I've added this topic to the agenda for the Wednesday ARCH call.
On Tue, Jun 7, 2016 at 8:14 AM, Oriol Arcas
wrote:
> Hello all,
>
> Sorry for joining the convesation late too. At Starflow we detected a
> regression at this exact commit. We use the RAW socket
Hi, Christophe and Mike
Thanks, it is great idea to collect all tests into the envolop "Test
Suite". I update the second diagram to reflect this and I will start coding
on this:
Dependencies Rearrangement Update
This patch needs a rebase:
bill@Ubuntu15:~/linaro/matiasprint$ git am --reject ~/Mail/Incoming/Matias/1
Applying: linux-gen: fix debug info print functions
Checking patch platform/linux-generic/include/odp_debug_internal.h...
Checking patch platform/linux-generic/odp_name_table.c...
error: while
In general this looks good by I wonder if --enable-compatibility-mode is
sufficiently descriptive. Compatible with what? I think Zoltan was playing
around with distinguishing between static vs. dynamic linking, which is
another dimension.
The desired distinction here is whether the code is to be
https://bugs.linaro.org/show_bug.cgi?id=2309
--- Comment #1 from Bill Fischofer ---
Given the nature of the reported failure, the question is whether this is an
issue with the timer implementation or with the timer test? Is there a
crossover point in CPU count above
Example for simple timer usage.
https://bugs.linaro.org/show_bug.cgi?id=2090
Signed-off-by: Maxim Uvarov
---
example/timer/.gitignore | 1 +
example/timer/Makefile.am| 12 ++-
example/timer/odp_timer_simple.c | 167
They can be updated separately so no need to bump both versions at the
same time, specially when there is no change in one of them.
Signed-off-by: Anders Roxell
---
configure.ac | 3 +++
helper/Makefile.am | 2 +-
2 files changed, 4 insertions(+), 1 deletion(-)
https://bugs.linaro.org/show_bug.cgi?id=2310
Bill Fischofer changed:
What|Removed |Added
Ever confirmed|0 |1
Hello all,
Sorry for joining the convesation late too. At Starflow we detected a
regression at this exact commit. We use the RAW socket interface
(socket_mmap PKTIO) and now the performance has dropped to kB/s. There is
connectivity, but the behavior is erratic and the performance very low.
Hi,
Just realised that I haven't sent these comment. Sorry for the late feedback.
Comments inline...
Regards,
Bala
On 2 June 2016 at 13:15, Matias Elo wrote:
> Don't allocate new packets inside of the internal
> classifier helpers _odp_packet_cls_enq() and
>
https://bugs.linaro.org/show_bug.cgi?id=2310
Bug ID: 2310
Summary: odp_packet_copy() does not copy packet user area
Product: OpenDataPlane - linux- generic reference
Version: v1.10
Hardware: Other
OS: Linux
Status:
Sounds like this should be considered a bug. I'll open one to track it.
On Tue, Jun 7, 2016 at 2:18 AM, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia.com> wrote:
> /**
> * Full copy of a packet
> *
> * Create a new copy of the packet. The new packet is exact copy of the
>
> + if (nb_txq == 0)
> + fwd_pktio->nb_txq = 1;
> + else if (nb_txq > (int)capa.max_output_queues)
> + fwd_pktio->nb_txq = capa.max_output_queues;
> +
> + in_queue_param.num_queues = fwd_pktio->nb_rxq;
> + out_queue_param.num_queues = fwd_pktio->nb_txq;
> +
On 7 June 2016 at 11:04, Xuelin Shi wrote:
> multi-thread, multi-queue and bi-directional forwarding.
>
> support (port, queue, core) arguments in cmdline.
> it means core will handle rx queue at port.
> forwarding logic will decide out port by this specification.
> if no
Helper's chksum.h API had a void* to uint16_t* casting, which is not
legal in C++ (Clang and GCC).
Signed-off-by: Oriol Arcas
Reviewed-by: Bill Fischofer
---
helper/include/odp/helper/chksum.h | 2 +-
1 file changed, 1 insertion(+), 1
multi-thread, multi-queue and bi-directional forwarding.
support (port, queue, core) arguments in cmdline.
it means core will handle rx queue at port.
forwarding logic will decide out port by this specification.
if no this argument in cmdline, default used.
Signed-off-by: Xuelin Shi
On 7 June 2016 at 10:22, Yi He wrote:
> Hi, team
>
> I send my thoughts on the ODP thread part:
>
> S1, S2, S3, S4
> Yi: Yes
> This set of statements defines ODP thread concept as a higher
> level abstraction of underlying concurrent execution context.
>
> S5, S6, S7, S8, S9,
https://bugs.linaro.org/show_bug.cgi?id=2309
Bug ID: 2309
Summary: Timer validation test failure
Product: OpenDataPlane - linux- generic reference
Version: unspecified
Hardware: Other
OS: Linux
Status: UNCONFIRMED
Hi, team
I send my thoughts on the ODP thread part:
S1, S2, S3, S4
Yi: Yes
This set of statements defines ODP thread concept as a higher
level abstraction of underlying concurrent execution context.
S5, S6, S7, S8, S9, S10:
Yi: Yes
This set of statements add several constraints upon ODP
/**
* Full copy of a packet
*
* Create a new copy of the packet. The new packet is exact copy of the source
* packet (incl. data and metadata). The pool must have been created with
* ODP_POOL_PACKET type.
*
All metadata should be copied into the new packet. User controls user_area
sizes in
> -Original Message-
> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
> Sent: Monday, June 06, 2016 9:16 PM
> To: Savolainen, Petri (Nokia - FI/Espoo) ;
> lng-odp@lists.linaro.org
> Subject: Re: [lng-odp] [PATCH v2 4/6] linux-gen: sched: add skeleton for
Hi Yi!
Your first diagram shows at least one thing very clearly, apart the
mess we have with dependencies: The location of the
platform//test directory is confusing!
This directory is *NOT* a part of the ODP implementation, despite its
location: it is a part of the test structure: that part
22 matches
Mail list logo