Re: [lng-odp] [RFC 1/2] api: classification: Add queue group to classification

[Radosław Biernacki]

Hi

I cannot find this patch either next nor api-next, so I assume that it was
not accepted yet.


2016-04-25 14:31 GMT+02:00 Balasubramanian Manoharan <
bala.manoha...@linaro.org>:

> Adds queue group to classification
>
> Signed-off-by: Balasubramanian Manoharan 
> ---
>  include/odp/api/spec/classification.h | 19 ++-
>  1 file changed, 18 insertions(+), 1 deletion(-)
>
> diff --git a/include/odp/api/spec/classification.h b/include/odp/api/spec/
> classification.h
> index 6eca9ab..cf56852 100644
> --- a/include/odp/api/spec/classification.h
> +++ b/include/odp/api/spec/classification.h
> @@ -126,6 +126,12 @@ typedef struct odp_cls_capability_t {
>
> /** A Boolean to denote support of PMR range */
> odp_bool_t pmr_range_supported;
> +
> +   /** A Boolean to denote support of queue group */
> +   odp_bool_t queue_group_supported;
> +
> +   /** A Boolena to denote support of queue */
> +   odp_bool_t queue_supported;
>  } odp_cls_capability_t;
>
>  /**
> @@ -162,7 +168,18 @@ typedef enum {
>   * Used to communicate class of service creation options
>   */
>  typedef struct odp_cls_cos_param {
> -   odp_queue_t queue;  /**< Queue associated with CoS */
> +   /** If True, odp_queue_t is linked with CoS,
> +* if False odp_queue_group_t is linked with CoS.
> +*/
> +   odp_bool_t enable_queue;
>

Since this flag defines either queue or queue_gropup field from union will
be used, it might be better to use some enum instead of bool.
Bool is OK for enable/disable flags but may be confusing when used as
selector type.


> +
> +   typedef union {
> +   /** Queue associated with CoS */
> +   odp_queue_t queue;
> +
> +   /** Queue Group associated with CoS */
> +   odp_queue_group_t queue_group;
> +   };
> odp_pool_t pool;/**< Pool associated with CoS */
> odp_cls_drop_t drop_policy; /**< Drop policy associated with
> CoS */
>  } odp_cls_cos_param_t;
> --
> 1.9.1
>
> ___
> lng-odp mailing list
> lng-odp@lists.linaro.org
> https://lists.linaro.org/mailman/listinfo/lng-odp
>

[lng-odp] [Bug 2510] ODP build system broken

[bugzilla-daemon]

https://bugs.linaro.org/show_bug.cgi?id=2510

Bill Fischofer  changed:

   What|Removed |Added

 CC||bill.fischo...@linaro.org

--- Comment #1 from Bill Fischofer  ---
This appears to be a duplicate of https://bugs.linaro.org/show_bug.cgi?id=2507
which has been resolved.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

Re: [lng-odp] [PATCHv3] test: linux-gen: add pcap playback test

[Anders Roxell]

On 2016-09-12 19:54, Bill Fischofer wrote:
> On Mon, Sep 12, 2016 at 10:24 AM, Mike Holmes 
> wrote:
> 
> > On 12 September 2016 at 10:43, Maxim Uvarov 
> > wrote:
> >
> > > if no comments than it can be merged?
> >
> >
> > Not without a review, I am interested in this capability but I have not
> > reviewed it.
> >
> 
> I had previously reviewed and tested v2 of this patch. v3 just incorporates
> some additional comments by Anders. If Anders has no objection, my review
> can be propagated to v3.

I don't think we resolved this issue raised [1], and you (Bill) had a
comment about not screwing up performance while tinkering with it [2].

Cheers,
Anders
[1] https://lists.linaro.org/pipermail/lng-odp/2016-August/025421.html
[2] https://lists.linaro.org/pipermail/lng-odp/2016-August/025407.html

> 
> 
> >
> > I really believe ODP shoudl have as part of the validation the ability to
> > pass packet streams though and confirm the exact result on the far side,
> > Krishna ran into this issue with crypto on odp-dpdk which was giving
> > different results.
> >
> >
> > >
> > >
> > > Maxim.
> > >
> > > On 08/29/16 10:50, Maxim Uvarov wrote:
> > >
> > >> add pcap play back test which takes 2 arguments: 1 - pcap file,
> > >> 2 - packet mask to match. Intend is to test odp with different
> > >> input traffic to check internal implementation functions. In
> > >> current case it's test for vlan tag instertion for packet mmap:
> > >> pkt_mmap_vlan_insert().
> > >>
> > >> Signed-off-by: Maxim Uvarov 
> > >> ---
> > >>   v3: fix comments noted by Anders
> > >>
> > >>   test/linux-generic/.gitignore   |   1 +
> > >>   test/linux-generic/Makefile.am  |  10 +-
> > >>   test/linux-generic/cls/.gitignore   |   1 +
> > >>   test/linux-generic/cls/Makefile.am  |  13 ++
> > >>   test/linux-generic/cls/cls.c| 351
> > ++
> > >> ++
> > >>   test/linux-generic/cls/cls_main.c   |  12 ++
> > >>   test/linux-generic/cls/cls_suites.h |   1 +
> > >>   test/linux-generic/cls/vlan.pcap| Bin 0 -> 9728 bytes
> > >>   test/linux-generic/cls/vlan_run.sh  |  49 +
> > >>   test/linux-generic/m4/configure.m4  |   1 +
> > >>   10 files changed, 435 insertions(+), 4 deletions(-)
> > >>   create mode 100644 test/linux-generic/cls/.gitignore
> > >>   create mode 100644 test/linux-generic/cls/Makefile.am
> > >>   create mode 100644 test/linux-generic/cls/cls.c
> > >>   create mode 100644 test/linux-generic/cls/cls_main.c
> > >>   create mode 100644 test/linux-generic/cls/cls_suites.h
> > >>   create mode 100644 test/linux-generic/cls/vlan.pcap
> > >>   create mode 100755 test/linux-generic/cls/vlan_run.sh
> > >>
> > >> diff --git a/test/linux-generic/.gitignore
> > b/test/linux-generic/.gitignor
> > >> e
> > >> index 5dabf91..f65c7c1 100644
> > >> --- a/test/linux-generic/.gitignore
> > >> +++ b/test/linux-generic/.gitignore
> > >> @@ -1,3 +1,4 @@
> > >>   *.log
> > >>   *.trs
> > >>   tests-validation.env
> > >> +test_out.pcap
> > >> diff --git a/test/linux-generic/Makefile.am
> > >> b/test/linux-generic/Makefile.am
> > >> index 4660cf0..2554f8e 100644
> > >> --- a/test/linux-generic/Makefile.am
> > >> +++ b/test/linux-generic/Makefile.am
> > >> @@ -1,5 +1,8 @@
> > >>   include $(top_srcdir)/test/Makefile.inc
> > >>   TESTS_ENVIRONMENT += TEST_DIR=${top_builddir}/test/
> > >> common_plat/validation
> > >> +TEST_EXTENSIONS = .sh
> > >> +
> > >> +dist_check_SCRIPTS = run-test tests-validation.env $(LOG_COMPILER)
> > >> ALL_API_VALIDATION_DIR = ${top_builddir}/test/common_pl
> > >> at/validation/api
> > >>   @@ -37,11 +40,14 @@ TESTS = validation/api/pktio/pktio_run.sh \
> > >> SUBDIRS += validation/api/pktio\
> > >>validation/api/shmem\
> > >> +  cls\
> > >>pktio_ipc\
> > >>ring
> > >> if HAVE_PCAP
> > >>   TESTS += validation/api/pktio/pktio_run_pcap.sh
> > >> +TESTS += cls/vlan_run.sh
> > >> +dist_check_SCRIPTS += cls/vlan_run.sh cls/vlan.pcap
> > >>   endif
> > >>   if netmap_support
> > >>   TESTS += validation/api/pktio/pktio_run_netmap.sh
> > >> @@ -61,10 +67,6 @@ SUBDIRS += validation/api/pktio
> > >>   endif
> > >>   endif
> > >>   -TEST_EXTENSIONS = .sh
> > >> -
> > >> -dist_check_SCRIPTS = run-test tests-validation.env $(LOG_COMPILER)
> > >> -
> > >>   test_SCRIPTS = $(dist_check_SCRIPTS)
> > >> tests-validation.env:
> > >> diff --git a/test/linux-generic/cls/.gitignore
> > >> b/test/linux-generic/cls/.gitignore
> > >> new file mode 100644
> > >> index 000..5a652b7
> > >> --- /dev/null
> > >> +++ b/test/linux-generic/cls/.gitignore
> > >> @@ -0,0 +1 @@
> > >> +cls_main
> > >> diff --git a/test/linux-generic/cls/Makefile.am
> > >> b/test/linux-generic/cls/Makefile.am
> > >> new file mode 100644
> > >> index 000..43fb0bc
> > >> --- /dev/null
> > >> +++ b/test/linux-generic/cls/Makefile.am
> > >> @@ -0,0 +1,13 @@
> > >> +include ../Makefile.inc
> > >> +
> 

[lng-odp] [API-NEXT PATCHv2] api: odp_time_sub

[Maxim Uvarov]

Previously we had odp_time_sum but missing call to subtract
some value from odp_time_t. Because on different platforms it
can be implemented differently we need this call.

Signed-off-by: Maxim Uvarov 
---
 include/odp/api/spec/time.h   | 10 ++
 platform/linux-generic/odp_time.c | 17 +
 2 files changed, 27 insertions(+)

diff --git a/include/odp/api/spec/time.h b/include/odp/api/spec/time.h
index fcc94c9..5e94b70 100644
--- a/include/odp/api/spec/time.h
+++ b/include/odp/api/spec/time.h
@@ -87,6 +87,16 @@ odp_time_t odp_time_diff(odp_time_t t2, odp_time_t t1);
 odp_time_t odp_time_sum(odp_time_t t1, odp_time_t t2);
 
 /**
+ * Time subtract
+ *
+ * @param t1Time stamp to subtract from
+ * @param t2Value to subtract
+ *
+ * @return Time stamp t1 - t2
+ */
+odp_time_t odp_time_sub(odp_time_t t1, odp_time_t t2);
+
+/**
  * Convert time to nanoseconds
  *
  * @param time  Time
diff --git a/platform/linux-generic/odp_time.c 
b/platform/linux-generic/odp_time.c
index 81e0522..7fef46f 100644
--- a/platform/linux-generic/odp_time.c
+++ b/platform/linux-generic/odp_time.c
@@ -81,6 +81,18 @@ static inline odp_time_t time_sum(odp_time_t t1, odp_time_t 
t2)
return time;
 }
 
+static inline odp_time_t time_sub(odp_time_t t1, odp_time_t t2)
+{
+   uint64_t ns;
+   odp_time_t time;
+
+   ns = time_to_ns(t1) - time_to_ns(t2);
+   time.tv_sec = ns / ODP_TIME_SEC_IN_NS;
+   time.tv_nsec = ns - time.tv_sec * ODP_TIME_SEC_IN_NS;
+
+   return time;
+}
+
 static inline odp_time_t time_local_from_ns(uint64_t ns)
 {
odp_time_t time;
@@ -152,6 +164,11 @@ odp_time_t odp_time_sum(odp_time_t t1, odp_time_t t2)
return time_sum(t1, t2);
 }
 
+odp_time_t odp_time_sub(odp_time_t t1, odp_time_t t2)
+{
+   return time_sub(t1, t2);
+}
+
 uint64_t odp_time_local_res(void)
 {
return time_local_res();
-- 
2.7.1.250.gff4ea60

[lng-odp] [API-NEXT PATCH] api: odp_time_sub

[Maxim Uvarov]

Previously we had odp_time_sum but missing call to subtract
some value from odp_time_t. Because on different platforms it
can be implemented differently we need this call.

Signed-off-by: Maxim Uvarov 
---
 include/odp/api/spec/time.h | 10 ++
 platform/linux-generic/odp_time.c   | 17 +
 test/common_plat/validation/api/time/time.c |  2 +-
 3 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/include/odp/api/spec/time.h b/include/odp/api/spec/time.h
index fcc94c9..5e94b70 100644
--- a/include/odp/api/spec/time.h
+++ b/include/odp/api/spec/time.h
@@ -87,6 +87,16 @@ odp_time_t odp_time_diff(odp_time_t t2, odp_time_t t1);
 odp_time_t odp_time_sum(odp_time_t t1, odp_time_t t2);
 
 /**
+ * Time subtract
+ *
+ * @param t1Time stamp to subtract from
+ * @param t2Value to subtract
+ *
+ * @return Time stamp t1 - t2
+ */
+odp_time_t odp_time_sub(odp_time_t t1, odp_time_t t2);
+
+/**
  * Convert time to nanoseconds
  *
  * @param time  Time
diff --git a/platform/linux-generic/odp_time.c 
b/platform/linux-generic/odp_time.c
index 81e0522..7fef46f 100644
--- a/platform/linux-generic/odp_time.c
+++ b/platform/linux-generic/odp_time.c
@@ -81,6 +81,18 @@ static inline odp_time_t time_sum(odp_time_t t1, odp_time_t 
t2)
return time;
 }
 
+static inline odp_time_t time_sub(odp_time_t t1, odp_time_t t2)
+{
+   uint64_t ns;
+   odp_time_t time;
+
+   ns = time_to_ns(t1) - time_to_ns(t2);
+   time.tv_sec = ns / ODP_TIME_SEC_IN_NS;
+   time.tv_nsec = ns - time.tv_sec * ODP_TIME_SEC_IN_NS;
+
+   return time;
+}
+
 static inline odp_time_t time_local_from_ns(uint64_t ns)
 {
odp_time_t time;
@@ -152,6 +164,11 @@ odp_time_t odp_time_sum(odp_time_t t1, odp_time_t t2)
return time_sum(t1, t2);
 }
 
+odp_time_t odp_time_sub(odp_time_t t1, odp_time_t t2)
+{
+   return time_sub(t1, t2);
+}
+
 uint64_t odp_time_local_res(void)
 {
return time_local_res();
diff --git a/test/common_plat/validation/api/time/time.c 
b/test/common_plat/validation/api/time/time.c
index 530d5c0..3885b65 100644
--- a/test/common_plat/validation/api/time/time.c
+++ b/test/common_plat/validation/api/time/time.c
@@ -1,4 +1,4 @@
-/* Copyright (c) 2015, Linaro Limited
+b/* Copyright (c) 2015, Linaro Limited
  * All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
-- 
2.7.1.250.gff4ea60

Re: [lng-odp] [buggyPATCH] linux-generic: add vlan insertion test

[Maxim Uvarov]

ping,

there was question about autotools inside the patch.
Can somebody replay please?

Thank you,
Maxim.

On 07/05/16 11:01, Maxim Uvarov wrote:

Use vlan tagged traffic for example validation test
to add code coverage for pkt_mmap_vlan_insert() internal
function.

Signed-off-by: Maxim Uvarov 
---
  In general this patch work and tests internal function. But there
  is one small issue related to autotools which I don't know how to
  fix better:
  make; make check works, but make check after clean fails 
example/packet/odp_pktio
  binary is not yet build while tests going to be executed. The only thing which
  cames to my mind is to reorder root directory Makefile.am and add "-" between 
tests
  and examples. In that case examples should be compiled before any test 
compilation/execution.
  Is that a right way to do?

  Thanks,
  Maxim.

  test/platform/linux-generic/Makefile.am|   6 ++-
  test/platform/linux-generic/pktio/.gitignore   |   1 +
  test/platform/linux-generic/pktio/Makefile.am  |   1 +
  .../platform/linux-generic/pktio/pktio_run_vlan.sh |  46 +
  test/platform/linux-generic/pktio/vlan.pcap| Bin 0 -> 9728 bytes
  5 files changed, 53 insertions(+), 1 deletion(-)
  create mode 100755 test/platform/linux-generic/pktio/pktio_run_vlan.sh
  create mode 100644 test/platform/linux-generic/pktio/vlan.pcap

diff --git a/test/platform/linux-generic/Makefile.am 
b/test/platform/linux-generic/Makefile.am
index 2d58c57..597a1c1 100644
--- a/test/platform/linux-generic/Makefile.am
+++ b/test/platform/linux-generic/Makefile.am
@@ -39,6 +39,9 @@ SUBDIRS = $(ODP_MODULES)
  
  if HAVE_PCAP

  TESTS += pktio/pktio_run_pcap.sh
+if test_example
+TESTS += pktio/pktio_run_vlan.sh
+endif
  endif
  if PKTIO_IPC
  TESTS += pktio_ipc/pktio_ipc_run.sh
@@ -54,7 +57,8 @@ endif
  
  TEST_EXTENSIONS = .sh
  
-dist_check_SCRIPTS = run-test tests-validation.env $(LOG_COMPILER)

+dist_check_SCRIPTS = run-test tests-validation.env $(LOG_COMPILER) \
+pktio/vlan.pcap
  
  test_SCRIPTS = $(dist_check_SCRIPTS)
  
diff --git a/test/platform/linux-generic/pktio/.gitignore b/test/platform/linux-generic/pktio/.gitignore

index 7e563b8..5f75e14 100644
--- a/test/platform/linux-generic/pktio/.gitignore
+++ b/test/platform/linux-generic/pktio/.gitignore
@@ -1,2 +1,3 @@
  *.log
  *.trs
+*.pcap
diff --git a/test/platform/linux-generic/pktio/Makefile.am 
b/test/platform/linux-generic/pktio/Makefile.am
index 4a14343..094a222 100644
--- a/test/platform/linux-generic/pktio/Makefile.am
+++ b/test/platform/linux-generic/pktio/Makefile.am
@@ -4,6 +4,7 @@ dist_check_SCRIPTS = pktio_env \
  
  if HAVE_PCAP

  dist_check_SCRIPTS += pktio_run_pcap.sh
+dist_check_SCRIPTS += pktio_run_vlan.sh
  endif
  if netmap_support
  dist_check_SCRIPTS += pktio_run_netmap.sh
diff --git a/test/platform/linux-generic/pktio/pktio_run_vlan.sh 
b/test/platform/linux-generic/pktio/pktio_run_vlan.sh
new file mode 100755
index 000..b58aacb
--- /dev/null
+++ b/test/platform/linux-generic/pktio/pktio_run_vlan.sh
@@ -0,0 +1,46 @@
+#!/bin/sh
+#
+# Copyright (c) 2016, Linaro Limited
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+#
+
+#
+# This test is intend to test pkt_mmap_vlan_insert() feature for
+# linux-generic packet mmap pktio.
+#
+#
+
+# any parameter passed as arguments to this script is passed unchanged to
+# the test itself (pktio_main)
+
+# directories where pktio_main binary can be found:
+# -in the validation dir when running make check (intree or out of tree)
+# -in the script directory, when running after 'make install', or
+# -in the validation when running standalone intree.
+# -in the current directory.
+# running stand alone out of tree requires setting PATH
+PATH=${TEST_DIR}/pktio:$PATH
+PATH=$(dirname $0):$PATH
+PATH=../../../example/packet:$PATH
+PATH=../../../../_inst/bin:$PATH
+PATH=.:$PATH
+
+odp_pktio_path=$(which odp_pktio${EXEEXT})
+if [ -x "$odp_pktio_path" ] ; then
+   echo "running with $odp_pktio_path"
+else
+   pwd
+   echo $PATH
+   echo "cannot find odp_pktio${EXEEXT}: please set you PATH for it."
+   exit 1
+fi
+
+PCAP_IN=`find . ${TEST_DIR} $(dirname $0) -name vlan.pcap -print -quit`
+echo "using PCAP_IN = ${PCAP_IN}"
+PCAP_OUT=vlan_out.pcap
+odp_pktio${EXEEXT} -ipcap:in=${PCAP_IN}:out=${PCAP_OUT} -t 1 -m 0
+ret=$?
+rm -f ${PCAP_OUT}
+exit $ret
diff --git a/test/platform/linux-generic/pktio/vlan.pcap 
b/test/platform/linux-generic/pktio/vlan.pcap
new file mode 100644
index 
..106ccb682e51495b4025337518a0bce63c2c7681
GIT binary patch
literal 9728
zcmeHNeQXrR6@Rikyh3QAN~9xK+}$
z4T)^-QnlAgC4W>Yk*X>mRlo92km}Nw!c|oiNTn)LBLcyK0BKSYLZiA0xI}f|@6FEc
z}Zu4i?dS2hYnK$pf`MuxF}9(>lN6zOYfipb#)S3BqRK6jeva#{)P%KZ4H
z2`Z!uf1f`})Rm#nP-l2Nx8d?iyv8ohr8v56pV)=ly+yzZvNmzS?$(tvQ({

Re: [lng-odp] [API-NEXT PATCHv2] linux-generic: _fdserver: allocating data table dynamicaly

[Maxim Uvarov]

Merged,
Maxim.

On 09/13/16 15:46, Christophe Milard wrote:

Hi Bill, and thanks for your review/comments:
I don't think that there any reason (at this time, at least) to worry about
table alignment: The server is -hopefully- used only at non critical times
(such as init...). It is based on socket communications which will be dead
slow compared to the alignment gain anyway.
So I don't think there is much to gain with the alignment.

Christophe.

On 13 September 2016 at 14:02, Bill Fischofer 
wrote:


This looks good now. Only question is whether you need to worry about
cache-aligning the table, but that's a tuning option that can be considered
later if needed.

On Tue, Sep 13, 2016 at 7:40 AM, Christophe Milard <
christophe.mil...@linaro.org> wrote:


The table containing the saved file-descriptors<->{context, key} couples
is
now dynamically malloc'd in the fd server process, hence avoiding
the memory waste which happened in other process when the table was
staticaly reserved in all processes.

Signed-off-by: Christophe Milard 


Reviewed-by: Bill Fischofer 



---
  platform/linux-generic/_fdserver.c | 14 +-
  1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/platform/linux-generic/_fdserver.c
b/platform/linux-generic/_fdserver.c
index bf36eb2..5cef806 100644
--- a/platform/linux-generic/_fdserver.c
+++ b/platform/linux-generic/_fdserver.c
@@ -73,7 +73,7 @@ typedef struct fdentry_s {
 uint64_t key;
 int  fd;
  } fdentry_t;
-static fdentry_t fd_table[FDSERVER_MAX_ENTRIES];
+static fdentry_t *fd_table;
  static int fd_table_nb_entries;

  /*
@@ -622,8 +622,20 @@ int _odp_fdserver_init_global(void)
 /* TODO: pin the server on appropriate service cpu mask */
 /* when (if) we can agree on the usage of service mask  */

+   /* allocate the space for the file descriptor<->key
table: */
+   fd_table = malloc(FDSERVER_MAX_ENTRIES *
sizeof(fdentry_t));
+   if (!fd_table) {
+   ODP_ERR("maloc failed!\n");
+   exit(1);
+   }
+
+   /* wait for clients requests */
 wait_requests(sock); /* Returns when server is stopped  */
 close(sock);
+
+   /* release the file descriptor table: */
+   free(fd_table);
+
 exit(0);
 }

--
2.7.4

Re: [lng-odp] [API-NEXT PATCH] linux-generic: _fdserver: fixing comment typo

[Maxim Uvarov]

Merged,
Maxim.

On 09/13/16 14:24, Bill Fischofer wrote:

On Tue, Sep 13, 2016 at 5:05 AM, Christophe Milard <
christophe.mil...@linaro.org> wrote:


Signed-off-by: Christophe Milard 


Reviewed-by: Bill Fischofer 



---
  platform/linux-generic/_fdserver.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/platform/linux-generic/_fdserver.c b/platform/linux-generic/_
fdserver.c
index bf36eb2..97661d0 100644
--- a/platform/linux-generic/_fdserver.c
+++ b/platform/linux-generic/_fdserver.c
@@ -103,7 +103,7 @@ typedef struct fd_server_msg {
   * Send a fdserver_msg, possibly including a file descriptor, on the
socket
   * This function is used both by:
   * -the client (sending a FD_REGISTER_REQ with a file descriptor to be
shared,
- *  or FD_LOOKUP_REQ/FD_DEREGISTER_REQ without a file descirptor)
+ *  or FD_LOOKUP_REQ/FD_DEREGISTER_REQ without a file descriptor)
   * -the server (sending FD_REGISTER_ACK/NACK, FD_LOOKUP_NACK,
   *  FD_DEREGISTER_ACK/NACK... without a fd or a
   *  FD_LOOKUP_ACK with a fd)
@@ -165,7 +165,7 @@ static int send_fdserver_msg(int sock, int command,
   * given socket.
   * This function is used both by:
   * -the server (receiving a FD_REGISTER_REQ with a file descriptor to be
shared,
- *  or FD_LOOKUP_REQ, FD_DEREGISTER_REQ without a file descirptor)
+ *  or FD_LOOKUP_REQ, FD_DEREGISTER_REQ without a file descriptor)
   * -the client (receiving FD_REGISTER_ACK...without a fd or a
FD_LOOKUP_ACK with
   * a fd)
   * This function make use of the ancillary data (control data) to pass and
--
2.7.4

Re: [lng-odp] [PATCH] example: generator: actually use specified default

[Maxim Uvarov]

Merged,
Maxim.

On 09/14/16 05:50, Bill Fischofer wrote:

On Tue, Sep 13, 2016 at 12:35 PM, Mike Holmes 
wrote:


The help states default is 1000ms. 0 for flood mode, however the
default was incorrectly set to zero.

Signed-off-by: Mike Holmes 


Reviewed-by: Bill Fischofer 



---
  example/generator/odp_generator.c | 1 +
  1 file changed, 1 insertion(+)

diff --git a/example/generator/odp_generator.c b/example/generator/odp_
generator.c
index b0053b9..48d7f5f 100644
--- a/example/generator/odp_generator.c
+++ b/example/generator/odp_generator.c
@@ -946,6 +946,7 @@ static void parse_args(int argc, char *argv[],
appl_args_t *appl_args)
 appl_args->number = -1;
 appl_args->payload = 56;
 appl_args->timeout = -1;
+   appl_args->interval = DEFAULT_PKT_INTERVAL;

 opterr = 0; /* do not issue errors on helper options */

--
2.7.4

Re: [lng-odp] [PATCH] linux-gen: build: de-couple abi compatibility from shared lib

[Savolainen, Petri (Nokia - FI/Espoo)]

> > +ODP_ABI_COMPAT=0
> > +AC_ARG_ENABLE([abi-compat],
> > +[  --enable-abi-compat build in ABI specification compatible
> mode],
> > +[if test "x$enableval" = "xyes"; then
> > +   ODP_ABI_COMPAT=1
> > + else
> > +   ODP_ABI_COMPAT=0
> 
> if we disable ABI compat mode I think we should set the SO-version to
> 0:0:0, to make
> it clear that we don't try to be ABI compatible.
> 
> Cheers,
> Anders
> 

Didn't address this in v2, since we have not been ABI compatible before either. 
Not sure how lib versioning should indicate ABI compatibility. Strictly 
speaking we are not ABI spec compatible yet, since there's no ABI spec defined 
anywhere.

A next step towards ABI compatibility would be to add new ABI spec header files 
e.g. under odp/include/api/abi_x86_64 and include those files instead/in 
addition of platform/xxx/include/api/plat/xxx_types.h files (== force ABI spec).


-Petri

[lng-odp] [PATCH v2] linux-gen: build: de-couple abi compatibility from shared lib

[Petri Savolainen]

Building ABI compatible or shared library are two different
targets. A shared library may be used also without ABI
compatibility. A new --enable-abi-compat configuration option
is introduced. By default libraries are not built in ABI compat
mode to enable function inlining. There is a noticeable
performance difference when e.g. odp_atomic_xxx calls
are not inlined.

Signed-off-by: Petri Savolainen 
---

v2:
  * ABI compat enabled by default
  * print static/shared/abi_compat selection in config results
  * added missing header file include guards


 configure.ac   | 26 -
 platform/linux-generic/.gitignore  |  1 -
 platform/linux-generic/include/odp/api/atomic.h|  2 +-
 platform/linux-generic/include/odp/api/byteorder.h |  2 +-
 .../linux-generic/include/odp/api/plat/inlines.h   | 30 
 .../include/odp/api/plat/inlines.h.in  | 33 --
 platform/linux-generic/include/odp/api/std_clib.h  |  2 +-
 platform/linux-generic/include/odp/api/sync.h  |  2 +-
 platform/linux-generic/m4/configure.m4 |  3 +-
 platform/linux-generic/odp_atomic.c|  2 +-
 platform/linux-generic/odp_byteorder.c |  2 +-
 platform/linux-generic/odp_std_clib.c  |  2 +-
 platform/linux-generic/odp_sync.c  |  2 +-
 13 files changed, 58 insertions(+), 51 deletions(-)
 delete mode 100644 platform/linux-generic/.gitignore
 create mode 100644 platform/linux-generic/include/odp/api/plat/inlines.h
 delete mode 100644 platform/linux-generic/include/odp/api/plat/inlines.h.in

diff --git a/configure.ac b/configure.ac
index 982aff7..e05d4dd 100644
--- a/configure.ac
+++ b/configure.ac
@@ -176,13 +176,6 @@ AM_CONDITIONAL([test_example], [test x$test_example = xyes 
])
 AM_CONDITIONAL([HAVE_DOXYGEN], [test "x${DOXYGEN}" = "xdoxygen"])
 AM_CONDITIONAL([user_guide], [test "x${user_guides}" = "xyes" ])
 AM_CONDITIONAL([HAVE_MSCGEN], [test "x${MSCGEN}" = "xmscgen"])
-if test x$enable_shared != xyes;
-then
-   _ODP_INLINES="_ODP_INLINES"
-else
-   _ODP_INLINES="_ODP_NO_INLINES"
-fi
-AC_SUBST(_ODP_INLINES)
 
 ##
 # Setup doxygen documentation
@@ -225,6 +218,22 @@ AC_ARG_ENABLE([debug],
 ODP_CFLAGS="$ODP_CFLAGS -DODP_DEBUG=$ODP_DEBUG"
 
 ##
+# Enable/disable ABI compatible build
+##
+ODP_ABI_COMPAT=1
+abi_compat=yes
+AC_ARG_ENABLE([abi-compat],
+[  --enable-abi-compat build all targets in ABI compatible mode 
(default=yes)],
+[if test "x$enableval" = "xyes"; then
+   ODP_ABI_COMPAT=1
+   abi_compat=yes
+ else
+   ODP_ABI_COMPAT=0
+   abi_compat=no
+fi])
+ODP_CFLAGS="$ODP_CFLAGS -DODP_ABI_COMPAT=$ODP_ABI_COMPAT"
+
+##
 # Default warning setup
 ##
 ODP_CFLAGS="$ODP_CFLAGS -W -Wall -Werror -Wstrict-prototypes 
-Wmissing-prototypes"
@@ -307,6 +316,9 @@ AC_MSG_RESULT([
am_ldflags: ${AM_LDFLAGS}
libs:   ${LIBS}
defs:   ${DEFS}
+   static libraries:   ${enable_static}
+   shared libraries:   ${enable_shared}
+   ABI compatible: ${abi_compat}
cunit:  ${cunit_support}
test_vald:  ${test_vald}
test_perf:  ${test_perf}
diff --git a/platform/linux-generic/.gitignore 
b/platform/linux-generic/.gitignore
deleted file mode 100644
index ec6ca37..000
--- a/platform/linux-generic/.gitignore
+++ /dev/null
@@ -1 +0,0 @@
-include/odp/api/plat/inlines.h
diff --git a/platform/linux-generic/include/odp/api/atomic.h 
b/platform/linux-generic/include/odp/api/atomic.h
index c18e68b..66cfef7 100644
--- a/platform/linux-generic/include/odp/api/atomic.h
+++ b/platform/linux-generic/include/odp/api/atomic.h
@@ -25,7 +25,7 @@ extern "C" {
  */
 
 #include 
-#ifdef _ODP_INLINES
+#if ODP_ABI_COMPAT == 0
 #include 
 #endif
 
diff --git a/platform/linux-generic/include/odp/api/byteorder.h 
b/platform/linux-generic/include/odp/api/byteorder.h
index 84d1173..37f5636 100644
--- a/platform/linux-generic/include/odp/api/byteorder.h
+++ b/platform/linux-generic/include/odp/api/byteorder.h
@@ -26,7 +26,7 @@ extern "C" {
  */
 
 #include 
-#ifdef _ODP_INLINES
+#if ODP_ABI_COMPAT == 0
 #include 
 #endif
 
diff --git a/platform/linux-generic/include/odp/api/plat/inlines.h 
b/platform/linux-generic/include/odp/api/plat/inlines.h
new file mode 100644
index 000..297a199
--- /dev/null
+++ b/platform/linux-generic/include/odp/api/plat/inlines.h
@@ -0,0 +1,30 @@
+/* Copyright (c) 2016, Linaro Limited
+ * All rights reserved.
+ *
+ * 

[lng-odp] [PATCH v2 1/2] test: perf: add new scheduling latency test

[Matias Elo]

Add new scheduling latency benchmark application. The application
measures delays (avg, min, max) for high and low priority events.

The test has a configurable number of TRAFFIC events and few SAMPLE events
(one common or one per priority). The scheduling latency is only measured
from the SAMPLE events to minimize measurement overhead.

The application's command line arguments enable configuring:
- Number of processing threads
- Number of high/low priority queues
- Number of high/low priority events
- Use separate SAMPLE events for each priority
- Scheduled queue type (PARALLEL, ATOMIC, ORDERED)

Signed-off-by: Matias Elo 
---

V2:
- Remove unnecessary 'num_workers' initialization (Maxim)

 test/common_plat/performance/.gitignore  |   1 +
 test/common_plat/performance/Makefile.am |   4 +
 test/common_plat/performance/odp_sched_latency.c | 767 +++
 3 files changed, 772 insertions(+)
 create mode 100644 test/common_plat/performance/odp_sched_latency.c

diff --git a/test/common_plat/performance/.gitignore 
b/test/common_plat/performance/.gitignore
index edcc832..1527d25 100644
--- a/test/common_plat/performance/.gitignore
+++ b/test/common_plat/performance/.gitignore
@@ -4,4 +4,5 @@ odp_atomic
 odp_crypto
 odp_l2fwd
 odp_pktio_perf
+odp_sched_latency
 odp_scheduling
diff --git a/test/common_plat/performance/Makefile.am 
b/test/common_plat/performance/Makefile.am
index d23bb3e..f5dd8dd 100644
--- a/test/common_plat/performance/Makefile.am
+++ b/test/common_plat/performance/Makefile.am
@@ -5,6 +5,7 @@ TESTS_ENVIRONMENT += TEST_DIR=${builddir}
 EXECUTABLES = odp_crypto$(EXEEXT) odp_pktio_perf$(EXEEXT)
 
 COMPILE_ONLY = odp_l2fwd$(EXEEXT) \
+  odp_sched_latency$(EXEEXT) \
   odp_scheduling$(EXEEXT)
 
 TESTSCRIPTS = odp_l2fwd_run.sh \
@@ -20,6 +21,8 @@ bin_PROGRAMS = $(EXECUTABLES) $(COMPILE_ONLY)
 
 odp_crypto_LDFLAGS = $(AM_LDFLAGS) -static
 odp_crypto_CFLAGS = $(AM_CFLAGS) -I${top_srcdir}/test
+odp_sched_latency_LDFLAGS = $(AM_LDFLAGS) -static
+odp_sched_latency_CFLAGS = $(AM_CFLAGS) -I${top_srcdir}/test
 odp_scheduling_LDFLAGS = $(AM_LDFLAGS) -static
 odp_scheduling_CFLAGS = $(AM_CFLAGS) -I${top_srcdir}/test
 
@@ -27,6 +30,7 @@ noinst_HEADERS = \
  $(top_srcdir)/test/test_debug.h
 
 dist_odp_crypto_SOURCES = odp_crypto.c
+dist_odp_sched_latency_SOURCES = odp_sched_latency.c
 dist_odp_scheduling_SOURCES = odp_scheduling.c
 dist_odp_pktio_perf_SOURCES = odp_pktio_perf.c
 
diff --git a/test/common_plat/performance/odp_sched_latency.c 
b/test/common_plat/performance/odp_sched_latency.c
new file mode 100644
index 000..063fb21
--- /dev/null
+++ b/test/common_plat/performance/odp_sched_latency.c
@@ -0,0 +1,767 @@
+/* Copyright (c) 2016, Linaro Limited
+ * All rights reserved.
+ *
+ * SPDX-License-Identifier: BSD-3-Clause
+ */
+
+/**
+ * @file
+ *
+ * @example odp_sched_latency.c  ODP scheduling latency benchmark application
+ */
+
+#include 
+#include 
+#include 
+
+#include 
+
+/* ODP main header */
+#include 
+
+/* ODP helper for Linux apps */
+#include 
+
+/* GNU lib C */
+#include 
+
+#define MAX_WORKERS  64/**< Maximum number of worker threads */
+#define MAX_QUEUES   4096  /**< Maximum number of queues */
+#define EVENT_POOL_SIZE  (1024 * 1024) /**< Event pool size */
+#define TEST_ROUNDS (4 * 1024 * 1024)  /**< Test rounds for each thread */
+#define MAIN_THREAD   1 /**< Thread ID performing maintenance tasks */
+
+/* Default values for command line arguments */
+#define SAMPLE_EVENT_PER_PRIO0 /**< Allocate a separate sample event for
+each priority */
+#define HI_PRIO_EVENTS   0 /**< Number of high priority events */
+#define LO_PRIO_EVENTS  32 /**< Number of low priority events */
+#define HI_PRIO_QUEUES  16 /**< Number of high priority queues */
+#define LO_PRIO_QUEUES  64 /**< Number of low priority queues */
+
+#define EVENTS_PER_HI_PRIO_QUEUE 0  /**< Alloc HI_PRIO_QUEUES x HI_PRIO_EVENTS
+events */
+#define EVENTS_PER_LO_PRIO_QUEUE 1  /**< Alloc LO_PRIO_QUEUES x LO_PRIO_EVENTS
+events */
+ODP_STATIC_ASSERT(HI_PRIO_QUEUES <= MAX_QUEUES, "Too many HI priority queues");
+ODP_STATIC_ASSERT(LO_PRIO_QUEUES <= MAX_QUEUES, "Too many LO priority queues");
+
+#define CACHE_ALIGN_ROUNDUP(x)\
+   ((ODP_CACHE_LINE_SIZE) * \
+(((x) + ODP_CACHE_LINE_SIZE - 1) / (ODP_CACHE_LINE_SIZE)))
+
+/* Test priorities */
+#define NUM_PRIOS 2 /**< Number of tested priorities */
+#define HI_PRIO  0
+#define LO_PRIO  1
+
+/** Test event types */
+typedef enum {
+   WARM_UP, /**< Warm up event */
+   TRAFFIC, /**< Event used only as traffic load */
+   SAMPLE   /**< Event used to measure latency */
+} event_type_t;
+
+/** Test event */
+typedef struct {
+   uint64_t ts;/**< Send 

[lng-odp] [PATCH v2 2/2] test: sched_latency: add test script

[Matias Elo]

Signed-off-by: Matias Elo 
---

V2:
- Exit on first failure (Maxim)

 test/common_plat/performance/Makefile.am   |  1 +
 .../performance/odp_sched_latency_run.sh   | 28 ++
 2 files changed, 29 insertions(+)
 create mode 100755 test/common_plat/performance/odp_sched_latency_run.sh

diff --git a/test/common_plat/performance/Makefile.am 
b/test/common_plat/performance/Makefile.am
index f5dd8dd..f184609 100644
--- a/test/common_plat/performance/Makefile.am
+++ b/test/common_plat/performance/Makefile.am
@@ -9,6 +9,7 @@ COMPILE_ONLY = odp_l2fwd$(EXEEXT) \
   odp_scheduling$(EXEEXT)
 
 TESTSCRIPTS = odp_l2fwd_run.sh \
+ odp_sched_latency_run.sh \
  odp_scheduling_run.sh
 
 TEST_EXTENSIONS = .sh
diff --git a/test/common_plat/performance/odp_sched_latency_run.sh 
b/test/common_plat/performance/odp_sched_latency_run.sh
new file mode 100755
index 000..6048f58
--- /dev/null
+++ b/test/common_plat/performance/odp_sched_latency_run.sh
@@ -0,0 +1,28 @@
+#!/bin/sh
+#
+# Copyright (c) 2016, Linaro Limited
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+#
+# Script that passes command line arguments to odp_sched_latency test when
+# launched by 'make check'
+
+TEST_DIR="${TEST_DIR:-$(dirname $0)}"
+ALL=0
+
+run()
+{
+   echo odp_sched_latency_run starts requesting $1 worker threads
+   echo ===
+
+   $TEST_DIR/odp_sched_latency${EXEEXT} -c $1 || exit $?
+}
+
+run 1
+run 5
+run 8
+run 11
+run $ALL
+
+exit 0
-- 
2.7.4

Re: [lng-odp] [PATCH] example: ipsec: Poll mode fix.

[Nikhil Agarwal Agarwal]

Does it matter. We are just saving this to local structure. Also we will start 
polling on these queues after launching worker threads post init.

Regards
Nikhil

-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Maxim 
Uvarov
Sent: Thursday, September 08, 2016 9:37 PM
To: lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [PATCH] example: ipsec: Poll mode fix.

Hello Nikhil,

I think it's better to add inq to poll before pktio start.

Maxim.

On 08/16/16 17:22, Nikhil Agarwal wrote:
> Adding pktio queues in poll mode to polled queues.
>
> Signed-off-by: gagandeep singh 
> Signed-off-by: Nikhil Agarwal 
> ---
>   example/ipsec/odp_ipsec.c | 6 ++
>   1 file changed, 6 insertions(+)
>
> diff --git a/example/ipsec/odp_ipsec.c b/example/ipsec/odp_ipsec.c 
> index 6a9a9fe..89ba1e9 100644
> --- a/example/ipsec/odp_ipsec.c
> +++ b/example/ipsec/odp_ipsec.c
> @@ -561,6 +561,12 @@ void initialize_intf(char *intf)
>  odp_queue_to_u64(inq),
>  mac_addr_str(src_mac_str, src_mac));
>   
> + if (pktio_param.in_mode == ODP_PKTIN_MODE_QUEUE) {
> + poll_queues[num_polled_queues++] = inq;
> + printf("%s: adding %"PRIu64"\n", __func__,
> +odp_queue_to_u64(inq));
> + }
> +
>   /* Resolve any routes using this interface for output */
>   resolve_fwd_db(intf, pktout, src_mac);
>   }

Re: [lng-odp] issue with odp_crypto validation suite

[Nikhil Agarwal Agarwal]

Krishna,

Following is the API documentation for "auth_cipher_text"

* @var odp_crypto_session_params_t::auth_cipher_text
 *
 *   Controls ordering of authentication and cipher operations,
 *   and is relative to the operation (encode vs decode).
 *   When encoding, @c TRUE indicates the authentication operation
 *   should be performed @b after the cipher operation else before.
 *   When decoding, @c TRUE indicates the reverse order of operation.

Logically speaking this seems correct as well. If in encode direction you did 
cipher first and then auth, then in decode direction you have to follow the 
reverse order. I hope this clarifies your doubt.

Regards
Nikhil

-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Krishna 
Garapati
Sent: Tuesday, September 06, 2016 6:34 PM
To: LNG ODP Mailman List 
Subject: [lng-odp] issue with odp_crypto validation suite

From odp crypto validation suite I see that the application is setting 
"auth_cipher_text" flag from odp_crypto_session_params to "false" all ways 
(Irrespective of ENCODE or DECODE). What this means from my understanding is 
application wants to do "cipher first and then authentication".  And in the 
odp_crypto implementation during session_create if the crypto operation is 
ENCODE, this flag is set as is against the 
"odp_crypto_generic_session_t->do_cipher_first" which actually instead leads to 
do authentication then ciphering during crypto operation. This feels wrong in 
some way. Any idea about this implementation ?.

/Krishna