06/06/2017 08:36, Jerin Jacob:
> Add a hook in generic rte.sdkbuild.mk file
> to include exec-env specific targets.
>
> Signed-off-by: Jerin Jacob
> ---
> Useful in integrating some custom targets in nonstandard execution
> environments.
> For example, a bare-metal-simulator exec execution envir
05/06/2017 23:43, Gabriel Carrillo:
> Commit aafaea3d3b70 ("devtools: add tags
> and cscope index generation") introduced
> new make targets. This change updates the
> help target output to reflect the additions.
>
> Signed-off-by: Gabriel Carrillo
> ---
> doc/build-sdk-quick.txt | 3 +++
> 1 fi
Add a hook in generic rte.sdkbuild.mk file
to include exec-env specific targets.
Signed-off-by: Jerin Jacob
---
Useful in integrating some custom targets in nonstandard execution environments.
For example, a bare-metal-simulator exec execution environment may need
a target to run the dpdk applica
Since libnuma is added as a general dependency for EAL,
it is safe to enable LIBRTE_VHOST_NUMA by default.
Signed-off-by: Ilya Maximets
---
config/common_base | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/config/common_base b/config/common_base
index c858769..db4cc1c 100644
Version 3:
* Implemented hybrid schema for allocation.
* Fixed not needed mempolicy change while remapping. (orig = 0)
* Added patch to enable VHOST_NUMA by default.
Version 2:
* rebased (fuzz in Makefile)
Ilya Maximets (2):
mem: balanced allocation of hugepages
Currently EAL allocates hugepages one by one not paying attention
from which NUMA node allocation was done.
Such behaviour leads to allocation failure if number of available
hugepages for application limited by cgroups or hugetlbfs and
memory requested not only from the first socket.
Example:
re-bind PCI function to igb_uio
after process termination
tested adapters: Intel 10G, Intel 40G
known issues: ENA adapters on AWS I3 instances fail to reset PCI function
Signed-off-by: Gregory Etelson
---
lib/librte_eal/linuxapp/igb_uio/igb_uio.c | 12
1 files changed, 12 inser
Hi Konstantin:
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Monday, June 5, 2017 7:05 AM
> To: Zhang, Qi Z ; Zhang, Helin
>
> Cc: Lu, Wenzhuo ; Kinsella, Ray
> ; dev@dpdk.org
> Subject: RE: [PATCH v2] net/ixgbe: enable PTYPE offload for x86 vector PMD
>
> Hi Qi,
>
> > -O
CRC bytes should be excluded, so rx/tx bytes of VF stats is aligned
with PF stats.
Fixes: 9aace75fc82e ("i40e: fix statistics")
Cc: sta...@dpdk.org
Signed-off-by: Qi Zhang
---
drivers/net/i40e/i40e_ethdev.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.
-Original Message-
> Date: Tue, 6 Jun 2017 08:37:22 +0530
> From: Jerin Jacob
> To: Santosh Shukla
> CC: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] event/octeontx: Fix err msg in
> mbox_wait_response
> User-Agent: Mutt/1.8.3 (2017-05-23)
>
> -Original Message-
> > Date: Mon,
-Original Message-
> Date: Fri, 2 Jun 2017 13:14:39 +
> From: "Eads, Gage"
> To: Jerin Jacob
> Subject: RE: [dpdk-dev] [PATCH] event/octeontx: add driver name in info get
>
>
>
> > -Original Message-
> > From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com]
> > Sent:
-Original Message-
> Date: Thu, 1 Jun 2017 16:45:54 +0100
> From: Harry van Haaren
> To: dev@dpdk.org
> CC: jerin.ja...@caviumnetworks.com, gage.e...@intel.com, Harry van Haaren
>
> Subject: [PATCH v2] event/sw: fix credit tracking in port dequeue
> X-Mailer: git-send-email 2.7.4
>
> Si
-Original Message-
> Date: Mon, 29 May 2017 12:59:40 +0530
> From: Santosh Shukla
> To: jerin.ja...@caviumnetworks.com
> Cc: dev@dpdk.org, Santosh Shukla
> Subject: [PATCH] event/octeontx: Fix err msg in mbox_wait_response
> X-Mailer: git-send-email 2.11.0
>
> (Fixes: 6da9d2457 event/oct
-Original Message-
> Date: Mon, 5 Jun 2017 14:55:55 +0200
> From: Gaëtan Rivet
> To: Jerin Jacob
> Cc: dev@dpdk.org, bruce.richard...@intel.com, harry.van.haa...@intel.com,
> hemant.agra...@nxp.com, gage.e...@intel.com, nipun.gu...@nxp.com
> Subject: Re: [dpdk-dev] [PATCH] eventdev: remo
On 5 June 2017 at 16:58, Jerin Jacob wrote:
> CC: Jianbo Liu
> Signed-off-by: Jerin Jacob
> ---
> v2:
> - Removed YEILD instruction comment, as it is an implementation
> specific(Jianbo)
> ---
> lib/librte_eal/common/include/arch/arm/rte_pause.h | 4 ++
> .../common/include/arch/arm/rte_pause
> -Original Message-
> From: Allain Legacy [mailto:allain.leg...@windriver.com]
> Sent: Monday, June 5, 2017 4:49 PM
> To: Ananyev, Konstantin
> Cc: dev@dpdk.org; Osman, Dahir (Wind River)
> Subject: [PATCH v2] ip_frag: free mbufs on reassembly table destroy
>
> From: Dahir Osman
>
>
Commit aafaea3d3b70 ("devtools: add tags
and cscope index generation") introduced
new make targets. This change updates the
help target output to reflect the additions.
Signed-off-by: Gabriel Carrillo
---
doc/build-sdk-quick.txt | 3 +++
1 file changed, 3 insertions(+)
diff --git a/doc/build-sd
12/05/2017 13:04, Burakov, Anatoly:
> > From: Alejandro Lucero [mailto:alejandro.luc...@netronome.com]
> >
> > Checking against VFIO_MAX_GROUPS goes beyond the maximum array
> > index which should be (VFIO_MAX_GROUPS - 1).
> >
> > Fixes: 94c0776b1bad("support hotplug")
> > Coverity issue: 144555
22/05/2017 10:18, Olivier Matz:
> On Sat, 20 May 2017 15:12:37 +0200, David Marchand
> wrote:
> > vfio is the kernel framework used by the vfio-pci kernel driver.
> > DPDK drivers do not rely solely on vfio, but rather on vfio-pci to gain
> > access to pci resources.
> >
> > Fixes: 0880c40113ef
There is no need for initializing the complete
packet buffer with zero as the packet data area will be
overwritten by the NIC Rx HW anyway.
The testpmd configures the packet mempool
with around 180k buffers with
2176B size. In existing scheme, the init routine
needs to memset around ~370MB vs the
12/05/2017 10:42, Bruce Richardson:
> On Thu, May 11, 2017 at 11:03:43PM -0700, Tonghao Zhang wrote:
> > If the socket_id is invalid (e.g. -2, -3), the
> > memzone_reserve_aligned_thread_unsafe should return the
> > EINVAL and not ENOMEM. To avoid it, we should check the
> > socket_id before callin
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Saturday, June 3, 2017 3:00 AM
> To: Verkamp, Daniel ; dev@dpdk.org
> Cc: Richardson, Bruce
> Subject: RE: [dpdk-dev] [PATCH v2] ring: use aligned memzone allocation
>
>
>
> >
> > The PROD/CONS_ALIGN values on x86-64 are set to 2
19/05/2017 12:57, Bruce Richardson:
> On Fri, May 19, 2017 at 08:13:48AM +0800, Tiwei Bie wrote:
> > Signed-off-by: Tiwei Bie
> > ---
> > examples/exception_path/Makefile | 9 -
> > examples/exception_path/main.c | 28 +++-
> > 2 files changed, 27 insertions(+),
25/05/2017 17:46, Andrew Rybchenko:
> Fix broken build with -Werror=unused-but-set-variable
>
> Signed-off-by: Andrew Rybchenko
Applied, thanks
From: Dahir Osman
The rte_ip_frag_table_destroy procedure simply releases the memory for the
table without freeing the packet buffers that may be referenced in the hash
table for in-flight or incomplete packet reassembly operations. To prevent
leaked mbufs go through the list of fragments and fr
In virtio_read_caps, rte_pci_read_config returns the number of bytes
read from PCI config or <0 on error.
If less than the expected number of bytes are read then log the
failure and return rather than carrying on with garbage.
Signed-off-by: Brian Russell
---
drivers/net/virtio/virtio_pci.c | 12
On Mon, 5 Jun 2017 01:09:19 +0200
Gaëtan Rivet wrote:
> On Thu, Jun 01, 2017 at 11:00:10AM -0700, Stephen Hemminger wrote:
> > On Thu, 1 Jun 2017 16:12:41 +0200
> > Gaëtan Rivet wrote:
> >
> > > On Wed, May 31, 2017 at 08:15:26AM -0700, Stephen Hemminger wrote:
> > > > On Mon, 29 May 2017 1
> -Original Message-
> From: Ananyev, Konstantin [mailto:konstantin.anan...@intel.com]
> Sent: Sunday, June 04, 2017 12:21 PM
<..>
> >
> > +/* delete fragmentation table */
> > +void
> > +rte_ip_frag_table_destroy(struct rte_ip_frag_tbl *tbl)
> > +{
> > + uint32_t i;
> > +
> > + for (i
31/05/2017 18:43, Ferruh Yigit:
> On 5/31/2017 4:40 PM, Lee Roberts wrote:
> > Fix kni compilation under RHEL 7.4.
> >
> > Signed-off-by: Lee Roberts
>
> Not tested on RHEL 7.4, but patch looks good to me except following
> nit-picking:
> - There should not be "." at the end of the patch title
>
11/05/2017 14:53, Remy Horton:
> Fixes memory access errors detected by Coverity.
> All cases are the maximum permissable value causing an
> out-by-one overrun.
>
> Coverity issue 143433: Memory - illigal access
> Coverity issue 143434: Memory - illigal access
> Coverity issue 143460: Memory - cor
02/06/2017 11:30, Bruce Richardson:
> On Fri, Jun 02, 2017 at 05:07:46AM +, Sangjin Han wrote:
> > When rte_lpm.h is used on x86, -O0 option (no optimization at all)
> > given to gcc causes a compile error like this:
> >
> > error: the last argument must be an 8-bit immediate
> >i24 = _mm_
02/06/2017 13:20, Jerin Jacob:
> This fixes compiler warnings with GCC 7.1.1
Series applied, thanks
Hope many of us will work with GCC 7.1 now.
Hi Jerin,
On Thu, Jun 01, 2017 at 10:11:46PM +0530, Jerin Jacob wrote:
> Remove the PCI dependency from generic data structures
> and moved the PCI specific code to rte_event_pmd_pci*
>
Thanks for working on this.
Do you plan on removing rte_pci.h in rte_eventdev_pmd.h? Do you think it
would be
Remove rte_pause() definition from rte_common.h and
switchover to architecture specific rte_pause.h
Signed-off-by: Jerin Jacob
---
examples/distributor/main.c | 1 +
examples/l2fwd-jobstats/main.c | 1 +
examples/performance-thread/l3fwd-t
The patch does not provide any functional change for ppc64
with respect to existing rte_pause() definition.
CC: Chao Zhu
Signed-off-by: Jerin Jacob
Acked-by: Chao Zhu
---
.../common/include/arch/ppc_64/rte_pause.h | 51 ++
1 file changed, 51 insertions(+)
create mo
CC: Jianbo Liu
Signed-off-by: Jerin Jacob
---
v2:
- Removed YEILD instruction comment, as it is an implementation specific(Jianbo)
---
lib/librte_eal/common/include/arch/arm/rte_pause.h | 4 ++
.../common/include/arch/arm/rte_pause_64.h | 52 ++
2 files changed, 56 i
The patch does not provide any functional change for x86
with respect to existing rte_pause() definition.
CC: Bruce Richardson
CC: Konstantin Ananyev
Signed-off-by: Jerin Jacob
---
lib/librte_eal/common/include/arch/x86/rte_pause.h | 59 ++
1 file changed, 59 insertions(+)
The patch does not provide any functional change for ARM32
with respect to existing rte_pause() definition.
CC: Jan Viktorin
CC: Jianbo Liu
Signed-off-by: Jerin Jacob
Acked-by: Jan Viktorin
Acked-by: Jianbo Liu
---
lib/librte_eal/common/include/arch/arm/rte_pause.h | 46 +++
Each architecture may have different instructions for optimized
and power consumption aware rte_pause() implementation.
Signed-off-by: Jerin Jacob
---
doc/api/doxy-api-index.md | 3 +-
lib/librte_eal/common/Makefile| 2 +-
lib/librte_eal/common/inclu
On Sun, Jun 04, 2017 at 05:25:20AM +, Shachar Beiser wrote:
> remove/add blank line according to convensions
>
> Shachar Beiser (1):
> net/mlx5: implement drop action in hardware classifier
>
> drivers/net/mlx5/Makefile| 5 +
> drivers/net/mlx5/mlx5_flow.c | 17 +
>
> -Original Message-
> From: Wodkowski, PawelX
> Sent: Monday, June 05, 2017 1:54 PM
> To: Wodkowski, PawelX
> Subject: [PATCH] vfio: allow to map other memory regions
>
> Currently it is not possible to use memory that is not owned by DPDK to
> perform DMA. This scenarion might be used i
On Fri, 2 Jun 2017 13:37:58 -0700, Stephen Hemminger
wrote:
> On Fri, 2 Jun 2017 18:37:40 +
> "Dumitrescu, Cristian" wrote:
>
> > > Why? Do you need to host a repo on dpdk.org to try a new idea?
> > >
> >
> > Prototyping new DPDK-related ideas and sharing them with DPDK
> > community,
-Original Message-
> Date: Fri, 2 Jun 2017 16:09:13 +
> From: "Van Haaren, Harry"
> To: "dev@dpdk.org"
> CC: Thomas Monjalon , Jerin Jacob
> , "Richardson, Bruce"
> , "Ananyev, Konstantin"
> , "Wiles, Keith"
> Subject: [dpdk-dev] [RFCv2] service core concept
>
> > From: dev [mail
43 matches
Mail list logo
