> -Original Message-
> From: Wu, Jingjing
> Sent: Thursday, February 9, 2017 6:42 AM
> To: dev@dpdk.org
> Cc: Wu, Jingjing ; Mcnamara, John
>
> Subject: [PATCH v2] doc: add known issue for i40e PMD
>
> Rx statistic is incorrect when packet is oversize.
>
> Signed-off-by: Jingjing Wu
Signed-off-by: Nelio Laranjeiro
---
doc/guides/nics/features/default.ini | 1 +
doc/guides/nics/features/mlx5.ini| 1 +
doc/guides/nics/mlx5.rst | 1 +
3 files changed, 3 insertions(+)
diff --git a/doc/guides/nics/features/default.ini
b/doc/guides/nics/features/default.ini
index
Biggest change is in the first patch to provide a better visibility for users
on which NIC has been tested on which platform.
Others two are just updates for Mellanox PMDs.
Nelio Laranjeiro (3):
doc: merge release notes sections
doc: update release notes for mlx4
doc: update release notes f
These sections do not provide the exact tests that were done nor whether
specific NICs are supported by all platforms.
Signed-off-by: Nelio Laranjeiro
---
doc/guides/rel_notes/release_17_02.rst | 60 +-
1 file changed, 30 insertions(+), 30 deletions(-)
diff --git
Signed-off-by: Nelio Laranjeiro
---
doc/guides/nics/mlx4.rst | 9 +++--
doc/guides/rel_notes/release_17_02.rst | 35 ++
2 files changed, 42 insertions(+), 2 deletions(-)
diff --git a/doc/guides/nics/mlx4.rst b/doc/guides/nics/mlx4.rst
index 49f4
Signed-off-by: Nelio Laranjeiro
---
doc/guides/nics/mlx5.rst | 37 --
doc/guides/rel_notes/release_17_02.rst | 202 +
2 files changed, 230 insertions(+), 9 deletions(-)
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 571a9
On Sat, Feb 04, 2017 at 03:57:47AM +, Pei, Yulong wrote:
> Hi Thomas,
>
> Did you notice this patch ? do you have any comments ?
>
> Thanks
> Yulong Pei
>
> -Original Message-
> From: Pei, Yulong
> Sent: Monday, January 23, 2017 4:38 PM
> To: dev@dpdk.org
> Cc: Mcnamara, John ; thom
Dear all,
Is there anyone using Intel XL710-QDA2? Can it achieve 64B line rate in
testpmd?
How many cores can achieve the maximum throughput?
Thanks.
--
Regards,
Royce
On Thu, Feb 09, 2017 at 12:08:54PM +0530, Karthikraj palanichamy wrote:
> Hi developers,
>
> For /`rte_pktmbuf_pool_create()`, /I normally provide
> RTE_MBUF_DEFAULT_BUF_SIZE as the data room size.
>
> I needed to change this dynamically based on the frame size to be
> transmitted.
> To transmit
Thanks for the help.
I couldn't stop or close the device/port since Rx is a continuous
process with a separate mempool.
I'm creating dynamic mempools only for Tx.
Tried stopping the Tx queue 'rte_eth_dev_tx_queue_stop()' after freeing
the mempool. Created mempool again and then started the tx
On 2/9/2017 12:40 AM, John Daley wrote:
> Hard coded mask values were being used for several of the IPv4 and IPv6
> fields. Use the values in the rte_eth_fdir_masks structure provided by the
> caller.
>
> Fixes: dfbd6a9cb504 ("net/enic: extend flow director support for 1300 series")
>
> Cc: sta..
Hi all,
When DPDK was a small project, it was easy to propose a major change,
get feedback from the few contributors and figure a decision.
It had the drawback of the lack of various point of views.
So we probably made some quick and wrong decisions.
As the community is growing, we need to improv
> -Original Message-
> From: Patrick MacArthur [mailto:patr...@patrickmacarthur.net]
> Sent: Thursday, January 26, 2017 11:05 PM
> To: dev@dpdk.org; Burakov, Anatoly
> Cc: Patrick MacArthur
> Subject: [PATCH] vfio: fix file descriptor leak in multi-process applications
>
> When a seconda
Fix segmentation fault happened when use QAT PMD's kasumi, snow3g or zug
algorithm to do cipher-then-auth performance test application.
The mentioned algorithms required authentication key data be set value
equal to cipher key data.
Fixes: f8be1786b1b8 ("app/crypto-perf: introduce performance test
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
>
> Hi all,
>
> When DPDK was a small project, it was easy to propose a major change,
> get feedback from the few contributors and figure a decision.
> It had the drawback of the lack of various point of views.
> So we probabl
Fix segmentation fault happened when use QAT PMD's kasumi, snow3g or zug
algorithm to do cipher-then-auth performance test application.
The mentioned algorithms required authentication key data be set.
This patch fix issue that gmac algorithm required authentication key data
be set value equal to c
On 2/9/2017 12:40 AM, John Daley wrote:
> This should replace
> http://www.dpdk.org/dev/patchwork/patch/20147/ which was an intermediate patch
> in a small patchset and had a compile error. Also, please reject the 2nd patch
> in the patchset http://www.dpdk.org/dev/patchwork/patch/20148/ which is
>
On Thu, Feb 09, 2017 at 12:11:39PM +0100, Thomas Monjalon wrote:
> Hi all,
>
Hi Thomas,
thanks for kicking off the discussion here.
> When DPDK was a small project, it was easy to propose a major change,
> get feedback from the few contributors and figure a decision.
> It had the drawback of the
On 2/9/2017 12:40 AM, John Daley wrote:
> Hard coded mask values were being used for several of the IPv4 and IPv6
> fields. Use the values in the rte_eth_fdir_masks structure provided by the
> caller.
>
> Fixes: dfbd6a9cb504 ("net/enic: extend flow director support for 1300
> series")
> Cc: sta.
L2fwd-crypto app was padding an incoming buffer,
to be aligned with the algorithm block size, in all cases.
This was not the right approach, as padding is only necessary
when using block cipher algorithms, such as AES-CBC.
In case of using a stream cipher algorithm, such as SNOW3G UEA2,
there is no
Trying to query the link status through the new ETHTOOL_GLINKSETTINGS
ioctl available since Linux 4.5 was always failing due to a kernel bug
fixed since version 4.9.
This commit also addresses a common issue where the headers version used
at compile time differs from that of the kernel on the targ
On 2/9/2017 4:05 AM, Jianbo Liu wrote:
> To get better performance, Rx bulk alloc recv function will scan 8 descs
> in one time, but the statuses are not consistent on ARM platform because
> the memory allocated for Rx descriptors is cacheable hugepages.
> This patch is to calculate the number of r
On 2/9/2017 12:39 PM, Ferruh Yigit wrote:
> On 2/9/2017 4:05 AM, Jianbo Liu wrote:
>> To get better performance, Rx bulk alloc recv function will scan 8 descs
>> in one time, but the statuses are not consistent on ARM platform because
>> the memory allocated for Rx descriptors is cacheable hugepage
On 2/9/2017 4:05 AM, Jianbo Liu wrote:
> vPMD will check 4 descs in one time, but the statuses are not consistent
> because the memory allocated for RX descriptors is cacheable huagepage.
> This patch is to calculate the number of received packets by scann DD bit
> sequentially, and stops when meet
On 2/8/2017 6:12 PM, Yong Wang wrote:
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ido Barnea (ibarnea)
>> Sent: Thursday, February 2, 2017 4:08 AM
>> To: dev@dpdk.org
>> Subject: [dpdk-dev] [PATCH] Fill speed_capa for vmxnet3
>>
>> From: Ido Barnea
>>
>> S
On 2/9/2017 7:06 AM, Rasesh Mody wrote:
> This patch bumps bnx2x PMD version to 1.0.5.1 to track the recent
> fixes to the driver.
>
> Signed-off-by: Rasesh Mody
Applied to dpdk-next-net/master, thanks.
2017-02-09 11:54, O'Driscoll, Tim:
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
> > I suggest that the technical board should check whether every new
> > proposed features are explained, discussed and approved enough in
> > the community.
>
> I assume you don't mean every
Fixes the wrong slave initialization issue on start-up
Fixes: 100e4f7("add round-robin mode")
Signed-off-by: Fan Zhang
---
drivers/crypto/scheduler/scheduler_roundrobin.c | 16 +++-
1 file changed, 7 insertions(+), 9 deletions(-)
diff --git a/drivers/crypto/scheduler/scheduler_roun
> -Original Message-
> From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com]
> Sent: Tuesday, February 7, 2017 6:25 AM
> To: Van Haaren, Harry
> Cc: dev@dpdk.org; Richardson, Bruce
> Subject: Re: [PATCH v2 02/15] eventdev: add APIs for extended stats
> How about adding the reset fu
On 2/6/2017 8:08 AM, David Hunt wrote:
> Given that the packet distributor library improvements (1) will
> not be in 17.02, I plan on doing some consolidation of the
> API for burst operation for 17.05, merging the two api's into
> one, with options for single or burst operation.
>
> (1) http://dp
After this change, the EAL CPU NUMA node resolution step can no longer
emit an rte_panic. This aligns with the code in rte_eal_init, which
expects failures to return an error code.
Signed-off-by: Aaron Conole
---
lib/librte_eal/common/eal_common_lcore.c | 7 ---
1 file changed, 4 insertions
When attempting to scan hugepages, signal to the eal.c that an error has
occurred, rather than performing a panic.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal
It's possible that the application could take a corrective action here,
and either prompt the user for different arguments, or at least perform
a better logging. Exiting this early prevents any useful information
gathering from the application layer.
Signed-off-by: Aaron Conole
---
lib/librte_e
If we fail to acquire hugepage information, simply signal an error to
the application. This clears the run_once counter, allowing the user or
application to take a corrective action and retry.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 8 ++--
1 file changed, 6 inse
There may be no way to gracefully recover, but the application
should be notified that a failure happened, rather than completely
aborting. This allows the user to proceed with a "slow-path" type
solution.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 7 +--
1 file cha
In many cases, it's enough to simply let the application know that the
call to initialize DPDK has failed. A complete halt can then be
decided by the application based on error returned (and the app could
even attempt a possible re-attempt after some corrective action by the
user or application).
It's now possible to gracefully exit the application, or for
applications which support non-dpdk datapaths working in concert with
DPDK datapaths, there no longer is the possibility of exiting for
unsupported CPUs.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 6 +-
1 f
When memzone initialization fails, report the error to the calling
application rather than panic(). Without a good way of detaching /
releasing hugepages, at this point the application will have to restart.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 7 +--
1 file ch
This adds a new API to check for the eal cpu versions.
Signed-off-by: Aaron Conole
---
lib/librte_eal/common/eal_common_cpuflags.c | 13 +++--
lib/librte_eal/common/include/generic/rte_cpuflags.h | 9 +
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/lib
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal.c
b/lib/librte_eal/linuxapp/eal/eal.c
index 8f9bce1..e0dff6e 100644
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librte_e
When log initialization fails, it's generally because the fopencookie
failed. While this is rare in practice, it could happen, and it is
likely because of memory pressure. So, flag the error, and allow the
user to retry.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 9 +++
This can only happen when access to hugepages (either as primary or
secondary process) fails (and that is usually permissions). Since the
manner of failure is not reversible, we cannot allow retry.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 7 +--
1 file changed, 5
There are some theoretical racy conditions in the system that _could_
cause early tailq init to fail; however, no need to panic the
application. While it can't continue using DPDK, it could make better
alerts to the user.
Signed-off-by: Aaron Conole
---
lib/librte_eal/common/eal_common_tailqs.
rte_eal_alarm_init() call uses the linux timerfd framework to create a
poll()-able timer using standard posix file operations. This could fail
for a few reasons given in the man-pages, but many could be
corrected by the user application. No need to panic.
Signed-off-by: Aaron Conole
---
lib/li
There could be some confusion as to why the call failed - this change
will always reflect the value of the error in rte_error.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal_interrupts.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/lib/librte_eal/linux
After code inspection, there is no way for eal_timer_init() to fail. It
simply returns 0 in all cases. As such, this test could either go-away
or stay here as 'future-proofing'.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 7 +--
1 file changed, 5 insertions(+), 2 de
When initializing the interrupt thread, there are a number of possible
reasons for failure - some of which are correctable by the application.
Do not panic() needlessly, and give the application a change to reflect
this information to the user.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linu
Plugins are useful and important. However, it seems crazy to abort
everything just because they don't initialize properly.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal.c
b/l
It may even be possible to simply log the error and continue on letting
the user check the logs and restart the application when things are failed.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/lib/lib
Even if one vdev should fail, there's no need to prevent further
processing. Log the error, and reflect it to the higher levels to
decide.
Signed-off-by: Aaron Conole
---
lib/librte_eal/common/eal_common_dev.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/lib/librte_e
Some devices may be inaccessible for a variety of reasons, or the
PCI-bus may be unavailable causing the whole thing to fail. Still,
better to continue attempts at probes.
Signed-off-by: Aaron Conole
---
lib/librte_eal/common/eal_common_pci.c | 15 ++-
1 file changed, 10 insertions(
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal.c
b/lib/librte_eal/linuxapp/eal/eal.c
index be03b63..57c41e0 100644
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librt
The rte_eal_init function will now pass failure reason hints to the
application. To help app developers deciper this, add some brief
information about what the codes are indicating.
Signed-off-by: Aaron Conole
---
lib/librte_eal/common/include/rte_eal.h | 27 ++-
1 file
Seems like it's possible to continue. At least, the error is reflected
properly in the logs. A user could then go and correct or investigate
the situation.
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/
On 2/8/2017 4:46 PM, Ananyev, Konstantin wrote:
>
>
>> -Original Message-
>> From: Wu, Jingjing
>> Sent: Wednesday, February 8, 2017 5:06 AM
>> To: dev@dpdk.org
>> Cc: Wu, Jingjing ; olivier.m...@6wind.com; Zhang,
>> Helin ; Ananyev, Konstantin
>>
>> Subject: [PATCH v4 0/4] fix bitmask
The stats register can rewind to zero when the port is running for a long
period.
So I am afraid that this check is not always correct.
Why not introduce a variable to directly indicate whether the resulted stats
should be updated or not.
> -Original Message-
> From: dev [mailto:dev-boun.
This will usually be an issue because of permissions. However, it could
also be caused by OOM. In either case, errno will contain the
underlying cause. It is safe to re-init the system here, so allow the
application to take corrective action and reinit.
Signed-off-by: Aaron Conole
---
lib/lib
Signed-off-by: Aaron Conole
---
lib/librte_eal/linuxapp/eal/eal.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal.c
b/lib/librte_eal/linuxapp/eal/eal.c
index 2a3d2f6..b21d715 100644
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/libr
On 2/9/2017 12:29 PM, Shahaf Shuler wrote:
> Trying to query the link status through the new ETHTOOL_GLINKSETTINGS
> ioctl available since Linux 4.5 was always failing due to a kernel bug
> fixed since version 4.9.
>
> This commit also addresses a common issue where the headers version used
> at c
2017-01-24 07:34, Jianfeng Tan:
> Patch 1: howto doc of virtio_user for container networking.
> Patch 2: howto doc of virtio_user as exceptional path.
> Patch 3: remove ABI changes in igb_uio
For the patch 3, we are waiting a new revision postponing the notice.
For the first 2 patches, the SVG fi
2016-12-06 11:54, Mcnamara, John:
> From: Baruch Siach [mailto:bar...@tkos.co.il]
> > > > --- a/doc/guides/prog_guide/source_org.rst
> > > > +++ b/doc/guides/prog_guide/source_org.rst
> > > > @@ -162,5 +162,5 @@ The examples directory contains sample
> > > > applications that show how libraries can
On 2/8/2017 1:57 PM, Shahaf Shuler wrote:
> This patchset adds support for rxq interrupts on MLX5 PMD.
> The first patch introduces changes on eal_interrupt to support external
> interrupt handlers.
> The second patch implements the support on mlx5 PMD and demonstrate the use
> of the changes mad
2016-12-06 17:53, Thomas Monjalon:
> 2016-12-06 14:38, Baruch Siach:
> > Extend the note at the end of Source Organization chapter to also mention
> > the app directory.
> >
> > Signed-off-by: Baruch Siach
> [...]
> > --- a/doc/guides/prog_guide/source_org.rst
> > +++ b/doc/guides/prog_guide/sour
Another way is to clear
hw_stats->last_vfgprc/last_vfgorc/last_vfgptc/last_vfmprc
at the same time PF is set down.
> -Original Message-
> From: Dai, Wei
> Sent: Thursday, February 9, 2017 10:38 PM
> To: 'Olivier Matz' ; dev@dpdk.org; Zhang, Helin
> ; Ananyev, Konstantin
>
> Cc: Guo Feng
2017-02-09 09:27, Nelio Laranjeiro:
> Signed-off-by: Nelio Laranjeiro
> ---
> doc/guides/nics/features/default.ini | 1 +
> doc/guides/nics/features/mlx5.ini| 1 +
> doc/guides/nics/mlx5.rst | 1 +
> 3 files changed, 3 insertions(+)
You missed to add the feature check for ixgbe a
2017-02-01 11:21, John Daley:
> Signed-off-by: John Daley
Applied, thanks
2017-02-06 15:10, Stephen Hemminger:
> On Mon, 6 Feb 2017 14:16:53 +0100
> Olivier Matz wrote:
>
> > +**Implication**:
> > + In this situation, the shared data will be modified by the driver,
> > + potentially causing race conditions with the other users of the mbuf
> > + data.
>
> This i
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Wenzhuo Lu
> Sent: Thursday, February 9, 2017 1:28 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo
> Subject: [dpdk-dev] [PATCH v2 2/2] doc: announce ixgbe MTU setting
> limitation
>
> Signed-off-by: Wenzhuo Lu
Acked-
> -Original Message-
> From: stable [mailto:stable-boun...@dpdk.org] On Behalf Of Wenzhuo Lu
> Sent: Thursday, February 9, 2017 1:28 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; sta...@dpdk.org
> Subject: [dpdk-stable] [PATCH v2 1/2] doc: fix framework of ixgbe doc
>
> The framework of the
> > Currently, extended statistics has been supported by virtio. But there
> > are no corresponding document updates. Therefore, this patch is to update
> > the document for virtio xstats feature.
> >
> > Signed-off-by: Jiayu Hu
>
> Acked-by: Yuanhan Liu
Fixes: 76d4c652e07d ("virtio: add exten
2017-02-07 10:44, Daniel Mrzyglod:
> This patch fixes error: implicit declaration of function 'getline'
>
> Fixes: f8be1786b1b8 ("app/crypto-perf: introduce performance test
> application")
>
> Signed-off-by: Daniel Mrzyglod
Applied, thanks
Signed-off-by: Nelio Laranjeiro
---
Changes in v2:
- add ixgbe and i40e.
---
doc/guides/nics/features/default.ini | 1 +
doc/guides/nics/features/i40e.ini| 1 +
doc/guides/nics/features/ixgbe.ini | 1 +
doc/guides/nics/features/mlx5.ini| 1 +
doc/guides/nics/mlx5.rst | 1
On 2/9/2017 8:26 AM, Mcnamara, John wrote:
>
>
>> -Original Message-
>> From: Wu, Jingjing
>> Sent: Thursday, February 9, 2017 6:42 AM
>> To: dev@dpdk.org
>> Cc: Wu, Jingjing ; Mcnamara, John
>>
>> Subject: [PATCH v2] doc: add known issue for i40e PMD
>>
>> Rx statistic is incorrect when
Hi Slawomir,
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Slawomir
> Mrozowicz
> Sent: Friday, February 03, 2017 3:55 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; Mrozowicz, SlawomirX
> Subject: [dpdk-dev] [PATCH v2] cryptodev: fix segmentation fault
>
> Thi
On 2/7/2017 3:25 PM, Wu, Jingjing wrote:
>
>
>> -Original Message-
>> From: Sexton, Rory
>> Sent: Thursday, January 19, 2017 6:38 PM
>> To: Wu, Jingjing
>> Cc: dev@dpdk.org; Marjanovic, Nemanja
>> Subject: RE: [PATCH v2] net/i40e: set no drop for traffic class
>>
>> Perhaps the best sol
On 2/9/2017 1:27 AM, Wenzhuo Lu wrote:
> 1, adjust the framework of the ixgbe doc.
> 2, announce the MTU setting limitation.
>
> v2:
> - split the patch to 2 to add more fixes.
>
> Wenzhuo Lu (2):
> doc: fix framework of ixgbe doc
> doc: announce ixgbe MTU setting limitation
>
> doc/guides/
> Signed-off-by: Nelio Laranjeiro
Applied, thanks
Hi,
On Mon, 30 Jan 2017 10:54:08 +0100, Thomas Monjalon
wrote:
> It is fixing the introduction of the new flag PKT_RX_VLAN_STRIPPED.
>
> Fixes: b37b528d957c ("mbuf: add new Rx flags for stripped VLAN")
>
> This patch is applying the flag to the software emulation case
> (currently only for virt
v2:
- Change svg files.
- Postpone instead of remove ABI changes in igb_uio.
Patch 1: howto doc of virtio_user for container networking.
Patch 2: howto doc of virtio_user as exceptional path.
Patch 3: postpone ABI changes in igb_uio
Signed-off-by: Jianfeng Tan
Jianfeng Tan (3):
doc: add g
Signed-off-by: Jianfeng Tan
---
.../howto/img/virtio_user_as_exceptional_path.svg | 386 +
doc/guides/howto/index.rst | 1 +
.../howto/virtio_user_as_exceptional_path.rst | 142
3 files changed, 529 insertions(+)
create mode 100644 doc
This ABI changes to remove iomem and ioport mapping in igb_uio. The
purpose of this changes was to fix a bug: when DPDK app crashes,
those devices by igb_uio are not stopped either DPDK PMD driver or
igb_uio driver.
Then it has been pointed out by Stephen Hemminger that it has
backward compatibili
Signed-off-by: Jianfeng Tan
---
.../use_models_for_running_dpdk_in_containers.svg | 574 ++
.../img/virtio_user_for_container_networking.svg | 638 +
doc/guides/howto/index.rst | 1 +
.../howto/virtio_user_for_container_networking.r
2017-02-02 12:05, Ido Barnea:
> From: Ido Barnea
>
> Signed-off-by: Ido Barnea
> ---
> drivers/net/virtio/virtio_ethdev.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/net/virtio/virtio_ethdev.c
> b/drivers/net/virtio/virtio_ethdev.c
> index d1ff234..1d572b5 100644
> --- a/
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, Konstantin
>
> >
> > The main changes are:
> > ...
> > - change port and nb_segs to 16 bits
>
> Not that I am completely against it,
> but changing nb_segs to 16 bits seems like an overkill to me.
> I think we can keep and extra 8bit
2017-02-08 14:56, Stephen Hemminger:
> I am trying to make DPDK more agnostic about bus type. The existing API still
> has hardwired into that ethernet devices are either PCI or not PCI (ie
> pci_dev == NULL).
> Jan, Jerin, and Shreyansh started the process but it hasn't gone far enough.
>
> It w
rte_eth_copy_pci_info no longer sets RTE_ETH_DEV_DETACHABLE
as per my previous patch, but it still does initialize
the dev_flags. In a number of cases, a later assignment to
dev_flags to set RTE_ETH_DEV_DETACHABLE was overwriting
the previously initialized value.
Pointed out by Michael Larsen.
Si
2016-11-06 22:09, Thomas Monjalon:
> 2016-09-29 18:34, Thomas Monjalon:
> > 2016-09-30 02:54, Nikhil Rao:
> > > The original code used movl instead of xchgl, this caused
> > > rte_atomic64_cmpset to use ebx as the lower dword of the source
> > > to cmpxchg8b instead of the lower dword of function a
> -Original Message-
> From: Morten Brørup [mailto:m...@smartsharesystems.com]
> Sent: Thursday, February 9, 2017 4:20 PM
> To: Ananyev, Konstantin
> Cc: Olivier Matz ; dev@dpdk.org
> Subject: RE: [dpdk-dev] [RFC 0/8] mbuf: structure reorganization
>
> > From: dev [mailto:dev-boun...@dp
rte_eal_pci_scan can be called repeatedly to re-scan the PCI
bus. If a device was removed from the system, the associated
driver will automatically be unloaded.
Signed-off-by: Ben Walker
---
Only code style changes compared to last submission. It is also rebased
onto the latest from master, whic
The user needs to register drivers before scanning, so
it makes the most sense to put the registration
functions above the scan function in the header file.
Signed-off-by: Ben Walker
---
Only rebased onto latest master with the rest of this series.
No changes.
lib/librte_eal/common/include/rte
There are now two functions - rte_eal_pci_attach_driver and
rte_eal_pci_detach_driver - that dynamically attempt to attach
and detach drivers from PCI devices. These only control
whether a registered PCI driver is loaded or not - they are
independent of whether the PCI device exists on the system.
On 9/2/2017 2:20 PM, Ferruh Yigit wrote:
On 2/6/2017 8:08 AM, David Hunt wrote:
Given that the packet distributor library improvements (1) will
not be in 17.02, I plan on doing some consolidation of the
API for burst operation for 17.05, merging the two api's into
one, with options for single o
Hi Fan,
> -Original Message-
> From: Zhang, Roy Fan
> Sent: Thursday, February 09, 2017 1:44 PM
> To: dev@dpdk.org
> Cc: De Lara Guarch, Pablo
> Subject: [PATCH] crypto/scheduler: fix initialization
>
> Fixes the wrong slave initialization issue on start-up
>
> Fixes: 100e4f7("add round-
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Tuesday, February 7, 2017 10:50 PM
> To: dev@dpdk.org
> Cc: De Lara Guarch, Pablo ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] crypto: fix incorrect key setting
>
> When ciphering and au
> > When a secondary process wants access to the VFIO container file descriptor,
> > the primary process calls vfio_get_container_fd() which always opens an
> > entirely new file descriptor on /dev/vfio/vfio.
> > However, once the file descriptor has been passed to the subprocess, it is
> > effecti
On 2/9/2017 4:40 PM, Ben Walker wrote:
> rte_eth_copy_pci_info no longer sets RTE_ETH_DEV_DETACHABLE
> as per my previous patch, but it still does initialize
> the dev_flags. In a number of cases, a later assignment to
> dev_flags to set RTE_ETH_DEV_DETACHABLE was overwriting
> the previously initi
On 2/9/2017 4:06 PM, Jianfeng Tan wrote:
> This ABI changes to remove iomem and ioport mapping in igb_uio. The
> purpose of this changes was to fix a bug: when DPDK app crashes,
> those devices by igb_uio are not stopped either DPDK PMD driver or
> igb_uio driver.
>
> Then it has been pointed out
> -Original Message-
> From: Ferruh Yigit [mailto:ferruh.yi...@intel.com]
> Sent: Thursday, February 09, 2017 2:16 AM
> To: John Daley (johndale)
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: Re: [PATCH v2] net/enic: fix hardcoding of some flow director masks
>
> On 2/9/2017 12:40 AM,
> -Original Message-
> From: Ferruh Yigit [mailto:ferruh.yi...@intel.com]
> Sent: Thursday, February 09, 2017 4:15 AM
> To: John Daley (johndale)
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v2] enic flow director fixes
>
> On 2/9/2017 12:40 AM, John Daley wrote:
> > This should replace
> >
1 - 100 of 163 matches
Mail list logo
