Re: [lng-odp] thread/shmem discussion summary V3

On Thu, Jun 2, 2016 at 11:17 AM, Ola Liljedahl wrote: > > > On 2 June 2016 at 11:08, Christophe Milard > wrote: > >> since V2: Update following Barry and Bill's comments >> since V1: Update following arch call 31 may 2016 >> >> This is a

Re: [lng-odp] [PATCH v2 0/6] Modular scheduler continuation

For parts 1-5 only: Reviewed-and-tested-by: Bill Fischofer We should merge parts 1-5 since they restructure the scheduler to add a framework that permits the development of alternate scheduler implementations. The current scheduler implementation remains the default

Re: [lng-odp] [PATCH] linux-gen: fix debug info print functions

The use of ODP_DBG() is intentional in all of these cases. These are diagnostic output which should be optional unless compiled with debug enabled. For normal operation these should not be seen. On Thu, Jun 2, 2016 at 4:28 AM, Matias Elo wrote: > ODP_DBG() macro should not

Re: [lng-odp] [RFC] Adding Multi variant for Async crypto operation.

This should be tagged RFC API-NEXT since it proposes an API change On Thu, Jun 2, 2016 at 10:07 AM, Nikhil Agarwal wrote: > Signed-off-by: Nikhil Agarwal > --- > include/odp/api/spec/crypto.h | 25 +++-- > 1 file

Re: [lng-odp] [API-NEXT PATCH v3 1/2] helper: table: add impl of ip lookup table

Just status to be clear here. Bill wanted to review/test this patch but fail to apply it. Please update to current master and send v4. Maxim. On 06/02/16 06:25, HePeng wrote: Ping. 在 2016年5月23日，下午12:26，Ru Jia 写道： This is an implementation of the 16,8,8 ip lookup (longest

Re: [lng-odp] [PATCH v2 1/4] linux-gen: netmap: handle modified _odp_packet_cls_enq() return values correctly

Merged, Thanks. On 06/02/16 10:45, Matias Elo wrote: Do not increase the count of received packets if classifier already handled the packet. Fixes odp_classifier example crashing with netmap pktio. Signed-off-by: Matias Elo Reviewed-and-tested-by: Bill Fischofer

Re: [lng-odp] NAT and FW for odp

Thank you, Maxim, Mike! I looked though the Venkatesh's works today. That's a good ipfw port, but original netmap-ipfw is quite outdated, so i was thinking is there something i have missed Thank you! Ivan On 02.06.2016 21:08, Mike Holmes wrote: Maxim Is any of this specifically the

Re: [lng-odp] thread/shmem discussion summary V3

On 2 June 2016 at 11:08, Christophe Milard wrote: > since V2: Update following Barry and Bill's comments > since V1: Update following arch call 31 may 2016 > > This is a tentative to sum up the discussions around the thread/process > that have been happening these

Re: [lng-odp] NAT and FW for odp

Maxim Is any of this specifically the POC that Venki did that you are referring to ? https://git.linaro.org/?a=project_list=ven=Search people/maxim.uvarov/venkatesh.vivekanandan/ipfwtest.git people/maxim.uvarov/venkatesh.vivekanandan/odp-dpdk-ipfw.git

Re: [lng-odp] [PATCH 1/3] linux-generic: shared memory: fix the name composition in odp_shm_free

Merged, Maxim. On 06/02/16 14:04, Yi He wrote: Fix the name composition in odp_shm_free() as the same algorithm in odp_shm_reserve(), use IPC namespace if provided or process ID in default. Signed-off-by: Yi He --- platform/linux-generic/odp_shared_memory.c | 10 --

Re: [lng-odp] NAT and FW for odp

ipfw On 2 June 2016 at 15:25, Ivan Koveshnikov wrote: > Good day everyone! > > Is there any implementations of NAT (network address translation) and > firewall running on top the ODP library? > > Best regards, > Ivan > ___ >

[lng-odp] [Bug 2102] make distcheck does not inherit original options

https://bugs.linaro.org/show_bug.cgi?id=2102 Mike Holmes changed: What|Removed |Added Status|IN_PROGRESS |RESOLVED

[lng-odp] [PATCH 3/3] linux-generic: pktio: ring: add _ring_destroy()

Add _ring_destroy() to release shared memory allocated by _ring_create() calls. previously this was accomplished by manually odp_shm_lookup() and odp_shm_free(). Signed-off-by: Yi He --- .../include/odp_packet_io_ring_internal.h | 8 +

[lng-odp] [PATCH 2/3] linux-generic: pktio: ipc: _ring_tailq_init() at global init

Fix the uninitialized Q and RW lock which may cause occasional segmentation fault in _ring_create() etc. Signed-off-by: Yi He --- platform/linux-generic/pktio/ipc.c | 8 +++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/platform/linux-generic/pktio/ipc.c

[lng-odp] [PATCH 1/3] linux-generic: shared memory: fix the name composition in odp_shm_free

Fix the name composition in odp_shm_free() as the same algorithm in odp_shm_reserve(), use IPC namespace if provided or process ID in default. Signed-off-by: Yi He --- platform/linux-generic/odp_shared_memory.c | 10 -- 1 file changed, 8 insertions(+), 2 deletions(-)

[lng-odp] [RFC] Adding Multi variant for Async crypto operation.

Signed-off-by: Nikhil Agarwal --- include/odp/api/spec/crypto.h | 25 +++-- 1 file changed, 23 insertions(+), 2 deletions(-) diff --git a/include/odp/api/spec/crypto.h b/include/odp/api/spec/crypto.h index d8123e9..1c2b384 100644 ---

[lng-odp] [PATCH] linux-gen: fix debug info print functions

ODP_DBG() macro should not be used inside *_print() functions. Since debug messages are disabled by default, the print functions won't output anything with ODP_DBG. The patch also removes the extra space previously added to the beginning of ODP_PRINT messages. Signed-off-by: Matias Elo

[lng-odp] thread/shmem discussion summary V3

since V2: Update following Barry and Bill's comments since V1: Update following arch call 31 may 2016 This is a tentative to sum up the discussions around the thread/process that have been happening these last weeks. Sorry for the formalism of this mail, but it seems we need accuracy here...

[lng-odp] [PATCH v2 3/4] linux-gen: pktio: remove old single queue recv and send functions

RX/TX locks are moved inside pktio implementations and pktios set packet header's input member. Signed-off-by: Matias Elo Reviewed-and-tested-by: Bill Fischofer --- .../linux-generic/include/odp_packet_io_internal.h | 12 ++--

[lng-odp] [PATCH v2 4/4] linux-gen: pktio: don't allocate new packets in classifier

Don't allocate new packets inside of the internal classifier helpers _odp_packet_cls_enq() and _odp_packet_classifier(). Instead, save destination queue to the parsed packet header and return correct packet pool. This enables zero-copy packet classification. Added new internal pktio helper

[lng-odp] [PATCH v2 2/4] linux-gen: packet: increase input flags bit field size to 64 bits

Make room for more input flags. Signed-off-by: Matias Elo Reviewed-and-tested-by: Bill Fischofer --- .../linux-generic/include/odp_packet_internal.h| 66 +++--- platform/linux-generic/odp_packet.c| 2 +- 2

[lng-odp] [PATCH v2 1/4] linux-gen: netmap: handle modified _odp_packet_cls_enq() return values correctly

Do not increase the count of received packets if classifier already handled the packet. Fixes odp_classifier example crashing with netmap pktio. Signed-off-by: Matias Elo Reviewed-and-tested-by: Bill Fischofer ---

[lng-odp] [PATCH v2 3/6] linux-gen: sched: add sched params to queue init

Added scheduler parameters into queue init scheduler interface call. Signed-off-by: Petri Savolainen --- platform/linux-generic/include/odp_schedule_if.h | 4 +++- platform/linux-generic/odp_queue.c | 3 ++- platform/linux-generic/odp_schedule.c

[lng-odp] [PATCH v2 6/6] linux-gen: sched: started SP scheduler implementation

Started to implement simple strict priority scheduler. This will not be optimized for throughput, but for low latency processing of high priority events. Implementation may have limited support on various scheduler features (e.g. ordered queues) and low number of priorities, groups, etc.

[lng-odp] [PATCH v2 4/6] linux-gen: sched: add skeleton for strict priority

Added configuration option to use strict priority scheduler implementation instead of the default. Signed-off-by: Petri Savolainen --- configure.ac | 1 + platform/linux-generic/Makefile.am| 1 +

[lng-odp] [PATCH v2 0/6] Modular scheduler continuation

Continued internal scheduler interface development with API side interface. Added configuration option to select strict priority scheduler instead of the default scheduler. Started SP scheduler implementation which already runs successfully odp_scheduling test, but does not pass validation test

[lng-odp] [PATCH v2 1/6] linux-gen: sched: add init functions to interface

Added global/local init and term functions to the scheduler interface. Signed-off-by: Petri Savolainen --- platform/linux-generic/include/odp_internal.h| 5 - platform/linux-generic/include/odp_schedule_if.h | 9 - platform/linux-generic/odp_init.c

Re: [lng-odp] [PATCH 4/4] linux-gen: pktio: don't allocate new packets in classifier

> > diff --git a/platform/linux-generic/pktio/socket_mmap.c b/platform/linux- > generic/pktio/socket_mmap.c > > index ce05775..6d7ec59 100644 > > --- a/platform/linux-generic/pktio/socket_mmap.c > > +++ b/platform/linux-generic/pktio/socket_mmap.c > > @@ -166,6 +166,10 @@ static inline unsigned

Re: [lng-odp] [PATCH 5/6] linux-gen: sched: bind default scheduler api calls

From: Bill Fischofer [mailto:bill.fischo...@linaro.org] Sent: Wednesday, June 01, 2016 11:11 PM To: Savolainen, Petri (Nokia - FI/Espoo) Cc: LNG ODP Mailman List Subject: Re: [lng-odp] [PATCH 5/6] linux-gen: sched: bind default scheduler