The patch introduces following commands.
- port attach [ident]
- port detach [port_id]
- attach: attaching a port
- detach: detaching a port
- ident: pci address of physical device.
Or device name and paramerters of virtual device.
(ex. :02:00.0, eth_pcap0,iface=eth0)
- p
This patch adds finalization code to free resources allocated by the
PMD.
v4:
- Change function name.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_pmd_pcap/rte_eth_pcap.c | 40 ++
1 file changed, 40 insertions(+)
diff --git a/lib/librte_pmd_pcap/rte_eth_pca
The patch enables CONFIG_RTE_LIBRTE_EAL_HOTPLUG in Linux configuration.
Signed-off-by: Tetsuya Mukawa
---
config/common_linuxapp | 5 +
1 file changed, 5 insertions(+)
diff --git a/config/common_linuxapp b/config/common_linuxapp
index 2f9643b..27d05be 100644
--- a/config/common_linuxapp
+++
These functions are used for attaching or detaching a port.
When rte_eal_dev_attach() is called, the function tries to realize the
device name as pci address. If this is done successfully,
rte_eal_dev_attach() will attach physical device port. If not, attaches
virtual devive port.
When rte_eal_dev_
This new parameter is needed to keep device type like physical or virtual.
Port detaching processes are different between physical and virtual.
This paramerter lets detaching function know a device type of the port.
v4:
- Fix comments of rte_eth_dev_type.
Signed-off-by: Tetsuya Mukawa
---
app/t
- Add rte_eal_pci_close_one_dirver()
The function is used for closing the specified driver and device.
- Add pci_invoke_all_drivers()
The function is based on pci_probe_all_drivers. But it can not only
probe but also close drivers.
- Add pci_close_all_drivers()
The function tries to find a
The function removes the specified devargs entry from devargs_list.
Also the patch adds sanity checking to rte_eal_devargs_add().
v4:
- Fix sanity check code
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/common/eal_common_devargs.c | 57 +
lib/librte_eal/common/i
The patch adds functions for unmapping igb_uio resources. The patch is only
for Linux and igb_uio environment. VFIO and BSD are not supported.
v4:
- Add paramerter checking.
- Add header file to determine if hotplug can be enabled.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/common/Makefil
The patch adds following functions.
- rte_eth_dev_save()
The function is used for saving current rte_eth_dev structures.
- rte_eth_dev_get_changed_port()
The function receives the rte_eth_dev structures, then compare
these with current values to know which port is actually
attached or deta
The patch adds function pointer to rte_pci_driver and eth_driver
structure. These function pointers are used when ports are detached.
Also the patch adds rte_eth_dev_uninit(). So far, it's not called
by anywhere, but it will be called when port hotplug function is
implemented.
v4:
- Add paramerter
This patch adds rte_eth_dev_free(). The function is used for changing a
attached status of the device that has specified name.
v4:
- Add paramerter checking.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_ether/rte_ethdev.c | 20
lib/librte_ether/rte_ethdev.h | 11 +++
This patch replaces pci_addr_comparison() and memcmp() of pci addresses by
eal_compare_pci_addr().
v4:
- Fix calculation method of eal_compare_pci_addr().
- Add parameter checking.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 16 +--
lib/librte_eal/c
To remove assumption, do like followings.
This patch adds "RTE_PCI_DRV_DETACHABLE" to drv_flags of rte_pci_driver
structure. The flags indicates the driver can detach devices at runtime.
Also remove assumption that port will not be detached.
To remove the assumption.
- Add 'attached' member to rt
This patch series adds a dynamic port hotplug framework to DPDK.
With the patches, DPDK apps can attach or detach ports at runtime.
The basic concept of the port hotplug is like followings.
- DPDK apps must have responsibility to manage ports.
DPDK apps only know which ports are attached or deta
On 2015/01/19 18:49, Qiu, Michael wrote:
> Hi Tetsuya,
>
> I see two lines log like:
>
> EAL: cannot open VFIO container, error 2 (No such file or directory)
> EAL: VFIO support could not be initialized
>
> It should not happen as I have a patch merged to master branch.
>
> There should have some
I would be happy with the original suggestion. If the ethdev data for a port in
use is in cache it removes the performance concern associated the current setup
and my fix. The original suggestion also fixes the crash that I was seeing
because of memory being reserved from a numa node with no "--
Hi,
On 01/19/2015 05:33 PM, Neil Horman wrote:
> On Mon, Jan 19, 2015 at 11:23:07AM +0800, Helin Zhang wrote:
>> As there are only 6 bit flags in ol_flags for indicating packet types,
>> which is not enough to describe all the possible packet types hardware
>> can recognize. For example, i40e hard
2015-01-12 19:16, Konstantin Ananyev:
> +/*
> + * Select highest avaialbe classify method as default one.
Typo here.
Actually, I've seen few typos when browsing your big patchset
but I don't remember exactly where.
Maybe that a tool like codespell could help.
[...]
> + else if (rte_cpu_get_f
2015-01-12 19:16, Konstantin Ananyev:
> /*
> + * ACL RT structure is a set of multibit tries (with stride == 8)
> + * represented by an array of transitions. The next position is calculated
> + * based on the current position and the input byte.
> + * Each transition is 64 bit value with the follo
2015-01-14 13:39, Neil Horman:
> On Mon, Jan 12, 2015 at 07:16:04PM +, Konstantin Ananyev wrote:
> > v2 changes:
> > - When build with the compilers that don't support AVX2 instructions,
> > make rte_acl_classify_avx2() do nothing and return an error.
> > - Remove unneeded 'ifdef __AVX2__' in a
Hi Doherty,
On 2015/01/19 17:53, Doherty, Declan wrote:
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tetsuya Mukawa
>> Sent: Monday, January 19, 2015 7:11 AM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] rte_eal_init() causes PANIC using latest DPDK code
Hi list,
I got " multiple definition of 'rte_power_freq_min' " link error, while
compiling my dpdk project.
I think extern keyword was missed in rte_power.h header for some reason:
/**
?* Scale down the frequency of a specific lcore to the lowest according to the
?* available frequencies.
?* Re
Hi,
2015-01-19 16:55, ?? :
> Hi list,
>
> I got " multiple definition of 'rte_power_freq_min' " link error, while
> compiling my dpdk project.
> I think extern keyword was missed in rte_power.h header for some reason:
You're right.
It can be seen in this commit:
http://dpd
> -Original Message-
> From: Zhang, Helin
> Sent: Monday, January 19, 2015 3:23 AM
> To: dev at dpdk.org
> Cc: Liang, Cunming; Liu, Jijiang; Ananyev, Konstantin; Zhang, Helin
> Subject: [RFC 01/17] mbuf: add definitions of unified packet types
>
> As there are only 6 bit flags in ol_flag
Hi,
I've got following error using latest DPDK code.
Does someone face the same issue?
In my environment, It seems almost all DPDK apps (at lease l2fwd and
testpmd) cannot work.
EAL: Support maximum 128 logical core(s) by configuration.
EAL: Detected 8
Any comments?
On 2015/1/15 22:45, Qiu, Michael wrote:
> For ixgbe, when queue start failure, for example, mbuf allocate
> failure, the device will still start success, which could be
> an issue.
>
> Add return status check of queue start to avoid this issue.
>
> Signed-off-by: Michael Qiu
> ---
>
Hi Konstantin,
On 01/19/2015 02:04 PM, Ananyev, Konstantin wrote:
>> case 2) calculate checksum of out_ip and out_udp
>>
>>mb->l2_len = len(out_eth)
>>mb->l3_len = len(out_ip)
>>mb->ol_flags |= PKT_TX_IPV4 | PKT_TX_IP_CSUM | PKT_TX_UDP_CKSUM
>>set out_ip checksum to 0 in the packet
RSS offloads supported 'ip' and 'udp' only, which did not demonstrate
all of the hardware capabilities. The modifications adds support of
new RSS offloads of 'tcp', 'sctp', 'ether' and 'all'.
Signed-off-by: Helin Zhang
---
app/test-pmd/cmdline.c | 19 +++
1 file changed, 15 inser
RSS offload types were defined seperately for 1/10G and 40G NICs,
and have no relationship with flow types. The modifications are to
unify all RSS offload types for all PMDs. Unified RSS offload types
have new and common names which can be used for any PMD or
applications, and decouple from specifi
Flow types was defined actually for i40e hardware specifically,
and wasn't able to be used for defining RSS offload types of all
PMDs. It removed the enum flow types, and uses macros instead
with new names. The new macros can be used for defining RSS
offload types later. Also modifications are made
It wrongly calculates the size of the flow type mask array. The fix
is to align the flow type maximum index ID with the number of
element bit width, and then divide the number of element bit width.
Signed-off-by: Helin Zhang
---
lib/librte_ether/rte_eth_ctrl.h | 6 +-
1 file changed, 5 inser
Added code style fixes.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_i40e/i40e_fdir.c | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/lib/librte_pmd_i40e/i40e_fdir.c b/lib/librte_pmd_i40e/i40e_fdir.c
index 68511c8..c9e535b 100644
--- a/lib/librte_pmd_i40e/i40e_fdir.c
Added code style fixes.
Signed-off-by: Helin Zhang
---
lib/librte_ether/rte_eth_ctrl.h | 32
1 file changed, 16 insertions(+), 16 deletions(-)
diff --git a/lib/librte_ether/rte_eth_ctrl.h b/lib/librte_ether/rte_eth_ctrl.h
index 5d9c387..4308eae 100644
--- a/lib/
Added code style fixes.
Signed-off-by: Helin Zhang
---
app/test-pmd/config.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 97b6525..87dedf9 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -1918,11 +
It unifies the flow types and RSS offload types for all PMDs. Previously
flow types are defined actually for i40e, and there has different RSS
offloads tyeps for 1/10G and 40G seperately. This is not so convenient
for application development, and not good for adding new PMDs. In
addition, it enable
Hi, Tetsuya
You see lots of places have below three lines:
if (rte_eth_dev_validate_port(port_id) == DEV_INVALID) {
PMD_DEBUG_TRACE("Invalid port_id=%d\n", port_id);
return -EINVAL;
}
They are all the same(only few has the print log different), so
Looks like "make test" was added recently? I have "1.8.0" version and when
I run "make test" after "make install T=x86_64-native-linuxapp-gcc" I get
following errors
root at user-PC:/home/rkerur/dpdk-new-5/dpdk# make test
No test found, please do a 'make build' first, or specify O=
root at user-PC
This patch incorporate fixes to support DCB in SRIOV mode for testpmd.
Signed-off-by: Pawel Wodkowski
---
app/test-pmd/cmdline.c |4 ++--
app/test-pmd/testpmd.c | 39 +--
app/test-pmd/testpmd.h | 10 --
3 files changed, 31 insertions(+), 22 del
Add support for DCB in SRIOV mode. When no PFC is enabled this feature
might be used as multiple queues for VF (up to 8 queues if VFs num is
less or equal 16 or 4 if FVs num is less or equal 32).
The PF must initializes RX in ETH_MQ_RX_VMDQ_DCB and TX in
ETH_MQ_TX_VMDQ_DCB.
VF should initialize Rx
If SRIOV is used and device configuration does not use MQ the
RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool is set to 1 in
rte_eth_dev_check_mq_mode().
This is bad becouse of two reasons:
1. Port reconfiguration from non-MQ mode to MQ mode is impossible
2. Confguring RX and TX side in different way is impos
Allow zero rx/tx queues to be passed to rte_eth_dev_configure(). This
way PF might be used only for configuration purpose when no receive
and/or transmit functionality is needed.
Rationale:
in SRIOV mode PF use first free VF to RX/TX (at least ixgbe based NICs).
For example: if using 82599EB based
v2:
- Split patch for easer review.
- Remove "pmd: add api version negotiation for ixgbe driver" and "pmd: extend
mailbox api to report number of RX/TX queues" patches as those are already
already marged from other patch
v1:
This patchset enables DCB in SRIOV (ETH_MQ_RX_VMDQ_DCB and ETH_MQ_TX
This patch adds unit tests for mode 4. It is split into separate
file to avoid problems with other modes that does not need to
look into packets payload.
This patch includes also a modification of maximum number of ports
used in their tests for bonding modes 0-3 from 16 to 6.
Signed-off-by: Tomasz
Signed-off-by: Tomasz Kulasek
Signed-off-by: Pawel Wodkowski
---
app/test/test.h | 112 ---
1 file changed, 66 insertions(+), 46 deletions(-)
diff --git a/app/test/test.h b/app/test/test.h
index 896f7db..5450986 100644
--- a/app/test/test.h
+
Update v2:
Patch split - separated test.h rework and unit test for mode 4 patch.
Tomasz Kulasek (2):
Unit tests - test.h rework
Unit tests for mode 4
app/test/Makefile |1 +
app/test/test.h| 112 +--
app/test/test_link_bonding.c |2 +-
app/
On Mon, Jan 19, 2015 at 06:16:02PM +0100, Thomas Monjalon wrote:
> 2015-01-14 13:39, Neil Horman:
> > On Mon, Jan 12, 2015 at 07:16:04PM +, Konstantin Ananyev wrote:
> > > v2 changes:
> > > - When build with the compilers that don't support AVX2 instructions,
> > > make rte_acl_classify_avx2()
Hi, Tetsuya
What I miss? I haven't got the first patch, just start with "PATCH 2/11".
Does this only happens with my mail client?
Thanks,
Michael
On 2015/1/19 18:42, Tetsuya Mukawa wrote:
> This patch series adds a dynamic port hotplug framework to DPDK.
> With the patches, DPDK apps can attach
On Mon, Jan 19, 2015 at 06:27:02PM +0100, Olivier MATZ wrote:
> Hi,
>
> On 01/19/2015 05:33 PM, Neil Horman wrote:
> > On Mon, Jan 19, 2015 at 11:23:07AM +0800, Helin Zhang wrote:
> >> As there are only 6 bit flags in ol_flags for indicating packet types,
> >> which is not enough to describe all t
Hi Olivier,
> -Original Message-
> From: Olivier MATZ [mailto:olivier.matz at 6wind.com]
> Sent: Friday, January 16, 2015 5:28 PM
> To: Ananyev, Konstantin; Liu, Jijiang
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/3] enhance TX checksum command and csum
> forwarding engine
On 2015/1/19 18:06, Tetsuya Mukawa wrote:
> On 2015/01/19 18:49, Qiu, Michael wrote:
>> Hi Tetsuya,
>>
>> I see two lines log like:
>>
>> EAL: cannot open VFIO container, error 2 (No such file or directory)
>> EAL: VFIO support could not be initialized
>>
>> It should not happen as I have a patch
Initialization procedure fix to allow per device MAC configuration.
Signed-off-by: Tomasz Kulasek
Signed-off-by: Pawel Wodkowski
---
lib/librte_pmd_ring/rte_eth_ring.c | 30 +-
1 file changed, 25 insertions(+), 5 deletions(-)
diff --git a/lib/librte_pmd_ring/rte_e
Added MAC addr add/remove dummy callbacks.
Signed-off-by: Tomasz Kulasek
Signed-off-by: Pawel Wodkowski
---
lib/librte_pmd_ring/rte_eth_ring.c | 16
1 file changed, 16 insertions(+)
diff --git a/lib/librte_pmd_ring/rte_eth_ring.c
b/lib/librte_pmd_ring/rte_eth_ring.c
index 9
Link up and down implementation.
Signed-off-by: Tomasz Kulasek
Signed-off-by: Pawel Wodkowski
---
lib/librte_pmd_ring/rte_eth_ring.c | 16
1 file changed, 16 insertions(+)
diff --git a/lib/librte_pmd_ring/rte_eth_ring.c
b/lib/librte_pmd_ring/rte_eth_ring.c
index 4f1b6ed..92
Patch split into smaller parts to separate features from previous version.
Tomasz Kulasek (3):
PMD Ring - Add link up/down functions
PMD Ring - Add MAC addr add/remove functions
PMD Ring - Fix for per device management
lib/librte_pmd_ring/rte_eth_ring.c | 62 +
On 12/23/14 19:08, Thomas Monjalon wrote:
> Hello all,
>
> We now have 2 new maintainers managing their own sub-tree for some
> specific areas of DPDK:
> - Stephen Hemminger will maintain the driver bnx2x
Cool!
> - Helin Zhang will maintain the driver i40e
>
> These repositories can
Hi All,
What are the various means available (as of now) to share a NIC port
between Linux and DPDK applications? Idea is to separate control packets
and data packets; using Linux userspace application to process control
packets while DPDK application to access/process data packets (same NIC
port
On 01/18/15 23:58, Thomas Monjalon wrote:
> 2015-01-08 11:56, Vlad Zolotarov:
>> I've acked PATCH1 and PATCH2 already before and since there are no
>> changes in them, pls., consider them ACKed... ;)
> Thank you for the great review.
>
> Note to everybody reading that:
> Vlad approved the 2 first
On Mon, Jan 19, 2015 at 11:23:07AM +0800, Helin Zhang wrote:
> As there are only 6 bit flags in ol_flags for indicating packet types,
> which is not enough to describe all the possible packet types hardware
> can recognize. For example, i40e hardware can recognize more than 150
> packet types. Unif
Hi everybody,
we quite recently updated one of our applications to DPDK 1.8.0 and are
now seeing a segmentation fault in ixgbe_rxtx_vec.c:444 after a few minutes.
I just did some quick debugging and I only have a very limited
understanding of the code in question but it seems that the 'continue'
i
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
lib/librte_mbuf/rte_mbuf.c | 6 --
lib/librte_mbuf/rte_mbuf.h | 10 ++
2 files changed, 2 insertion
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
examples/l3fwd/main.c | 64 +--
1 file changed, 37 insertions(+)
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
examples/l3fwd-power/main.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/examples/l3
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
examples/l3fwd-acl/main.c | 19 ++-
1 file changed, 6 insertions(+), 13 deletions(-)
diff --git
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
examples/ip_reassembly/main.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/example
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
examples/ip_fragmentation/main.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/exam
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
app/test/packet_burst_generator.c | 10 --
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
app/test-pmd/csumonly.c | 6 +++---
app/test-pmd/rxonly.c | 9 +++--
2 files changed, 6 insertions(+), 9 d
To unify packet types among all PMDs, bit masks and relevant macros
of packet type for ol_flags are replaced by unified packet type and
relevant macros.
Signed-off-by: Helin Zhang
---
app/test-pipeline/pipeline_hash.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/app/te
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_vmxnet3/vmxnet3_rxtx.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/librte_pmd_vmxnet3/vmxnet3_rxtx.c
b/l
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_enic/enic_main.c | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/lib/librte_pmd_enic/enic_main.c b/lib
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_bond/rte_eth_bond_pmd.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/lib/librte_pmd_bond/rte_eth_bond_pmd
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type.
Signed-off-by: Helin Zhang
Signed-off-by: Jijiang Liu
---
lib/librte_pmd_i40e/i40e_rxtx.c | 778 ++--
1 file changed, 504 insertions(+), 274 delet
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type for Vector PMD.
Signed-off-by: Cunming Liang
Signed-off-by: Helin Zhang
---
lib/librte_pmd_ixgbe/ixgbe_rxtx_vec.c | 39 +++
1 file changed, 21 insertio
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 141 +-
1 file changed, 107 insertions(+), 34 deletions(-)
diff --git a/lib/lib
To unify packet types among all PMDs, bit masks of packet type for
ol_flags are replaced by unified packet type.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_e1000/igb_rxtx.c | 95 ++---
1 file changed, 80 insertions(+), 15 deletions(-)
diff --git a/lib/libr
As there are only 6 bit flags in ol_flags for indicating packet types,
which is not enough to describe all the possible packet types hardware
can recognize. For example, i40e hardware can recognize more than 150
packet types. Unified packet type is composed of tunnel type, L3 type,
L4 type and inne
Currently only 6 bits which are stored in ol_flags are used to indicate
the packet types. This is not enough, as some NIC hardware can recognize
quite a lot of packet types, e.g i40e hardware can recognize more than 150
packet types. Hiding those packet types hides hardware offload capabilities
whi
Hello,
Following from a previous thread, I took up Bruce Richardson's advice to
modify the RSS redirection table(RETA) to exclude entries pointing to a
specific queue:
For e.g, this is the RETA table before any modifications are done:
*Default *Reta table:
0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1
On Sun, Jan 18, 2015 at 07:12:31PM +0800, XU Liang wrote:
> I configured the 82599 ports to work in multi-queue mode and flow director to
> assign different TCP connections to different queues. A multi-process
> application receive packets from queues and each?process reads a queue. When
> I kil
On Mon, Jan 19, 2015 at 10:30 AM, Qiu, Michael
wrote:
>
> This could be, but I hope every contributor could do basic check like
> build and run of at lease one app like test-pmd, then it will reduce
> most of failure we faced.
>
testpmd is fine, but make test should be preferred as it requires n
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ouyang Changchun
> Sent: Monday, January 12, 2015 6:59 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v6 4/6] ether: Check VMDq RSS mode
>
> Check mq mode for VMDq RSS, handle it correctly instead of retu
2015-01-19 08:21, David Marchand:
> On Mon, Jan 19, 2015 at 3:45 AM, Zhang, Helin
> wrote:
>
> > It seems that your fix result in cannot launching applications.
> > I don't suspect the correction of your fix, but somewhere else needs to be
> > corrected together with your fix.
> >
> > Logs:
> >
On 2015/1/19 17:50, David Marchand wrote:
On Mon, Jan 19, 2015 at 10:30 AM, Qiu, Michael mailto:michael.qiu at intel.com>> wrote:
This could be, but I hope every contributor could do basic check like
build and run of at lease one app like test-pmd, then it will reduce
most of failure we faced.
te
2015-01-19 17:56, Tetsuya Mukawa:
> On 2015/01/19 17:53, Doherty, Declan wrote:
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tetsuya Mukawa
> >> Hi,
> >>
> >> I've got following error using latest DPDK code.
> >> Does someone face the same issue?
> >> In my environment, It seems almo
On 2015/1/19 17:50, David Marchand wrote:
On Mon, Jan 19, 2015 at 10:30 AM, Qiu, Michael mailto:michael.qiu at intel.com>> wrote:
This could be, but I hope every contributor could do basic check like
build and run of at lease one app like test-pmd, then it will reduce
most of failure we faced.
te
2015-01-18 19:26, Doherty, Declan:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-01-16 15:10, Declan Doherty:
> > > In commit 2fc8d6d the behaviour of function rte_is_power_of_2 was
> > > changed to not return true for 0. memzone_reserve_aligned_thread_unsafe
> > > and rte_
Main code changes:
1. Differentiate architectural features based on CPU flags
a. Implement separated move functions for SSE/AVX/AVX2 to make full
utilization of cache bandwidth
b. Implement separated copy flow specifically optimized for target
architecture
2. Rewrite the memcpy functi
Main code changes:
1. Added more typical data points for a thorough performance test
2. Added unaligned test cases since it's common in DPDK usage
Signed-off-by: Zhihong Wang
---
app/test/test_memcpy_perf.c | 238 +---
1 file changed, 156 insertions(+),
Removed unnecessary test cases for base move functions since the function
"func_test" covers them all.
Signed-off-by: Zhihong Wang
---
app/test/test_memcpy.c | 52 +-
1 file changed, 1 insertion(+), 51 deletions(-)
diff --git a/app/test/test_memc
VTA is for debugging only, it increases compile time and binary size,
especially when there're a lot of inlines.
So disable it since memcpy test contains a lot of inline calls.
Signed-off-by: Zhihong Wang
---
app/test/Makefile | 6 ++
1 file changed, 6 insertions(+)
diff --git a/app/test/M
This patch set optimizes memcpy for DPDK for both SSE and AVX platforms.
It also extends memcpy test coverage with unaligned cases and more test points.
Optimization techniques are summarized below:
1. Utilize full cache bandwidth
2. Enforce aligned stores
3. Apply load address alignment based
Hi Tetsuya,
I see two lines log like:
EAL: cannot open VFIO container, error 2 (No such file or directory)
EAL: VFIO support could not be initialized
It should not happen as I have a patch merged to master branch.
There should have something wrong, could you please help me to debug this?
All
On 2015/1/19 17:14, Thomas Monjalon wrote:
> 2015-01-19 08:21, David Marchand:
>> On Mon, Jan 19, 2015 at 3:45 AM, Zhang, Helin
>> wrote:
>>
>>> It seems that your fix result in cannot launching applications.
>>> I don't suspect the correction of your fix, but somewhere else needs to be
>>> corre
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of David Marchand
> Sent: Monday, January 19, 2015 7:21 AM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] Fix rte_is_power_of_2
>
> Hello,
>
> On Mon, Jan 19, 2015 at 3:45 AM, Zhan
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ouyang Changchun
> Sent: Monday, January 12, 2015 6:59 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v6 3/6] ixgbe: Get VF queue number
>
> Get the available Rx and Tx queue number when receiving
> IXG
Sorry for the inconvenience. I was not aware of "make test" utility which
does more elaborate testing. I will make a note of it and make sure future
patches will go through it. I had done basic testing with testpmd as the
changes were minimal.
Thanks,
Ravi
On Mon, Jan 19, 2015 at 1:13 AM, Thomas
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tetsuya Mukawa
> Sent: Monday, January 19, 2015 7:11 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] rte_eal_init() causes PANIC using latest DPDK code.
>
> Hi,
>
> I've got following error using latest DPDK co
On Sun, Jan 18, 2015 at 09:48:33PM +, O'driscoll, Tim wrote:
> > -Original Message-
> > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > Sent: Sunday, January 18, 2015 6:25 PM
> > To: O'driscoll, Tim
> > Cc: Thomas Monjalon; dev at dpdk.org
> > Subject: Re: [dpdk-dev] Why nothing
Hello,
On Mon, Jan 19, 2015 at 3:45 AM, Zhang, Helin wrote:
> It seems that your fix result in cannot launching applications.
> I don't suspect the correction of your fix, but somewhere else needs to be
> corrected together with your fix.
>
> Logs:
> /
On Mon, Jan 19, 2015 at 09:53:30AM +0800, zhihong.wang at intel.com wrote:
> This patch set optimizes memcpy for DPDK for both SSE and AVX platforms.
> It also extends memcpy test coverage with unaligned cases and more test
> points.
>
> Optimization techniques are summarized below:
>
> 1. Utili
1 - 100 of 103 matches
Mail list logo
