On Fri, Jun 02, 2017 at 05:29:48PM +0200, Thomas Monjalon wrote:
> 01/06/2017 07:07, Tiwei Bie:
> > Hello everyone,
> >
> > We'd like to make a proposal of making DPDK.org allow hosting
> > some personal/project repos, which could be very useful when
> > someone wants to try some experimental proj
For PCI prefetchable resources, Linux will create a
write combined file as well, the library will try
to map resourceX_wc file first, if the file does
not exist, then it will map resourceX as usual.
Signed-off-by: Changpeng Liu
---
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 19 ++---
The PROD/CONS_ALIGN values on x86-64 are set to 2 cache lines, so members of
struct rte_ring are 128 byte aligned, and therefore the whole struct needs
128-byte alignment according to the ABI so that the 128-byte alignment of the
fields can be guaranteed.
If the allocation is only 64-byte align
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Daniel Verkamp
> Sent: Friday, June 2, 2017 9:12 PM
> To: dev@dpdk.org
> Cc: Verkamp, Daniel
> Subject: [dpdk-dev] [PATCH v2] ring: use aligned memzone allocation
>
> rte_memzone_reserve() provides cache line al
On Thu, 1 Jun 2017 01:03:36 +
"Wang, Yipeng1" wrote:
> Hi Vincent,
>
> Thanks for the comments and some quick responses below:
>
> - DPDK Bloom Filter is derived from the libbloom (as shown by the included
> BSD license), but optimized for performance with various DPDK goodness such
> as
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, with
> some of them likely to eventually make their way into DPDK once accepted
rte_memzone_reserve() provides cache line alignment, but
struct rte_ring may require more than cache line alignment: on x86-64,
it needs 128-byte alignment due to PROD_ALIGN and CONS_ALIGN, which are
128 bytes, but cache line size is 64 bytes.
Fixes runtime warnings with UBSan enabled.
Fixes: d9f
rte_memzone_reserve() provides cache line alignment, but
struct rte_ring may require more than cache line alignment: on x86-64,
it needs 128-byte alignment due to PROD_ALIGN and CONS_ALIGN, which are
128 bytes, but cache line size is 64 bytes.
Fixes runtime warnings with UBSan enabled.
Fixes: d9f
> 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, with
some of them likely to eventually make their way into DPDK once accepted and
mature enough.
> I am against adding some user repos in this list:
>
> -Original Message-
> From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com]
> Sent: Friday, June 2, 2017 12:21 PM
> To: dev@dpdk.org
> Cc: Richardson, Bruce ;
> tho...@monjalon.net; Dumitrescu, Cristian
> ; yuanhan@linux.intel.com;
> maxime.coque...@redhat.com; Jerin Jacob
>
> S
On 06/02/2017 05:31 PM, Nirmoy Das wrote:
> disable truncation check to ignore below warning
> dpdk/x86_64-native-linuxapp-gcc-default/build/lib/librte_eal/linuxapp/kni/igb_main.c:2476:30:
> error: '%d' directive output may be truncated writing between 1 and 5 bytes
> into a region of size betwee
disable truncation check to ignore below warning
dpdk/x86_64-native-linuxapp-gcc-default/build/lib/librte_eal/linuxapp/kni/igb_main.c:2476:30:
error: '%d' directive output may be truncated writing between 1 and 5 bytes
into a region of size between 0 and 11 [-Werror=format-truncation=]
Signed-of
Suppose we have 2 virtio devices for a VM, with only the first one,
virtio0, binding to igb_uio. Start a primary DPDK process, driving
only virtio0. Then start a secondary DPDK process, it encounters
segfault at eth_virtio_dev_init() because hw is NULL, when trying
to initialize the 2nd virtio devi
Hi Wei,
Have you had a chance to look at the debug output yet? Is there any
additional data you need?
Regards,
Roger
On 5/24/17 2:38 PM, Roger B. Melton wrote:
Hi Wei,
I am using ixgbe:
00:02.0 Ethernet controller: Intel Corporation 82599ES 10-Gigabit
SFI/SFP+ Network Connection (rev 01
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Van Haaren, Harry
> Sent: Thursday, May 25, 2017 2:28 PM
> Subject: Re: [dpdk-dev] [RFC] service core concept header implementation
>
> Thanks everybody for the input on the RFC - appreciated! From an application
> point-of-view, I
> see meri
01/06/2017 07:07, Tiwei Bie:
> Hello everyone,
>
> We'd like to make a proposal of making DPDK.org allow hosting
> some personal/project repos, which could be very useful when
> someone wants to try some experimental projects in DPDK. Many
> formal/mature opensource communities allow this. Such as
02/06/2017 15:34, Yuanhan Liu:
> Signed-off-by: Yuanhan Liu
> ---
>
> - I left Intel today.
>
> ---
> MAINTAINERS | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
Applied, thanks and good luck in your new job :)
-Original Message-
> Date: Thu, 18 May 2017 19:45:44 +0530
> From: Hemant Agrawal
> To: Jerin Jacob , dev@dpdk.org
> CC: bruce.richard...@intel.com, harry.van.haa...@intel.com,
> gage.e...@intel.com, nipun.gu...@nxp.com, narender.vang...@intel.com
> Subject: Re: [PATCH] eventdev: define t
> The queue allocation was changed, from allocating one queue-pair at a
> time to one queue at a time. Most of the changes have been done, but
> just with one being missed: the size of coping the old queue is still
> based on queue-pair at numa_realloc(), which leads to overwritten issue.
> As a re
The error return code for rte_ring_sc_dequeue_bulk() and
rte_ring_mc_dequeue_bulk() function should be -ENOENT rather
than -ENOBUFS as described in the function description.
Signed-off-by: Anand B Jyoti
---
lib/librte_ring/rte_ring.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
dif
Signed-off-by: Yuanhan Liu
---
- I left Intel today.
---
MAINTAINERS | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index afb4cab..f6095ef 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -424,7 +424,7 @@ F: doc/guides/nics/vmxnet3.rst
F: doc/g
-Original Message-
> Date: Thu, 1 Jun 2017 21:04:15 +
> From: "Eads, Gage"
> To: Jerin Jacob , "dev@dpdk.org"
>
> CC: "Richardson, Bruce" , "Van Haaren, Harry"
> , "hemant.agra...@nxp.com"
> , "nipun.gu...@nxp.com" ,
> "Vangati, Narender" , "Rao, Nikhil"
> , "gprathyu...@caviumne
-Original Message-
> Date: Thu, 1 Jun 2017 21:11:29 +
> From: "Eads, Gage"
> To: Jerin Jacob , "dev@dpdk.org"
>
> CC: "Richardson, Bruce" , "Van Haaren, Harry"
> , "hemant.agra...@nxp.com"
> , "nipun.gu...@nxp.com" ,
> "Vangati, Narender" , "Rao, Nikhil"
> , "gprathyu...@caviumne
Fixing the below error by returning from the function early
when count == 0.
Issue flagged by GCC 7.1.1
examples/vhost/virtio_net.c:370:38: error: ‘desc_indexes[0]’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]
rte_prefetch0(&vr->desc[desc_indexes[0]]);
Fixes: ca059f
This one is more of a compiler issue as application
checks the app_parse_opt_vals() return value.
Since this code is in slow path, adding a memset
to fix following "maybe-uninitialized" warning.
qos_sched/args.c: In function ‘app_parse_args’:
examples/qos_sched/args.c:254:32: error: ‘vals[0]’ may
This fixes compiler warnings with GCC 7.1.1
Fixes: d48415e1fee3 ("examples/performance-thread: add l3fwd-thread app")
Signed-off-by: Jerin Jacob
---
examples/performance-thread/l3fwd-thread/main.c | 8
1 file changed, 8 insertions(+)
diff --git a/examples/performance-thread/l3fwd-threa
This fixes compiler warnings with GCC 7.1.1
Fixes: 26b5b020 ("examples/l3fwd: modularize")
Fixes: 94c54b4158d5 ("examples/l3fwd: rework exact-match")
Signed-off-by: Jerin Jacob
---
examples/l3fwd/l3fwd_lpm_sse.h | 2 ++
examples/l3fwd/l3fwd_sse.h | 8
2 files changed, 10 inserti
Hi Fiona,
On Mon, May 29, 2017 at 02:51:11PM +, Trahe, Fiona wrote:
> Hi Umesh,
>
> > -Original Message-
> > From: Umesh Kartha [mailto:umesh.kar...@caviumnetworks.com]
> > Sent: Friday, May 26, 2017 8:18 AM
> > To: Trahe, Fiona
> > Cc: dev@dpdk.org; Jerin Jacob ;
> > Balasubramania
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_srli_si128(i24, sizeof(uint64_t));
>
> -O0 opti
Hi all,
I'm going to announce the 17.02.1 stable release:
http://fast.dpdk.org/rel/dpdk-17.02.1.tar.xz
The git tree is at:
http://dpdk.org/browse/dpdk-stable/
Thanks.
--yliu
---
app/test-crypto-perf/cperf_test_throughput.c | 13 +
app/test-crypto-perf/cperf_test_vectors
On Fri, Jun 02, 2017 at 09:54:46AM +0530, santosh wrote:
> Ping?
>
> On Wednesday 24 May 2017 09:41 PM, Santosh Shukla wrote:
>
> > Some NPU hardware like OCTEONTX follows push model to get
> > the packet from the pktio device. Where packet allocation
> > and freeing done by the HW. Since HW can
Signed-off-by: Jerin Jacob
---
drivers/event/octeontx/ssovf_evdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/event/octeontx/ssovf_evdev.c
b/drivers/event/octeontx/ssovf_evdev.c
index c80a44379..f1b7eee3d 100644
--- a/drivers/event/octeontx/ssovf_evdev.c
+++ b/drivers/event/oct
On Fri, Jun 02, 2017 at 08:14:46AM +0800, Yuanhan Liu wrote:
> The queue allocation was changed, from allocating one queue-pair at a
> time to one queue at a time. Most of the changes have been done, but
> just with one being missed: the size of coping the old queue is still
s/coping/copying/ ?
>
Hi,
> -Original Message-
> From: Zhao1, Wei
> Sent: Friday, June 2, 2017 2:36 PM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo; Zhao1, Wei
> Subject: [PATCH v2 04/11] net/e1000: restore flex type filter
>
> Add support for restoring flex type filter in SW.
>
> Signed-off-by: Wei Zhao
Acked-by: W
Hi,
> -Original Message-
> From: Zhao1, Wei
> Sent: Friday, June 2, 2017 2:36 PM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo; Zhao1, Wei
> Subject: [PATCH v2 03/11] net/e1000: restore ether type filter
>
> Add support for restoring ether type filter in SW.
>
> Signed-off-by: Wei Zhao
Acked-by:
Hi, wenzhuo
> -Original Message-
> From: Lu, Wenzhuo
> Sent: Friday, June 2, 2017 3:57 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Subject: RE: [PATCH v2 02/11] net/e1000: restore n-tuple filter
>
> Hi Wei,
>
> > -Original Message-
> > From: Zhao1, Wei
> > Sent: Friday, June 2, 2017 2:3
Hi Wei,
> -Original Message-
> From: Zhao1, Wei
> Sent: Friday, June 2, 2017 2:36 PM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo; Zhao1, Wei
> Subject: [PATCH v2 02/11] net/e1000: restore n-tuple filter
>
> Add support for restoring n-tuple
> filter in SW.
>
> Signed-off-by: Wei Zhao
> ---
>
Hi,
> -Original Message-
> From: Zhao1, Wei
> Sent: Friday, June 2, 2017 2:36 PM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo; Zhao1, Wei
> Subject: [PATCH v2 01/11] net/e1000: store and restore TCP SYN filter
>
> Add support for storing and restoring TCP SYN filter in SW.
>
> Signed-off-by: Wei
38 matches
Mail list logo
