Hi,
According to "21.2.4.6.6.2", it seems that the implementation is supposed
to achieve max-min fairness. In your example, the effective cap of single
active pipe should be 1Mbps, given the total demand of other 1999 pipes is
less than 999Mbps.
Sangjin
On Mon, Apr 20, 2015 at 9:40 AM Greg Smith
This patch removes the use of rte_fdir_filter from enic_clsf.
This also takes care of modifying the version and copyright string.
Signed-off-by: Sujith Sankar
---
lib/librte_pmd_enic/enic.h | 4 ++--
lib/librte_pmd_enic/enic_clsf.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
d
> -Original Message-
> From: Richardson, Bruce
> Sent: Monday, April 20, 2015 4:03 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org; Wiles, Keith
> Subject: Re: [dpdk-dev] [RFC PATCH 1/4] Add example pktdev implementation
>
> On Mon, Apr 20, 2015 at 12:26:43PM +0100, Ananyev, Konstantin
On 04/20/15 13:50, Bruce Richardson wrote:
> On Mon, Apr 20, 2015 at 01:07:59PM +0300, Vlad Zolotarov wrote:
>> Hi,
>> I would like to ask if there is any reason why DPDK doesn't have support for
>> DCA feature?
>>
>> thanks,
>> vlad
> With modern platforms with DDIO the data written by the NIC a
On Tue, Apr 21, 2015 at 09:40:05AM +0100, Ananyev, Konstantin wrote:
> > -Original Message-
> > From: Richardson, Bruce
> > Sent: Monday, April 20, 2015 4:03 PM
> > To: Ananyev, Konstantin
> > Cc: dev at dpdk.org; Wiles, Keith
> > Subject: Re: [dpdk-dev] [RFC PATCH 1/4] Add example pktdev i
On Tue, Apr 21, 2015 at 11:51:40AM +0300, Vlad Zolotarov wrote:
>
>
> On 04/20/15 13:50, Bruce Richardson wrote:
> >On Mon, Apr 20, 2015 at 01:07:59PM +0300, Vlad Zolotarov wrote:
> >>Hi,
> >>I would like to ask if there is any reason why DPDK doesn't have support for
> >>DCA feature?
> >>
> >>th
Hi Stephen,
> The VFIO_PRESENT #define was a landmine and we hit it.
> The DPDK has a config system and it should be used rather than silently
> dropping a feature during build only to have it fail at run time.
>
> If VFIO is configured, and the kernel headers are not present the build
> should f
On 04/21/15 12:27, Bruce Richardson wrote:
> On Tue, Apr 21, 2015 at 11:51:40AM +0300, Vlad Zolotarov wrote:
>>
>> On 04/20/15 13:50, Bruce Richardson wrote:
>>> On Mon, Apr 20, 2015 at 01:07:59PM +0300, Vlad Zolotarov wrote:
Hi,
I would like to ask if there is any reason why DPDK doesn
The first objective of this series is to fix the support of indirect
mbufs when the application reserves a private area in mbufs. It also
removes the limitation that rte_pktmbuf_clone() is only allowed on
direct (non-cloned) mbufs. The series also contains some enhancements
and fixes in the mbuf ar
Deduct the mbuf data room size from mempool->elt_size and priv_size,
instead of using an hardcoded value that is not related to the real
buffer size.
To use rte_pktmbuf_pool_init(), the user can either:
- give a NULL parameter to rte_pktmbuf_pool_init(): in this case, the
private size is assumed
The mbuf pool private area must always be populated in a mbuf pool.
The applications or drivers may expect that for a mbuf pool, the mbuf
pool private area (mbuf_data_room_size and mbuf_priv_size) are
properly filled.
Signed-off-by: Olivier Matz
---
examples/ip_fragmentation/main.c | 4 ++--
exa
This code retrieving the pool private area is duplicated in many
places, we can use of function for it.
Signed-off-by: Olivier Matz
---
lib/librte_ether/rte_ethdev.c| 4 +--
lib/librte_mbuf/rte_mbuf.h | 41
lib/librte_pmd_af_packet/
Allow the user to use the default rte_pktmbuf_init() function even
if the mbuf private size is not 0.
Signed-off-by: Olivier Matz
---
lib/librte_mbuf/rte_mbuf.c | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf
The rte_pktmbuf_pool_init() and rte_pktmbuf_init() functions now
support to have a non-hardcoded buffer length. We can remove the
specific functions used in testpmd and replace them by the standard
ones.
Signed-off-by: Olivier Matz
---
app/test-pmd/testpmd.c | 74 +---
Add a new wrapper to rte_mempool_create() to simplify the creation
of a packet mbuf pool.
This wrapper can be used if there is no specific mempool flags, and
no specific mbuf or pool constructor function, which is most of the
use cases.
Signed-off-by: Olivier Matz
---
doc/guides/rel_notes/updat
When it's possible, use the new helper to create the mbuf pools.
Most of the patch is trivial, except for the following files that
have some specifics (indirect mbufs):
- ip_fragmentation
- ip_pipeline
- ipv4_multicast
- vhost
Signed-off-by: Olivier Matz
---
app/test-pipeline/init.c
Add a new private_size field in mbuf structure that should
be initialized at mbuf pool creation. This field contains the
size of the application private data in mbufs.
Introduce new static inline functions rte_mbuf_from_indirect()
and rte_mbuf_to_baddr() to replace the existing macros, which
take
Remove one limitation of rte_pktmbuf_attach(): "mbuf we're attaching to
must be direct".
Now, when we attach to an indirect mbuf:
- copy the all relevant fields (addr, len, offload, ...) as before
- get the pointer to the mbuf that embeds the data buffer (direct mbuf),
and increase the reference
It's better to name the mbuf 'm' instead of 'mc' as it's not a clone.
Signed-off-by: Olivier Matz
---
app/test/test_mbuf.c | 25 -
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index 4774263..2614598 100644
---
Check that the data in the cloned mbuf is the same than in the
reference mbuf.
Check that the reference counter is incremented for each segment.
Signed-off-by: Olivier Matz
---
app/test/test_mbuf.c | 37 +
1 file changed, 37 insertions(+)
diff --git a/app/tes
Signed-off-by: Olivier Matz
---
app/test/test_mbuf.c | 26 ++
1 file changed, 26 insertions(+)
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index 01838c6..b5ae5b7 100644
--- a/app/test/test_mbuf.c
+++ b/app/test/test_mbuf.c
@@ -324,6 +324,7 @@ testclone_testup
On Mon, Apr 20, 2015 at 12:26:15PM -0700, Eric Kinzie wrote:
> From: Eric Kinzie
>
> Provide functions to allow an external 802.3ad state machine to transmit
> and recieve LACPDUs and to set the collection/distribution flags on
> slave interfaces.
>
> Signed-off-by: Eric Kinzie
> ---
> l
Hi Dor,
What you are looking for seems straight forward to implement and it
should not really affect the driver's performance at all. Even adding
the full timestamp (seconds plus microseconds). However, I don't see too
much people looking for that feature to make it to mainline. I could be
wro
On Tue, Apr 21, 2015 at 11:55:10AM +0200, Olivier Matz wrote:
> The first objective of this series is to fix the support of indirect
> mbufs when the application reserves a private area in mbufs. It also
> removes the limitation that rte_pktmbuf_clone() is only allowed on
> direct (non-cloned) mbuf
Aside from testing, this also has the benefit of being able to run a
capture file through your application without having to send it
through another NIC (if you have only one, that'd be impossible, for
example). I can see this being needed if you had, for instance, a DPI
app in DPDK and wanted to r
On Mon, Apr 20, 2015 at 02:54:10PM -0700, Stephen Hemminger wrote:
> The hyper-v device driver forces the base EAL code to change
> to support multiple bus types. This is done changing the pci_device
> in ether driver to a generic union.
>
> As much as possible this is done in a backwards source c
> -Original Message-
> From: Tetsuya Mukawa [mailto:mukawa at igel.co.jp]
> Sent: Friday, March 27, 2015 8:39 AM
> To: dev at dpdk.org
> Cc: Iremonger, Bernard; Richardson, Bruce; david.marchand at 6wind.com;
> Tetsuya Mukawa
> Subject: [PATCH v3 0/5] Clean up pci uio implementations
>
>
When I'm run testpmd with DPDK 2.0 and the Mellanox ConnectX-3. I get a error.
Here are my questions:
1> lspci |grep Mell
02:00.0 Network controller: Mellanox Technologies MT27500 Family [ConnectX-3]
2> compile libibverbs-1.1.7mlnx1 && libmlx4-1.0.5mlnx1
export EXTRA_CFLAGS=-I/DPDK/mlx4/inst
Hi Kang,
You probably compiled the downloaded libraries using dynamic linkage, prior to
the static one you specified here.
You need to remove them from /usr/local/libs and also please delete
/usr/local/include/infiniband
Let me know if it solves your problem.
Best Regards,
Olga
-Orig
On Sat, Apr 18, 2015 at 12:43:06PM -0700, Ravi Kerur wrote:
> Changes in v6
> Remove RTE_EXEC_ENV_BSDAPP from eal_common_thread.c file.
> Add pthread_setname_np/pthread_set_name_np for Linux/FreeBSD
> respectively. Plan to use _getname_ in RTE_LOG when available.
> Use existing rte_get_systid() in
On Sat, Apr 18, 2015 at 12:43:07PM -0700, Ravi Kerur wrote:
> Changes in v6
> Split eal_common_system.c and eal_common_runtime.c into
> eal_common_sysfs.c
> eal_common_mem_cfg.c
> eal_common_proc_type.c
> eal_comm_app_usage.c
> based on functionality.
>
> Changes in v5
> Rebase to latest code.
>
Hi Olivier,
> -Original Message-
> From: Olivier Matz [mailto:olivier.matz at 6wind.com]
> Sent: Tuesday, April 21, 2015 10:55 AM
> To: dev at dpdk.org
> Cc: Ananyev, Konstantin; zoltan.kiss at linaro.org; Richardson, Bruce;
> nhorman at tuxdriver.com; olivier.matz at 6wind.com
> Subject:
> -Original Message-
> From: Olivier Matz [mailto:olivier.matz at 6wind.com]
> Sent: Tuesday, April 21, 2015 10:55 AM
> To: dev at dpdk.org
> Cc: Ananyev, Konstantin; zoltan.kiss at linaro.org; Richardson, Bruce;
> nhorman at tuxdriver.com; olivier.matz at 6wind.com
> Subject: [PATCH v5
Hi Konstantin,
On 04/21/2015 05:01 PM, Ananyev, Konstantin wrote:
> Hi Olivier,
>
>> -Original Message-
>> From: Olivier Matz [mailto:olivier.matz at 6wind.com]
>> Sent: Tuesday, April 21, 2015 10:55 AM
>> To: dev at dpdk.org
>> Cc: Ananyev, Konstantin; zoltan.kiss at linaro.org; Richardso
Hyper-V Poll Mode Driver.
Only change from v3 is addition of pieces that were missing
for rte_vmbus.h and rte_vmbus.c
Stephen Hemminger (7):
ether: add function to query for link state interrupt
pmd: change drivers initialization for pci
hv: add basic vmbus support
hv: uio driver
hv: po
From: Stephen Hemminger
Allow application to query whether link state will work.
This is also part of abstracting dependency on PCI.
Signed-off-by: Stephen Hemminger
---
lib/librte_ether/rte_ethdev.c | 14 ++
lib/librte_ether/rte_ethdev.h | 12
2 files changed, 26 inse
From: Stephen Hemminger
The change to generic ether device structure to support multiple
bus types requires a change to all existing PMD but only in the
initialization (and the change is backwards compatiable).
Signed-off-by: Stephen Hemminger
---
lib/librte_pmd_e1000/em_ethdev.c| 2 +-
The hyper-v device driver forces the base EAL code to change
to support multiple bus types. This is done changing the pci_device
in ether driver to a generic union.
As much as possible this is done in a backwards source compatiable
way. It will break ABI for device drivers.
Signed-off-by: Stephen
From: Stephen Hemminger
Add new UIO driver in kernel to support DPDK Poll Mode Driver.
Signed-off-by: Stas Egorov
Signed-off-by: Stephen Hemminger
---
lib/librte_eal/linuxapp/Makefile| 3 +
lib/librte_eal/linuxapp/hv_uio/Makefile | 57 ++
lib/librte_eal/linuxapp/hv_uio/hv_u
From: Stephen Hemminger
This is new Poll Mode driver for using hyper-v virtual network
interface.
Signed-off-by: Stas Egorov
Signed-off-by: Stephen Hemminger
---
lib/Makefile |1 +
lib/librte_pmd_hyperv/Makefile| 28 +
lib/librte_pmd_hyperv/hyperv.h
From: Stephen Hemminger
Add hyperv driver config to enable it.
Signed-off-by: Stephen Hemminger
---
config/common_linuxapp | 9 +
1 file changed, 9 insertions(+)
diff --git a/config/common_linuxapp b/config/common_linuxapp
index 0078dc9..58cc352 100644
--- a/config/common_linuxapp
+++
From: Stephen Hemminger
For users using non latest kernels, put kernel patch in for
them to use.
Signed-off-by: Stephen Hemminger
---
.../linuxapp/hv_uio/vmbus-get-pages.patch | 55 ++
1 file changed, 55 insertions(+)
create mode 100644 lib/librte_eal/linuxapp/hv_
On Tue, 21 Apr 2015 09:34:27 +
"Burakov, Anatoly" wrote:
> Hi Stephen,
>
> > The VFIO_PRESENT #define was a landmine and we hit it.
> > The DPDK has a config system and it should be used rather than silently
> > dropping a feature during build only to have it fail at run time.
> >
> > If VF
On Tue, Apr 21, 2015 at 10:27:48AM +0100, Bruce Richardson wrote:
> Can you perhaps comment on the use-case where you find this binding
> limiting? Modern platforms have multiple NUMA nodes, but they also generally
> have PCI slots connected to those multiple NUMA nodes also, so that you can
> h
On Tue, Apr 21, 2015 at 7:36 AM, Bruce Richardson <
bruce.richardson at intel.com> wrote:
> On Sat, Apr 18, 2015 at 12:43:07PM -0700, Ravi Kerur wrote:
> > Changes in v6
> > Split eal_common_system.c and eal_common_runtime.c into
> > eal_common_sysfs.c
> > eal_common_mem_cfg.c
> > eal_common_proc_
On Tue, Apr 21, 2015 at 7:25 AM, Bruce Richardson <
bruce.richardson at intel.com> wrote:
> On Sat, Apr 18, 2015 at 12:43:06PM -0700, Ravi Kerur wrote:
> > Changes in v6
> > Remove RTE_EXEC_ENV_BSDAPP from eal_common_thread.c file.
> > Add pthread_setname_np/pthread_set_name_np for Linux/FreeBSD
>
Hi Stephen
Will you have documentation to go along with these changes?
Thanks
Siobhan
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Tuesday, April 21, 2015 6:33 PM
> To: alexmay at microsoft.com
> Cc: dev at dpdk.org; Stas Egoro
On Tue, 21 Apr 2015 19:34:39 +
"Butler, Siobhan A" wrote:
> Hi Stephen
> Will you have documentation to go along with these changes?
> Thanks
> Siobhan
Unlikely. Microsoft or other contributors might add something
in a later version.
The documentation that exists in DPDK related drivers ju
On Apr 21, 2015 6:23 PM, "Zhang, Helin" wrote:
>
> Hi Vlad
>
> I have a concern about the code changes you added in ixgbe_type.h.
> For ixgbe, all source files in librte_pmd_ixgbe/ixgbe, except
ixgbe_osdep.h were called as "base driver", which was not developed by DPDK
developers, and released by
Linux forced Intel to change away from the base driver model,
and at this point as many patches come from community as Intel.
The "base driver" model is not sustainable over the long term in my
admittedly biased opinion.
Just look how much dead code is in the base driver
Any comments/suggestions?
On Tue, Apr 7, 2015 at 3:58 AM, Takuya ASADA
wrote:
> This is 3rd version of OSv support patchset.
>
> Changelist:
> - sync with latest master
> - updated Capstan build script for OSv v0.19
> - rte_eal_version.map fix
> - CONFIG_RTE_BUILD_SHARED_LIB=y and CONFIG_RTE
52 matches
Mail list logo