/librte_compat/rte_compat.h b/lib/librte_compat/rte_compat.h
inde
On Wed, Jun 24, 2015 at 11:23:26AM +, Mcnamara, John wrote:
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> > Sent: Tuesday, June 23, 2015 8:34 PM
> > To: dev at dpdk.org
> > Subject: [dpdk-dev] [PATCH 1/2] rte_compat.h : Clean up some ty
When stopping the bond device we don't need to try and free up the LACPDU's
from deactivated devices since this is covered by
bond_mode_8023ad_deactivate_slave().
This fixes the following:
[0.100569] PANIC in bond_ethdev_stop():
[0.100589] line 1172assert "port->rx_ring != NULL" fa
On 6/17/15, 3:22 PM, "Liang-Min Larry Wang"
wrote:
>This implementation is designed to provide a familar interface for
>applications that rely on kernel-space driver to support ethtool_op and
>net_device_op for device management. The initial implementation focuses
>on ops that can be implemented
On Tue, Jun 23, 2015 at 01:38:14PM +0200, Maciej Gajdzica wrote:
> Previous version of versioning macros didn't work properly. When
> building dpdk as static library, external application couldn't use
> versioned function. It happened, because versioning macros for static
> library were empty. Now
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Michal Jastrzebski
> Sent: Sunday, June 7, 2015 9:12 AM
> To: dev at dpdk.org
> Cc: =tomaszx.kulasek.intel.com at dpdk.org
> Subject: [dpdk-dev] [PATCH] vpmd: rss hash ol_flag for unified packet type
>
> From: T
2015-06-23 17:36, Olivier MATZ:
> On 06/22/2015 08:34 PM, Cyril Chemparathy wrote:
> > This series contains a few improvements that allow the DPDK code base
> > to build properly on machines that enforce strict pointer cast
> > alignment constraints.
> >
> > When dealing with packet data which coul
This patch adds support for new x550 PHY IDs:
0x0154 0x0223
0x0154 0x0221
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_phy.c | 4 +++-
drivers/net/ixgbe/base/ixgbe_type.h | 4 +++-
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_phy.c
b
This patch adds max_link_up_time parameter to mac structure.
This parameter is used to control maximum link polling time in
ixgbe_check_mac_link functions. It is required to prevent long wait
time on x550 PHY that have no external link.
Since x550 is handled by software, we have to reset internal
According to the hardware, the LINE side of the cs4227 needs to be
set to 10 Gbps SR mode regardless of the configuration of the link.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_phy.h | 2 +
drivers/net/ixgbe/base/ixgbe_x550.c | 143
2 file
This patch fixes comment description for setting 2.5G and 5G speeds in
ixgbe_setup_phy_link_generic.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_phy.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_phy.c
b/drivers/net/ixgbe/ba
The code in the if statement is specific to the KR PHY. However,
the conditional was only checking for backplane. If the driver
was using the KX4 backplane PHY, this code would execute but
would write to the wrong PHY.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 7 +++
This patch skips the PCI transactions pending check in
ixgbe_disable_pcie_master. The PCI transactions pending bit sticks high
when there were pending transactions, we should wait and then continue
with our reset flow.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_common.c | 3 +++
This path changes ((P == 0) ? () : ())
register definition into ((P) ? : )
style.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers/net/ixgbe/bas
This patch removes the clearing of the FEC(Forward Error Correction)
bits in ixgbe_setup_kr_speed_x550em. FEC default enablement is
configured via the NVM and SW should not be overriding these defaults
in this function.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 2 --
1
This patch disables SW LPLU on x557 V2. It also sets the
enter_lplu function pointer to NULL on x557 V2. LPLU will be
implemented in FW for all x557 V2 interfaces. The SW LPLU
implementation must be disabled on V2 to avoid conflicts with
FW. SW LPLU support is still required for x557 V1.
Signed-of
Add new values that vary by MAC type that are introduced in
x540, x550.
And remove some meaningless comments BTW.
Please note this patch hits a checkpatch errror.
"ERROR: Macros with complex values should be enclosed in parentheses
#105: FILE: drivers/net/ixgbe/base/ixgbe_type.h:3337:"
We think i
Use the mvals array to hold differing values use for CIAA and
CIAD symbols.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 18 --
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers/net/ixgbe/base/ixgb
Use the mvals array to hold differing values used for
IXGBE_*_GPI* symbols.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 41 +++--
1 file changed, 26 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers
Use the mvals array to hold differing values used for
IXGBE_I2C_* symbols.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 63 -
1 file changed, 49 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers/
Add a const u32 *mvals pointer to the ixgbe_hw struct to point to an
array of mac-type-dependent values. These can include register
offsets, masks, whatever can be in a u32. When the ixgbe_hw struct
is initialized, a pointer to the appropriate array must be set.
The IXGBE_I2CCTL register references
EEE(Energy Efficient Ethernet) is not supported on the initial revision
of IXGBE_DEV_ID_X550EM_X_KR. We determine the revision by reading a fuse
register.
Also, the requirements for FEC(Forward Error Correction) have changed
slightly. Now, we don't change the "request" bit at all. When EEE is
enab
This patch implements ixgbe_led_on_t_X550em and ixgbe_led_off_t_X550em
function for turning on and off LEDs on X557 external PHY.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 3 +++
drivers/net/ixgbe/base/ixgbe_x550.c | 52 +
drivers/ne
The MDIO clock needs to be configured for a specific speed for
x550em. We expected this to be done automatically, but in
the early days of the project this was not happening. We put
code in to do this ourselves.
Eventually, we decided that there is no harm in having SW do
this all the time, so we m
Ensure link is still up after getting the speed, to ensure that the
speed read is valid.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 79 ++---
1 file changed, 57 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_x550
This patch adds support for x550em KR/iXFI internal link modes. The initial
x550em-10GBASET and x550em-SFP designs use iXFI internal link mode between
the internal PHY and the external PHY.
However future designs will use a KR internal link. This patch is intended
to future proof the driver by addi
During init, check the ucode running in the CS4227. If
it is not responding correctly, reset the part. This is
a global reset so it must only be done the first time a
driver loads after power-on.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 78 +
Because each IOSF access requires the use of multiple registers,
use a semaphore to serialize those accesses.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 30 +-
1 file changed, 21 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ixgbe/ba
Add a helper function to wait for IOSF accesses to complete. Also
perform this wait before each access, as specified in the datasheet.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 85 ++---
1 file changed, 50 insertions(+), 35 deletions(-)
Because we acquire two semaphore bits before setting the SWFW_SYNC
register, we should release them in the reverse order that they
were acquired.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x540.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/
The FEC(Forward Error Correction) feature had been disabled
because it increases power consumption. However, some customers
want to use it. This patch enables FEC when EEE(Energy Efficient
Ethernet) is disabled; FEC was already being disabled when EEE
was enabled, but now both are done in the same
Since this 82599 device supports WoL(Wake on LAN), driver needs
a define for the sub device ID.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers/net/ixgbe/base/ixgbe_type.h
index c
x550em has two different bus speeds, 300 MHz and 400 MHz.
A bit in a fuse register tells which it is, with a 1 meaning
the bus is running at 300 MHz.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ixgbe/base/ixg
In UniPHY we have 2 IOSF targets that are UniPHY related. We can
write to PHY and PCS. In earlier times I've been told that there
were 2 separate PCS targets for IOSF commands and that's why I
implemented it with 2 defines and adding hw->bus.lan_id, but lately
I confirmed with HW that FW takes care
On systems that support LPLU in the firmware, the driver wouldn't be
aware of the LPLU speed change, and it wouldn't cache the new value when
the driver resumes. This patch emulates the same behavior by restoring
the previous autoneg settings to autoneg_advertised.
Signed-off-by: Wenzhuo Lu
---
This patch powers down the x550em PHY on over-temp events. The
PHY firmware is supposed to do this autonomously but that isn't
implemented. The short-term stop-gap solution is for SW to power
down the PHY when it reports an overtemp event.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixg
This patch updates ixgbe_supported_sfp_modules_X550em to return
IXGBE_ERR_SFP_NOT_PRESENT when SFP module is not present and updates
ixgbe_setup_mac_link_sfp_x550em to handle this return value.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 9 -
1 file changed, 8 ins
Fix the warning caused by an uninitialized variable which might have
been used.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c
b/drivers/net/ixgbe/base/ixgbe_x550.c
index e
This patch adds identify SFP module support for x550em. x550em
support returns an error for SFP module types not supported by
x550em design.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 101 +++-
drivers/net/ixgbe/base/ixgbe_x550.h | 1 +
Adds define for the number of bits needed to shift the EEE_SU register
in order to get the value of TEEE_DLY.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_type.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers/net/ixgbe/base/ixgbe_
This patch updates the x550em SFP link setup by adding
ixgbe_sfp_type_unknown and ixgbe_sfp_type_not_present case expression
to the ixgbe_setup_mac_link_sfp_x550em SFP type switch statement. This
handles the case when no module is present.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixg
Configure the CS4227 correctly for both 1G and 10G operation,
by moving the code to ixgbe_setup_mac_link_sfp_x550em(). It
needs to be in this function because we need both the module
type and the speed, and this is the only function in the init
flow that knows the speed. In contrast,
ixgbe_setup_sf
This patch updates x550em LPLU to use the MAC ops setup_link
function pointer. This removes redundant code and provides iXFI
and KR support between internal and external PHY.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/ixgbe_x550.c | 44 ++---
1 file chan
Update readme file to show the version of the base code.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/base/README | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/base/README b/drivers/net/ixgbe/base/README
index fa71d85..ac38708 100644
--- a/drivers/net/i
Short summary:
*update Low Power Link Up to use MAC ops link
*fix 1G and 10G link stability for x550em SFP+
*update x550em SFP link setup
*add shift define for EEE_SU.TEEE_DLY
*add x550em identify SFP module support
*return err when SFP module is not present
*power down the x550em PHY on overtemp e
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> Sent: Tuesday, June 23, 2015 8:34 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 1/2] rte_compat.h : Clean up some typos
>
> Clean up some macro definition typos and comments
Hi,
There is
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Tuesday, June 23, 2015 8:34 PM
> To: dev at dpdk.org
> Cc: Neil Horman; Mcnamara, John; thomas.monjalon at 6wind.com
> Subject: [PATCH 2/2] ABI: Add some documentation
>
> People have been asking for ways to
On Wed, Jun 24, 2015 at 10:04:53AM +0300, Vladimir Medvedkin wrote:
> I published changes to rte_lpm_tbl24_entry only because it was just an idea
> :)
Understood. Just wanted to be sure I understood it right to convert it into C
code. :)
> different defines and checks should be modified too.
Ye
On Fortville NIC, link status change interrupt callback is not executed when
slave in bonding is (re-)started. It causes that slave's NIC is inactive even
if its link status is up on the start.
This patch invokes lsc callback, just after port's start, to check its initial
link status and manage pr
Newer kernels make this unreadable for security reasons for non-roots.
Running the application will then fill the logs with
rte_mem_virt2phy: cannot open /proc/self/pagemap
messages.
However, there are cases when DPDK is and should be run as non-root,
without the need for virtual-to-physical a
Hi Matthew,
I published changes to rte_lpm_tbl24_entry only because it was just an idea
:) So rte_lpm_tbl8_entry should look like:
struct rte_lpm_tbl8_entry {
uint32_t next_hop :24; /**< next hop. */
uint32_t valid :1; /**< Validation flag. */
uint32_t valid_group :1;
On Mon, 22 Jun 2015 11:58:32 -0700
Cyril Chemparathy wrote:
> This series adds support for the EZchip TILE-Gx family of SoCs. The
> architecture port in itself is fairly straight forward due to its
> reliance on generics for the most part.
>
> In addition to adding TILE-Gx architecture specific
Kernel mmap syscall and VFIO kernel driver expect unsigned long for offset.
The BAR index to work with inside VFIO kernel driver will be the wrong one
except for BAR 0.
The patch solves the issue.
On 6/5/2015 12:26 AM, Bernard Iremonger wrote:
> This patch depends on the Port Hotplug Framework.
> It implements the eth_dev_uninit functions for rte_i40e_pmd and
> rte_i40evf_pmd.
>
> Signed-off-by: Bernard Iremonger
> ---
Tested-by: Michael Qiu
> drivers/net/i40e/i40e_ethdev.c| 79
>
On 6/9/2015 12:01 AM, Bernard Iremonger wrote:
> This patch depends on the Port Hotplug Framework.
> It implements the eth_dev_uninit functions for rte_ixgbe_pmd and
> rte_ixgbevf_pmd.
>
> Changes in V4:
> Release rx and tx queues in dev_uninit() functions.
> Replace TRUE and FALSE with 1 and 0.
>
On 6/9/2015 12:01 AM, Bernard Iremonger wrote:
> This patch depends on the Port Hotplug Framework.
> It implements the eth_dev_uninit functions for rte_ixgbe_pmd and
> rte_ixgbevf_pmd.
>
> Changes in V4:
> Release rx and tx queues in dev_uninit() functions.
> Replace TRUE and FALSE with 1 and 0.
>
Hello,
How do I get the latest code for DPDK 2 with all fixes and patches since
DPDK 2.0.0? Are they already integrated into the master version?
Many thanks,
Chui-hui
Hi,
I am running DPDP KNI application on linux (3.18 kernel) VM (ESXi 5.5),
directly connected to another linux box to measure throughput using iperf
tool. Link speed: 1Gbps. Maximum throughput I get is 50% with 1470 Bytes.
With 512B pkt sizes, throughput drops to 282 Mbps.
Tried using KN
70 matches
Mail list logo
