2016-05-03 18:12, Jerin Jacob:
> Elements of struct rte_eth_dev used in the fast path.
> Make struct rte_eth_dev cache aligned to avoid the cases where
> rte_eth_dev elements share the same cache line with other structures.
>
> Signed-off-by: Jerin Jacob
Let's try it in real tests.
Applied,
2016-06-18 02:27, Hiroyuki Mikita:
> This commit fixes some functions missing in API documentation.
>
> Signed-off-by: Hiroyuki Mikita
Applied, thanks
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, June 22, 2016 5:18 PM
> To: Lu, Wenzhuo; Jerin Jacob
> Cc: dev at dpdk.org; Ananyev, Konstantin; Stephen Hemminger; Richardson,
> Bruce; Chen, Jing D; Liang, Cunming; Wu,
Hi Jerin,
> -Original Message-
> From: Jerin Jacob [mailto:jerin.jacob at caviumnetworks.com]
> Sent: Wednesday, June 22, 2016 7:07 PM
> To: Thomas Monjalon
> Cc: Lu, Wenzhuo; dev at dpdk.org; Ananyev, Konstantin; Stephen Hemminger;
> Richardson, Bruce; Chen, Jing D; Liang, Cunming; Wu,
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, June 23, 2016 1:02 AM
> To: Lu, Wenzhuo
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3] ethdev: fix DCB config issue on ixgbe
>
> 2016-05-06 05:33, Wenzhuo Lu:
> >
Bifurcated driver is a mechanism which depends the advanced
Ethernet device to split traffic between queues. It provides
the capability to let the kernel driver and DPDK driver co-exist
and take their advantage.
It is achieved by using SRIOV and NIC's advanced filtering. This
patch describes it
This patch describes how bifurcated driver is done and
adds the user guide on i40e NICs.
Signed-off-by: Jingjing Wu
---
doc/guides/nics/i40e.rst | 140 +++
1 file changed, 140 insertions(+)
diff --git a/doc/guides/nics/i40e.rst
Hi,
Do we have assumption each core will have a separate crypto device?
I have only one crypto device with several queues which are shared by
multi-cores.
As I run the l2fwd-crypto, only one cparam is valid and has a good dev_id.
The other cparam is crashing.
The reason is the other qconf
Nelio,
I'll check. Thanks!
-Original Message-
From: N?lio Laranjeiro [mailto:nelio.laranje...@6wind.com]
Sent: 2016?6?22? 20:31
To: Cristian Dumitrescu ; Chao Zhu
Cc: dev at dpdk.org
Subject: librte_meter compilation fails on IBM Power8
Hi Cristian, Chao,
I have encountered a
Update base driver for ixgbe, mainly work on new features and bug fixes.
v4 changes:
Update some commit logs.
Merge related patched.
v3 changes:
Fix some commit log issues.
Beilei Xing (29):
ixgbe/base: add new VF requests for mailbox API
ixgbe/base: add sgmii link for X550
ixgbe/base:
This patch adds two new VF requests of IXGBE_VF_GET_RETA and
IXGBE_VF_GET_RSS_KEY for mailbox API.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_mbx.h | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.h
This patch adds new phy type and media type to support
sgmii link for X550, and add ixgbe_setup_sgmii to support
sgmii link setup.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_type.h | 9 +++
drivers/net/ixgbe/base/ixgbe_x550.c | 127 +---
2
An error code indicating that the PF rejects the MAC address change
should be returned, in case that the PF has already assigned a MAC
for the VF.
Fixes: af75078fece3 ("first public release")
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_vf.c | 4 +++-
1 file changed, 3
When software sends commands to firmware using the host
slave command interface, firmware fails to recieve the
request command with a checksum failed error.
This patch sets command checksum to the default value of
0xFF according to datasheet, therefore the checksum won't
be checked by firmware.
Break ixgbe_setup_eee_X550 down to better handle a change from if
statements to switch statements needed to add X550em_a KR support.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_x550.c | 174 ++--
1 file changed, 105 insertions(+), 69 deletions(-)
Use the method pointers instead of direct function calls so that
the right thing will happen on X550EM_a.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_x550.c | 76 ++---
1 file changed, 38 insertions(+), 38 deletions(-)
diff --git
This patch updates ixgbe_setup_mac_link_sfp_x550a for X550 SFP+.
ixgbe_set_lan_id_multi_port_pcie has been updated to set the MAC
instance(0/1) which is needed when configuring the external PHY,
since X550a has two instances of MGPK. The MAC instance is read
from the EEPROM.
Signed-off-by: Beilei
This patch adds clearing the pool mappings when configuring default
MAC addresses for the interface. Without this there will be the risk
of leaking an address into pool 0 which really belongs to VF 0 when
SR-IOV is enabled.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_82599.c |
This patch adds ixgbe_setup_mac_link_sfp_x550a for X550a SFP.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_type.h | 4 +++
drivers/net/ixgbe/base/ixgbe_x550.c | 64 -
drivers/net/ixgbe/base/ixgbe_x550.h | 3 ++
3 files changed, 70
Implement KR support for X550em_a devices.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_x550.c | 53 +
1 file changed, 42 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c
b/drivers/net/ixgbe/base/ixgbe_x550.c
When there is an error getting the PHY token, the error path
fails to release the locks that it has taken. Release those
locks in that failure case.
Fixes: 86b8fb293fdf ("ixgbe/base: add sw-firmware sync for resource sharing on
X550em_a")
Signed-off-by: Beilei Xing
---
This patch renames IXGBE_PVFTTDLEN to IXGBE_PVFTDLEN according to
abbreviation of Transmit Descriptor Length in datasheet.
Signed-off-by: Beilei Xing
---
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
The ixgbe_vf.h file did not use __ and instead used
which is not the standard used in every other file.
Fixes: af75078fece3 ("first public release")
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_vf.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
NW_MNG_IF_SEL register is critical PHY link configuration, add
ixgbe_read_mng_if_sel_x550em to read NW_MNG_IF_SEL, validate
register values and save fields such as PHY MDIO address, optimize
internal PHY mode determination.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_type.h | 2
It adds new phy definitions.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_phy.c | 16 +-
drivers/net/ixgbe/base/ixgbe_type.h | 14 +-
drivers/net/ixgbe/base/ixgbe_x550.c | 303 ++--
drivers/net/ixgbe/base/ixgbe_x550.h | 43 +
4 files
There're two device IDs changed from 15C6/15C7 to 15E4/15E5 cause
PHY info changes. Make the change and use 15C6/15C7 for the backplane
SGMII. Clean up some discovery kludges from the previous shared ID,
and also add 15C6/15C7 to ixgbe_set_mdio_speed just for paranoia
to control MDIO speed even
For X540 and forward it is possible if a system reset occur at the
right time to leave the SWFW semaphore high. This new function will
attempt to grab and release the semaphore. If the grab times out it
will still release the semaphore placing it in a known good state.
The idea is to call this
This patch fixes possible race issue between ports when issuing host
interface command by acquiring/releasing the management host interface
semaphore in ixgbe_host_interface_command.
Fixes: 36f43e8679ae ("ixgbe/base: refactor manageability block communication")
Signed-off-by: Beilei Xing
---
This patch simplifies the adding and removing VLANs from
VFTA/VLVF/VLVFB registers. The logic to determine registers has
been simplified to (vid / 32) and (1 - vid / 32). Many conditional
paths and checks are no longer needed with this patch.
Signed-off-by: Beilei Xing
---
This patch adds support for the VLVF to be bypassed when adding or
removing a VFTA entry. The PF can utilize the default pool while
preserving the VLVF for the VFs use.
Meanwhile, update corresponding VF ops and drivers where corresponding
ops is invoked.
Signed-off-by: Beilei Xing
---
Since PCIe standard defines maximum of 8 functions per device lan_id
is a value 0..7. Because of that, lan_id don't need to be u16.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_common.c | 2 +-
drivers/net/ixgbe/base/ixgbe_type.h | 2 +-
2 files changed, 2 insertions(+), 2
This patch fixes endianness issues about host interface command.
Fixes: ad66a85dce9a ("ixgbe/base: new FW values")
Fixes: 0790adeb5675 ("ixgbe/base: support X550em_a device")
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_osdep.h | 1 +
drivers/net/ixgbe/base/ixgbe_type.h | 17
Make ixgbe_set_mac_anti_spoofing() consistent with the other
functions that deal with setting VLAN and Ethertype spoofing by
changing the prototype to accept a VF parameter.
Also change the logic for writing the PFVFSPOOF register to be similar
to the MAC and Ethertype functions.
Signed-off-by:
This patch changes static keyword to STATIC definition, which can be
redefined depending on the compiler used.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_x550.c | 38 ++---
1 file changed, 19 insertions(+), 19 deletions(-)
diff --git
This patch adds X550a flow control auto negotiation support.
ixgbe_setup_fc_x550a and ixgbe_fc_autoneg_X550a functions where
added to setup and enable flow control. MAC ops function pointer
fc_autoneg was added so that hardware specific fc autoneg functions
can be called from
The ixgbe base driver was updated refer to version
cid-10g-shared-code.2016.04.12 released by ND.
The changes include:
Added sgmii link for X550.
Added mac link setup for X550a SFP and SFP+.
Added KR support for X550em_a.
Added new phy definitions for M88E1500.
Added support for the VLVF to be
A work around for a new crosstalk erratum that causes link flap in
entry cages has been introduced. So this patch defines the bit in
NVM that will tell software if this work around is needed.
Signed-off-by: Beilei Xing
---
drivers/net/ixgbe/base/ixgbe_type.h | 1 +
1 file changed, 1
Tested-by: Chen, Zhaoyan
* Commit: 3901ed99c2f82d3e979bb1bea001d61898241829
* Patch Apply: Success
* Compilation: Success
* Kernel/OS: 3.11.10-301.fc20.x86_64
* GCC: 4.8.3 20140911
* Case 1
./app/test -cf -n4
cryptodev_sw_kasumi_autotest
KASUMI Unittest execution is successful.
Thanks,
Joey
>
Implicit int to enum conversion is not allowed when icc is used as
the compiler. It raises the compiling error like,
/.../dpdk/drivers/net/virtio/virtio_user/vhost_user.c(257):
error #188: enumerated type mixed with another type
msg.request = req;
^
The fix
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Azarewicz, PiotrX T
> Sent: Thursday, May 12, 2016 1:20 PM
> To: Mrozowicz, SlawomirX ; Richardson,
> Bruce
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] lpm: unchecked return value
>
> Hi,
>
> I
On Thu, Jun 23, 2016 at 09:13:22AM +0100, Jastrzebski, MichalX K wrote:
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Azarewicz, PiotrX T
> > Sent: Thursday, May 12, 2016 1:20 PM
> > To: Mrozowicz, SlawomirX ; Richardson,
> > Bruce
> > Cc: dev at
On Wed, Jun 22, 2016 at 05:32:37PM +0200, Olivier Matz wrote:
> On 06/22/2016 02:20 PM, Thomas Monjalon wrote:
> > 2016-06-22 13:57, Olivier Matz:
> >> Hi Thomas,
> >>
> >> On 06/22/2016 01:49 PM, Thomas Monjalon wrote:
> >>> 2016-06-22 14:34, Panu Matilainen:
> --- a/lib/librte_port/Makefile
Hi Bruce,
On 06/23/2016 10:53 AM, Bruce Richardson wrote:
>>> If we want to specifically build this directory, why preventing us to do
>>> so with CONFIG_RTE_LIBRTE_PORT?
>>
>> If we call foo_sub with CONFIG_FOO=n, it will generate a library and
>> install headers in the build directory, however
On 6/15/2016 10:03 AM, Jianfeng Tan wrote:
...
>
> diff --git a/config/common_linuxapp b/config/common_linuxapp
> index 7e698e2..2483dfa 100644
> --- a/config/common_linuxapp
> +++ b/config/common_linuxapp
> @@ -43,3 +43,4 @@ CONFIG_RTE_LIBRTE_VHOST=y
> CONFIG_RTE_LIBRTE_PMD_VHOST=y
>
> -Original Message-
> From: Richardson, Bruce
> Sent: Thursday, June 23, 2016 10:48 AM
> To: Jastrzebski, MichalX K
> Cc: Azarewicz, PiotrX T ; Mrozowicz,
> SlawomirX ; dev at dpdk.org
> Subject: Re: [PATCH] lpm: unchecked return value
>
> On Thu, Jun 23, 2016 at 09:13:22AM +0100,
Update l3fwd documentation with -E, -L and --eth-dest options.
Signed-off-by: Beilei Xing
---
doc/guides/sample_app_ug/l3_forward.rst | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/doc/guides/sample_app_ug/l3_forward.rst
b/doc/guides/sample_app_ug/l3_forward.rst
Add the missing braces to the 'if' statements to fix the misleading
identation. This also fixes the following build errors when building
with gcc >= 6:
drivers/net/e1000/base/e1000_phy.c:4156:2:
error: this 'if' clause does not guard... [-Werror=misleading-indentation]
if (locked)
^~
On Thu, Jun 09, 2016 at 03:23:43PM +0100, Bruce Richardson wrote:
> On Mon, May 23, 2016 at 01:33:42AM +, Wu, Jingjing wrote:
> >
> >
> > > -Original Message-
> > > From: Xing, Beilei
> > > Sent: Friday, May 20, 2016 11:17 PM
> > > To: Wu, Jingjing
> > > Cc: dev at dpdk.org; Xing,
hi markos,
please see : cba50f6be0db9efdf694dcf4bce4a6945a275182, which should already
fix this.
--
thanks
anupam
On Thu, Jun 23, 2016 at 2:55 PM, Markos Chandras wrote:
> Add the missing braces to the 'if' statements to fix the misleading
> identation. This also fixes the following build
Hi Anupam,
I have seen your commit, but my patch fixes a different file (although
the fix is similar).
Am I missing something?
On 2016-06-23 11:26, Anupam Kapoor wrote:
> hi markos,
>
> please see : cba50f6be0db9efdf694dcf4bce4a6945a275182, which should
> already
> fix this.
>
> --
> thanks
On Thu, Jun 23, 2016 at 4:04 PM, Markos Chandras wrote:
> I have seen your commit, but my patch fixes a different file (although the
> fix is similar).
>
> Am I missing something?
>
?ah no. my bad. sorry about that.
?--
thanks
anupam?
On Mon, Apr 18, 2016 at 12:06:06AM -0300, Ricardo Salveti wrote:
> Fix build failure since PAGE_SIZE is not defined on ARM (multiple values
> are possible, so it needs to dynamically get the page size used).
>
> Signed-off-by: Ricardo Salveti
> ---
> drivers/net/xenvirt/rte_eth_xenvirt.c | 3
To prevent double 0x in logs
Signed-off-by: Ferruh Yigit
---
drivers/net/vmxnet3/vmxnet3_rxtx.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c
b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index ccafc0c..1e0fde0 100644
---
On 6/22/2016 8:08 AM, Simon K?gstr?m wrote:
> On 2016-06-21 15:11, Ferruh Yigit wrote:
>> On 6/20/2016 5:19 PM, Stephen Hemminger wrote:
>>> On Mon, 20 Jun 2016 12:44:35 +0200
>>> Simon Kagstrom wrote:
>>>
I.e., avoid dump messages with double 0x0x, e.g.,
dump mbuf at
On Wed, Jun 15, 2016 at 08:35:30AM +, Wu, Jingjing wrote:
>
>
> > -Original Message-
> > From: Tao, Zhe
> > Sent: Tuesday, June 14, 2016 1:24 PM
> > To: dev at dpdk.org
> > Cc: Tao, Zhe ; Wu, Jingjing
> > Subject: [PATCH v3] i40e: fix olflags for vector Rx
> >
> > Problem:
> > The
On Tue, Jun 14, 2016 at 03:20:44PM +0800, Beilei Xing wrote:
> The patch enables configuring the ether types of both inner and
> outer VLANs.
>
This patch also changes the behaviour for configuring the ether type of a single
vlan tag (as with the i40e patch we previously discussed). That must
On Tue, Jun 14, 2016 at 04:52:28PM -0700, Nelson Escobar wrote:
> The enic_clsf_init() function is called once per enic instance, but it
> used a static name to create the hash table. Consequently when using
> more than one enic instance, there was a name collision which caused
> errors:
>
>
On Tue, Jun 14, 2016 at 04:55:34PM -0700, Nelson Escobar wrote:
> When trying to release the mbufs, the function was incorrectly
> iterating over the max size configured instead of the actual size
> of the ring.
>
> Fixes: cbb44dddcade ("enic: receive path performance improvements")
>
>
Hi,
Thank you for providing more doc.
I am just not convinced it is the right place
and it must be discussed.
Some questions below to help deciding where it must be.
2016-06-23 11:02, Jingjing Wu:
> Bifurcated driver is a mechanism which depends the advanced
> Ethernet device to split traffic
From: Zyta Szpak
Version 4 of fixing the fixed register width assumption.
The app was allocating too little space for 64-bit registers
which resulted in memory corruption. This commit resolves
this by getting the number of registers and size of register
by
Hi Bruce,
> -Original Message-
> From: Richardson, Bruce
> Sent: Thursday, June 23, 2016 6:14 PM
> To: Wu, Jingjing
> Cc: Xing, Beilei ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4] i40e: configure MTU
>
> On Thu, Jun 09, 2016 at 03:23:43PM +0100, Bruce Richardson wrote:
> > On
On Thu, Jun 23, 2016 at 02:37:42PM +0100, Xing, Beilei wrote:
> Hi Bruce,
>
> > -Original Message-
> > From: Richardson, Bruce
> > Sent: Thursday, June 23, 2016 6:14 PM
> > To: Wu, Jingjing
> > Cc: Xing, Beilei ; dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH v4] i40e: configure MTU
On Wed, Jun 08, 2016 at 11:26:40AM +, Mcnamara, John wrote:
>
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Chen Jing D(Mark)
> > Sent: Wednesday, June 8, 2016 9:44 AM
> > To: thomas.monjalon at 6wind.com
> > Cc: dev at dpdk.org; Chen, Jing D
>
On Thu, Jun 16, 2016 at 12:56:09AM +, Lu, Wenzhuo wrote:
> Hi,
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Konstantin Ananyev
> > Sent: Wednesday, June 15, 2016 8:59 PM
> > To: dev at dpdk.org
> > Cc: Ananyev, Konstantin
> > Subject: [dpdk-dev]
replaced strncpy with snprintf for safely
copying the strings.
Cverity issue 127350: string overflow
Fixes: 278f945402c5 ("pdump: add new library for packet capture")
Signed-off-by: Reshma Pattan
---
lib/librte_pdump/rte_pdump.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
This patchset includes listed fixes
1)fix default socket path in pdump library.
2)fix coverity issues in pdump library.
3)fix coverity issues in pdump tool.
v3:
added new patch for fixing default socket paths "HOME" and "/var/run".
reworked coverity fixes on top of the above change.
v2:
fixed
replaced strncpy with snprintf for safely
copying the strings.
Coverity issue 127351: string overflow
Fixes: caa7028276b8 ("app/pdump: add tool for packet capturing")
Signed-off-by: Reshma Pattan
---
app/pdump/main.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
SOCKET_PATH_HOME is to specify environment variable "HOME",
so it should not contain "/pdump_sockets" in the macro.
So remove "/pdump_sockets" from SOCKET_PATH_HOME
and create new macro for "/pdump_sockets". Similary removed
"/pdump_sockets" from SOCKET_PATH_VAR_RUN.
Changes are done in
Using gcc 6.1, in some cases, kni fails to compile
because of unused variables:
build/lib/librte_eal/linuxapp/kni/ixgbe_main.c:82:19:
error: ?ixgbe_copyright?
defined but not used [-Werror=unused-const-variable=]
static const char ixgbe_copyright[] =
^~~
2016-06-23 15:38, Pablo de Lara:
> Using gcc 6.1, in some cases, kni fails to compile
> because of unused variables:
Curiosity: is it only with gcc 6.1? Have you tested other versions?
> lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_main.c | 4
> 1 file changed, 4 deletions(-)
We all
On Wed, Jun 22, 2016 at 11:30:05AM +0200, Adrien Mazarguil wrote:
> On Wed, Jun 22, 2016 at 10:19:14AM +0100, Bruce Richardson wrote:
> > On Wed, Jun 22, 2016 at 10:20:54AM +0200, Adrien Mazarguil wrote:
> > > On Tue, Jun 21, 2016 at 05:42:29PM +0100, Ferruh Yigit wrote:
> > > > On 6/21/2016 8:23
On Wed, Jun 15, 2016 at 10:47:07PM -0700, Rasesh Mody wrote:
> From: Sony Chacko
>
> Add support for setting the MTU.
>
> Signed-off-by: Sony Chacko
> ---
Patch is missing an update to overview.rst for the new feature support.
If there is no objection, I'll add that into it on apply for you.
On Wed, Jun 15, 2016 at 10:47:02PM -0700, Rasesh Mody wrote:
> Hi Bruce,
>
> v2 series:
> - made subject and description of the patches verbose where needed.
> - extended stats patch: some of the stats may sound similar, however,
>they are different stats collected by different hw
2016-05-16 11:33, Simon Kagstrom:
> This allows releasing RX/TX queue memory.
> ---
> We're using DPDK 16.04 and have a test suite which performs a sequence
> of separate tests of the type
>
>allocate mempool
>rte_eth_dev_configure(port, n_rxq, n_txq, ...)
>setup rx/tx queues
>
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, June 23, 2016 3:59 PM
> To: De Lara Guarch, Pablo
> Cc: dev at dpdk.org; Zhang, Helin
> Subject: Re: [dpdk-dev] [PATCH] kni: fix compilation with gcc 6.1
>
> 2016-06-23 15:38, Pablo de
Enhance mlx5 with a data path that bypasses Verbs.
The first half of this patchset removes support for functionality completely
rewritten in the second half (scatter/gather, inline send), while the data
path is refactored without Verbs.
The PMD remains usable during the transition.
This
Fixes: 701c8d80c820 ("pci: support class id probing")
Signed-off-by: Nelio Laranjeiro
---
drivers/crypto/qat/rte_qat_cryptodev.c | 5 +
drivers/net/mlx4/mlx4.c| 18 ++
drivers/net/mlx5/mlx5.c| 24
Except for the first time when memory registration occurs, the lkey is
always cached. Since memory registration is slow and performs system calls,
performance can be improved by moving that code to its own function outside
of the data path so only the lookup code is left in the original inlined
This is done in preparation of bypassing Verbs entirely for the data path
as a performance improvement. TX gather cannot be maintained during the
transition and will be reimplemented later.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_ethdev.c | 2
This is done in preparation of bypassing Verbs entirely for the data path
as a performance improvement. RX scatter cannot be maintained during the
transition and will be reimplemented later.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_ethdev.c |
There is no scatter/gather support anymore, CONFIG_RTE_LIBRTE_MLX5_SGE_WR_N
has no purpose and can be removed.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
config/common_base | 1 -
doc/guides/nics/mlx5.rst | 7 ---
drivers/net/mlx5/Makefile| 4
Inline TX will be fully managed by the PMD after Verbs is bypassed in the
data path. Remove the current code until then.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
config/common_base | 1 -
doc/guides/nics/mlx5.rst | 10 --
drivers/net/mlx5/Makefile
To keep the data path as efficient as possible, move fields only useful to
the control path into new structure txq_ctrl.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5.c| 21 +++--
drivers/net/mlx5/mlx5_ethdev.c | 28 +++---
To keep the data path as efficient as possible, move fields only useful to
the control path into new structure rxq_ctrl.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5.c | 6 +-
drivers/net/mlx5/mlx5_fdir.c | 8 +-
drivers/net/mlx5/mlx5_rxq.c
The latest version of Mellanox OFED exposes hardware definitions necessary
to implement data path operation bypassing Verbs. Update the minimum
version requirement to MLNX_OFED >= 3.3 and clean up compatibility checks
for previous releases.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien
These structures and macros extend those exposed by libmlx5 (in mlx5_hw.h)
to let the PMD manage work queue and completion queue elements directly.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_prm.h | 163
Hi,
I got an connection failure to my stmp server in the middle.
If you want a v6 tell me.
Regards,
--
N?lio Laranjeiro
6WIND
> -Original Message-
> From: Panu Matilainen [mailto:pmatilai at redhat.com]
> Sent: Wednesday, June 22, 2016 12:34 PM
> To: dev at dpdk.org
> Cc: Dumitrescu, Cristian ;
> zhuangwj at gmail.com
> Subject: [PATCH] port: fix build when KNI support is not enabled
>
> Commit 9fc37d1c071c is
> -Original Message-
> From: N?lio Laranjeiro [mailto:nelio.laranjeiro at 6wind.com]
> Sent: Wednesday, June 22, 2016 1:31 PM
> To: Dumitrescu, Cristian ; Chao Zhu
>
> Cc: dev at dpdk.org
> Subject: librte_meter compilation fails on IBM Power8
>
> Hi Cristian, Chao,
>
> I have
This commit adds Python script for generating diagram of the application
configuration file. This script requires graphviz package to be installed
on the machine. The input config file is translated to an output file in
DOT syntax, which is then used to create the image file using graphviz.
To
2016-06-02 02:39, Zhe Tao:
> On Tue, May 03, 2016 at 01:51:10PM +0800, Jingjing Wu wrote:
> > NSH packet can be recognized by Intel X710/XL710 series. This
> > patch set enables it.
> >
> > Jingjing Wu (2):
> > mbuf: new NSH packet type
> > i40e: NSH packet type support
> >
> Acked-by: Zhe
> -Original Message-
> From: Simon Kagstrom [mailto:simon.kagstrom at netinsight.net]
> Sent: Friday, June 10, 2016 7:29 AM
> To: Dumitrescu, Cristian ;
> stephen at networkplumber.org; dev at dpdk.org;
> thomas.monjalon at 6wind.com
> Subject: [PATCH / RFC] sched: Correct subport
> On Jun 23, 2016, at 3:52 AM, Ferruh Yigit wrote:
>
> To prevent double 0x in logs
>
> Signed-off-by: Ferruh Yigit
> ?
Acked-by: Yong Wang
> drivers/net/vmxnet3/vmxnet3_rxtx.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c
>
On Thu, 23 Jun 2016 07:40:20 +
Jianfeng Tan wrote:
> Implicit int to enum conversion is not allowed when icc is used as
> the compiler. It raises the compiling error like,
> /.../dpdk/drivers/net/virtio/virtio_user/vhost_user.c(257):
> error #188: enumerated type mixed with
The compilation for 32-bit fails when CONFIG_RTE_VIRTIO_USER is enabled:
drivers/net/virtio/virtio_user_ethdev.c:84:47:
error: format ?%llu? expects argument of type ?long long unsigned int?,
but argument 5 has type ?size_t {aka unsigned int}?
Fixes: e9efa4d93821 ("net/virtio-user: add
2016-06-23 22:53, Thomas Monjalon:
> The compilation for 32-bit fails when CONFIG_RTE_VIRTIO_USER is enabled:
>
> drivers/net/virtio/virtio_user_ethdev.c:84:47:
> error: format ?%llu? expects argument of type ?long long unsigned int?,
> but argument 5 has type ?size_t {aka unsigned
The enic_clsf_init() function is called once per enic instance, but it
used a static name to create the hash table. Consequently when using
more than one enic instance, there was a name collision which caused
errors:
EAL: memzone_reserve_aligned_thread_unsafe():
memzone already exists
RING:
enic_alloc_consistent() allocated memory, but enic_free_consistent()
was an empty function, so allocated memory was never freed.
This commit adds a list and lock to the enic structure to keep track
of the memzones allocated in enic_alloc_consistent(), and
enic_free_consistent() uses that
98 matches
Mail list logo