On 02/14 14:17:04, Bill Fischofer wrote:
> I thought Petri's arguments this morning were articulate.
Agreed. That is part of why I put such things into a shared document.
> From a
> performance tuning standpoint cycle count provides a time-independent
> means of measuring algorithmic efficiency s
Clarify usage information around packet references in cases where
implementations simulate references by returning copies of packets.
Signed-off-by: Bill Fischofer
---
doc/users-guide/users-guide-packet.adoc | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/doc/user
that was merged,
Closing this thread.
Maxim.
On 02/10/17 05:22, Bill Fischofer wrote:
> For this series:
>
> Reviewed-and-tested-by: Bill Fischofer
>
> On Thu, Feb 9, 2017 at 3:02 AM, Balasubramanian Manoharan
> wrote:
>> Add implementation for outer and inner vlan ID packet matching rule
>>
Signed-off-by: Maxim Uvarov
---
.travis.yml | 5 +
1 file changed, 5 insertions(+)
diff --git a/.travis.yml b/.travis.yml
index ef9c3283..d7f4c66e 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -100,6 +100,9 @@ before_install:
- make > /dev/null
- sudo insmod ./netmap.ko
Github may allow administrators in Linaro github group post
codecov from their local repositories. That happens if
same token is used for personal git and Linaro group. There is
one small side affect that history of tests fill web page of
Linaro official project. To make history more clean post cod
make doxygen-doc does not return negative code on
errors. Do it manually.
Signed-off-by: Maxim Uvarov
---
.travis.yml | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/.travis.yml b/.travis.yml
index 9e7471c2..1ab943fe 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -119,7 +11
ubuntu trusty has updated doxygen 1.8.4 which is the same
stable version 1.8 which is described in DEPENDANCIES file.
Now doxygen do not generate any warnings. Compiler version
stays the same 4.8 for gcc.
Signed-off-by: Maxim Uvarov
---
.travis.yml | 8 +---
1 file changed, 5 insertions(+),
Signed-off-by: Maxim Uvarov
---
.travis.yml | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/.travis.yml b/.travis.yml
index a2e0a46a..9e7471c2 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -47,8 +47,11 @@ matrix:
- compiler: clang
addons:
apt:
+
in current patchset:
- switch to ubuntu trustu to make doxygen undestand odp syntax and not generate
reports;
- in unbuntu trusty libcunit appears in packages, but it works is old clang,
with
3.8 it fails to compile const to not const pointer assigment in odp code. So
as before compile
I thought Petri's arguments this morning were articulate. From a
performance tuning standpoint cycle count provides a time-independent
means of measuring algorithmic efficiency since the number of cycles
is simply a function of the ISA and compiler optimization level rather
than the specific CPU fr
if cpu goes for some small time to turbo mode then what you will measure
with timer?
btw, is cpu timers have only single propose - to measure functions time?
Maybe we need some tracing functionality for that case? Not just api for
current counters?
Maxim.
On 14 February 2017 at 22:37, Francois
Not at all. But if we are going to start adding additional terms it may be
better to look at a longer term strategies as well. If we have immediate
need for other PMR terms then we should add them, but I think as looking
for a "next gen" approach is also worthwhile.
On Tue, Feb 14, 2017 at 1:42 PM
If we can find a SW P4 implementation for linux-generic...
Do you mean that we should drop all existing pre-defined pattern matching
rules for P4?
On 14 February 2017 at 18:40, Bill Fischofer
wrote:
> While it is designed to be realized in HW, I believe P4 can also be
> realized in SW. That w
On 02/14/17 19:50, Joe Savage wrote:
>> I can not compile your code with clang-3.8, gcc works well. But looks
>> like clang does not support libatomic so -latomic -mcx16 does not work
>> for functions ending on _16"
>> ./test.c:(.text+0x61): undefined reference to `__atomic_load_16'
>> ./test.c:(.t
Thanks for taking the lead ;-)
I fully agree on deprecating cpu cycles API in ODP. Better use a true PMU
profiling API: we will not be able to upstream a special driver and if we
have to spend engineering resources, better off on ODP itself.
FF
On 14 February 2017 at 19:37, Brian Brooks wrot
On today's call, it became clear to me that some users are extremely
interested in measuring cycles. Since this is not as simple as it
sounds, I started documenting...
https://docs.google.com/a/linaro.org/document/d/1sY7rOxqCNu-bMqjBiT5_keAIohrX1ZW-eL0oGLAQ4OM/edit?usp=sharing
Please have a
While it is designed to be realized in HW, I believe P4 can also be
realized in SW. That would provide a generic means of having a unified
parser/classification strategy that is portable across all platforms, in
keeping with ODP's design goals.
On Tue, Feb 14, 2017 at 11:29 AM, Francois Ozog
wrot
> I can not compile your code with clang-3.8, gcc works well. But looks
> like clang does not support libatomic so -latomic -mcx16 does not work
> for functions ending on _16"
> ./test.c:(.text+0x61): undefined reference to `__atomic_load_16'
> ./test.c:(.text+0x92): undefined reference to `__atomi
did you test that in travis CI?
Maxim.
On 14 February 2017 at 19:00, Ravineet Singh
wrote:
> The wrapper script; odp_run_app_isolated.sh can be used to run ODP
> testcases
> in a isolated environment. Background noise can also be generated.
>
> Signed-off-by: Ravineet Singh
> ---
> .../perfor
The wrapper script; odp_run_app_isolated.sh can be used to run ODP testcases
in a isolated environment. Background noise can also be generated.
Signed-off-by: Ravineet Singh
---
.../performance/odp_run_app_isolated.sh| 108
test/linux-generic/task-isolation/README
Le 02/13/2017 à 01:49 PM, Bogdan Pricope a écrit :
> Signed-off-by: Bogdan Pricope
> ---
> example/generator/odp_generator.c | 131
> +-
> 1 file changed, 114 insertions(+), 17 deletions(-)
>
> diff --git a/example/generator/odp_generator.c
> b/example/gene
Joe,
I can not compile your code with clang-3.8, gcc works well. But looks
like clang does not support libatomic so -latomic -mcx16 does not work
for functions ending on _16"
./test.c:(.text+0x61): undefined reference to `__atomic_load_16'
./test.c:(.text+0x92): undefined reference to `__atomic_st
Hi,
I am not sure about who you are, what you are trying to do, and how
you are trying to do it. I am not even sure I understand fully your
concerns.
ODP itself does not specify what concurrency model is used. Some ODP
implementations may use pthreads, some others processes, or whatever
they want
Sorry for another cross but this is also interesting to ODP
-- Forwarded message --
From: Janne Peltonen
Date: 14 February 2017 at 08:24
Subject: [openfastpath] [PATCH 0/1] Add limited IPsec support
To: openfastp...@list.openfastpath.org
Hi,
Here is an initial version of an IPs
I am trying to apply odp to our project,
But I found odp is a multi-threaded model .
So I began to worry about the stability of the work-thread,
If a thread crashes, what happens to other threads?
Why do not odp use a multi-process model?
Who can answer my question?
thanks
Cross post from ofp team that is interesting discussion
-- Forwarded message --
From: Andriy Berestovskyy
Date: 14 February 2017 at 03:48
Subject: [openfastpath] Cavium's OFP IPsec Implementation Proposal
To: openfastp...@list.openfastpath.org
Hey guys,
Cavium is implementing IP
Hi,
Any mechanism that will prevent from copying data is a good thing.
This change can be done after the new API is merged in master branch.
Br,
Bogdan
On 13 February 2017 at 21:04, Bill Fischofer wrote:
> On Mon, Feb 13, 2017 at 6:49 AM, Bogdan Pricope
> wrote:
>> Signed-off-by: Bogdan Prico
27 matches
Mail list logo