On Tue, Feb 7, 2017 at 12:06 PM, Dmitry Eremin-Solenikov
wrote:
> In preparation to update crypto code for OpenSSL 1.1.0 move locks out of
> global data.
>
> Signed-off-by: Dmitry Eremin-Solenikov
> ---
> platform/linux-generic/odp_crypto.c | 67
> ++---
> 1 file
Given these issues with obsolete compiler levels, it may be simpler to
go with Brian's suggestion and just use some simple macros. I'll post
a v4 along those lines.
On Tue, Feb 7, 2017 at 7:51 AM, Maxim Uvarov wrote:
> On 02/07/17 16:39, Bill Fischofer wrote:
>> The errors
On Tue, Feb 7, 2017 at 7:39 AM, Nikhil Agarwal
wrote:
>
>
> On 7 February 2017 at 18:58, Bill Fischofer
> wrote:
>>
>> On Tue, Feb 7, 2017 at 7:22 AM, Nikhil Agarwal
>> wrote:
>> > On 3 February 2017 at 19:20, Ola Liljedahl
>> > wrote:
>>
start tag `constantgroups' found in
scope='namespace/memberdecl/'!
error: Unexpected start tag `constantgroups' found in scope='file/memberdecl/'!
So do we also have a doxygen version issue on these systems?
On Tue, Feb 7, 2017 at 7:30 AM, Maxim Uvarov wrote:
> I
On Tue, Feb 7, 2017 at 7:22 AM, Nikhil Agarwal
wrote:
> On 3 February 2017 at 19:20, Ola Liljedahl wrote:
>
>> Do we have any performance comparison between using this new API
>> compared to using the existing API and the SW behind
>> odp_schedule_multi() sorts the events (if necessary) and only
On Tue, Feb 7, 2017 at 2:03 AM, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>
>
>> -Original Message-----
>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>> Sent: Tuesday, February 07, 2017 2:10 AM
>> To: Petri Savolainen
>> Cc: lng-odp-forward
On Tue, Feb 7, 2017 at 1:50 AM, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>> -Original Message-
>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>> Sent: Tuesday, February 07, 2017 2:05 AM
>> To: Petri Savolainen
>> Cc: lng-odp-forward
>>
On Mon, Feb 6, 2017 at 8:35 PM, Yi He wrote:
> Fixed configure script when enables schedule option
> schedule-iquery=yes or schedule-sp=yes: command not found
>
> Signed-off-by: Yi He
Reviewed-and-tested-by: Bill Fischofer
> ---
> platform/linux-generic/m4/odp_schedule.m4
On Thu, Dec 22, 2016 at 8:32 PM, Yi He wrote:
> SP scheduler hangs in process mode performance test
> due to global data structure were not created in shared
> memory region.
>
> Signed-off-by: Yi He
Reviewed-and-tested-by: Bill Fischofer
> ---
> since v1: rebased u
checking for doxygen... doxygen
checking for asciidoctor... asciidoctor
...
This isn't related to this patch, but something to look into?
On Mon, Feb 6, 2017 at 7:04 PM, Bill Fischofer
wrote:
> This looks reasonable, but I'm having trouble trying to test it. I've
> specified --
This looks reasonable, but I'm having trouble trying to test it. I've
specified --enable-test-perf-proc on ./configure but it the various
perf tests still seem to be using pthreads. Did I miss some
configuration flag?
On Mon, Feb 6, 2017 at 1:37 AM, Yi He wrote:
> Ping, this patch is still good f
For this series:
Reviewed-and-tested-by: Bill Fischofer
On Fri, Feb 3, 2017 at 6:45 AM, Balasubramanian Manoharan
wrote:
> Adds test case for ipv6 source and destination address matching
>
> Signed-off-by: Balasubramanian Manoharan
> ---
> v3: performance enhancement using bit
On Mon, Feb 6, 2017 at 5:15 PM, Dmitry Eremin-Solenikov
wrote:
> On 07.02.2017 00:57, Bill Fischofer wrote:
>>
>> The problem is not what level of OpenSSL ODP might be compiled
>> against, but what level is installed on the system the ODP application
>> is running on
It's not clear why this unrelated patch is part of this series. The
only parts that seem relevant to this series are Parts 1, 2, 3, 9, and
10. The rest should either be their own patches or else be part of a
general "miscellaneous cleanups" series.
On Mon, Feb 6, 2017 at 6:37 AM, Petri Savolainen
These are reasonable changes, but they conflict with the code in
api-next so I suggest this series be rebased onto api-next and then
moved to master as part of the next release (which should be later
this month) to avoid a huge merge headache for Maxim.
On Mon, Feb 6, 2017 at 6:37 AM, Petri Savo
The problem is not what level of OpenSSL ODP might be compiled
against, but what level is installed on the system the ODP application
is running on, since we don't distribute OpenSSL with ODP. OpenSSL
v1.1.0 is backward-compatible with the older callback structure (they
become no-ops) so there's no
I'm all for keeping things simple and just using printf() or other
standard C library functions in validation tests. The purpose of the
macros in ODP code was to allow for transparent redirection of log
messages to an application-supplied logger, and that's not something
that we need for the valida
Ping. Brian can you please verify that this now works on your system.
It works on my Ubuntu 16.04 system. A review would also be nice. :)
On Thu, Feb 2, 2017 at 4:51 PM, Bill Fischofer
wrote:
> The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
> used in C++ program
mentation whether that processing is done in SW on visible or
invisible CPUs or some other HW path that doesn't involve any CPUs.
> On 2 February 2017 at 20:01, Brian Brooks wrote:
>>
>> On Wed, Feb 1, 2017 at 9:41 AM, Bill Fischofer
>> wrote:
>> > On Wed, Feb
In the various table lookup routines, check that odp_shm_lookup() returns
a valid handle before attempting to dereference it via odp_shm_addr().
Signed-off-by: Bill Fischofer
---
helper/cuckootable.c | 7 +--
helper/hashtable.c | 7 +--
helper/iplookuptable.c | 5 -
helper
Resolve Bug https://bugs.linaro.org/show_bug.cgi?id=2829 by avoiding
unused variable.
Signed-off-by: Bill Fischofer
---
helper/iplookuptable.c | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/helper/iplookuptable.c b/helper/iplookuptable.c
index 845125b3..9abdab22
Ensure that malloced storage areas are freed on error paths.
This resolves Bug https://bugs.linaro.org/show_bug.cgi?id=2830
Signed-off-by: Bill Fischofer
---
helper/test/cuckootable.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/helper/test/cuckootable.c b/helper
The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
used in C++ programs this needs to expand to static_assert().
This resolves Bug https://bugs.linaro.org/show_bug.cgi?id=2852
Reported-by: Moshe Tubul
Signed-off-by: Bill Fischofer
---
Changes for v4:
- Remove extraneous
The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
used in C++ programs this needs to expand to static_assert().
This resolves Bug https://bugs.linaro.org/show_bug.cgi?id=2852
Reported-by: Moshe Tubul
Signed-off-by: Bill Fischofer
---
Changes for v3:
- Support older
Resolve Bug https://bugs.linaro.org/show_bug.cgi?id=2865 by adding
missing doxygen documentation for helper table functions
Signed-off-by: Bill Fischofer
---
helper/include/odp/helper/odph_cuckootable.h | 80 ++---
helper/include/odp/helper/odph_hashtable.h | 75
On Wed, Feb 1, 2017 at 7:14 AM, Mike Holmes wrote:
> Signed-off-by: Mike Holmes
> ---
> helper/Makefile.am | 1 +
> helper/include/odp/helper/ip.h | 48 +
> helper/include/odp/helper/ip_inlines.h | 65
> ++
> h
On Wed, Feb 1, 2017 at 7:14 AM, Mike Holmes wrote:
> Signed-off-by: Mike Holmes
> ---
> helper/Makefile.am | 1 +
> helper/chksum.c| 4 ++
> helper/include/odp/helper/chksum.h | 59 ++-
> helper/include/odp/helper/
On Wed, Feb 1, 2017 at 7:14 AM, Mike Holmes wrote:
> Signed-off-by: Mike Holmes
> ---
> configure.ac | 6 +
> helper/Makefile.am | 1 +
> helper/include/odp/helper/.gitignore | 1 +
> helper/include/odp/helper/static_inline
On Wed, Feb 1, 2017 at 1:14 PM, Brian Brooks wrote:
> On 02/01 08:30:22, Bill Fischofer wrote:
>> The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
>> used in C++ programs this needs to expand to static_assert().
>>
>> This resolves Bug https://b
On Wed, Feb 1, 2017 at 9:29 AM, Maxim Uvarov wrote:
> I have some theoretical question to think...
>
>
> rstp on hardware switches supported with moving port into disabled mode
> where it do not accept packets but only stp packets can pass-thought to
> rstp daemon app which decided which where pat
On Wed, Feb 1, 2017 at 8:24 AM, Elo, Matias (Nokia - FI/Espoo)
wrote:
>>
>> Thanks, Matias. For your benchmarking it would be good to get a
>> comparison run without those options to better quantify the overhead
>> of ABI compatibility mode. Right now we're taking a strict approach to
>> ABI so as
The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
used in C++ programs this needs to expand to static_assert().
This resolves Bug https://bugs.linaro.org/show_bug.cgi?id=2852
Signed-off-by: Bill Fischofer
---
Changes for v2:
- Update C++ test case to include helper apis to
low those ABI rules. :)
>
> Maxim.
>
> On 1 February 2017 at 01:34, Bill Fischofer
> wrote:
>>
>> On Thu, Jan 26, 2017 at 9:31 PM, Brian Brooks
>> wrote:
>> > Signed-off-by: Brian Brooks
>>
>> Reviewed-by: Bill Fischofer
>>
>> &g
Thanks, Petri. This now seems fully bisectable. For this series:
Reviewed-and-tested-by: Bill Fischofer
On Tue, Jan 31, 2017 at 2:07 AM, Petri Savolainen
wrote:
> Added initial version of couple of ABI specification files. These files define
> types/values that are implementation speci
For this series:
Reviewed-and-tested-by: Bill Fischofer
On Tue, Jan 31, 2017 at 12:42 AM, Balasubramanian Manoharan
wrote:
> Adds test case for ipv6 source and destination address matching
>
> Signed-off-by: Balasubramanian Manoharan
> ---
> v2: clang compilation fix
On Thu, Jan 26, 2017 at 9:31 PM, Brian Brooks wrote:
> Signed-off-by: Brian Brooks
Reviewed-by: Bill Fischofer
> ---
> platform/linux-generic/arch/arm/odp/api/cpu_arch.h | 6 ++
> 1 file changed, 6 insertions(+)
>
> diff --git a/platform/linux-generic/arch/arm/odp/a
On Tue, Jan 31, 2017 at 9:30 AM, Elo, Matias (Nokia - FI/Espoo)
wrote:
>
>> Thanks, with this configuration it is working.
>> My doubt here is:
>> Why we are not using shared lib, and abi-compat ??
>
> With --disable-shared and --disable-abi-compat you get the best performance
> but they are not
git bisect shows this is the offending patch:
88ed645e7378b955c580ec2b8ce9b846a5fb48f0 is the first bad commit
commit 88ed645e7378b955c580ec2b8ce9b846a5fb48f0
Author: Mike Holmes
Date: Mon Jan 23 14:46:06 2017 -0500
helper: use odph_api.h for test include for unexported files
Adding t
On Mon, Jan 30, 2017 at 4:32 AM, Joe Savage wrote:
> Add an example application implementing lock-free IPv4 fragmentation
> and reassembly functionality using ODP's packet "concat" and "split".
>
> Signed-off-by: Joe Savage
Reviewed-and-tested-by
Checkpatch still has some issues with this:
ill@Ubuntu15:~/linaro/review$ ./scripts/checkpatch.pl
0001-example-add-IPv4-fragmentation-reassembly-example.patch
WARNING: 'DONT' may be misspelled - perhaps 'DON'T'?
#1056: FILE: example/ipfragreass/odp_ipfragreass_ip.h:20:
+#define IP_FRAG_DONT 0x4000
Add a section to the ODP User Guide discussing application portability
considerations and the use of source and binary portability options
provided by ODP.
Signed-off-by: Bill Fischofer
---
doc/users-guide/users-guide.adoc | 120 +++
1 file changed, 120
On Mon, Jan 30, 2017 at 6:56 AM, Balasubramanian Manoharan
wrote:
> Adds test case for ipv6 source and destination address matching
>
> Signed-off-by: Balasubramanian Manoharan
> ---
> .../validation/api/classification/classification.h | 21 ++
> .../api/classification/odp_classification_common
ub also and not have to point at a personal location ?
If we're going to fork, agree we should do so "officially" rather than
in a personal repo.
>
> Mike
>
> On 29 January 2017 at 17:03, Bill Fischofer
> wrote:
>>
>> On Fri, Jan 27, 2017 at 1:21
Resolve Bug https://bugs.linaro.org/show_bug.cgi?id=2862 by checking
pointer validity before dereferencing.
Signed-off-by: Bill Fischofer
---
helper/iplookuptable.c | 20 ++--
1 file changed, 14 insertions(+), 6 deletions(-)
diff --git a/helper/iplookuptable.c b/helper
On Mon, Jan 30, 2017 at 6:40 AM, Ola Liljedahl wrote:
> On 30 January 2017 at 05:18, Honnappa Nagarahalli
> wrote:
>> On 26 January 2017 at 15:10, Bill Fischofer
>> wrote:
>>> On Wed, Jan 25, 2017 at 11:13 PM, Honnappa Nagarahalli
>>> wrote:
>>>&
On Sat, Jan 28, 2017 at 1:40 PM, Honnappa Nagarahalli
wrote:
> Hi,
> The current documentation is as follows:
>
> /**
> * Schedule group destroy
> *
> * Destroys a schedule group. All queues belonging to the schedule group must
> * be destroyed before destroying the group. Other operations o
On Fri, Jan 27, 2017 at 1:21 PM, Maxim Uvarov wrote:
> some time curl fails to download cunit from sf.net or it's
> mirrors. Try 10 times with 1 second delay.
>
> Signed-off-by: Maxim Uvarov
> ---
> .travis.yml | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/.travis.yml
On Sun, Jan 29, 2017 at 7:57 AM, Vamsi Krishna wrote:
> Hi,
>
> In the current ODP PKTIO APIs, there are calls to get mtu and mac
> information of the pktio interfaces. I did not find any APIs to configure
> the mtu and mac address of the pktio interface from the application.
>
> Could someone com
Thanks. I'll look this over and I've added this to the agenda for
Monday's ARCH call.
On Fri, Jan 27, 2017 at 7:24 AM, Ola Liljedahl wrote:
> On 27 January 2017 at 14:22, Ola Liljedahl wrote:
>
>> Anyone with a linaro.org email address should now be able to access the
>> document and make commen
Ok, if you want to handle it that way.
For this series:
Reviewed-and-tested-by: Bill Fischofer
On Fri, Jan 27, 2017 at 8:12 AM, Maxim Uvarov wrote:
> git am -3 *.patch
>
> -3 correctly applies on top of current master.
>
> Maxim.
>
> On 01/27/17 17:04, Bill Fischofer wro
Maxim, I tried to apply but part 2 fails with:
Applying: test: linux-gen: vlan insertion: reduce memory pool
Checking patch test/linux-generic/mmap_vlan_ins/mmap_vlan_ins.c...
error: while searching for:
#include
#include
#define POOL_NUM_PKT 8192
#define POOL_SEG_LEN 1856
#define MAX_PKT_BURST
edule_multi() call.
>
> On 25 January 2017 at 04:16, Ola Liljedahl wrote:
>>
>>
>> On 25 January 2017 at 06:34, Honnappa Nagarahalli
>> wrote:
>>>
>>> On 24 January 2017 at 19:16, Bill Fischofer
>>> wrote:
>>> > On Tue, Jan 24, 2
NIC (from the DPDK PMD) in bursts bigger than
> 4 bytes if the stack is very highly loaded. The bursts seen in the
> ODP API is then another matter.
>
> Janne
>
>
>> -Original Message-
>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>
end to move that capability to odp_init_global() time
at some point.
>
> Thanks,
> Geoffrey Blake
>
> On 1/25/17, 1:48 PM, "Bill Fischofer" wrote:
>
> On Wed, Jan 25, 2017 at 10:16 AM, Mike Holmes
> wrote:
> > Relevant to ODP community CC'ed
Additional review comments based on testing:
On Wed, Jan 25, 2017 at 8:26 AM, Joe Savage wrote:
> Add an example application implementing lock-free IPv4 fragmentation
> and reassembly functionality using ODP's packet "concat" and "split".
>
> Signed-off-by: Joe Savage
> ---
> (This code contribu
On Wed, Jan 25, 2017 at 2:03 PM, Bill Fischofer
wrote:
> I agree with all of Maxim's comments. In addition, attempting to
> compile with clang gives these errors:
>
> Making all in fragreass
> make[2]: Entering directory '/home/bill/linaro/review/example/fragreass
I agree with all of Maxim's comments. In addition, attempting to
compile with clang gives these errors:
Making all in fragreass
make[2]: Entering directory '/home/bill/linaro/review/example/fragreass'
CC odp_fragreass-odp_fragreass.o
In file included from odp_fragreass.c:21:
In file includ
On Wed, Jan 25, 2017 at 10:16 AM, Mike Holmes wrote:
> Relevant to ODP community CC'ed
>
> On 25 January 2017 at 11:14, Sorin Vultureanu
> wrote:
>> Hi,
>>
>>
>>
>> It looks like your issue is with ODP API
>>
>>
>>
>> IMO odp_schedule_multi() should look like this:
>>
>> odp_schedule_multi() sho
I also think this cleanup is a step in the right direction. If the
helpers are to be an adjunct ODP package that can be distributed
alongside ODP then this is what's needed. Otherwise they are just a
disjoint set of unrelated individual files that we should not try to
package at all. But that would
On Wed, Jan 25, 2017 at 6:58 AM, Ola Liljedahl wrote:
>
>
> On 25 January 2017 at 13:39, Bill Fischofer
> wrote:
>>
>> On Wed, Jan 25, 2017 at 4:22 AM, Ola Liljedahl
>> wrote:
>> >
>> >
>> > On 24 January 2017 at 15:30, Nikhil Agarwal
On Wed, Jan 25, 2017 at 4:22 AM, Ola Liljedahl wrote:
>
>
> On 24 January 2017 at 15:30, Nikhil Agarwal wrote:
>>
>>
>>
>> -Original Message-
>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Bill
>> Fischofer
>> Se
On Wed, Jan 25, 2017 at 12:57 AM, Maxim Uvarov wrote:
>
>
> On 25 January 2017 at 04:46, Bill Fischofer
> wrote:
>>
>> The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
>> used in C++ programs this needs to expand to static_assert().
On Mon, Jan 23, 2017 at 2:25 AM, Christophe Milard
wrote:
> Hence fixing CID 174663
> (Fixes https://bugs.linaro.org/show_bug.cgi?id=2827)
>
> Signed-off-by: Christophe Milard
Reviewed-by: Bill Fischofer
> ---
>
> since V1:
> -print strerror(errno) and sets __odp_e
ODP Bug https://bugs.linaro.org/show_bug.cgi?id=2852 opened to track this
issue and Patch http://patches.opendataplane.org/patch/7973/ posted to
address it. It's a simple fix, but a testing oversight on our part. Thanks
for finding this.
On Tue, Jan 24, 2017 at 7:22 PM, Bill Fischofer
The ODP_STATIC_ASSERT() macro expands to _Static_assert(), however when
used in C++ programs this needs to expand to static_assert().
This resolves Bug https://bugs.linaro.org/show_bug.cgi?id=2852
Signed-off-by: Bill Fischofer
---
platform/linux-generic/include/odp/api/debug.h | 6 --
1
Copying this to the ODP thread since these are issues with the ODP Helpers,
not OFP. ODP APIs are designed to be callable from C++ but it appears that
the ODP helpers have not been properly tested in a C++ environment and
presumably OFP makes use of them which results in C++ callers attempting to
u
On Tue, Jan 24, 2017 at 8:30 AM, Nikhil Agarwal wrote:
>
>
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Bill
> Fischofer
> Sent: Tuesday, January 24, 2017 1:15 AM
> To: Nikhil Agarwal
> Cc: Kevin Wang ; lng-odp-forwa
Mike, the adoc rendering doesn't appear to properly support includes.
If you look at the User Guide, for example, it stops after rendering
the initial section and then the include files are separate links.
This is in addition to the known issue with .svg diagrams.
On Tue, Jan 24, 2017 at 1:01 PM,
Ping. This still needs a review.
On Sat, Jan 14, 2017 at 12:24 PM, Bill Fischofer
wrote:
> As reported in Bug https://bugs.linaro.org/show_bug.cgi?id=2816 packet
> alloc/free performance for non-references appears significantly degraded
> after the introduction of packet reference sup
> pre-reserved area), which is guaranteed to succeed, and finally removing
> the non-fixed mapping.
> The non-fix mapping just acts as a failure guard when the proper maping
> is done.
> Fixes https://bugs.linaro.org/show_bug.cgi?id=2834
> (but very hard to trigger i.e. to prove)
&
OK. For the v2 series:
Reviewed-and-tested-by: Bill Fischofer
On Mon, Jan 23, 2017 at 5:09 AM, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>> > +#ifndef ODP_ABI_EVENT_H_
>> > +#define ODP_ABI_EVENT_H_
>> > +
>> > +#ifdef __cplusplus
>> > +ex
Moving this discussion on the ODP mailing list rather than the
Internal list as that way it will be archived.
The existing ODP controls over scheduling include schedule groups as
well as queue priorities. The former is a strict requirement (threads
can only receive events from queues that belong t
kes it easy to track closing a bug
>
>>
>> Signed-off-by: Christophe Milard
Reviewed-and-tested-by: Bill Fischofer
>> ---
>> platform/linux-generic/_ishm.c | 7 ++-
>> 1 file changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/platform/linux
Signed-off-by: Christophe Milard
Reviewed-and-tested-by: Bill Fischofer
> ---
> platform/linux-generic/_ishm.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/platform/linux-generic/_ishm.c b/platform/linux-generic/_ishm.c
> index f889834..3797f
In general these look good, but still have some problems:
1. The are checkpatch formatting issues in parts 1, 2, and 3. Please
run ./scripts/checkpatch on your patches before submitting and ensure
that they are clean.
2. Each new revision of a patch should increment the version number.
This one r
On Fri, Jan 20, 2017 at 8:11 AM, Petri Savolainen
wrote:
> Used event API as the first example of an ABI spec file. Used
> the same default architecture file initially for all
> architectures. Default ABI files avoid multiple copies
> of the same definition.
>
> Signed-off-by: Petri Savolainen
>
On Fri, Jan 20, 2017 at 4:28 AM, Yi He wrote:
> For this patch series, I've reviewed and tested good against master branch:
>
> Reviewed-and-tested-by: Yi He
>
> On api-next branch it encounters compilation error, what's our typical
> solution for this kind of problem?
> _ishmpool.c:209:3: error:
For the v2 series:
Reviewed-and-tested-by: Bill Fischofer
On Fri, Jan 20, 2017 at 3:14 AM, Petri Savolainen
wrote:
> Implement round up to power of two as a macro. Macro
> enables preprocessor to calculate the constant value and
> avoid code generation. Also dependency to __builtin_c
On Fri, Jan 20, 2017 at 8:22 AM, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>
>
>> -Original Message-----
>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>> Sent: Friday, January 20, 2017 4:13 PM
>> To: Savolainen, Petri (Nokia - FI/Espoo) >
On Fri, Jan 20, 2017 at 3:33 AM, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>> -Original Message-
>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>> Sent: Thursday, January 19, 2017 11:53 PM
>> To: Savolainen, Petri (Nokia - FI/Espoo) > labs.
gt;> wrote:
>>>> -Original Message-
>>>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>>>> Sent: Thursday, January 19, 2017 11:53 PM
>>>> To: Savolainen, Petri (Nokia - FI/Espoo) >>> labs.com>
>>>> Cc: l
This looks good, however it should be against api-next. Use git
format-patch origin/api-next --subject-prefix="API-NEXT PATCH"
Also, part 2 should not have the same commit short log. Part 1 is
fine, though by convention ODP uses all lower case in short logs so it
should be "api: timer: add odp_tim
On Thu, Jan 19, 2017 at 9:28 AM, Petri Savolainen
wrote:
> Implement round up to power of two as a macro. Macro
> enables preprocessor to calculate the constant value and
> avoid code generation. Also dependency to __builtin_clz()
> is not needed anymore.
>
> Signed-off-by: Petri Savolainen
> ---
This series represents a significant structural change and as such
should go against api-next, not master. Patches aimed at master should
be bug fixes, not something that introduces new structure development.
While the goals here seem good, I'm concerned that at this stage
applying this removes st
On Thu, Jan 19, 2017 at 7:23 AM, Joe Savage wrote:
> Hey,
>
> I'm just finalising a patch to add a lock-free IPv4 fragmentation and
> reassembly example to ODP, and it has occurred to me that as the whole
> examples directory is compiled with odp-linux by default, I ought to be
> careful about sup
On Thu, Jan 19, 2017 at 10:03 AM, Stanislaw Kardach wrote:
>
>
> Best Regards,
> Stanislaw Kardach
>
>
> On 01/19/2017 04:57 PM, Bill Fischofer wrote:
>>
>> On Thu, Jan 19, 2017 at 7:17 AM, Stanislaw Kardach
>> wrote:
>>>
>>> Hi all,
&
On Thu, Jan 19, 2017 at 7:17 AM, Stanislaw Kardach wrote:
> Hi all,
>
> While going through thread and scheduler APIs I've stumbled on one
> uncertainty in ODP API that I could not find straight solution to.
>
> Does ODP allow scheduling packets from an ODP_THREAD_CONTROL thread?
When a thread ca
Resolve Bug https://bugs.linaro.org/show_bug.cgi?id=2826 by adding
an explicit check of the rc from odp_pool_capability() for consistency
with other code that calls this routine.
Signed-off-by: Bill Fischofer
---
platform/linux-generic/odp_pool.c | 3 ++-
1 file changed, 2 insertions(+), 1
Petri, did Part 1/8 of this series not get posted? It doesn't seem to
be on either the mailing list or patchworks.
Bill
On Wed, Jan 18, 2017 at 6:56 AM, Petri Savolainen
wrote:
> Added initial version of couple of ABI specification files. These files define
> types/values that are implementation
+cc ODP mailing list. These sort of discussions belong on the mailing
list as that's where those most knowledgeable and impacted by API
changes will see it.
ODP implementations are free to batch/burst individual
odp_crypto_operation() calls but extensions to support application
configuration of th
First off, thanks for this submission, however this series has a
number of problems:
1. It does not compile due to the incorrect specification of
There is no /api/ level in these
includes.
2. Numerous checkpatch issues throughout. Please run
./scripts/checkpatch *.patch before submitting and ens
On Mon, Jan 16, 2017 at 2:45 AM, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>
>
>> -Original Message-----
>> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
>> Sent: Sunday, January 15, 2017 5:09 PM
>> To: Savolainen, Petri (Nokia - FI/Esp
tions (v1
= 538, v3 = 268). There's probably still some room for improvement
here, but clearly this is a tractable problem at this point.
I'll work my experiments with odp_bench_packet into a patch set that
can be used to improve this very worthwhile framework.
On Sat, Jan 14, 2017 at 2:0
Please ignore v2. It had a memory leak that Matias' excellent
odp_bench_packet test found. A corrected v3 is posted at
http://patches.opendataplane.org/patch/7879/
On Sat, Jan 14, 2017 at 10:04 AM, Bill Fischofer
wrote:
> Hi Petri,
>
> I've posted v2 of the tun
As reported in Bug https://bugs.linaro.org/show_bug.cgi?id=2816 packet
alloc/free performance for non-references appears significantly degraded
after the introduction of packet reference support. Add fastpaths for
non-reference packet frees to optimize this.
Signed-off-by: Bill Fischofer
i: Benchmark index to run indefinitely
>
> The application prints benchmark results as a table, which can be copied
> as is to a spreadsheet application.
>
> Signed-off-by: Matias Elo
Reviewed-and-tested-by: Bill Fischofer
> ---
> test/common_plat/performance/.gitigno
ee routines now call buffer_free_to_pool() directly.
If you can confirm that these are heading in the right direction we
can see what else might be needed from a tuning perspective.
Thanks.
On Thu, Jan 12, 2017 at 4:15 PM, Bill Fischofer
wrote:
> I've posted patch http://patches.opend
As reported in Bug https://bugs.linaro.org/show_bug.cgi?id=2816 packet
alloc/free performance for non-references appears significantly degraded
after the introduction of packet reference support. Add fastpaths for
non-reference packet frees to optimize this.
Signed-off-by: Bill Fischofer
Signed-off-by: Bill Fischofer
---
CHANGELOG | 178 ++
1 file changed, 178 insertions(+)
diff --git a/CHANGELOG b/CHANGELOG
index 17afe44..72bf225 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,181 @@
+== OpenDataPlane (1.13.0.0
On Fri, Jan 13, 2017 at 9:09 AM, Maxim Uvarov wrote:
> Hello,
>
> Bill found interesting tool to print structures layout for some elf
> binary. Please find output in text file attached to that email. Hope
> it's useful for everybody who is working on performance things.
>
> Best regards,
> Maxim.
701 - 800 of 4369 matches
Mail list logo