Re: [lng-odp] [PATCH v2] linux-generic: fixup cache line size API

On 07/22 08:39:38, Christophe Milard wrote: > On 19 July 2016 at 17:23, Brian Brooks wrote: > > On 07/21 18:36:22, Christophe Milard wrote: > >> Also... Can we do the same with the other symbols: for instance > >> odp_cpu_pause() as _odp_cpu_pause() in /arch/XXX, and the definition > >> of odp_cpu

Re: [lng-odp] [PATCH v2] linux-generic: fixup cache line size API

On 07/22 15:00:49, Christophe Milard wrote: > an another strange thing: > Shouldn't the statement: > #include > defining _ODP_CACHE_LINE_SIZE be at the beginning of > platform/linux-generic/include/odp/api/align.h whichb uses it? > > Seems to work, though. But I really wonder how :-) This is wha

Re: [lng-odp] [API-NEXT PATCH] helper: cuckootable: fix func with no return

ping On 15 July 2016 at 07:43, Mike Holmes wrote: > fix non-void function with the possibility of no return statement > > Signed-off-by: Mike Holmes > --- > helper/cuckootable.c | 10 +++--- > 1 file changed, 7 insertions(+), 3 deletions(-) > > diff --git a/helper/cuckootable.c b/helper/cu

Re: [lng-odp] [PATCH] configure.ac: fix have_pcap AM_CONDITIONAL

On 18 July 2016 at 10:40, Zoltan Kiss wrote: > This causes "./configure: line 20048: test: =: unary operator expected" > error > messages, while the generated code was like this: > > if test $pktio_dpdk_support = yes ; then > PKTIO_DPDK_TRUE= > PKTIO_DPDK_FALSE='#' > else > ... > > Signed-of

Re: [lng-odp] [PATCH] configure.ac: fix have_pcap AM_CONDITIONAL

Ping On 18/07/16 15:40, Zoltan Kiss wrote: This causes "./configure: line 20048: test: =: unary operator expected" error messages, while the generated code was like this: if test $pktio_dpdk_support = yes ; then PKTIO_DPDK_TRUE= PKTIO_DPDK_FALSE='#' else ... Signed-off-by: Zoltan Kiss --

Re: [lng-odp] [API-NEXT PATCH v2] doc: driver-guide: initial revision

Not sure I really like the term "driver-api" for "driver interfce", but we'll see how it evolves. If we understand each-other, it is fine :-) Reviewed-by: Christophe Milard On 11 July 2016 at 18:46, Mike Holmes wrote: > Add an initial driver interface document structure for the existing driver

Re: [lng-odp] [PATCH v2] linux-generic: fixup cache line size API

an another strange thing: Shouldn't the statement: #include defining _ODP_CACHE_LINE_SIZE be at the beginning of platform/linux-generic/include/odp/api/align.h whichb uses it? Seems to work, though. But I really wonder how :-) On 19 July 2016 at 12:18, Brian Brooks wrote: > Define the ODP API

Re: [lng-odp] [PATCH v2] linux-generic: fixup cache line size API

Another comment: Do we have now any reason to keep these paths: platform/linux-generic/arch/powerpc/odp/api/* I mean, if we intend to suppress the api definition things from this place, should this simply be: platform/linux-generic/arch/powerpc/* ? Christophe On 19 July 2016 at 12:18, Brian Broo

[lng-odp] [Bug 2436] CI FAIL: pktio_ipc/pktio_ipc_run.sh

https://bugs.linaro.org/show_bug.cgi?id=2436 Mike Holmes changed: What|Removed |Added Status|UNCONFIRMED |CONFIRMED Ever confirmed|0

[lng-odp] [Bug 2436] CI FAIL: pktio_ipc/pktio_ipc_run.sh

https://bugs.linaro.org/show_bug.cgi?id=2436 Mike Holmes changed: What|Removed |Added Priority|--- |High -- You are receiving this mail because: Yo

[lng-odp] [PATCH] helper: test: fixed gitignore

Obviously a miss from commit c4aefb88d31452b3add8cf16f9eef152525c3e93 Signed-off-by: Christophe Milard --- helper/test/.gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/helper/test/.gitignore b/helper/test/.gitignore index 482fdb5..e5b6a0f 100644 --- a/helper/test/.gitignore +++ b/h

Re: [lng-odp] [API-NEXT PATCH v2] doc: driver-guide: initial revision

Ping - Christophe, is this ok for you, Bill did review already On 11 July 2016 at 12:46, Mike Holmes wrote: > Add an initial driver interface document structure for the existing driver > framework. > > Signed-off-by: Mike Holmes > Reviewed-and-tested-by: Bill Fischofer > --- > v2 > use seper

Re: [lng-odp] [PATCHv3] linux-gen: Makefile: Move EXTRA_DIST to to platform/Makefile.inc

Merged, Maxim. On 07/22/16 13:52, Anders Roxell wrote: Make it easier for other platforms like odp-dpdk to reuse arch specific files in EXTRA_DIST for "free". Reported-by: Fathi Boudra Signed-off-by: Anders Roxell --- platform/Makefile.inc | 17 + platform/linu

[lng-odp] [PATCHv3] test: update in readme file

mostly updating paths following new test structure Signed-off-by: Christophe Milard --- test/README | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/test/README b/test/README index 857c366..f4886d3 100644 --- a/test/README +++ b/test/README @@ -3,14 +3,15 @@ All right

Re: [lng-odp] [PATCHv2] test: update in readme file

Forget it! sent the old contents! sorry! On 22 July 2016 at 14:47, Christophe Milard wrote: > mostly updating paths following new test structure > > Signed-off-by: Christophe Milard > --- > test/README | 9 + > 1 file changed, 5 insertions(+), 4 deletions(-) > > diff --git a/test/README

[lng-odp] [PATCHv2] test: update in readme file

mostly updating paths following new test structure Signed-off-by: Christophe Milard --- test/README | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/test/README b/test/README index 857c366..ad82943 100644 --- a/test/README +++ b/test/README @@ -3,14 +3,15 @@ All right

[lng-odp] [PATCH v2] dox:application: add examples to guide

Signed-off-by: Mike Holmes --- v2 alphabetical order (Maxim) doc/application-api-guide/examples.dox | 37 +++--- 1 file changed, 34 insertions(+), 3 deletions(-) diff --git a/doc/application-api-guide/examples.dox b/doc/application-api-guide/examples.dox index 2dc

Re: [lng-odp] [PATCH] dox:application: add examples to guide

The output is order independent when doxygen uses this, I don't see a big need for it to be ordered, but if it bothers you I will sort the existing list and make it part of this as well, no worries :) I do think we should have a file per example that contains a useful description, the current pla

[lng-odp] [PATCH] example: odp_l2fwd place readme with symlink

odp_l2fwd is standalone application which can be used as example or as performance test. Hiding it inside performance tests might be not new users friendly way. Create symlink with small readme file for better finding it. Signed-off-by: Maxim Uvarov --- example/l2fwd/README | 8 ex

[lng-odp] [PATCHv3] linux-gen: Makefile: Move EXTRA_DIST to to platform/Makefile.inc

Make it easier for other platforms like odp-dpdk to reuse arch specific files in EXTRA_DIST for "free". Reported-by: Fathi Boudra Signed-off-by: Anders Roxell --- platform/Makefile.inc | 17 + platform/linux-generic/Makefile.am | 17 - 2 files change

Re: [lng-odp] [PATCH] dox:application: add examples to guide

On 07/20/16 13:49, Mike Holmes wrote: Signed-off-by: Mike Holmes --- doc/application-api-guide/examples.dox | 25 + 1 file changed, 25 insertions(+) diff --git a/doc/application-api-guide/examples.dox b/doc/application-api-guide/examples.dox index 2dccd7f..2398d15 10

Re: [lng-odp] [PATCHv2] linux-gen: Makefile: Move EXTRA_DIST to to platform/Makefile.inc

On 07/21/16 16:02, Anders Roxell wrote: Make it easier for other platforms like odp-dpdk to reuse arch specific files in EXTRA_DIST for "free". Reported-by: Fathi Boudra Signed-off-by: Anders Roxell --- platform/Makefile.inc | 17 + platform/linux-generic/Makefi

Re: [lng-odp] [PATCH] test: update in readme file

On 07/22/16 12:21, Christophe Milard wrote: mostly updating paths following new test structure Signed-off-by: Christophe Milard --- test/README | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/test/README b/test/README index 857c366..ad82943 100644 --- a/test/READ

Re: [lng-odp] [PATCH] linux-dpdk/m4/configure.m4: move m4_include above ac_check_*

Applied to ODP-DPDK (with subject line formatted a bit) On 21/07/16 15:42, Anders Roxell wrote: Signed-off-by: Anders Roxell --- platform/linux-dpdk/m4/configure.m4 | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/platform/linux-dpdk/m4/configure.m4 b/platform/

Re: [lng-odp] [PATCH] linux-generic: pool: set maximum number of pools to 64

Patch merged, tests we can add later. Maxim. On 07/19/16 15:36, Mike Holmes wrote: On 19 July 2016 at 08:30, Maxim Uvarov > wrote: That increase sizeof (pool_table_t) from 136192 to 544768 bytes, ie. takes additional 400kb stack memory. I think t

[lng-odp] [PATCH v3 1/4] linux-gen: pool: optimize thread local buffer cache

Optimize local buffer cache performance which is critical to many use cases - including packet IO. Main parts of the optimization are: * Local cache implemented as an array of buf_hdr pointers, instead of a linked list (which causes a lot of cache misses) * Alloc and free N buffers per operat

[lng-odp] [PATCH v3 2/4] linux-gen: pktio: use multi alloc

Modify dpdk pktio to take advantage of multi-alloc/free. Others pktios do alloc/free still one packet at a time. For example, l2fwd test application packet throughput is increased about 10% (with dpdk pktio). Signed-off-by: Petri Savolainen Signed-off-by: Matias Elo --- V2: - Allocate MTU sized

[lng-odp] [PATCH v3 3/4] linux-gen: netmap: use multi alloc/free

Modify netmap pktio to take advantage of multi-alloc/free. Signed-off-by: Matias Elo --- platform/linux-generic/include/odp_packet_netmap.h | 6 + platform/linux-generic/pktio/netmap.c | 126 - 2 files changed, 78 insertions(+), 54 deletions(-) diff --git a/p

[lng-odp] [PATCH v3 4/4] performance: odp_scheduling: use multi alloc/free buffer functions

Use odp_buffer_alloc_multi() and odp_buffer_free_multi() functions where appropriate for improved performance. Signed-off-by: Matias Elo --- test/common_plat/performance/odp_scheduling.c | 50 ++- 1 file changed, 33 insertions(+), 17 deletions(-) diff --git a/test/common

[lng-odp] [PATCH] test: update in readme file

mostly updating paths following new test structure Signed-off-by: Christophe Milard --- test/README | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/test/README b/test/README index 857c366..ad82943 100644 --- a/test/README +++ b/test/README @@ -3,14 +3,15 @@ All right