This discussion is about improving performance of specific use cases
by moving the mbuf fields when needed.
We could consider how to configure it and how complicated it would be to
write applications or drivers (especially vector ones) for such a moving
structure.
But it is simpler to say that havi
2015-11-02 10:19, Harry van Haaren:
> + {"rx_managment_packets", offsetof(struct ixgbe_hw_stats, mngprc)},
> + {"rx_managment_dropped", offsetof(struct ixgbe_hw_stats, mngpdc)},
> + {"tx_managment_packets", offsetof(struct ixgbe_hw_stats, mngptc)},
Looks to be a typo. Management?
> > Make vhost-user virtio 1.0 compatible by adding it to the
> > supported features and keeping the header length
> > the same as for mergeable RX buffers.
> >
> > Signed-off-by: Marcel Apfelbaum
>
> Looks good to me
>
> Acked-by: Michael S. Tsirkin
Applied, thanks
2015-11-02 23:09, Thomas Monjalon:
> > Acked-by: Jianfeng Tan
>
> Applied with the modifications discussed in this thread, thanks.
Please Huawei,
Could you share some numbers for these optimizations?
> Acked-by: Jianfeng Tan
Applied with the modifications discussed in this thread, thanks.
Test commands are added to support selecting differnt length of GRE key.
Signed-off-by: Helin Zhang
Signed-off-by: Andrey Chilikin
---
app/test-pmd/cmdline.c | 52 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 11 ++
2 files changed, 63 ins
By default, only first 3 bytes of GRE key will be used for hash or FD
calculation. With these changes, it can select 3 or 4 bytes of GRE key
for hash or FD calculation.
Signed-off-by: Helin Zhang
Signed-off-by: Andrey Chilikin
---
doc/guides/rel_notes/release_2_2.rst | 2 +
drivers/net/i40e/i4
By default, only 3 bytes of GRE key will be used for hash or FD calculation.
Here adds selecting 3 or 4 bytes of GRE key for those purpose.
Note that these patches are based on another patch set of
"[PATCH v8 0/2] i40e: RSS/FD granularity configuration".
v2 changes:
Updated with RSS/FD granularity
Test commands are added to support selecting input set, or adding
new input set per different pctype.
Signed-off-by: Helin Zhang
Signed-off-by: Andrey Chilikin
---
app/test-pmd/cmdline.c | 218
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 33 +
The default input set of fields of a received packet are loaded from
firmware, which cannot be modified even users want to use different
fields for RSS or flow director. Here adds more flexibilities of
selecting packet fields for hash calculation or flow director for
users.
Signed-off-by: Helin Zh
The default input set of fields of a received packet are loaded from firmware,
which cannot be modified even users want to use different fields for RSS or flow
director. Here adds more flexibilities of selecting or adding packet fields for
hash calculation or flow director to users. It also include
Ok, You are saying re-order the fields based on the configurations params. I
took word "NO" in the param to eliminate. Sure, this does not require and
change in the code that uses it. Will it not now boil down to same as having
completely different layout definition and be more messier ?
For ex
On Mon, Nov 02, 2015 at 03:49:17PM +, Hunt, David wrote:
> On 02/11/2015 15:36, Jan Viktorin wrote:
> >On Mon, 2 Nov 2015 15:26:19 +
> --snip--
> >It was looking like we can share a lot of common code for both
> >architectures. I didn't know how much different are the cpuflags.
>
> CPU fla
On Mon, Nov 02, 2015 at 04:39:37PM +0100, Jan Viktorin wrote:
> On Mon, 2 Nov 2015 19:48:40 +0530
> Jerin Jacob wrote:
>
> > Signed-off-by: Jerin Jacob
> > ---
> > app/test-acl/main.c | 4 +
> > lib/librte_acl/Makefile | 5 +
> > lib/librte_acl/acl.h | 4 +
> > li
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bagh Fares
> Sent: Monday, November 2, 2015 6:03 PM
> To: Dave Neary; CHIOSI, MARGARET T; Stephen Hemminger
> Cc: dev at dpdk.org; Pradeep Kathail (pkathail at cisco.com)
> Subject: Re: [dpdk-dev] Proposals from
If NO_TX_OFFLOAD only changes the layout in terms of relative field
location in cache lines, and does not eliminate the fields themselves
why should the using code be affected?
On Mon, Nov 2, 2015 at 8:30 PM, shesha Sreenivasamurthy (shesha) <
shesha at cisco.com> wrote:
> One issue I see with op
Signed-off-by: Jerin Jacob
---
config/defconfig_arm64-armv8a-linuxapp-gcc | 1 -
1 file changed, 1 deletion(-)
diff --git a/config/defconfig_arm64-armv8a-linuxapp-gcc
b/config/defconfig_arm64-armv8a-linuxapp-gcc
index 79a9533..3e3bf12 100644
--- a/config/defconfig_arm64-armv8a-linuxapp-gcc
+++
Signed-off-by: Jerin Jacob
---
app/test-acl/main.c | 4 +
lib/librte_acl/Makefile | 5 +
lib/librte_acl/acl.h | 4 +
lib/librte_acl/acl_run_neon.c | 46 +++
lib/librte_acl/acl_run_neon.h | 290 ++
lib/librte_acl/rte_acl.c
added the definition of rte_xmm and xmm_t for acl noen implementation
removed the emulated _mm_* functions
Signed-off-by: Jerin Jacob
---
lib/librte_eal/common/include/arch/arm/rte_vect.h | 58 +++
1 file changed, 17 insertions(+), 41 deletions(-)
diff --git a/lib/librte_eal
ARM64 Neon based ACL implementation
This patch is based on David's armv8 v3 patch set
Jerin Jacob (3):
arm: ret_vector.h improvements
arm64: acl: add neon based acl implementation
arm64: defconfig: enabled CONFIG_RTE_LIBRTE_ACL
app/test-acl/main.c | 4 +
c
On Mon, Nov 02, 2015 at 09:52:17AM -0800, Stephen Hemminger wrote:
> On Fri, 30 Oct 2015 19:52:38 +0100
> Adrien Mazarguil wrote:
>
> > +static int
> > +mlx5_link_update_unlocked(struct rte_eth_dev *dev, int wait_to_complete)
> > +{
> > + struct priv *priv = dev->data->dev_private;
> > + stru
On Mon, Nov 02, 2015 at 11:51:23PM +0100, Thomas Monjalon wrote:
> But it is simpler to say that having an API depending of some options
> is a "no-design" which could seriously slow down the DPDK adoption.
What about something similar to how Java JNI works? It needed to support
multiple Java JRE
On Mon, Nov 02, 2015 at 05:26:21PM +0100, Thomas Monjalon wrote:
> 2015-11-02 15:23, Ferruh Yigit:
> > +ifeq ($(COMBINED_BUILD),1)
> > include $(RTE_SDK)/mk/rte.sharelib.mk
> > +endif
>
> Why this ifeq?
> rte.sharelib.mk is always used for combined lib.
>
When combine lib config selected, other
From: Nelio Laranjeiro
ConnectX-4 adapters to not have a constant indirection table size, which is
set at runtime from the number of RX queues. The maximum size is retrieved
using a hardware query and is normally 512.
Since the current RETA API cannot handle a variable size, any query/update
com
Hi,
It is really great to see various companies desiring to be part of the project.
2015-11-02 17:44, Bagh Fares:
> As SOC vendor we will contribute heavily to the project. Example crypto
> acceleration.
> We already contribute a lot to the linux community.
> So not sure why the doubt about of c
On Mon, Nov 02, 2015 at 10:47:53AM +, Hunt, David wrote:
> On 02/11/2015 06:32, Jerin Jacob wrote:
> >On Fri, Oct 30, 2015 at 04:28:25PM +, Hunt, David wrote:
>
> --snip--
>
> >
> >Hi Jan and Dave,
> >
> >I have reviewed your patches for arm[64] support. Please check the
> >review comments.
On Nov 2, 2015 6:23 PM, "Ananyev, Konstantin"
wrote:
> >>Probably user deliberately doesn't want to call dev_start() for PF
device -
> >>as he doesn't plan to use it for RX/TX.
> >>Or might be dev_stop() was called just to do some re-configuration
> >>(allow to TX scattered packets on the PF queue
2015-11-02 16:44, Iremonger, Bernard:
> Hi Thomas,
>
>
>
> > Subject: Re: [dpdk-dev] [PATCH v7 02/28] librte_ether: add fields from
> > rte_pci_driver to rte_eth_dev_data
> >
> > 2015-11-02 10:36, Iremonger, Bernard:
> > > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > > > 2015
On Mon, Nov 02, 2015 at 06:31:07PM +0100, Adrien Mazarguil wrote:
> From: Nelio Laranjeiro
>
> ConnectX-4 is able to use indirection tables size of power of two, but
> with the current API it is impossible to predict its size, so to simplify,
> for any query/update RETA command, the indirection t
On Mon, 2 Nov 2015 21:49:54 +0530
Jerin Jacob wrote:
> On Mon, Nov 02, 2015 at 04:39:37PM +0100, Jan Viktorin wrote:
> > On Mon, 2 Nov 2015 19:48:40 +0530
> > Jerin Jacob wrote:
> >
> > > Signed-off-by: Jerin Jacob
> > > ---
> > > app/test-acl/main.c | 4 +
> > > lib/librte_acl/
From: Nelio Laranjeiro
ConnectX-4 is able to use indirection tables size of power of two, but
with the current API it is impossible to predict its size, so to simplify,
for any query/update RETA command, the indirection table is modified to use
256 entries.
A port stop/start must be done to appl
One issue I see with optimization config options such as NO_TX_OFFLOAD,
NO_MULTISEG, NO_REFCOUNT is: It is not sufficient to have those "Ifdefs" inside
mbuf structure, but should be sprinkled all over the code where corresponding
fields are used. This may make the code messier.
--
- Thanks
char
On Mon, 2 Nov 2015 21:59:12 +0530
Jerin Jacob wrote:
> On Mon, Nov 02, 2015 at 03:49:17PM +, Hunt, David wrote:
> > On 02/11/2015 15:36, Jan Viktorin wrote:
> > >On Mon, 2 Nov 2015 15:26:19 +
> > --snip--
> > >It was looking like we can share a lot of common code for both
> > >archi
On Mon, Nov 02, 2015 at 12:22:47PM +, Hunt, David wrote:
> On 02/11/2015 04:57, Jerin Jacob wrote:
> >On Fri, Oct 30, 2015 at 01:49:14PM +, David Hunt wrote:
> >>Signed-off-by: David Hunt
> --snip--
> >>+#ifndef _RTE_MEMCPY_ARM_64_H_
> >>+#define _RTE_MEMCPY_ARM_64_H_
> >>+
> >>+#include
On Mon, Nov 2, 2015 at 5:41 PM, Ananyev, Konstantin <
konstantin.ananyev at intel.com> wrote:
> > If you don't want rx/tx, don't poll the port.
>
> Well, the question is why to add an extra restriction here?
>
Well, until I start a port, I would expect it to do nothing.
Probably user deliberatel
Yes. Thank you. What we like is to get to a point where we discuss API and
align on APIs for SOC as Margaret mention. As you know Arm has been driving ODP
as the API for SOC.
What we like to do is to drive the APIs under DPDK even for Arm SOC. Long term,
and based on shrinking silicon geometries
Fix minor, and non critical, copy and paste error in strncmp() of eth-dest
commandline argument.
Fixes: bd785f6f6791 ("examples/l3fwd: make destination mac address
configurable")
Signed-off-by: John McNamara
---
examples/l3fwd/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
As SOC vendor we will contribute heavily to the project. Example crypto
acceleration. We already contribute a lot to the linux community.
So not sure why the doubt about of contribution?
-Original Message-
From: Dave Neary [mailto:dne...@redhat.com]
Sent: Monday, November 02, 2015 11:3
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Martin Weiser
> Sent: Wednesday, October 21, 2015 9:38 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] ixgbe: ierrors counter spuriously increasing in DPDK 2.1
> with DPDK 2.1 we are seeing the ierrors counter increasing for 82599ES
> ports
Hello,
On Mon, Nov 2, 2015 at 8:21 AM, Lu, Wenzhuo wrote:
>
> > @@ -2502,7 +2503,14 @@ static int igbvf_set_vfta(struct e1000_hw *hw,
> > uint16_t vid, bool on)
> > if (on)
> > msgbuf[0] |= E1000_VF_SET_VLAN_ADD;
> >
> > - return (mbx->ops.write_posted(hw, msgbuf, 2, 0));
2015-11-02 10:36, Iremonger, Bernard:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-10-30 15:08, Bernard Iremonger:
> > > add dev_flags to rte_eth_dev_data, add macros for dev_flags.
> > > add kdrv to rte_eth_dev_data.
> > > add numa_node to rte_eth_dev_data.
> > > add drv_
I think it is very important for the first version of governance that we have
ARM/SOC vendor/future contributors to be part of TSC.
If based on historical contribution - they will be at a disadvantage.
We need to have the DPDK organization support an API which supports a broader
set of chips.
--
2015-11-02 15:23, Ferruh Yigit:
> +ifeq ($(COMBINED_BUILD),1)
> include $(RTE_SDK)/mk/rte.sharelib.mk
> +endif
Why this ifeq?
rte.sharelib.mk is always used for combined lib.
>>> If you don't want rx/tx, don't poll the port.
>>Well, the question is why to add an extra restriction here?
>Well, until I start a port, I would expect it to do nothing.
>>Probably user deliberately doesn't want to call dev_start() for PF device -
>>as he doesn't plan to use it for RX/TX.
2015-11-02 10:17, Van Haaren, Harry:
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Sent: Monday, November 2, 2015 7:59 AM
> > > + /* if xstats_get() is implemented by the PMD, the Q stats are
> > > done */
> > > + if (dev->dev_ops->xstats_get != NULL)
> > > +
On Mon, Nov 2, 2015 at 5:09 PM, Ananyev, Konstantin <
konstantin.ananyev at intel.com> wrote:
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of David Marchand
> > Sent: Monday, November 02, 2015 4:03 PM
> > To: Liang, Cunming
> > Cc: dev at dpdk.org
> > S
On Fri, Oct 30, 2015 at 6:27 AM, Cunming Liang
wrote:
> As ixgbe vf doesn't support lsc, the patch removes those unused code.
> In addition, it does some tiny cleanup.
>
Please, can you describe this tiny cleanup ?
Did it trigger some bug ?
> diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c
> b/d
On Thu, Sep 24, 2015 at 7:33 AM, Cunming Liang
wrote:
> When ixgbe runs as a PF, mbox interrupt is prerequisite to make VF start
> normally.
> And PF sometimes won't 'dev_start', so the mbox interrupt register during
> 'dev_init' is required.
>
Can you describe the cases/situations where you wou
On Fri, Oct 30, 2015 at 6:27 AM, Cunming Liang
wrote:
> VFIO allows multiple MSI-X vector, others doesn't, but maybe will allow it
> in the future.
> Device drivers need to be aware of the capability.
> It's better to avoid condition check on interrupt type(VFIO) everywhere,
> instead
> a capabil
Hi Jacob,
> diff --git a/lib/librte_acl/rte_acl.c b/lib/librte_acl/rte_acl.c
> index d60219f..e2fdebd 100644
> --- a/lib/librte_acl/rte_acl.c
> +++ b/lib/librte_acl/rte_acl.c
> @@ -55,11 +55,32 @@ rte_acl_classify_avx2(__rte_unused const struct
> rte_acl_ctx *ctx,
> return -ENOTSUP;
> }
>
Hello,
On Fri, Oct 30, 2015 at 6:27 AM, Cunming Liang
wrote:
[snip]
> diff --git a/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h
> b/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h
> index 45071b7..b8fd318 100644
> --- a/lib/librte_eal/linuxapp/eal/include/exec-env/
Hi Thomas,
> Subject: Re: [dpdk-dev] [PATCH v7 02/28] librte_ether: add fields from
> rte_pci_driver to rte_eth_dev_data
>
> 2015-11-02 10:36, Iremonger, Bernard:
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > > 2015-10-30 15:08, Bernard Iremonger:
> > > > add dev_flags to
>
> From: David Marchand [mailto:david.marchand at 6wind.com]
> Sent: Monday, November 02, 2015 4:22 PM
> To: Ananyev, Konstantin
> Cc: Liang, Cunming; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v1 06/11] ixgbe: fix rx intr compatible issue
> with PF mbox
>
> On Mon, Nov 2, 2015 at 5:09
On Mon, 2 Nov 2015 19:48:40 +0530
Jerin Jacob wrote:
> Signed-off-by: Jerin Jacob
> ---
> app/test-acl/main.c | 4 +
> lib/librte_acl/Makefile | 5 +
> lib/librte_acl/acl.h | 4 +
> lib/librte_acl/acl_run_neon.c | 46 +++
> lib/librte_acl/acl_run_neon.h | 290
On Mon, 2 Nov 2015 15:26:19 +
"Hunt, David" wrote:
> On 02/11/2015 12:57, Jerin Jacob wrote:
> > On Mon, Nov 02, 2015 at 12:22:47PM +, Hunt, David wrote:
> >> Jerin,
> >> I've just benchmarked the libc version against the hand-coded version
> >> of
> >> the memcpy routines, and the
On Mon, 2 Nov 2015 10:47:53 +
"Hunt, David" wrote:
> On 02/11/2015 06:32, Jerin Jacob wrote:
> > On Fri, Oct 30, 2015 at 04:28:25PM +, Hunt, David wrote:
>
> --snip--
>
> >
> > Hi Jan and Dave,
> >
> > I have reviewed your patches for arm[64] support. Please check the
> > review comme
On Mon, 2 Nov 2015 15:04:14 +
"Hunt, David" wrote:
> On 02/11/2015 13:17, Jerin Jacob wrote:
> -snip--
> > If am not wrong existing rte_cpu_get_flag_enabled() implementation
> > should be broken in your platform also for arm64. as I could see only
> > AT_HWCAP
> > not AT_HWCAP2 and AT_HWCAP
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of David Marchand
> Sent: Monday, November 02, 2015 4:03 PM
> To: Liang, Cunming
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v1 06/11] ixgbe: fix rx intr compatible issue
> with PF mbox
>
> On Thu, Se
To add to Bagh's comments - AT&T is very interested in the governance being
proposed in expanding to allow more equal voice including from the SOC vendors.
We think it is important to rally around one API for data plane acceleration
which allows innovation to continue at the chip level.
From: Ba
On 02/11/2015 15:36, Jan Viktorin wrote:
> On Mon, 2 Nov 2015 15:26:19 +
--snip--
> It was looking like we can share a lot of common code for both
> architectures. I didn't know how much different are the cpuflags.
CPU flags for ARMv8 are looking like this now. Quite different to the
ARMv7 on
Hi Thomas,
2015-10-27 19:00 GMT+01:00 Thomas Monjalon :
> 2015-10-27 18:33, Matej Vido:
> > 2015-10-26 16:09 GMT+01:00 Thomas Monjalon :
> > > SZEDATA2 is not a vdev. Is it possible to probe it as a standard PCI
> > > device?
> > >
> > It would be possible to probe it as a standard PCI device, bu
On 02/11/2015 12:57, Jerin Jacob wrote:
> On Mon, Nov 02, 2015 at 12:22:47PM +, Hunt, David wrote:
>> Jerin,
>> I've just benchmarked the libc version against the hand-coded version of
>> the memcpy routines, and the libc wins in most cases. This code was just an
>> initial attempt at optim
Fixes following error:
LD libdpdk.so
/usr/bin/ld: /root/dpdk/build/lib/libdpdk.so: version node not found
for symbol @DPDK_x.y
Defines version symbols in a fixed path libdpdk.map file and this
value hardcoded into makefile
Signed-off-by: Ferruh Yigit
---
drivers/net/Makefile | 3 +++
lib
On 02/11/2015 15:13, Jan Viktorin wrote:
> On Mon, 2 Nov 2015 15:04:14 +
> "Hunt, David" wrote:
>
>> On 02/11/2015 13:17, Jerin Jacob wrote:
>> -snip--
>>> If am not wrong existing rte_cpu_get_flag_enabled() implementation
>>> should be broken in your platform also for arm64. as I could see o
Tim and Dave,
I agree that an architecture board membership should be based on
technical standing and contribution but at the same time,
if you are trying to bring a new hardware paradigm into a project, you
need to give a chance to some of those experts to
participate and gain the standing.
If
On 02/11/2015 13:17, Jerin Jacob wrote:
-snip--
> If am not wrong existing rte_cpu_get_flag_enabled() implementation
> should be broken in your platform also for arm64. as I could see only AT_HWCAP
> not AT_HWCAP2 and AT_HWCAP is 0x7 that means your platform also
> follows
>
> http://lxr.free-elec
Signed-off-by: Wenzhuo Lu
---
doc/guides/rel_notes/release_2_2.rst | 110 +++
1 file changed, 110 insertions(+)
diff --git a/doc/guides/rel_notes/release_2_2.rst
b/doc/guides/rel_notes/release_2_2.rst
index cf510e9..dc03905 100644
--- a/doc/guides/rel_notes/relea
On Mon, 2 Nov 2015 11:23:05 +0530
Jerin Jacob wrote:
--snip--
> > +#ifndef _RTE_ATOMIC_ARM_H_
> > +#define _RTE_ATOMIC_ARM_H_
> > +
> > +#ifdef __cplusplus
> > +extern "C" {
> > +#endif
> > +
> > +#include "generic/rte_atomic.h"
> > +
> > +/**
> > + * General memory barrier.
> > + *
> > + * Guara
On Mon, 2 Nov 2015 11:23:05 +0530
Jerin Jacob wrote:
--snip--
> > +/*- 16 bit atomic operations
> > -*/
> > +
> > +#ifndef RTE_FORCE_INTRINSICS
> > +static inline int
> > +rte_atomic16_cmpset(volatile uint16_t *dst, uint16_t exp, uint16_t src)
> >
Signed-off-by: Konstantin Ananyev
---
app/test-pmd/cmdline.c | 3 +++
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 7 +++
2 files changed, 10 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 0afdc96..ae9be81 100644
--- a/app/test-pmd/cmdl
On Mon, 2 Nov 2015 12:22:47 +
"Hunt, David" wrote:
> On 02/11/2015 04:57, Jerin Jacob wrote:
> > On Fri, Oct 30, 2015 at 01:49:14PM +, David Hunt wrote:
> >> Signed-off-by: David Hunt
> --snip--
> >> +#ifndef _RTE_MEMCPY_ARM_64_H_
> >> +#define _RTE_MEMCPY_ARM_64_H_
> >> +
> >> +#inc
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Sunday, November 01, 2015 11:16 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org; Mcnamara, John
> Subject: Re: [dpdk-dev] [PATCHv7 8/9] testpmd: add new command to display
> RX/TX queue informat
Signed-off-by: Chas Williams <3chas3 at gmail.com>
---
lib/librte_eal/common/include/rte_pci_dev_ids.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/lib/librte_eal/common/include/rte_pci_dev_ids.h
b/lib/librte_eal/common/include/rte_pci_dev_ids.h
index 93aa38a..c722e64
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost. It means librte_vhost is also needed to compile the PMD.
The vhost messages will be handled only when a port is started. So start
a port first, then invoke QEMU.
The PMD has 2 parameters.
- iface: The parame
These variables are needed to be able to manage one of virtio devices
using both vhost library APIs and vhost PMD.
For example, if vhost PMD uses current callback handler and private data
provided by vhost library, A DPDK application that links vhost library
cannot use some of vhost library APIs. T
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost. The patch will work on below patch series.
- [PATCH v7 00/28] remove pci driver from vdevs
* Known issue.
We may see issues while handling RESET_OWNER message.
These handlings are done in vhost library, so no
Hi,
On the contrary! I am aware that Freescale has been engaged for some
time in DPDK. I was responding to Margaret's contention that future
contributors (and she called out ARM and SOC vendors) should have a voice.
I hope that clarifies my position and meaning.
Thanks,
Dave.
On 11/02/2015 12:4
Hi Margaret,
On 11/02/2015 12:28 PM, CHIOSI, MARGARET T wrote:
> I think it is very important for the first version of governance that we have
> ARM/SOC vendor/future contributors to be part of TSC.
> If based on historical contribution - they will be at a disadvantage.
> We need to have the DPDK
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
MAINTAINERS |5 +
doc/guides/rel_notes/release_2_2.rst |4
2 files changed, 9 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 7b1e079..284d93c 10064
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
doc/guides/nics/index.rst |1 +
doc/guides/nics/nfp.rst | 189 +
2 files changed, 190 insertions(+)
create mode 100644 doc/guides/nics/nfp.rst
diff -
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
config/common_linuxapp |6 ++
drivers/net/Makefile |1 +
mk/rte.app.mk |1 +
3 files changed, 8 insertions(+)
diff --git a/config/common_linuxapp b/config/common_linuxapp
ind
From: "Alejandro.Lucero"
- nfp_net_promisc_enable
- nfp_net_promisc_disable
- nfp_net_infos_get
- nfp_net_dev_mtu_set
- nfp_net_vlan_offload_set
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
drivers/net/nfp/nfp_net.c | 181 +++
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
drivers/net/nfp/nfp_net.c | 94 +
1 file changed, 94 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index e1badc3..9cc777
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
drivers/net/nfp/nfp_net.c | 171 +
1 file changed, 171 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 51855b1..e1bad
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
drivers/net/nfp/nfp_net.c | 214 +
1 file changed, 214 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 8339c2d..51855
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
drivers/net/nfp/nfp_net.c | 921 +
1 file changed, 921 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index a33ed37..8339c
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
drivers/net/nfp/Makefile | 88 +
drivers/net/nfp/nfp_net.c | 785
drivers/net/nfp/nfp_net_ctrl.h | 290 +++
drivers/net/nfp/nfp_ne
From: "Alejandro.Lucero"
This patchset adds a new PMD for Netronome nfp-6xxx cards.
Just PCI Virtual Functions supported.
v5:
- Submitting code piece by piece as requested
v4:
- Just submitting PMD
v3:
- Making all patches independent for applying and building
- changing commits messages f
On 11/01/2015 08:17 PM, Thomas Monjalon wrote:
> 2015-10-19 08:36, Eric Kinzie:
>>Size of struct rte_eth_bond_8023ad_conf changed. Increment LIBABIVER
>>and version bond_mode_8023ad_setup and bond_mode_8023ad_conf_get
>>functions.
> [...]
>> +VERSION_SYMBOL(bond_mode_8023ad_setup, _v20
On 02/11/2015 04:57, Jerin Jacob wrote:
> On Fri, Oct 30, 2015 at 01:49:14PM +, David Hunt wrote:
>> Signed-off-by: David Hunt
--snip--
>> +#ifndef _RTE_MEMCPY_ARM_64_H_
>> +#define _RTE_MEMCPY_ARM_64_H_
>> +
>> +#include
>> +#include
>> +
>> +#ifdef __cplusplus
>> +extern "C" {
>> +#endif
>
On 2015/10/31 3:30, Thomas Monjalon wrote:
> 2015-10-27 16:54, Tetsuya Mukawa:
>> Below patch has been submitted as a separate patch.
>>
>> - [dpdk-dev,1/3] vhost: Fix wrong handling of virtqueue array index
>> (http://dpdk.org/dev/patchwork/patch/8038/)
> Please could you rebase only the two
On 2015/10/31 2:49, Loftus, Ciara wrote:
>> These variables are needed to be able to manage one of virtio devices
>> using both vhost library APIs and vhost PMD.
>> For example, if vhost PMD uses current callback handler and private data
>> provided by vhost library, A DPDK application that links v
On Sun, Nov 01, 2015 at 11:32:06AM +0100, Thomas Monjalon wrote:
> 2015-10-30 19:57, Adrien Mazarguil:
> > Mellanox PMDs (mlx4 and mlx5) support link status notification with
> > libibverbs through a file descriptor, in a manner reminiscent of UIO/VFIO.
> >
> > It still requires a new "external" i
On Fri, Oct 30, 2015 at 04:28:25PM +, Hunt, David wrote:
> On 30/10/2015 16:11, Jan Viktorin wrote:
> >Hmm, I see. It's good to fix this in the generated e-mails between
> >format-patch
> > and send-email calls. I always review those to be sure they meet my
> expectations ;).
> >Anyway, it is
On Mon Nov 02 12:23:47 +0200 2015, Panu Matilainen wrote:
> On 11/01/2015 08:17 PM, Thomas Monjalon wrote:
> >2015-10-19 08:36, Eric Kinzie:
> >> Size of struct rte_eth_bond_8023ad_conf changed. Increment LIBABIVER
> >> and version bond_mode_8023ad_setup and bond_mode_8023ad_conf_get
> >> fu
On Fri, Oct 30, 2015 at 01:25:28AM +0100, Jan Viktorin wrote:
> From: Vlastimil Kosar
>
> This patch adds architecture specific atomic operation file
> for ARM architecture. It utilizes compiler intrinsics only.
>
> Signed-off-by: Vlastimil Kosar
> Signed-off-by: Jan Viktorin
> ---
> v1 -> v2:
>
This patch removes the mac local fault count and
mac remote fault count from rx errors. The mac
fault count registers count faults, not packets,
and hence should not be added to packet counters.
Signed-off-by: Harry van Haaren
---
drivers/net/ixgbe/ixgbe_ethdev.c | 2 --
1 file changed, 2 deleti
On 02/11/2015 06:32, Jerin Jacob wrote:
> On Fri, Oct 30, 2015 at 04:28:25PM +, Hunt, David wrote:
--snip--
>
> Hi Jan and Dave,
>
> I have reviewed your patches for arm[64] support. Please check the
> review comments.
Hi Jerin,
I'm looking at the comments now, and working on getting the su
On Fri, Oct 30, 2015 at 01:49:16PM +, David Hunt wrote:
> Signed-off-by: David Hunt
> ---
> .../common/include/arch/arm/rte_cycles.h | 4 ++
> .../common/include/arch/arm/rte_cycles_64.h| 77
> ++
> 2 files changed, 81 insertions(+)
> create mode 10064
On Mon, 2 Nov 2015 19:27:40 +0100
Adrien Mazarguil wrote:
> Thanks for pointing this out, I'm used to avoiding C extensions whenever
> possible but will stick to DPDK style next time.
>
> Still, it would be nice if we could steer DPDK away from such extensions as
> much as possible. As a library
1 - 100 of 158 matches
Mail list logo