Re: [lng-odp] [API-NEXT PATCHv5 00/10] add extend, trunc, and prefetch support

ping. We've already merged these API changes so the implementation and CUnit tests need review so they can be merged as well. On Thu, Apr 14, 2016 at 7:25 PM, Bill Fischofer wrote: > This patch series implements the packet extend, trunc, and prefetch APIs. > > This

Re: [lng-odp] [API-NEXT PATCHv3 1/6] api: pktio: add pktio index conversion API

ping. Needs review. On Sat, Apr 16, 2016 at 12:55 PM, Bill Fischofer wrote: > Add the odp_pktio_index() API to enable applications to convert packet I/O > handles to indices for managing user contexts and related associated data. > > Signed-off-by: Bill Fischofer

Re: [lng-odp] [API-NEXT PATCH 1/2] linux-generic: pktio: add missing pktio names

For this series; Reviewed-and-tested-by: Bill Fischofer On Tue, Apr 19, 2016 at 7:15 AM, Matias Elo wrote: > Add names to ipc and tap pktio types. > > Signed-off-by: Matias Elo > --- > platform/linux-generic/pktio/ipc.c

Re: [lng-odp] [API-NEXT PATCH v2 08/11] validation: pktio: add pktin timestamping test

Rebase needed? Attempting to apply part 8 of the v2 patch yields: bill@Ubuntu15:~/linaro/matias$ git am --reject ~/Mail/Incoming/Matias/8 Applying: validation: pktio: add pktin timestamping test Checking patch test/validation/pktio/pktio.c... error: while searching for: #define TX_BATCH_LEN

Re: [lng-odp] [API-NEXT PATCH 3/3] api: packet: rename and add copy functions

General comment: These are a lot of new APIs to push to the list 1 week before API freeze, especially as the implementation and tests are being left as an exercise for the reader. Specific comments inline. On Tue, Apr 19, 2016 at 8:07 AM, Petri Savolainen < petri.savolai...@nokia.com> wrote: >

Re: [lng-odp] [API-NEXT PATCH 1/3] api: packet: add packet concat and split

On Tue, Apr 19, 2016 at 8:07 AM, Petri Savolainen < petri.savolai...@nokia.com> wrote: > Added odp_packet_concat() and _split(). Two packets may be > concatenated one after another, and split at any offset. This > suites the common use cases of packet reassembly and > fragmentation. Reassembly is

Re: [lng-odp] [API-NEXT PATCHv7] api: make only the API visible

Ok. v8 sent to correct that. On Tue, Apr 19, 2016 at 11:49 AM, Mike Holmes wrote: > > > On 19 April 2016 at 10:16, Bill Fischofer > wrote: > >> From: Anders Roxell >> >> Internal functions should not be part of

[lng-odp] [API-NEXT PATCHv8] api: make only the API visible

From: Anders Roxell Internal functions should not be part of symbols that are visible outside the library. Using -fvisibility=hidden hides all internal functions from the public ABI. Suggested-by: Ricardo Salveti Signed-off-by: Anders

Re: [lng-odp] [API-NEXT PATCH 2/3] api: packet: add segment level head and tailroom

On Tue, Apr 19, 2016 at 10:00 AM, Savolainen, Petri (Nokia - FI/Espoo) < petri.savolai...@nokia.com> wrote: > > > > > *From:* EXT Bill Fischofer [mailto:bill.fischo...@linaro.org] > *Sent:* Tuesday, April 19, 2016 5:39 PM > *To:* Savolainen, Petri (Nokia - FI/Espoo) >

Re: [lng-odp] [API-NEXT PATCHv7] api: make only the API visible

On 19 April 2016 at 10:16, Bill Fischofer wrote: > From: Anders Roxell > > Internal functions should not be part of symbols that are visible > outside the library. Using -fvisibility=hidden hides all internal > functions from the public ABI.

Re: [lng-odp] [API-NEXT PATCHv6] api: make only the API visible

v7 of the patch has been posted that does this. The #includes are now visibility_begin.h and visibility_end.h On Tue, Apr 19, 2016 at 10:39 AM, Mike Holmes wrote: > This topic has had a good discussion in calls. > > Petri has reservations that putting this in the spec

[lng-odp] building from a tar source

Hi, Changes went into the ODP release around v1.4 to use the git information to set the version information for the build via scripts/git_hash.sh (rather than scripts/odp_version.sh). With the latest scripts/git_hash.sh, I cannot build from a tar source. This is interesting, since there was a

Re: [lng-odp] odp_cpumask_default_worker() and availability of the returned CPU's

On 19.04.16 18:31, Zoltan Kiss wrote: Hi, A quick question: my understanding is that odp_cpumask_default_worker() returns the available CPU's, but it doesn't guarantee that when you start your thread's/processes they will be still available, right? It should be the applications

Re: [lng-odp] [API-NEXT PATCHv6] api: make only the API visible

This topic has had a good discussion in calls. Petri has reservations that putting this in the spec file will prove to be insufficient when we take on the bigger ABI issues beyond the packaging problems we are solving here. But Petri agreed that the API white list for what symbols appear in the

[lng-odp] odp_cpumask_default_worker() and availability of the returned CPU's

Hi, A quick question: my understanding is that odp_cpumask_default_worker() returns the available CPU's, but it doesn't guarantee that when you start your thread's/processes they will be still available, right? It should be the applications responsibility to not start more threads than it

Re: [lng-odp] [API-NEXT PATCH 2/3] api: packet: add segment level head and tailroom

From: EXT Bill Fischofer [mailto:bill.fischo...@linaro.org] Sent: Tuesday, April 19, 2016 5:39 PM To: Savolainen, Petri (Nokia - FI/Espoo) Cc: LNG ODP Mailman List Subject: Re: [lng-odp] [API-NEXT PATCH 2/3] api: packet: add segment level

Re: [lng-odp] [PATCH v2] validation: time: shorten test duration

On 19.04.16 17:13, Petri Savolainen wrote: Time test duration was almost a minute on a 2.6GHz CPU. Combined local and global time monotonity tests, so that the there is only single run of the long wait loop. Shortened wait test time to 3 seconds and long loop into roughly 14 seconds on a 2.6GHz

Re: [lng-odp] [API-NEXT PATCH 2/3] api: packet: add segment level head and tailroom

General question: How is segment-level headroom/tailroom specified and/or configured? Also, there still hasn't been a concrete use-case made as to why these APIs are needed, or why they are needed specifically in Monarch and cannot be deferred to Tiger Moth. On Tue, Apr 19, 2016 at 8:07 AM,

Re: [lng-odp] [PATCH] validation: time: shorten test duration

> -Original Message- > From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org] > Sent: Monday, April 18, 2016 3:53 PM > To: Savolainen, Petri (Nokia - FI/Espoo) ; > lng-odp@lists.linaro.org > Subject: Re: [lng-odp] [PATCH] validation: time: shorten test

Re: [lng-odp] [GIT PULL ODP] running things in process mode

On 19 April 2016 at 14:30, Mike Holmes wrote: > I have a thought about odph_linux_odpthreads_create() and odph_linux_ > odpthreads_join() > > Should these be > > odph_odpthreads_create() and odph_odpthreads_join() > > > That way we create the notion of an odpthread that

[lng-odp] [API-NEXT PATCH 2/3] api: packet: add segment level head and tailroom

Segment level data length may be modified on bounds of segment level head- and tailrooms. Segment level head/tailroom content is not preserved when segmentation changes. Signed-off-by: Petri Savolainen --- include/odp/api/spec/packet.h | 106

[lng-odp] [API-NEXT PATCH 3/3] api: packet: rename and add copy functions

Added packet copy functions: * partial packet copy * data copy data from another packet * data copy within packet * data move within packet (allow overlap) Renamed copydata_in and _out functions, so that "to/from memory" functions are clearly distinctive from the new packet copy functions.

[lng-odp] [API-NEXT PATCH 1/3] api: packet: add packet concat and split

Added odp_packet_concat() and _split(). Two packets may be concatenated one after another, and split at any offset. This suites the common use cases of packet reassembly and fragmentation. Reassembly is done on complete packets, while fragmentation may be needed on any offset. Signed-off-by:

Re: [lng-odp] [GIT PULL ODP] running things in process mode

I have a thought about odph_linux_odpthreads_create() and odph_linux_ odpthreads_join() Should these be odph_odpthreads_create() and odph_odpthreads_join() That way we create the notion of an odpthread that is OS independent and then the tests possibly extend to another OS without

[lng-odp] [API-NEXT PATCH 2/2] validation: pktio: add test for odp_pktio_info()

Signed-off-by: Matias Elo --- test/validation/pktio/pktio.c | 28 test/validation/pktio/pktio.h | 1 + 2 files changed, 29 insertions(+) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index d52a520..415f8be 100644

[lng-odp] [API-NEXT PATCH 1/2] linux-generic: pktio: add missing pktio names

Add names to ipc and tap pktio types. Signed-off-by: Matias Elo --- platform/linux-generic/pktio/ipc.c | 1 + platform/linux-generic/pktio/tap.c | 1 + 2 files changed, 2 insertions(+) diff --git a/platform/linux-generic/pktio/ipc.c b/platform/linux-generic/pktio/ipc.c

Re: [lng-odp] [API-NEXT PATCHv6] api: make only the API visible

From: EXT Mike Holmes [mailto:mike.hol...@linaro.org] Sent: Tuesday, April 19, 2016 2:13 PM To: Bill Fischofer Cc: Savolainen, Petri (Nokia - FI/Espoo) ; lng-odp-forward Subject: Re: [lng-odp] [API-NEXT PATCHv6]

Re: [lng-odp] [API-NEXT PATCHv6] api: make only the API visible

On 19 April 2016 at 06:39, Bill Fischofer wrote: > > > On Tue, Apr 19, 2016 at 2:42 AM, Savolainen, Petri (Nokia - FI/Espoo) < > petri.savolai...@nokia.com> wrote: > >> I think we should not add #include odp/api/abi_begin.h/_end.h >> speculatively. I think that abi

[lng-odp] Fwd: Re: [PATCH] configure: default ODP_DEPUG_PRINT to false

Merged, Maxim. Forwarded Message Subject:Re: [lng-odp] [PATCH] configure: default ODP_DEPUG_PRINT to false Date: Fri, 15 Apr 2016 15:19:11 -0300 From: Ricardo Salveti To: Mike Holmes CC: LNG ODP Mailman

Re: [lng-odp] [API-NEXT PATCH 1/3] api: pktio: add user context support for interfaces

On 13 April 2016 at 21:47, Bill Fischofer wrote: > > On Wed, Apr 13, 2016 at 2:40 PM, Ola Liljedahl > wrote: > >> >> >> On 13 April 2016 at 21:26, Bill Fischofer >> wrote: >> >>> We already have odp_packet_input()

Re: [lng-odp] [API-NEXT PATCHv6] api: make only the API visible

On Tue, Apr 19, 2016 at 2:42 AM, Savolainen, Petri (Nokia - FI/Espoo) < petri.savolai...@nokia.com> wrote: > I think we should not add #include odp/api/abi_begin.h/_end.h > speculatively. I think that abi definition will go into their own spec > files parallel to API spec, like this > > //

[lng-odp] [API-NEXT PATCH] validation: pktio: add tests for odp_pktin_recv_tmo() and odp_pktin_recv_mq_tmo()

Add validation tests for odp_pktin_recv_tmo() and odp_pktin_recv_mq_tmo() function. Adds also a new helper function recv_packets_tmo() for testing both functions. Signed-off-by: Matias Elo --- Applies on top of the 'add pktin timestamping support' patch set.

Re: [lng-odp] [GIT PULL ODP] running things in process mode

Actually, it’s not the control of thread/process creation but the representation of those. With pthread create, user expects to give void *(*start)(void *) function pointer for thread starting point (with void* return type) and with processes there’s no entry point, but current thread of

Re: [lng-odp] [GIT PULL ODP] running things in process mode

On 19 April 2016 at 09:46, Maxim Uvarov wrote: > On 19.04.2016 10:40, Christophe Milard wrote: > >> Thanks for your comment, Petri. You are welcome to point out the coding >> guideline faults you have seen, so I don't miss them. I will review for the >> variable

Re: [lng-odp] [GIT PULL ODP] running things in process mode

On 19 April 2016 at 09:39, Maxim Uvarov wrote: > Similar changes can go to one patch. Like the same patches that change > args for unit tests. > Not sure what you mean here. If you mean that the test changes could be gathered in one single patch I could possibly agree

Re: [lng-odp] [GIT PULL ODP] running things in process mode

On 19.04.2016 10:40, Christophe Milard wrote: Thanks for your comment, Petri. You are welcome to point out the coding guideline faults you have seen, so I don't miss them. I will review for the variable declaration position. Regarding your other comments, wouldn't it be better to add a fifth

Re: [lng-odp] [API-NEXT PATCHv6] api: make only the API visible

I think we should not add #include odp/api/abi_begin.h/_end.h speculatively. I think that abi definition will go into their own spec files parallel to API spec, like this // Current API spec file // - Misses some typedefs e.g. odp_foo_t include/odp/api/spec/foo.h // ABI specification for 64bit

Re: [lng-odp] [GIT PULL ODP] running things in process mode

Thanks for your comment, Petri. You are welcome to point out the coding guideline faults you have seen, so I don't miss them. I will review for the variable declaration position. Regarding your other comments, wouldn't it be better to add a fifth parameter to odph_linux_odpthreads_create() like

Re: [lng-odp] [GIT PULL ODP] running things in process mode

Similar changes can go to one patch. Like the same patches that change args for unit tests. Is that patches sequence buildable? I think we also need to send this to mailing list. But it's huge number of small patches which probably might be reasonable to split on 2 or 3 logical patch sets if

Re: [lng-odp] [GIT PULL ODP] running things in process mode

Hi, Looks reasonable. There are some style / coding guideline faults (like should introduce all variables in the beginning of a code block). Also pure process and pthread helpers should remain (suggesting to leave out "helper: removing dead code"), since not all apps want to obscure whether

Re: [lng-odp] [PATCH] doc: api: gitignore output directory

already have this change, it's already in next and will be merged to master before next release. commit 3b409242fdd8516255c980cce4f8d16c1933ca81 Author: Bill Fischofer Date: Sun Mar 27 11:10:04 2016 -0500 doc: application: add missing .gitignore file for doc