On 9 September 2015 at 09:29, Maxim Uvarov wrote:
> On 09/08/15 16:04, Alexandru Badicioiu wrote:
>
>> I agree, at least in my case some CPUs are assigned at boot time for
>> dataplane work and cannot be changed.
>>
>> Alex
>>
>>
> Alex, how do you run validation tests
On 09/09/15 10:24, Nicolas Morey-Chaisemartin wrote:
On 09/08/2015 01:05 PM, Maxim Uvarov wrote:
ODP has api to request available number of workers. Now
no need limit that inside application.
Signed-off-by: Maxim Uvarov
---
v2: timer.c: do not substract -1 from
Merged,
corrected patch description to:
"when debug-print is disabled"
Maxim.
On 09/09/15 10:25, Nicolas Morey-Chaisemartin wrote:
Ping?
On 08/31/2015 01:52 PM, Christophe Milard wrote:
On 2015-08-27 18:47, Nicolas Morey-Chaisemartin wrote:
log_fn is never called when --disable-debug-print
Thanks Bob, having all OPNFV setup up is definitely interesting
project. I guess it even can be part of lng openembedded image or some
debian repository.
So that everyone can download and start paying with it. And of course
more people will learn what is ODP.
Best regards,
Maxim.
On
On 09/08/15 15:09, Savolainen, Petri (Nokia - FI/Espoo) wrote:
+#ifndef ODP_API_RWLOCK_RECURSIVE_H_
+#define ODP_API_RWLOCK_RECURSIVE_H_
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** @addtogroup odp_synchronizers
+ * Operations on recursive rwlocks.
+ * @{
+ */
+
+/**
+ * @typedef
On 09/08/2015 01:05 PM, Maxim Uvarov wrote:
> ODP has api to request available number of workers. Now
> no need limit that inside application.
>
> Signed-off-by: Maxim Uvarov
> ---
>
> v2: timer.c: do not substract -1 from number of workers.
>
>
Ping?
On 08/31/2015 01:52 PM, Christophe Milard wrote:
> On 2015-08-27 18:47, Nicolas Morey-Chaisemartin wrote:
>> log_fn is never called when --disable-debug-print was passed to
>> configure which causes replacement_logging_used to stay at 0
>>
>> Signed-off-by: Nicolas Morey-Chaisemartin
On 09/09/15 11:02, Nicolas Morey-Chaisemartin wrote:
On 09/09/2015 09:49 AM, Maxim Uvarov wrote:
On 09/09/15 10:24, Nicolas Morey-Chaisemartin wrote:
On 09/08/2015 01:05 PM, Maxim Uvarov wrote:
ODP has api to request available number of workers. Now
no need limit that inside application.
ODP has api to request available number of workers. Now
no need limit that inside application.
Signed-off-by: Maxim Uvarov
Reviewed-by: Nicolas Morey-Chaisemartin
---
v4: looks like I did not ammend timer change for v2 patch, do it now.
v3: account
Merged.
On 09/08/15 23:34, Bill Fischofer wrote:
Maxim: I believe this one should be ready to merge. I added my
reviewed to the v2 version of this patch last week.
Yes. Not it's good. I did not include it to 1.3 to not break anything
which we can miss.
Maxim.
On Tue, Sep 8, 2015 at
Petri, do you want to review these patches?
Thanks,
Maxim.
On 08/31/15 16:05, Ivan Khoronzhuk wrote:
This series contains fixes and is required before changing time API.
It includes previous patches:
- "[lng-odp] [Patch v2 0/2] performance: odp_pktio_perf: fix potentials ticks
overflows"
On 09/09/2015 09:49 AM, Maxim Uvarov wrote:
> On 09/09/15 10:24, Nicolas Morey-Chaisemartin wrote:
>>
>> On 09/08/2015 01:05 PM, Maxim Uvarov wrote:
>>> ODP has api to request available number of workers. Now
>>> no need limit that inside application.
>>>
>>> Signed-off-by: Maxim Uvarov
On 09/09/15 09:51, Alexandru Badicioiu wrote:
On 9 September 2015 at 09:29, Maxim Uvarov > wrote:
On 09/08/15 16:04, Alexandru Badicioiu wrote:
I agree, at least in my case some CPUs are assigned at boot
time for
On 09/09/2015 11:00 AM, Maxim Uvarov wrote:
> ODP has api to request available number of workers. Now
> no need limit that inside application.
>
> Signed-off-by: Maxim Uvarov
Reviewed-by: Nicolas Morey-Chaisemartin
> ---
> v3: account that thread_tbl
Merged,
Maxim.
On 09/08/15 18:08, Nicolas Morey-Chaisemartin wrote:
On 09/07/2015 01:53 PM, Christophe Milard wrote:
Renaming module libs in /test/validation/. Module test libs are
renamed from lib.la to libtestmodule.la.
This is done to avoid libname clashes, such as libcrypto, which seems
Fix codestyle errors reported by checkpatch.pl
Signed-off-by: Maxim Uvarov
---
platform/linux-generic/test/ring/ring_test.c | 32 +---
1 file changed, 15 insertions(+), 17 deletions(-)
diff --git a/platform/linux-generic/test/ring/ring_test.c
Add api for buffer reference count support. Which is useful in case:
- multicast support
- TCP retransmission
- traffic generator
odp_packet_free() function should relay on reference count and do not
free packet with reference count > 1. Implementation for pktio send()
function should be also
On 09/07/2015 10:50 AM, Ivan Khoronzhuk wrote:
> The resolution of schedule time can be more than 1ns. As result
> can happen that 1ns corresponds 0 ticks of timer counter, but if
> passed 1ns it's obvious that user wanted to schedule at least once.
> Currently it can lead to wait forever, as 0
On 09.09.15 15:34, Ivan Khoronzhuk wrote:
On 09.09.15 15:00, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
ext Ivan Khoronzhuk
Sent: Monday, September 07, 2015 11:51 AM
To:
On 09.09.15 15:00, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
ext Ivan Khoronzhuk
Sent: Monday, September 07, 2015 11:51 AM
To: lng-odp@lists.linaro.org
Subject: [lng-odp] [Patch] linux-generic:
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> ext Ivan Khoronzhuk
> Sent: Monday, September 07, 2015 11:51 AM
> To: lng-odp@lists.linaro.org
> Subject: [lng-odp] [Patch] linux-generic: odp_schedule: fix
> odp_schdule_wait_time
>
> The
Move odph_pause inside linux-generic. Validation test is also useless
with such move.
Signed-off-by: Maxim Uvarov
---
helper/Makefile.am | 4 +---
helper/ring.c | 2 +-
Signed-off-by: Maxim Uvarov
---
platform/linux-generic/test/Makefile.am | 2 +-
platform/linux-generic/test/ring/.gitignore | 2 +-
platform/linux-generic/test/ring/Makefile.am | 8
These patches were submitted before in v8 IPC pktio patchset. Because we did
not yet agree on api for IPC it's reasonable to consider that patches as
preparation
steps for pktio IPC support (packet exchange between different processes).
Because
of odp ring will be used as ipc pktio for
Added recursive read-write lock. It allows threads to
read lock or write lock the rwlock multiple times without
deadlocking. Mixing of read and write lock operations is
not supported.
ODP version of recursive rwlock enables porting legacy
applications, which use these kind of locks.
Applications can use recursive spinlocks to avoid deadlock from
single thread acquiring the same lock multiple times. Recursive
locks are used in legacy applications. ODP version of recursive
spinlock enable porting of those applications.
Signed-off-by: Petri Savolainen
> -Original Message-
> From: ext Hongbo Zhang [mailto:hongbo.zh...@linaro.org]
> Sent: Tuesday, September 08, 2015 2:21 PM
> To: Savolainen, Petri (Nokia - FI/Espoo)
> Cc: ext Bill Fischofer; ext hongbo.zh...@freescale.com;
> stuart.has...@arm.com; lng-odp@lists.linaro.org
> Subject: Re:
The patch in question is transformed to new one:
[lng-odp] [Patch] linux-generic: odp_schedule: fix odp_schdule_wait_time
Please, review.
On 07.09.15 17:19, Savolainen, Petri (Nokia - FI/Espoo) wrote:
*From:*ext Ola Liljedahl [mailto:ola.liljed...@linaro.org]
*Sent:* Monday, September 07, 2015
> -Original Message-
> From: EXT Maxim Uvarov [mailto:maxim.uva...@linaro.org]
> Sent: Wednesday, September 09, 2015 1:17 PM
> To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
> Subject: Re: [lng-odp] [API-NEXT PATCH v2 2/2] api: rwlock_recursive:
> added recursive
Signed-off-by: Maxim Uvarov
---
platform/linux-generic/include/odp_ring_internal.h | 20 +++---
platform/linux-generic/pktio/ring.c| 80 +++---
platform/linux-generic/test/ring/odp_ring_test.c | 16 ++---
3 files changed, 58
Signed-off-by: Maxim Uvarov
---
configure.ac | 1 -
helper/Makefile.am| 4 +---
platform/linux-generic/Makefile.am| 1 +
> -Original Message-
> From: ext Maxim Uvarov [mailto:maxim.uva...@linaro.org]
> Sent: Wednesday, September 09, 2015 10:46 AM
> To: Alexandru Badicioiu
> Cc: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
> Subject: Re: [lng-odp] [API-NEXT PATCH 3/4] api:
This is the easiest thing to try out
// #define QUEUES_PER_PRIO 4
#define QUEUES_PER_PRIO 1
The default value expects that ODP is running at least 4 cores. When you run
cores less than that, unfairness starts to appear. However, unfairness and
overload are two different things. Even the
On 9 September 2015 at 13:41, Maxim Uvarov wrote:
> Add api for buffer reference count support. Which is useful in case:
> - multicast support
> - TCP retransmission
> - traffic generator
>
I can understand that a traffic generator wants to reuse existing
initialised
On 09.09.15 15:44, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 09, 2015 3:35 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp]
On 09.09.15 16:27, Ivan Khoronzhuk wrote:
On 09.09.15 15:44, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 09, 2015 3:35 PM
To: Savolainen, Petri (Nokia - FI/Espoo);
Hi Maxim,
did you manually edit patch before sending? For some reason I can not
apply it:
nope, I will resend the patch serie. I guess my MUA messed things up.
By the way I am already on v5. What is the policy for resend? Should I
increment the version to v6? Or shjould I put [RESEND] in the
Why shm_ring_xxx rather than odp_ring_xxx (or _odp_ring_xxx if these are
internal APIs?) That would seem to be consistent with other internal APIs.
On Wed, Sep 9, 2015 at 5:06 AM, Maxim Uvarov
wrote:
> Drop odph_ prefix due to it's not helper anymore and do not
> add
Are these routines restricted to use by pktio routines? If not, perhaps
they should be at their own level (platform/linux-generic/odp_ring.c) ?
I can see more general use for these in odp_netmap or other performance
implementations.
On Wed, Sep 9, 2015 at 5:06 AM, Maxim Uvarov
Hi,
Data plane applications use tables publically to indicate the packet
processing. For example, the control part of the application add or remove or
update some entries in the table and the data part use some keys (i.e. five
tuples) extracted from the meta data or context of the packets to
Ping to Petri. You indicated on Tuesday that these changes were OK by
you. Please review so they can be merged to API-NEXT.
On Sat, Sep 5, 2015 at 9:50 PM, Bill Fischofer
wrote:
> Revise the ordered lock APIs per recent ARCH discussions. Ordered locks
> are now
I thought we decided we were going to resolve this issue by permitting
split quoted strings? That seems to be a more flexible solution. The
problem with allowing special cases for exceeding the agreed-to line limits
is where to you draw the line? Do we really want to allow a 200+ character
line
42 matches
Mail list logo