> -Original Message-
> From: Jerin Jacob Kollanukkaran
> Sent: Monday, July 15, 2019 2:35 PM
[...]
> Subject: RE: [dpdk-dev] [PATCH] vfio: fix interrupts race condition
>
> > > >
> > > > This is a real bug which should be fixed in this release.
> > > > As the patch is quite big and needs
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Monday, July 15, 2019 12:06 PM
> To: Zhang, Xiao
> Cc: dev@dpdk.org; Zhang, Qi Z ; Wang, Xiao W
> ; Xing, Beilei ; Lu, Wenzhuo
> ; Yang, Qiming ; Ananyev,
> Konstantin ; Wu, Jingjing
>
> Subject:
Hi, Stephen
>mlx5_get_ifname uses priv->ctx which is not a valid pointer in a secondary
>process.
1. Sorry, mlx5_priv structure does not contain ctx field. Do you mean priv->sh
(shared context) ?
This one is allocated with rte_zmalloc(), that supposes shared memory
allocation and shared
cont
This patch introduces an internal representation of the PCI device
which will be used to store the internal information that don't have
to be exposed, e.g. the VFIO region sizes/offsets.
In this patch, the internal structure is simply a wrapper of the
rte_pci_device structure. More fields will be
This patch adds a helper for reading string from sysfs.
Signed-off-by: Cunming Liang
Signed-off-by: Tiwei Bie
---
lib/librte_eal/common/eal_filesystem.h | 10 ++
lib/librte_eal/freebsd/eal/eal.c | 22 ++
lib/librte_eal/linux/eal/eal.c | 22 +
Hi everyone,
This is a draft implementation of the mdev (Mediated device [1])
support in DPDK PCI bus driver. Mdev is a way to virtualize devices
in Linux kernel. Based on the device-api (mdev_type/device_api),
there could be different types of mdev devices (e.g. vfio-pci).
In this RFCv2, the PCI
The value 40 used in VFIO_GET_REGION_ADDR() is a private value
(VFIO_PCI_OFFSET_SHIFT) defined in Linux kernel source [1]. It
is not part of VFIO API, and we should not depend on it.
[1]
https://github.com/torvalds/linux/blob/6fbc7275c7a9/drivers/vfio/pci/vfio_pci_private.h#L19
Signed-off-by: Ti
The MMIO regions may not be mmap-able for mediated PCI device.
In this case, the application should explicitly do read and write
to access these regions.
Signed-off-by: Tiwei Bie
---
drivers/bus/pci/bsd/pci.c | 22
drivers/bus/pci/linux/pci.c | 46 +
This patch adds the mdev (Mediated device) support in PCI bus
driver. With this patch, the PCI bus driver will be able to scan
and probe the mediated PCI devices (i.e. the Mediated devices
whose device API is "vfio-pci") in the system.
There are several things different between physical PCI device
Currently PKT_TX_IP_CKSUM is being set into mbuf->ol_flags
during fragmentation and reassemble operation implicitly.
Because of this, application is forced to use checksum offload
whether it is supported by platform or not.
Also documentation does not provide any expected value of ol_flags
in retu
Changes since v2: remove rte_experimental from c file
This patch series makes few enhancements w.r.t functionality and usabilty of
FSLMC bus.
1. Applications like OVS has mandated the use of hotplug of dpdk ports. The
fslmc bus shall support hotplug now to support dpaa2 ports with OVS.
2. DPAA
From: Shreyansh Jain
Recent (18.11+), devargs structure was changed and so was DPDK port
usage in applications like OVS. Applications are now allowed to
plug/unplug ports (eth) using APIs (hotplug) based on device
arguments.
This patch enables the plug/unplug function (which are dummy for
FSLMC)
From: Shreyansh Jain
In case an incorrect device is found, it was quiting further search
rather than skipping it.
Signed-off-by: Shreyansh Jain
---
drivers/bus/fslmc/fslmc_bus.c | 24 ++--
1 file changed, 14 insertions(+), 10 deletions(-)
diff --git a/drivers/bus/fslmc/fsl
From: Nipun Gupta
LS1088 platform CENA operation are causing issues
at high load. CINH (cache inhibited) mode is working
fine with minor performance impact.
This patch enables CINH mode selectively on LS1088 platform
Signed-off-by: Nipun Gupta
---
drivers/bus/fslmc/portal/dpaa2_hw_dpio.h
From: Shreyansh Jain
This patch adds following:
1. 'g_container' variable name is not right way to represent the
FSLMC container. Renaming it to fslmc_container.
2. dynamic selection of IOMMU mode based on run environment
Signed-off-by: Shreyansh Jain
---
drivers/bus/fslmc/fslmc_vfio.c | 38 ++
From: Sachin Saxena
This patch help in adding a routine to help memory map
the user provided memory via VFIO.
Signed-off-by: Sachin Saxena
---
drivers/bus/fslmc/fslmc_vfio.c | 39 +
drivers/bus/fslmc/fslmc_vfio.h | 5 ++-
drivers/bus/fslmc/rte_bus
On 13/07/2019 00:23, Stephen Hemminger wrote:
On Fri, 12 Jul 2019 10:06:51 +0100
David Hunt wrote:
Samples apps that depend on the power library should not build if
the power library is not present. So now the following apps will
check to see if it's built, else skip building.
* l3fwd-pwer
* v
When pattern has ETH, maybe it will contain
MAC and ethertype two kinds of lookup parameters,
so add more item number for memory malloc.
Signed-off-by: wei zhao
---
drivers/net/ice/ice_switch_filter.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ice/ice_switch_filter.c
b/dr
On 15-Jul-19 5:54 AM, Jerin Jacob Kollanukkaran wrote:
(also, i don't really like the name NO_PAGE_BOUND since in
memzone API there's a "bounded memzone" allocation API, and this
flag's name reads like objects would not be bounded by page size,
not that they won't cross page
boundary)
No strong
In the past, to distinguish whether the input set is outer or inner
by check the item appear once or twice. But this way doesn't work when user
don't configure the outer input set. This patch fix the issue.
Fixes: d76116a4678f ("net/ice: add generic flow API")
Cc: sta...@dpdk.org
Signed-off-by: Q
> -Original Message-
> From: Trybula, ArturX
> Sent: Monday, July 8, 2019 7:16 PM
> To: dev@dpdk.org; Trahe, Fiona ; shal...@marvell.com;
> Dybkowski, AdamX
> ; Trybula, ArturX ;
> akhil.go...@nxp.com
> Subject: [PATCH v8 7/7] app/test-compress-perf: 'magic numbers' removed
>
> This p
Thanks Arek. Response inline.
Sent from Workspace ONE Boxer
On 12-Jul-2019 6:45 PM, "Kusztal, ArkadiuszX"
wrote:
>
> In future we could unify tests even more (i.e. passing padding type as a
> param, or even information if it is encryption or signature generation) like
> Fan once done with bloc
Added telemetry to EAL long options so that when
--telemetry is passed as an EAL arg that there is
no unrecognized argument error message printed.
Fixes: 8877ac688b52 ("telemetry: introduce infrastructure")
Cc: ciara.po...@intel.com
Cc: sta...@dpdk.org
Signed-off-by: Sean Morrissey
---
lib/libr
+Harry Van Haaren
--
Intel Research and Development Ireland Limited
Registered in Ireland
Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
Registered Number: 308263
This e-mail and any attachments may contain confi
On 7/12/2019 5:29 PM, Vamsi Krishna Attunuru wrote:
>
>
>> -Original Message-
>> From: Ferruh Yigit
>> Sent: Friday, July 12, 2019 4:40 PM
>> To: Vamsi Krishna Attunuru ; dev@dpdk.org
>> Cc: olivier.m...@6wind.com; arybche...@solarflare.com; Kiran Kumar
>> Kokkilagadda
>> Subject: Re: [
From: Nipun Gupta
Sunil Kori is no longer with NXP
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
MAINTAINERS | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 4534e3e2d..d3e2b2c53 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1067,7
From: Sachin Saxena
Shreyansh is no longer with NXP.
Pankaj is no longer working on DPDK.
Signed-off-by: Sachin Saxena
Acked-by: Hemant Agrawal
---
MAINTAINERS | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index d3e2b2c53..a984ab194 1006
From: Rastislav Cernay
This patch adds timestamping support to nfb driver.
Signed-off-by: Rastislav Cernay
---
v2: rewrite timestamp enable to devargs
v3: check valid devargs
move timestamp part of doc to configuration
fix shared build
doc/guides/nics/nfb.rst | 20
This patch adds checking if device support ZUC
algorithms before running ZUC test cases.
It also removes unnecessary checks of digest
appended space and fixes some comments wording.
Signed-off-by: Damian Nowak
---
app/test/test_cryptodev.c | 28
1 file changed, 20 in
This patch updates the constant names and function names used
in code snippets in the Compression Device Library documentation.
Signed-off-by: Adam Dybkowski
---
doc/guides/prog_guide/compressdev.rst | 62 ++-
1 file changed, 33 insertions(+), 29 deletions(-)
diff --git
This patch introduces two new messages VHOST_USER_GET_INFLIGHT_FD
and VHOST_USER_SET_INFLIGHT_FD to support transferring a shared
buffer between qemu and backend. And the example code shows how
these APIs work.
JinYu (2):
vhost: support inflight share memory protocol feature
vhost: Add vhost-u
A vhost-user-blk example that support inflight feature. It uses the
new APIs that introduced in the first patch, so It can show how there
APIs work to support inflight feature.
Signed-off-by: JinYu
---
V1 - add the case.
V2 - add the rte_vhost prefix.
---
examples/vhost_blk/Makefile |
This patch introduces two new messages VHOST_USER_GET_INFLIGHT_FD
and VHOST_USER_SET_INFLIGHT_FD to support transferring a shared
buffer between qemu and backend.
Firstly, qemu uses VHOST_USER_GET_INFLIGHT_FD to get the
shared buffer from backend. Then qemu should send it back
through VHOST_USER_S
Hi Fan/Liron,
Could you please send the incremental patch ASAP. Otherwise I need to remove
this patch from the subtree.
I don't have environment to test this.
Regards,
Akhil
>
> Hi Fan,
>
> >
> > 08/07/2019 12:26, Thomas Monjalon:
> > > 07/07/2019 07:41, Liron Himi:
> > > > From: Akhil Goyal
> -Original Message-
> From: Artur Trybula
> Sent: Friday, July 12, 2019 4:13 PM
> To: dev@dpdk.org; fiona.tr...@intel.com; Shally Verma
> ; adamx.dybkow...@intel.com;
> arturx.tryb...@intel.com; akhil.go...@nxp.com
> Subject: [EXT] [PATCH v4 1/1] app/test-compress-perf: report header
>
> -Original Message-
> From: Ferruh Yigit
> Sent: Monday, July 15, 2019 4:57 PM
> To: Vamsi Krishna Attunuru ; dev@dpdk.org
> Cc: olivier.m...@6wind.com; arybche...@solarflare.com; Kiran Kumar
> Kokkilagadda
> Subject: Re: [EXT] Re: [PATCH v6 4/4] kernel/linux/kni: add IOVA support in
On 7/15/2019 1:03 PM, Rastislav Cernay wrote:
> From: Rastislav Cernay
>
> This patch adds timestamping support to nfb driver.
>
> Signed-off-by: Rastislav Cernay
Applied to dpdk-next-net/master, thanks.
>
> This patchset adds multiple cores feature to compression perf tool.
> All structures have been aligned and are consistent with crypto perf tool.
> All test cases have constructor, runner and destructor and can use more
> cores and compression devices at the same time.
>
> v8 changes:
> -
Some example applications have references to rte_eth_dev structure,
which is not a part of public API. These patches replace them with
calls to API functions.
Marcin Zapolski (3):
examples/l3fwd*: remove references to rte_eth_devices
examples/ip_fragmentation: remove usage of internal struct
Modify ip_fragmentation example app to use rte_eth_info_get instead
of global rte_eth_devices structure.
Signed-off-by: Marcin Zapolski
---
examples/ip_fragmentation/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/examples/ip_fragmentation/main.c b/examples/ip_fragment
Modify ipsec-secgw example app to use rte_eth_dev_info_get instead of
rte_eth_dev.
Signed-off-by: Marcin Zapolski
---
examples/ipsec-secgw/ipsec.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/examples/ipsec-secgw/ipsec.c b/examples/ipsec-secgw/ipsec.c
index 7b8533
Modify l3fwd and related example apps to use locally defined port_conf
instead of global rte_eth_devices which is not a part of public API.
Signed-off-by: Marcin Zapolski
---
examples/l3fwd-acl/main.c | 6 +-
examples/l3fwd-power/main.c | 6 +-
e
Hi ,
I am trying to verify Virtio Crypto use case on DPDK 19.05 . While running
DPDK test app "cryptodev_virtio_autotest " in the VM ,I could see control
connection created on the vhost backend(vhost-crypto) but functionality does
not work/test case fails.
Below is the configuration used in
On ESXi setups when we have SR-IOV and E-Switch enabled there is the
problem to receive VLAN traffic on VF interfaces. The NIC driver
in ESXi hypervisor does not setup E-Switch vport setting correctly
and VLAN traffic targeted to VF is dropped.
The patch provides the temporary workaround - if the
> -Original Message-
> From: Dybkowski, AdamX
> Sent: Monday, July 15, 2019 1:20 PM
> To: dev@dpdk.org; Trahe, Fiona
> Cc: Dybkowski, AdamX
> Subject: [PATCH] doc: update compression device docs
>
> This patch updates the constant names and function names used
> in code snippets in th
This patch removes the existing Tx datapath code
as preparation step before introducing the new
implementation. The following entities are being
removed:
- obsolete devargs
- tx_burst() routines
- related PRM definitions
- SQ configuration code
- Tx routine selection code
- incompatible Tx complet
This patchset introduces the new implementation of tx_burst
routine of mlx5 PMD. The existing implementation is based on the
several branches of tx_burst routines optimizied for most
common sets of Tx hardware offload, supported by Mellanox NICs.
It was not very easy to update, support and develop
This patch introduces new mlx5 PMD devarg options:
- txq_inline_min - specifies minimal amount of data to be inlined into
WQE during Tx operations. NICs may require this minimal data amount
to operate correctly. The exact value may depend on NIC operation mode,
requested offloads, etc.
- tx
This patch updates Tx datapath definitions, mostly hardware related.
The Tx descriptor structures are redefined with required fields,
size definitions are renamed to reflect the meanings in more
appropriate way. This is a preparation step before introducing
the new Tx datapath implementation.
Sign
This patch updates the Tx datapath control and configuration
structures and code for mananging Tx datapath settings.
Signed-off-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_rxtx.c | 4 +-
drivers/net/mlx5/mlx5_rxtx.h | 55 +
drivers/net/mlx5/mlx5_txq.c | 182 +++
Mellanox NICs support the wide set of Tx offloads. The supported
offloads are reported by the mlx5 PMD in rte_eth_dev_info tx_offload_capa
field. An application may choose any combination of supported offloads
and configure the device appropriately. Some of Tx offloads may be
not requested by appli
This patch adds the implementation of tx_burst routine template.
The template supports all Tx offloads and multiple optimized
tx_burst routines can be generated by compiler from this one.
Signed-off-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_rxtx.c | 2897
Tx data packet data may be inlined into transmit descriptor.
At some circumstances ConnectX NICs may require data to be
inlined for correct operation. The exact data amount may
depend on NIC operation mode, requested Tx offloads,
E-Switch configuration, etc.
The number of data bytes to inline may
-Original Message-
From: Shally Verma [mailto:shal...@marvell.com]
Sent: Monday, July 15, 2019 14:47
To: Trybula, ArturX ; dev@dpdk.org; Trahe, Fiona
; Dybkowski, AdamX ;
akhil.go...@nxp.com
Subject: RE: [EXT] [PATCH v4 1/1] app/test-compress-perf: report header
improvement
> -Or
On 7/12/2019 11:01 AM, kka...@marvell.com wrote:
> From: Krzysztof Kanas
>
> Print system error to make easier diagnosis of errors with af_packet.
>
> Signed-off-by: Krzysztof Kanas
Reviewed-by: Ferruh Yigit
Applied to dpdk-next-net/master, thanks.
(Whitespace updated, to use a tab instead
Current implementation only outputs 3 numbers for busyness, 0, 50 and 100.
Fix this so that the 50% is replaced by a curve, more meaningful.
This can be replaced in each use case by a suitable calculation for
that use case.
Fixes: 609e79841fcf ("examples/l3fwd-power: add telemetry mode")
Signed-o
> -Original Message-
> From: Nowak, DamianX
> Sent: Monday, July 15, 2019 1:15 PM
> To: dev@dpdk.org
> Cc: akhil.go...@nxp.com; Trahe, Fiona ; Kusztal,
> ArkadiuszX
> ; Nowak, DamianX
> Subject: [PATCH] test/crypto: add capability check for ZUC test cases
>
> This patch adds checking
Subject: [EXT] RE: [PATCH] config: fix meson build failure on
> > ThunderX2
> > > > Since gcc-8.3(I tried 8.3 and 9.1), the meson build failed on ThunderX2.
> > > > It got the following compiling errors:
> > > > /usr/lib/gcc/aarch64-linux-gnu/9/include/arm_neon.h:26493:1: error:
> > > > inlining
> > > +
> > > +IOVA Mode is selected by considering what the current usable
> > > +Devices on the system requires and/or supports.
> > > +
> > > +Below is the 2-step heuristic for this choice.
> > > +
> > > +For the first step, EAL asks each bus its requirement in terms of
> > > +IOVA mode and deci
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Monday, July 15, 2019 7:55 PM
> To: Jerin Jacob Kollanukkaran ; Gavin Hu (Arm
> Technology China) ; dev@dpdk.org
> Cc: nd ; tho...@monjalon.net; Pavan Nikhilesh Bhagavatula
> ; sta...@dpdk.org; Honnappa Nagarahalli
> ; nd
> Subje
On Mon, Jul 15, 2019 at 11:56:22AM +0200, Marcin Zapolski wrote:
> Modify ip_fragmentation example app to use rte_eth_info_get instead
> of global rte_eth_devices structure.
>
> Signed-off-by: Marcin Zapolski
> ---
> examples/ip_fragmentation/main.c | 2 +-
> 1 file changed, 1 insertion(+), 1 de
On Mon, Jul 15, 2019 at 03:39:56PM +0100, Bruce Richardson wrote:
> On Mon, Jul 15, 2019 at 11:56:22AM +0200, Marcin Zapolski wrote:
> > Modify ip_fragmentation example app to use rte_eth_info_get instead
> > of global rte_eth_devices structure.
> >
> > Signed-off-by: Marcin Zapolski
> > ---
> >
On Mon, Jul 15, 2019 at 11:56:23AM +0200, Marcin Zapolski wrote:
> Modify ipsec-secgw example app to use rte_eth_dev_info_get instead of
> rte_eth_dev.
>
> Signed-off-by: Marcin Zapolski
> ---
Reword title to be "fix use of ..." and include:
Fixes: a4677f78368b ("examples/ipsec-secgw: add targe
On 7/12/2019 9:32 AM, A.McLoughlin wrote:
> The 'Accumulated forward statistics for all ports' incorrectly displayed
> double the actual value for 'total_tx_dropped'. This was because 2
> lines in the same function both incremented total_tx_dropped every time
> a packet was dropped. I removed one
On Mon, Jul 15, 2019 at 11:56:21AM +0200, Marcin Zapolski wrote:
> Modify l3fwd and related example apps to use locally defined port_conf
> instead of global rte_eth_devices which is not a part of public API.
>
> Signed-off-by: Marcin Zapolski
As with the other patches, please treat this as a bu
15/07/2019 16:26, Jerin Jacob Kollanukkaran:
> > > > +
> > > > +IOVA Mode is selected by considering what the current usable
> > > > +Devices on the system requires and/or supports.
> > > > +
> > > > +Below is the 2-step heuristic for this choice.
> > > > +
> > > > +For the first step, EAL asks eac
> -Original Message-
> From: Thomas Monjalon
> Sent: Monday, July 15, 2019 8:34 PM
> To: Jerin Jacob Kollanukkaran
> Cc: Burakov, Anatoly ; David Marchand
> ; dev@dpdk.org; John McNamara
> ; Marko Kovacevic
> ; Igor Russkikh
> ; Pavel Belous ;
> Ajit Khaparde ; Somnath Kotur
> ; Wenzhuo L
If the timer subsystem is not initialized before rte_timer_manage (for
example) is invoked, a pointer to a shared hugepage memory region will
still be null and dereferenced when it is checked for validity; handle
this case.
Fixes: c0749f7096c7 ("timer: allow management in shared memory")
Cc: sta..
A crash occurs in the examples/performance-thread/l3fwd-thread when it calls
into rte_timer_manage() from the LThread library it utilizes. This happens
because the application omitted a call to rte_timer_subsystem_init(), which
leaves a pointer set to null in the timer library. This pointer is
de
The timer subsystem should be initialized in the l3fwd-thread app before
the L-thread subsystem can be used.
Fixes: d48415e1fee3 ("examples/performance-thread: add l3fwd-thread app")
Cc: sta...@dpdk.org
Cc: ian.be...@intel.com
Signed-off-by: Erik Gabriel Carrillo
---
examples/performance-thread
On Mon, 15 Jul 2019 10:39:31 -0500
Erik Gabriel Carrillo wrote:
> If the timer subsystem is not initialized before rte_timer_manage (for
> example) is invoked, a pointer to a shared hugepage memory region will
> still be null and dereferenced when it is checked for validity; handle
> this case.
>
A rough patch for the approach mentioned earlier. It is only for
discussion.
http://mails.dpdk.org/archives/dev/2019-July/138113.html
To try this out, first revert the following then apply.
commit 89aac60e0be9 ("vfio: fix interrupts race condition")
Signed-off-by: Hyong Youb Kim
---
.../common/
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger
> Sent: Monday, July 15, 2019 10:49 AM
> To: Carrillo, Erik G
> Cc: tho...@monjalon.net; dev@dpdk.org; sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/2] timer: fix null pointer dereference
>
> On Mon, 15 Jul 2019 10:39:31 -0
> >
> > > > Jerin, any reason for not enabling LSE on ThunderX2?
> > >
> > > LSE is enabled when -mcpu=thunderx2t99 selected.
> > >
> > > $ aarch64-linux-gnu-gcc -mcpu=thunderx2t99 -dM -E - > > grep __ARM_FEATURE_ATOMICS #define __ARM_FEATURE_ATOMICS 1
> >
> > This does not work with older GCC ve
15/07/2019 17:35, Jerin Jacob Kollanukkaran:
> From: Thomas Monjalon
> > 15/07/2019 16:26, Jerin Jacob Kollanukkaran:
> > > > > Is there any specific reason why we always prefer PA if physical
> > > > > addresses are available? Since we're already assuming that all
> > > > > devices support PA and
On 7/11/2019 9:03 PM, Stephen Hemminger wrote:
> The rte_vdev_driver is declared twice.
> The first one is not necessary.
>
> Signed-off-by: Stephen Hemminger
Acked-by: Ferruh Yigit
On 7/11/2019 9:03 PM, Stephen Hemminger wrote:
> Many drivers declare the local rte_vdev_driver structure
> multiple times. This is unnecessary.
>
> Surprised it is so common, I guess everyone copies the same code.
>
> Stephen Hemminger (9):
> net/tap: remove redundant declarations
> net/af_
> -Original Message-
> From: Thomas Monjalon
> Sent: Monday, July 15, 2019 9:36 PM
> To: Jerin Jacob Kollanukkaran
> Cc: Burakov, Anatoly ; David Marchand
> ; dev@dpdk.org; John McNamara
> ; Marko Kovacevic
> ; Igor Russkikh
> ; Pavel Belous ;
> Ajit Khaparde ; Somnath Kotur
> ; Wenzhuo L
> -Original Message-
> From: Hyong Youb Kim
> Sent: Monday, July 15, 2019 9:28 PM
> To: David Marchand ; Thomas Monjalon
> ; Jerin Jacob Kollanukkaran ;
> Ferruh Yigit
> Cc: dev@dpdk.org; John Daley ; Hyong Youb Kim
>
> Subject: [EXT] [RFC PATCH] vfio: avoid re-installing irq handler
>
On 7/15/2019 5:18 PM, Ferruh Yigit wrote:
> On 7/11/2019 9:03 PM, Stephen Hemminger wrote:
>> Many drivers declare the local rte_vdev_driver structure
>> multiple times. This is unnecessary.
>>
>> Surprised it is so common, I guess everyone copies the same code.
It seems it is common because a pa
12/07/2019 19:03, Ferruh Yigit:
> On 6/24/2019 5:47 PM, Stephen Hemminger wrote:
> > While testing KNI with netvsc, saw lots of places more code
> > could be safely removed from KNI kernel driver.
> >
> > v6 - more updates to documentation
> > v5 - add minimal ethtool, fix checkpath author complai
When I clone the repositories at http://dpdk/org/git/dpdk or
git://dpdk.org/dpdk I'm only seeing commits up through June 24 in the
log, but the web interface to the development repo at
https://git.dpdk.org/dpdk/ shows commits through today. I see the same
behavior on when accessed from the off
https://bugs.dpdk.org/show_bug.cgi?id=321
Bug ID: 321
Summary: BPF: Errors compiling example BPF programs
Product: DPDK
Version: 19.08
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
P
Hi David,
I believe that this is just confusion because of the way git tracks commit
date. Even if last commit is" June-24", this is the date of when the commit was
created, not when it was merged. You can see July dates in history.
This looks fine to me basically and latest is on master.
Chee
> -Original Message-
> From: Carrillo, Erik G
> Sent: Monday, July 15, 2019 11:04 AM
> To: Stephen Hemminger
> Cc: tho...@monjalon.net; dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH 1/2] timer: fix null pointer dereference
>
> Hi Stephen,
>
> > -Original Message-
Since the "eal-intr-thread" was introduced in versions >18.05 there seems to be
a memory leak and processor utilization issue with the i40e driver.
Just running testpmd and watching the threads in top shows the issue:
2075 root 99 0 4110M 3063M CPU0 0 0:20 99.87% testpmd{eal-intr-thread}
11/07/2019 09:46, Ferruh Yigit:
> On 7/10/2019 9:09 PM, Ferruh Yigit wrote:
> > On 6/25/2019 4:04 PM, Yangchao Zhou wrote:
> >> va2pa depends on the physical address and virtual address offset of
> >> current mbuf. It may get the wrong physical address of next mbuf which
> >> allocated in another h
15/07/2019 10:44, Hemant Agrawal:
> Nipun Gupta (1):
> bus/fslmc: use cinh read for eqcr ci on ls1088 platform
>
> Sachin Saxena (1):
> mempool/dpaa2: vfio dmamap for user allocated memory
>
> Shreyansh Jain (3):
> bus/fslmc: support device iteration
> bus/fslmc: enhance error handling fo
15/07/2019 10:17, Sunil Kumar Kori:
> Current patch is part of following patchset
> http://patches.dpdk.org/patch/53475/
> which is missed to be merged into upstream because of accepted version of this
> patch does not contain this change.
>
> Accepted patchset is given below:
> http://patches.dpd
15/07/2019 13:38, Hemant Agrawal:
> From: Sachin Saxena
>
> Shreyansh is no longer with NXP.
> Pankaj is no longer working on DPDK.
>
> Signed-off-by: Sachin Saxena
> Acked-by: Hemant Agrawal
Series applied
11/07/2019 10:16, Jerin Jacob Kollanukkaran:
> From: Stephen Hemminger
> > Putting color escape sequences in the log look pretty for the developer but
> > fails in real world DPDK usage. A real application will put DPDK log to
> > syslog,
> > and syslog does not handle escape sequences.
> >
> >
12/07/2019 13:20, Jerin Jacob Kollanukkaran:
> From: Gavin Hu
> > Since gcc-8.3(I tried 8.3 and 9.1), the meson build failed on ThunderX2.
> > It got the following compiling errors:
> > /usr/lib/gcc/aarch64-linux-gnu/9/include/arm_neon.h:26493:1: error:
> > inlining failed in call to always_inline
If DPDK is built as a shared library, then any application linked
with rte.app.mk will not find any PCI devices. When the application
is started no ethernet devices are found.
This is because the link order of libraries on the command line matters.
And PCI is before EAL. That causes there to be no
> -Original Message-
> From: Singh, Jasvinder
> Sent: Friday, July 12, 2019 11:57 AM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian ; Tovar, AbrahamX
> ; Krakowiak, LukaszX
>
> Subject: [PATCH v4 01/11] sched: remove wrr from strict priority tc queues
>
> All higher priority traffic cla
On Mon, 15 Jul 2019 16:41:36 -0700
Stephen Hemminger wrote:
> If DPDK is built as a shared library, then any application linked
> with rte.app.mk will not find any PCI devices. When the application
> is started no ethernet devices are found.
>
> This is because the link order of libraries on the
On Mon, 15 Jul 2019 16:41:36 -0700
Stephen Hemminger wrote:
> If DPDK is built as a shared library, then any application linked
> with rte.app.mk will not find any PCI devices. When the application
> is started no ethernet devices are found.
>
> This is because the link order of libraries on the
In ppc64le, expanding DMA areas always fail because we cannot remove
a DMA window. As a result, we cannot allocate more than one memseg in
ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap all
the mapped DMA before removing the window. This patch fixes this
incorrect behavior.
I als
> -Original Message-
> From: Singh, Jasvinder
> Sent: Friday, July 12, 2019 11:57 AM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian ; Tovar, AbrahamX
> ; Krakowiak, LukaszX
>
> Subject: [PATCH v4 02/11] sched: add config flexibility to tc queue sizes
>
> Add support for zero queue sizes
> -Original Message-
> From: Singh, Jasvinder
> Sent: Friday, July 12, 2019 11:57 AM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian ; Tovar, AbrahamX
> ; Krakowiak, LukaszX
>
> Subject: [PATCH v4 06/11] sched: improve doxygen comments
>
> Improve doxygen comments.
>
> Signed-off-by: Ja
> -Original Message-
> From: Singh, Jasvinder
> Sent: Friday, July 12, 2019 11:57 AM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian ; Tovar, AbrahamX
> ; Krakowiak, LukaszX
>
> Subject: [PATCH v4 06/11] sched: improve doxygen comments
>
> Improve doxygen comments.
>
> Signed-off-by: Ja
1 - 100 of 116 matches
Mail list logo
