> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Tuesday, September 20, 2016 5:36 AM
> To: dev at dpdk.org
> Cc: Trahe, Fiona; Jain, Deepak K; De Lara Guarch, Pablo; Griffin, John;
> Kusztal, ArkadiuszX
> Subject: [PATCH v2 0/3] add GMAC capability Intel QuickAssist driver
>
> Th
2016-09-19 14:22, Matthew Hall:
> On Mon, Sep 19, 2016 at 11:18:48PM +0200, Nikita Kozlov wrote:
> > I have submitted a patch that, among other things, increase this size.
> > But it needs some reviews: http://dpdk.org/dev/patchwork/patch/15295/
>
> A whole ton of us submitted patches to fix LPM l
The rte_eth_tap.c PMD creates a device using TUN/TAP interfaces
on the local host. The PMD allows for DPDK and the host to
communicate using a raw device interface on the host and in
the DPDK application. The device created is a Tap device with
a L2 packet header.
v3 - fix includes by removing ifd
Hi Arek,
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Thursday, August 25, 2016 5:03 AM
> To: dev at dpdk.org
> Cc: Trahe, Fiona; Jain, Deepak K; De Lara Guarch, Pablo; Griffin, John;
> Kusztal, ArkadiuszX
> Subject: [PATCH 1/3] crypto/aesni_gcm: move pre-counter block to GCM
>
Allow binding KNI thread to specific core in single threaded mode
by setting core_id and force_bind config parameters.
Signed-off-by: Vladyslav Buslov
---
doc/guides/prog_guide/kernel_nic_interface.rst | 3 +
lib/librte_eal/linuxapp/kni/kni_misc.c | 101 -
2 fil
On Tue, Sep 20, 2016 at 04:58:48PM +0100, Bruce Richardson wrote:
> On Tue, Sep 20, 2016 at 02:51:27PM +0200, Adrien Mazarguil wrote:
> > On Mon, Sep 19, 2016 at 04:26:05PM +0100, Bruce Richardson wrote:
> > > On Mon, Sep 19, 2016 at 04:59:59PM +0200, Adrien Mazarguil wrote:
> > > > Hi Bruce,
> > >
On Tue, Sep 20, 2016 at 02:16:48PM +0300, Dror Birkman wrote:
> Cool!
> Adding fixline:
You might want to send a v2: the following diff is malformed.
>
> If rte_pktmbuf_alloc() fails on any segment that is not the initial
> segment, previously allocated mbufs are not freed.
>
> Fixes: 6db141c91
On Tue, Sep 20, 2016 at 05:36:45PM +0800, Zhiyong Yang wrote:
> +enum vhost_xstats_pkts {
> + VHOST_UNDERSIZE_PKT = 0,
> + VHOST_64_PKT,
> + VHOST_65_TO_127_PKT,
> + VHOST_128_TO_255_PKT,
> + VHOST_256_TO_511_PKT,
> + VHOST_512_TO_1023_PKT,
> + VHOST_1024_TO_1522_PKT,
>
On Tue, Sep 20, 2016 at 01:39:23PM +0300, Dror Birkman wrote:
> If rte_pktmbuf_alloc() fails on any segment that is not the initial
> segment, previously allocated mbufs are not freed.
You should put a "fixline" here.
Besides that, I think it's a good candidate for stable branch. Thus,
stable at
On Tue, Sep 20, 2016 at 05:36:45PM +0800, Zhiyong Yang wrote:
> diff --git a/drivers/net/vhost/rte_eth_vhost.c
> b/drivers/net/vhost/rte_eth_vhost.c
> index 9157bf1..c3f404d 100644
> --- a/drivers/net/vhost/rte_eth_vhost.c
> +++ b/drivers/net/vhost/rte_eth_vhost.c
> @@ -78,6 +78,7 @@ struct vhost_
Thanks so much for your reply! Usually how did you test lpm performance with
variety of destination addresses? use which tool send the traffic? how many
flows rules will you add? what's the performance you get?
At 2016-09-20 17:41:13, "Andriy Berestovskyy" wrote:
>Hey,
>You are correct.
On Tue, Sep 20, 2016 at 05:36:44PM +0800, Zhiyong Yang wrote:
> The patch moves all stats counters to a new defined struct vhost_stats
> as follows, in order to manage all stats counters in a unified way and
> simplify the subsequent function implementation(vhost_dev_xstats_reset).
>
> struct vhos
I have already taken care of this in v5 of the patch , If possible please
review the same .
-Original Message-
From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
Sent: Tuesday, September 20, 2016 3:12 AM
To: Kavanagh, Mark B
Cc: Dey, Souvik ; dev at dpdk.org; stephen at
networkplum
From: Jan Viktorin
Now that rte_device is available, drivers can start using its members
(numa, name) as well as link themselves into another rte_device list.
As of now no one is using this list, but can be used for moving over all
devices (pdev/vdev/Xdev) and perform bulk actions (like cleanup)
From: Jan Viktorin
Signed-off-by: Jan Viktorin
Signed-off-by: Shreyansh Jain
---
lib/librte_eal/common/eal_common_dev.c | 13 +
lib/librte_eal/common/include/rte_dev.h | 31 +++
2 files changed, 44 insertions(+)
diff --git a/lib/librte_eal/common/eal_c
From: Jan Viktorin
To register both vdev and pci drivers into the list of all rte_driver,
we have to call rte_eal_driver_register explicitly.
Signed-off-by: Jan Viktorin
Signed-off-by: Shreyansh Jain
---
lib/librte_eal/common/eal_common_pci.c | 2 ++
lib/librte_eal/common/eal_common_vdev.c |
From: Jan Viktorin
Remove the 'name' member from rte_pci_driver and move to generic
rte_driver.
Most of the PMD drivers were initially using DRIVER_REGISTER_PCI(..)
as well as assigning a name to eth_driver.pci_drv.name member.
In this patch, only the original DRIVER_REGISTER_PCI(..) name has
be
From: Jan Viktorin
There is no need to have a custom memory resource representation for
each infrastructure (PCI, ...) as it would always have the same members.
Signed-off-by: Jan Viktorin
Signed-off-by: Shreyansh Jain
---
drivers/net/szedata2/rte_eth_szedata2.c | 4 ++--
lib/librte_eal/comm
From: Jan Viktorin
Further refactoring and generalization of PCI infrastructure will
require access to the rte_dev.h contents.
Signed-off-by: Jan Viktorin
Signed-off-by: Shreyansh Jain
---
lib/librte_eal/common/include/rte_pci.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_e
From: Jan Viktorin
- All devices register themselfs by calling a kind of DRIVER_REGISTER_XXX.
The PMD_REGISTER_DRIVER is not used anymore.
- PMD_VDEV type is also not being used - can be removed from all VDEVs.
Signed-off-by: Jan Viktorin
Signed-off-by: Shreyansh Jain
---
drivers/crypto/aes
From: Jan Viktorin
All PMD_VDEV drivers can now use rte_vdev_driver instead of the
rte_driver (which is embedded in the rte_vdev_driver).
Signed-off-by: Jan Viktorin
Signed-off-by: Shreyansh Jain
---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 10 ++
drivers/crypto/aesni_mb/rte_aesni
From: Jan Viktorin
- Remove checks for VDEV from rte_eal_vdev_(init/uninint) as all devices
are inherently virtual here.
- PDEVs perform PCI specific inits - rte_eal_dev_init() need not call
rte_driver->init();
Signed-off-by: Jan Viktorin
[Shreyansh: Reword commit log]
Signed-off-by: Shreya
From: Jan Viktorin
Move all PMD_VDEV-specific code into a separate module and header
file to not polute the generic code anymore. There is now a list
of virtual devices available.
The rte_vdev_driver integrates the original rte_driver inside
(C inheritance). The rte_driver will be however change
From: David Marchand
Now that hotplug has been moved to eal, there is no reason to keep the
device type in this layer.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
---
app/test/virtual_pmd.c| 2 +-
drivers/net/af_packet/rte_eth_af_packet.c | 2 +-
drivers/
From: David Marchand
Remove bus logic from ethdev hotplug by using eal for this.
Current api is preserved:
- the last port that has been created is tracked to return it to the
application when attaching,
- the internal device name is reused when detaching.
We can not get rid of ethdev hotplug
From: David Marchand
Hotplug invocations, which deals with devices, should come from the layer
that already handles them, i.e. EAL.
For both attach and detach operations, 'name' is used to select the bus
that will handle the request.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
From: David Marchand
No need to scan all devices, we only need to update the device being
attached.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
---
lib/librte_eal/common/eal_common_pci.c | 12 +---
lib/librte_ether/rte_ethdev.c | 3 ---
2 files changed, 9 ins
From: David Marchand
- Move rte_eth_dev_create_unique_device_name() from ether/rte_ethdev.c to
common/include/rte_pci.h as rte_eal_pci_device_name(). Being a common
method, can be used across crypto/net PCI PMDs.
- Remove crypto specific routine and fallback to common name function.
- Introdu
From: David Marchand
Now that all pdev are pci drivers, we don't need to register crypto and
ethdev drivers through a dedicated channel.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
---
lib/librte_cryptodev/rte_cryptodev.c | 23
lib/librte_cryptod
From: David Marchand
Simplify crypto and ethdev pci drivers init by using newly introduced
init macros and helpers.
Those drivers then don't need to register as "rte_driver"s anymore.
Exceptions:
- virtio and mlx* use RTE_INIT directly as they have custom initialization
steps.
- VDEV devices a
From: David Marchand
crypto and ethdev drivers aligned to PCI probe/remove. These wrappers are
mapped directly to PCI resources.
Existing handlers for init/uninit can be easily reused for this.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
---
lib/librte_cryptodev/rte_cryptodev.
From: David Marchand
Introduce a RTE_INIT macro used to mark an init function as a constructor.
Current eal macros have been converted to use this (no functional impact).
DRIVER_REGISTER_PCI is added as a helper for pci drivers.
Suggested-by: Jan Viktorin
Signed-off-by: David Marchand
[Shreyan
From: David Marchand
Pure coding style, but it might make it easier later if we want to move
fields in rte_cryptodev_driver and eth_driver structures.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
---
drivers/crypto/qat/rte_qat_cryptodev.c | 2 +-
drivers/net/ena/ena_ethdev.c
From: David Marchand
This information is not used and just adds noise.
Signed-off-by: David Marchand
Signed-off-by: Shreyansh Jain
Reviewed-by: Jan Viktorin
---
lib/librte_cryptodev/rte_cryptodev.c | 8 +++-
lib/librte_cryptodev/rte_cryptodev.h | 2 --
lib/librte_cryptodev/rte_cr
Probe and Remove are more appropriate names for PCI init and uninint
operations. This is a cosmetic change.
Only MLX* uses the PCI direct registeration, bypassing PMD_* macro. The
calls backs for this too have been updated.
VDEV are left out. For them, init/uninit are more appropriate.
Suggested
From: David Marchand
These lists can be initialized once and for all at build time.
With this, those lists are only manipulated in a common place
(and we could even make them private).
A nice side effect is that pci drivers can now register in constructors.
Signed-off-by: David Marchand
Review
From: David Marchand
rte_eal_dev_init is declared in both eal_private.h and rte_dev.h since its
introduction.
This function has been exported in ABI, so remove it from eal_private.h
Fixes: e57f20e05177 ("eal: make vdev init path generic for both virtual and pci
devices")
Signed-off-by: David M
Based on master (e15922d75)
Background:
===
It includes two different patch-sets floated on ML earlier:
* Original patch series is from David Marchand [1], [2].
`- This focused mainly on PCI (PDEV) part
`- v7 of this was posted by me [8] in August/2016
* Patch series [4] from Jan Vi
This feature adds vhost pmd extended statistics from per queue perspective
in order to meet the requirements of the applications such as OVS etc.
The statistics counters are based on RFC 2819 and RFC 2863 as follows:
rx/tx_good_packets
rx/tx_total_bytes
rx/tx_missed_pkts
rx/tx_broadcast_packets
r
The patch moves all stats counters to a new defined struct vhost_stats
as follows, in order to manage all stats counters in a unified way and
simplify the subsequent function implementation(vhost_dev_xstats_reset).
struct vhost_stats {
uint64_t rx_pkts;
uint64_t tx_pkts;
ui
This patch set adds the vhost pmd xstats support.
Patch 1 moves all stats counters to a new defined struct vhost_stats,
in order to manage all stats counters in a unified way.
Patch 2 adds the pmd xstats support.
Changes in v3:
1. rework the vhost_update_packet_xstats and separate it into two pa
Fixes: ce94a51ff05c ("mempool: add flag for removing phys contiguous
constraint")
Signed-off-by: Ferruh Yigit
---
lib/librte_mempool/rte_mempool.c | 4 +++-
lib/librte_mempool/rte_mempool.h | 3 +++
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/lib/librte_mempool/rte_mempool.c b
Fixes: 85226f9c526b ("mempool: introduce a function to create an empty pool")
Fixes: d1d914ebbc25 ("mempool: allocate in several memory chunks by default")
Signed-off-by: Ferruh Yigit
---
lib/librte_mempool/rte_mempool.c | 4 +---
lib/librte_mempool/rte_mempool.h | 4 +---
2 files changed, 2 ins
On Tue, Sep 20, 2016 at 02:51:27PM +0200, Adrien Mazarguil wrote:
> On Mon, Sep 19, 2016 at 04:26:05PM +0100, Bruce Richardson wrote:
> > On Mon, Sep 19, 2016 at 04:59:59PM +0200, Adrien Mazarguil wrote:
> > > Hi Bruce,
> > >
> > > On Mon, Sep 19, 2016 at 03:36:54PM +0100, Bruce Richardson wrote:
Hi,
> +
> + if (dev_data_id == RTE_MAX_ETHPORTS) {
> + RTE_PMD_DEBUG_TRACE("Reached maximum number of
> Ethernet ports by all "
> + "the processes\n");
> + return NULL;
> + }
> +
>
Can the log message be changed to ("Cannot allocate mor
AFAIR Intel hardware should do the 10Gbit/s line rate (i.e. ~14,8
MPPS) with one flow and LPM quite easily. Sorry, I don't have numbers
to share at hand.
Regarding the tool please see the pktgen-dpdk or TRex. Regarding the
number of flows and overall benchmarking methodology - please see
RFC2544.
Added lookup for pool name because secondary process should attach to
mempool created by primary process rather than create new one.
Added function free_shared_dev_data() used at the exit of the testpmd.
This causes detach devices data from array rte_eth_dev_data[] shared
between all processes. Th
Added prevention not overwrite device data in array rte_eth_dev_data[].
Secondary process appends in the first free place rather than at the
beginning. This behavior prevents overwriting devices of primary process
by secondary process.
Signed-off-by: Marcin Kerlin
---
lib/librte_ether/rte_ethdev
This patch ensure not overwrite device data in the multiprocess application.
1)Changes in the library introduces continuity in array rte_eth_dev_data[]
shared between all processes. Secondary process adds new entries in free
space instead of overwriting existing entries.
2)Changes in application
Hi Marcin,
> /**
> * @internal
> + * Returns a shared device data slot specified by the unique identifier name.
> + *
> + * @paramname
> + * The pointer to the Unique identifier name for each shared Ethernet
> +device
> + * between multiple processes.
> + * @return
> + * - The pointer
This patch ensure not overwrite device data in the multiprocess application.
1)Changes in the library introduces continuity in array rte_eth_dev_data[]
shared between all processes. Secondary process adds new entries in free
space instead of overwriting existing entries.
2)Changes in application
Now, run the example/ethtool, the drvinfo can not show the fireware
information. From customer point of view, it should be better if we
can have the same as kernel version ethtool show the bus-info and
firmware-version. We need to add a variable in struct rte_eth_dev_info
to get the fw version.
I?m
Hi Wenzhuo,
On 9/8/2016 2:07 AM, Lu, Wenzhuo wrote:
> Hi Wei,
>
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of wang wei
>> Sent: Tuesday, September 6, 2016 8:05 PM
>> To: thomas.monjalon at 6wind.com
>> Cc: dev at dpdk.org
>> Subject: [dpdk-dev] [RFC]
On Wed, Sep 14, 2016 at 12:15:37PM +, Kavanagh, Mark B wrote:
> >
> >>+{
> >>+?? struct rte_eth_dev_info dev_info;
> >>+?? uint32_t ether_hdr_len = ETHER_HDR_LEN + ETHER_CRC_LEN +
> >>+VLAN_TAG_LEN;
> >>+?? uint32_t frame_size = mtu + ether_hdr_len;
> >>+
> >>+?? virtio_dev_info
If rte_pktmbuf_alloc() fails on any segment that is not the initial
segment, previously allocated mbufs are not freed.
Fixes: 6db141c91e1f ("pcap: support jumbo frames")
Cc:
Signed-off-by: Dror Birkman
---
drivers/net/pcap/rte_eth_pcap.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
On Mon, Sep 19, 2016 at 04:26:05PM +0100, Bruce Richardson wrote:
> On Mon, Sep 19, 2016 at 04:59:59PM +0200, Adrien Mazarguil wrote:
> > Hi Bruce,
> >
> > On Mon, Sep 19, 2016 at 03:36:54PM +0100, Bruce Richardson wrote:
> > > With recent gcc versions, e.g. gcc 6.1, compilation of mlx drivers wit
On 8/26/2016 5:54 PM, Kamil Rytarowski wrote:
> From: Kamil Rytarowski
>
> Signed-off-by: Maciej Czekaj
> Signed-off-by: Kamil Rytarowski
> Signed-off-by: Zyta Szpak
> Signed-off-by: Slawomir Rosek
> Signed-off-by: Radoslaw Biernacki
> Signed-off-by: Jerin Jacob
> ---
Is it possible to giv
On 8/26/2016 5:54 PM, Kamil Rytarowski wrote:
> From: Kamil Rytarowski
>
> In case of the multiprocess mode a shared nicvf struct between processes
> cannot point with the eth_dev pointer to master device, therefore remove it
> allong with references to it refactoring the code where needed.
Patc
On 8/26/2016 5:53 PM, Kamil Rytarowski wrote:
> From: Kamil Rytarowski
>
> Changes:
> - add new message sqs_alloc in mailbox
> - add a queue container to hold secondary qsets.
> - add nicvf_mbox_request_sqs
> - handle new mailbox messages for secondary queue set support
> - register secondar
On 8/26/2016 5:53 PM, Kamil Rytarowski wrote:
> From: Kamil Rytarowski
>
> Refactored features:
> - enable nicvf_qset_rbdr_precharge to handle handle secondary queue sets
double "handle"
> - rte_free already handles NULL pointer
> - check mempool flags to predict being contiguous in memory
>
On 9/19/2016 8:13 PM, Yuanhan Liu wrote:
> Firstly, sorry for being late on this discussion: I just got a chance
> to follow what you guys were talking about.
>
> On Tue, Sep 13, 2016 at 02:51:31PM +0800, Tan, Jianfeng wrote:
>>> (2) we'd better not differentiate phys device and virtual
>
> Agreed.
Cool!
Adding fixline:
If rte_pktmbuf_alloc() fails on any segment that is not the initial
segment, previously allocated mbufs are not freed.
Fixes: 6db141c91e1f ("pcap: support jumbo frames")
Signed-off-by: Dror Birkman
---
drivers/net/pcap/rte_eth_pcap.c | 4 +++-
1 file changed, 3 insertions
Hi Dror,
On 9/20/2016 1:08 PM, Dror Birkman wrote:
> If rte_pktmbuf_alloc() fails on any segment that is not the initial
> segment, previously allocated mbufs are not freed.
>
> Fixes: 6db141c91e1f ("pcap: support jumbo frames")
>
> Cc:
> Signed-off-by: Dror Birkman
Acked-by: Ferruh Yigit
If rte_pktmbuf_alloc() fails on any segment that is not the initial
segment, previously allocated mbufs are not freed.
Signed-off-by: Dror Birkman
---
drivers/net/pcap/rte_eth_pcap.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/pcap/rte_eth_pcap.c b/drivers/
File rte_crypto_sym.h GMAC API comments need to be changed
to comply to the GMAC specification. Main areas of changes is
aad pointer and aad len which now will be used to
provide plaintext.
Signed-off-by: Arek Kusztal
Acked-by: Deepak Kumar Jain
---
lib/librte_cryptodev/rte_crypto_sym.h | 27 ++
Added Galois Message Authentication Code (GMAC) tests to cryptodev tests
Signed-off-by: Arek Kusztal
Acked-by: Deepak Kumar Jain
---
app/test/test_cryptodev.c | 265 +
app/test/test_cryptodev_gcm_test_vectors.h | 150
2 files changed
Added Galois Message Authentication Code (GMAC) capability to
QuickAssist Technology symmetric cryptographic driver.
GMAC is authentication only variant of Galois Counter Mode (GCM)
where all plaintext is provided with AAD pointer only.
Signed-off-by: Arek Kusztal
Acked-by: Deepak Kumar Jain
---
This patches add AES GMAC capability to Intel(R) QuickAssist
Technology driver and corresponsing test cases.
Changes in v2:
* Updated API comments librte_cryptodev
* Removed unnecessary rte_crypto_sym_op reference in the driver for GMAC
* Changed GMAC aad capability values
* Rebased against next-c
On Tue, 20 Sep 2016 06:46:31 +
Shreyansh Jain wrote:
> Hi Jan,
>
> > -Original Message-
> > From: Jan Viktorin [mailto:viktorin at rehivetech.com]
> > Sent: Monday, September 19, 2016 5:04 PM
> > To: Shreyansh Jain
> > Cc: dev at dpdk.org; Hemant Agrawal
> > Subject: Re: [PATCH v3
Hi,
I need to go over all the nics in my machine that are connected to the net
throw cable.
link_status of rte_eth_link returns link up on all nics, also those who are
not connected.
Is there any way to check it?
Thanks, Keren
This patch adds two sample configuration files to ipsec-secgw sample
application. The sample configuration files shows how to set-up systems
back-to-back that would forward traffic through an IPsec tunnel.
Signed-off-by: Fan Zhang
---
examples/ipsec-secgw/ep0.cfg | 160 ++
This patch adds the configuration file support to ipsec_secgw
sample application. Instead of hard-coded rules, the users can
specify their own SP, SA, and routing rules in the configuration
file. An command line option "-f" is added to pass the
configuration file location to the application.
Confi
This patchset adds the configuration file supported to ipsec_secgw
sample application. Two sample configuration files, ep0.cfg and ep1.cfg
are also added to show how to configure two systems back-to-back that
would forward traffic through an IPsec tunnel
v7 change:
- updated release note
v6 chan
[ just got few more comments after some fiddling with the build issue ]
On Fri, Sep 16, 2016 at 11:22:13AM -0500, Keith Wiles wrote:
> diff --git a/drivers/net/tap/Makefile b/drivers/net/tap/Makefile
> new file mode 100644
> index 000..442a2fe
> --- /dev/null
> +++ b/drivers/net/tap/Makefile
>
On Thu, 8 Sep 2016 19:32:09 +0100
Ferruh Yigit wrote:
> Linux kernel v4.8 removes macro DEFINE_PCI_DEVICE_TABLE
>
> Linux: 7e9321599011 ("treewide: remove references to the now unnecessary
> DEFINE_PCI_DEVICE_TABLE")
>
> Replaced macro with its value in kni ethtool drivers.
>
> Signed-off-by:
On Mon, Sep 19, 2016 at 03:56:03PM +, Wiles, Keith wrote:
> All of the below errors are from Linux header files and not the Tap
> driver.
>
>
> Yes, but you are referencing them, so ...
>
>
> Yes I am referencing them, but still they are not interacting with the tap
>
Hey,
You are correct. The LPM might need just one (TBL24) or two memory
reads (TBL24 + TBL8). The performance also drops once you have a
variety of destination addresses instead of just one (cache misses).
In your case for the dst IP 192.168.1.2 you will have two memory reads
(TBL24 + TBL8), becau
On Tue, 20 Sep 2016 21:16:37 +0300
Vladyslav Buslov wrote:
> @@ -123,6 +125,9 @@ static int __net_init kni_init_net(struct net *net)
> /* Clear the bit of device in use */
> clear_bit(KNI_DEV_IN_USE_BIT_NUM, &knet->device_in_use);
>
> + mutex_init(&knet->kni_kthread_lock);
> +
Tx function was handling a double loop to send segmented packets, it can be
done in a single one.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Vasily Philipov
---
drivers/net/mlx5/mlx5_rxtx.c | 312 ++-
1 file changed, 158 insertions(+), 154 deletions(-
PMD uses only power of two number of Work Queue Elements (aka WQE), storing
the number of elements in log2 helps to reduce the size of the container to
store it.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.c | 23 ---
drivers/net/mlx5/mlx5_rxtx.h | 2 +-
d
Blue Flame (aka BF) is a buffer allocated with a power of two value, its
size is returned by Verbs in log2.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.c | 2 +-
drivers/net/mlx5/mlx5_rxtx.h | 2 +-
drivers/net/mlx5/mlx5_txq.c | 2 +-
3 files changed, 3 insertions(+), 3 delet
PMD uses only power of two number of Completion Queue Elements (aka CQE),
storing the number of elements in log2 helps to reduce the size of the
container to store it.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxq.c | 2 +-
drivers/net/mlx5/mlx5_rxtx.c | 8
drivers/net/
PMD uses only power of two number of descriptors, storing the number of
elements in log2 helps to reduce the size of the container to store it.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_ethdev.c | 4 ++--
drivers/net/mlx5/mlx5_rxq.c| 10 +
Rework Work Queue Element (aka WQE) structures to fit PMD needs.
A WQE is an aggregation of 16 bytes elements known as "data segments"
(aka dseg).
The only common part is the first two elements i.e. the control one to
define the job type, and the Ethernet segment which embed offload requests
with
- Rework structure elements to reduce their size.
- Removes a second useless loop in Tx burst function.
This series should be applied on top of "net/mlx5: various fixes".
Changes in v2:
- rework serie to apply it on top of "net/mlx5: various fixes".
Changes in v3:
- Rework some commits log
Hi Cristian,
My comments inline prefixed with [nikhil].
On 19 September 2016 at 21:21, Dumitrescu, Cristian <
cristian.dumitrescu at intel.com> wrote:
>
>
>
> > -Original Message-
> > From: Nikhil Jagtap [mailto:nikhil.jagtap at gmail.com]
> > Sent: Wednesday, September 7, 2016 7:15 AM
>
Add git log capitalization check for LRO and NIC
Signed-off-by: Ferruh Yigit
---
scripts/check-git-log.sh | 2 ++
1 file changed, 2 insertions(+)
diff --git a/scripts/check-git-log.sh b/scripts/check-git-log.sh
index 1e05cf2..6db84c5 100755
--- a/scripts/check-git-log.sh
+++ b/scripts/check-git
On Mon, Sep 19, 2016 at 05:17:34PM +0100, Bruce Richardson wrote:
> On Wed, Sep 14, 2016 at 02:18:01PM +0200, Nelio Laranjeiro wrote:
> > - Rework structure elements to reduce their size.
> > - Removes a second useless loop in Tx burst function.
> >
> > This series should be applied on top of "n
On Mon, Sep 19, 2016 at 05:14:26PM +0100, Bruce Richardson wrote:
> On Wed, Sep 14, 2016 at 02:18:02PM +0200, Nelio Laranjeiro wrote:
> > Rework Work Queue Element (aka WQE) structures to fit PMD needs.
> > A WQE is an aggregation of 16 bytes elements known as "data segments"
> > (aka dseg).
> >
>
Hi Jerin,
> > > >
> >
> > [...]
> >
> > > > +
> > > > +#ifdef RTE_ETHDEV_TX_PREP
> > >
> > > Sorry for being a bit late on that discussion, but what the point of
> > > having that config macro (RTE_ETHDEV_TX_PREP ) at all?
> > > As I can see right now, if driver doesn't setup tx_pkt_prep, then
>
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Monday, September 19, 2016 4:24 PM
> To: Kulasek, TomaszX ; dev at dpdk.org
> Cc: jerin.jacob at caviumnetworks.com
> Subject: RE: [dpdk-dev] [PATCH v2 5/6] ixgbe: add Tx preparation
>
> >
> > [...]
> >
> > > >
> > > >
Hi Jan,
> -Original Message-
> From: Jan Viktorin [mailto:viktorin at rehivetech.com]
> Sent: Monday, September 19, 2016 5:04 PM
> To: Shreyansh Jain
> Cc: dev at dpdk.org; Hemant Agrawal
> Subject: Re: [PATCH v3 06/15] eal/soc: implement probing of drivers
>
> On Mon, 19 Sep 2016 12:17
Hi all,
Does anyone test IPv4 performance? If so, what's the throughput? I can get
almost 10Gb with 64 byte packets. But before the test, I would expect it will
be less than 10G. I thought the performance will not be affected by the
number of rule entires. But the throughput will be relate
Hi all,
I am using a memory safe tool safecode http://safecode.cs.illinois.edu/
compiling dpdk application. If I do not do the memory safety checking, it works
correctly. But my main aim is to evaluate if using safecode protects the
memory, what's the overhead?
The related compiling o
From: Eoin Breen
Adding the support to bind/unbind crypto devices with
dpdk-devbind.py script, as now it is not restricted
to network devices anymore.
Signed-off-by: Eoin Breen
Signed-off-by: Pablo de Lara
---
Changes since v3:
* Modified setup.sh script to show crypto details
* Added script u
> -Original Message-
> From: Jain, Deepak K
> Sent: Tuesday, September 13, 2016 3:42 AM
> To: dev at dpdk.org
> Cc: De Lara Guarch, Pablo; Jain, Deepak K
> Subject: [PATCH v4] crypto/qat: add Intel QuickAssist C62x device
>
> From: Deepak Kumar JAIN
>
> Signed-off-by: Deepak Kumar Jain
> -Original Message-
> From: Jain, Deepak K
> Sent: Monday, September 19, 2016 9:37 AM
> To: dev at dpdk.org
> Cc: De Lara Guarch, Pablo; Jain, Deepak K
> Subject: [PATCH v3] crypto/qat: add Intel(R) QuickAssist C3xxx device
>
> From: Deepak Kumar JAIN
>
> Signed-off-by: Deepak Kumar J
97 matches
Mail list logo
