Bala/Alex: Please review the new odp_schedule_order_lock() and
odp_schedule_order_unlock() routines (parts 12 and 13 of this series). I
believe they now provide the functionality you want, and do so quite
efficiently.
Thanks.
On Wed, Aug 5, 2015 at 5:55 PM, Bill Fischofer
wrote:
> Changes in v
Signed-off-by: Bill Fischofer
---
.../include/odp/plat/schedule_types.h | 2 -
.../linux-generic/include/odp_buffer_internal.h| 5 ++-
.../linux-generic/include/odp_queue_internal.h | 2 +
platform/linux-generic/odp_queue.c | 48 --
4 fi
Signed-off-by: Bill Fischofer
---
include/odp/api/schedule.h | 78 --
1 file changed, 41 insertions(+), 37 deletions(-)
diff --git a/include/odp/api/schedule.h b/include/odp/api/schedule.h
index ea16583..2762bdd 100644
--- a/include/odp/api/schedule.h
Signed-off-by: Bill Fischofer
---
.../linux-generic/include/odp_buffer_internal.h| 1 +
platform/linux-generic/odp_queue.c | 43 +++---
platform/linux-generic/odp_schedule.c | 24
3 files changed, 62 insertions(+), 6 deletions(-)
dif
Signed-off-by: Bill Fischofer
---
include/odp/api/schedule.h | 51 ++
1 file changed, 51 insertions(+)
diff --git a/include/odp/api/schedule.h b/include/odp/api/schedule.h
index f27bf78..ea16583 100644
--- a/include/odp/api/schedule.h
+++ b/include/odp
Signed-off-by: Bill Fischofer
---
.../linux-generic/include/odp_buffer_internal.h| 4 +
platform/linux-generic/odp_queue.c | 151 +++--
2 files changed, 115 insertions(+), 40 deletions(-)
diff --git a/platform/linux-generic/include/odp_buffer_internal.h
b/p
Signed-off-by: Bill Fischofer
---
include/odp/api/queue.h | 23 +++
1 file changed, 23 insertions(+)
diff --git a/include/odp/api/queue.h b/include/odp/api/queue.h
index ce04eb4..5a6dbc0 100644
--- a/include/odp/api/queue.h
+++ b/include/odp/api/queue.h
@@ -173,6 +173,29 @@ i
Signed-off-by: Bill Fischofer
---
platform/linux-generic/odp_schedule.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/platform/linux-generic/odp_schedule.c
b/platform/linux-generic/odp_schedule.c
index d595375..4d594a0 100644
--- a/platform/linux-generic/odp_schedule.c
+++ b/platform/
Signed-off-by: Bill Fischofer
---
platform/linux-generic/odp_queue.c | 169 +
1 file changed, 152 insertions(+), 17 deletions(-)
diff --git a/platform/linux-generic/odp_queue.c
b/platform/linux-generic/odp_queue.c
index 4d0e1b4..a2460e7 100644
--- a/platform/
Signed-off-by: Bill Fischofer
---
include/odp/api/schedule.h | 38 --
1 file changed, 24 insertions(+), 14 deletions(-)
diff --git a/include/odp/api/schedule.h b/include/odp/api/schedule.h
index bd858b3..f27bf78 100644
--- a/include/odp/api/schedule.h
+++ b/in
Signed-off-by: Bill Fischofer
---
.../linux-generic/include/odp_buffer_internal.h| 5 +
.../linux-generic/include/odp_queue_internal.h | 4 +
.../linux-generic/include/odp_schedule_internal.h | 2 +-
platform/linux-generic/odp_pool.c | 3 +
platform/linux-generic
Signed-off-by: Bill Fischofer
---
include/odp/api/config.h | 5 +
.../include/odp/plat/schedule_types.h | 4 +
platform/linux-generic/odp_schedule.c | 156 +
platform/linux-generic/odp_thread.c| 25 +++-
Signed-off-by: Bill Fischofer
---
test/validation/scheduler/scheduler.c | 4
1 file changed, 4 insertions(+)
diff --git a/test/validation/scheduler/scheduler.c
b/test/validation/scheduler/scheduler.c
index fe03ab7..bbe829e 100644
--- a/test/validation/scheduler/scheduler.c
+++ b/test/valid
Signed-off-by: Bill Fischofer
---
include/odp/api/schedule.h | 63 +++---
1 file changed, 59 insertions(+), 4 deletions(-)
diff --git a/include/odp/api/schedule.h b/include/odp/api/schedule.h
index 36e52cd..bd858b3 100644
--- a/include/odp/api/schedule.h
+
Changes in v8
- Miscellaneous comment improvements
- Delete odp_schedule_order_sync() in favor of two new APIs.
odp_schedule_order_lock() and odp_schedule_order_unlock() permit in-order
locking within an ordered context in a serially reusable manner. Locks
obtained by these routines are inten
The word "cycle" is left from old API time names. The "cycle" is
ambiguous word, especially when it can be used with software cycles.
So better to use "time" word or just "t" symbol, as no matter in
which units time is measured, in sec, hours, cycles or counts.
Signed-off-by: Ivan Khoronzhuk
---
The time API were changed from *_cycles* to *_counts*, so use
appropriate names.
Signed-off-by: Ivan Khoronzhuk
---
platform/linux-generic/arch/linux/odp_time.c | 2 +-
platform/linux-generic/arch/mips64/odp_time.c | 2 +-
platform/linux-generic/arch/x86/odp_time.c| 2 +-
platform/linux-gen
The word "cycle" is left from old API time names. The "cycle" is
ambiguous word, especially when it can be used with software cycles.
So better to use "time" word or just "t" symbol, as no matter in
which units time is measured, in sec, hours, cycles or counts.
Signed-off-by: Ivan Khoronzhuk
---
Current time API supposes that frequency of counter is equal
to CPU frequency. But that's not always true, for instance,
in case if no access to CPU cycle counter, another hi-resolution
timer can be used, and it`s frequency can be different from CPU
frequency. There is no big difference in which cy
This seres is intedent to change time API names from *_cycles*
to *_counts*. Also remove usage of word "cycle" from appropriate
places as it's no more valid.
Based on api-next.
Ivan Khoronzhuk (4):
api: time: unbind CPU cycles from time API
test/example: avoid "cycle" word usage
linux-gener
There hasn't been any interaction between these two projects to my
knowledge. As far as I can tell from the SAI web site, their goal seems to
be more on the switch management side, whereas ODP is focused on data plane
applications.
On Wed, Aug 5, 2015 at 2:50 PM, gyanesh patra
wrote:
> Hi All,
Hi All,
I have seen ODP and SAI have good traction in networking community with
very rapid contribution and development. Can anyone please explain how they
are different from each other? What is the fundamental difference in their
goal?
Thanks
P Gyanesh Kumar Patra
On 5 August 2015 at 14:55, Bill Fischofer wrote:
> linux-generic wouldn't introduce ARCH variants except in those cases where
> it truly matters (e.g., some of the atomics, Barry's proposed AES-based
> hashing for TM, etc.). For everything else the standard generic C code
> should be fine.
>
Ag
linux-generic wouldn't introduce ARCH variants except in those cases where
it truly matters (e.g., some of the atomics, Barry's proposed AES-based
hashing for TM, etc.). For everything else the standard generic C code
should be fine.
On Wed, Aug 5, 2015 at 1:52 PM, Mike Holmes wrote:
>
>
> On 5
On 5 August 2015 at 14:39, Bill Fischofer wrote:
> linux-generic serves several purposes and one of them is to serve as a
> model for "production" implementations. So it's reasonable for
> linux-generic to have an ARCH structure for other implementations to use as
> a model. For example, if we
linux-generic serves several purposes and one of them is to serve as a
model for "production" implementations. So it's reasonable for
linux-generic to have an ARCH structure for other implementations to use as
a model. For example, if we wanted to support odp-dpdk on ARM-based
platforms that do t
Fixed the comments in the new version.
On 5 August 2015 at 19:13, Stuart Haslam wrote:
> On Thu, Jul 30, 2015 at 01:24:01PM +0200, Balakrishna.Garapati wrote:
> > Signed-off-by: Balakrishna.Garapati
> > ---
> > changed the command line argument conventions.
> > example/generator/odp_generator
On Wed, Aug 5, 2015 at 10:26 AM, Stuart Haslam
wrote:
> Signed-off-by: Stuart Haslam
> ---
> platform/linux-generic/include/odp_packet_io_internal.h | 4 +++-
> platform/linux-generic/odp_packet_io.c | 9 +
> 2 files changed, 8 insertions(+), 5 deletions(-)
>
> diff --g
Signed-off-by: Balakrishna.Garapati
---
Fixed the comments
example/generator/odp_generator.c | 59 ++-
1 file changed, 40 insertions(+), 19 deletions(-)
diff --git a/example/generator/odp_generator.c
b/example/generator/odp_generator.c
index bdee222..e364af
Doxygen can create a perl module that is a hash of the entire API.
Set the flag to generate this so that tools may analyse differences in
the API between different repositories branches and tags.
The benefits are:
- This can allow CI to generate nightly data such as the difference
between api-ne
On Thu, Jul 30, 2015 at 01:24:01PM +0200, Balakrishna.Garapati wrote:
> Signed-off-by: Balakrishna.Garapati
> ---
> changed the command line argument conventions.
> example/generator/odp_generator.c | 57
> ++-
> 1 file changed, 39 insertions(+), 18 deletions
On Tue, Aug 04, 2015 at 11:07:57AM +0200, Balakrishna.Garapati wrote:
> Signed-off-by: Balakrishna.Garapati
> ---
> Added packet rate stats
> example/generator/odp_generator.c | 67
> ---
> 1 file changed, 56 insertions(+), 11 deletions(-)
>
> diff --git a/e
On 5 August 2015 at 12:04, Maxim Uvarov wrote:
> Building generated this warning message:
> "copying selected object files to avoid basename conflicts..."
That happed due to we have files with same name:
>
happened
> ./platform/linux-generic/odp_time.c
> and
> platform/linux-generic/arch/.../
On 5 August 2015 at 12:04, Maxim Uvarov wrote:
> Compile all arch depandant code to library, then link that
> library to main libodp.a
>
>
We don’t want two libraries for one implementation that is confusing if it
takes this level of complexity I wonder if we need to remove all ARCH
support from
Building generated this warning message:
"copying selected object files to avoid basename conflicts..."
That happed due to we have files with same name:
./platform/linux-generic/odp_time.c
and
platform/linux-generic/arch/.../odp_time.c
Autotools builds 2 objects with corresponding name odp_time.o.
Compile all arch depandant code to library, then link that
library to main libodp.a
Signed-off-by: Maxim Uvarov
---
platform/linux-generic/Makefile.am | 6 --
platform/linux-generic/arch/linux/Makefile.am | 5 +
platform/linux-generic/arch/mips64/Makefile.am | 5 +
platf
Try to orginize arch code compilation.
For some reason these patches break ./scripts/debian packaging. But build in
tree and
out of tree pass, make dist also pass. Will send v2 when will figure out what
is that,
any help how to debug debian build is welcome.
Maxim.
Maxim Uvarov (2):
linux-ge
Reviewed-by: Balasubramanian Manoharan
Regards,
Bala
> On 05-Aug-2015, at 8:54 pm, Stuart Haslam wrote:
>
> When the string containing the number of packets processed by one of the
> queues gets for be longer than a tab the alignment is messed up and the
> fields overlap with part of a previou
If pcap is enabled in the build, add a validation test for it.
Signed-off-by: Stuart Haslam
---
platform/linux-generic/test/Makefile.am | 5
platform/linux-generic/test/pktio/Makefile.am| 4 +++
platform/linux-generic/test/pktio/pktio_run_pcap | 33
3
Create a new pktio type that allows for reading from and writing to a
pcap capture file. This is intended to be used as a simple way of
injecting test packets into an application for functional testing and
can be used as it is with some of the existing example applications.
To use this interface t
cos_name will overflow if args->if_name is long, so change the sprintf to
snprintf. For consistency the other uses of sprintf are also removed.
Signed-off-by: Stuart Haslam
---
example/classifier/odp_classifier.c | 16 +---
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git
Signed-off-by: Stuart Haslam
---
platform/linux-generic/include/odp_packet_io_internal.h | 4 +++-
platform/linux-generic/odp_packet_io.c | 9 +
2 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/platform/linux-generic/include/odp_packet_io_internal.h
b/platf
When the string containing the number of packets processed by one of the
queues gets for be longer than a tab the alignment is messed up and the
fields overlap with part of a previous value.
Signed-off-by: Stuart Haslam
---
example/classifier/odp_classifier.c | 10 ++
1 file changed, 6 i
On Fri, Jul 31, 2015 at 09:56:44PM +0200, Christophe Milard wrote:
> Signed-off-by: Christophe Milard
Reviewed-by: Stuart Haslam
> ---
> test/README| 8 +---
> test/validation/README | 50
> ++
> 2 files changed, 51 insertions(+
The API names even can be leaved as is.
Only description change (remove CPU cycles)
On 05.08.15 15:01, Ivan Khoronzhuk wrote:
On 04.08.15 15:19, Mike Holmes wrote:
In the platform call just now the discussion came to this patch.
Ivan can you propose an API to get the timer resolution for the
p
On 04.08.15 15:19, Mike Holmes wrote:
In the platform call just now the discussion came to this patch.
Ivan can you propose an API to get the timer resolution for the
platform, then the test suite can call that and run the test appropriately.
I remember someone proposed to rename cycles to so
Yes. I'll reword in v8. Thanks.
On Wed, Aug 5, 2015 at 1:09 AM, Bala Manoharan
wrote:
> Hi,
>
> On 5 August 2015 at 04:11, Bill Fischofer
> wrote:
>
>> Signed-off-by: Bill Fischofer
>> ---
>> include/odp/api/schedule.h | 38 --
>> 1 file changed, 24 inser
47 matches
Mail list logo
