Re: [lng-odp] [API-NEXT PATCH 1/3] api: add cuckoo hash api

Perhaps a generic hash table APIs (not cuckoo hash) could be put in the main APIs? > 在 2016年4月6日，下午10:18，Ru Jia 写道： > > Last year, there was a discussion about cuckoo hash. I saw the last reply > (https://lists.linaro.org/pipermail/lng-odp/2015-December/017921.html >

Re: [lng-odp] [API-NEXT PATCH 5/7] api: packet: add segment level head and tailroom

I think these APIs are clearly out of scope for Monarch. I don't believe we can get a consensus on these since platforms that involve a lot of HW assists for buffer allocation and management are likely to have a lot of problems with these proposed concepts and semantics. We need to develop clear ap

Re: [lng-odp] [API-NEXT PATCH 3/7] api: packet: add packet concat and split

I think these general segment-based operations may be a bit ambitious for Monarch. Might we simplify for staging purposes by saying that we support concatenating two packets rather than segments of packets? In the main use-case we're looking at it seems that the application would construct a packe

Re: [lng-odp] [API-NEXT PATCH 6/7] api: packet: add packet data prefetch

On Thu, Apr 7, 2016 at 9:22 AM, Petri Savolainen wrote: > For improved cache hit rate, prefetch may be used to prefetch packet > data before accessing it the first time. Largest benefit may be expected > for the data that is deep in the packet. Typically, packet head and > metadata are automatica

Re: [lng-odp] [API-NEXT PATCH 7/7] api: pool: add pool capability

This patch seems orthogonal to the rest of this series. Not sure why it's part of it since it seems to be serving a entirely different purpose. Perhaps should be its own independent patch and combined with other parts that deprecate the corresponding odp_config_xxx() APIs that this seems designed t

Re: [lng-odp] [API-NEXT PATCH 4/7] api: packet: remove seg_buf_addr and seg_buf_len

+1 on the deprecation of these APIs. I don't think they were ever used much and I agree that their semantics are problematic. Review for this part only. On Thu, Apr 7, 2016 at 9:22 AM, Petri Savolainen wrote: > Removed definitions to access segment memory outside of segment > data. A tighter se

Re: [lng-odp] [API-NEXT PATCH 2/7] api: packet: re-organize file content

No problems with this part as it's just moving the declarations around. Review is for this part only. On Thu, Apr 7, 2016 at 9:22 AM, Petri Savolainen wrote: > Grouped packet allocation, segmentation and packet head/tail > manipulation functions together. Rest of the file are metadata > read/wri

Re: [lng-odp] [API-NEXT PATCH 1/7] api: packet: add extend and trunc

I'm glad to see these patches posted to start the discussion, however I feel that this needs to be split up a bit into what we can reasonably get into Monarch vs. what needs to wait until Tiger Moth. This entire series seems far too ambitious for Monarch at this date. I'll comment on each part in

[lng-odp] [API-NEXT PATCH 2/3] linux-generic: pktio: implement pktio support for user contexts

Signed-off-by: Bill Fischofer --- .../linux-generic/include/odp_packet_io_internal.h | 1 + platform/linux-generic/odp_packet_io.c | 22 ++ 2 files changed, 23 insertions(+) diff --git a/platform/linux-generic/include/odp_packet_io_internal.h b/platform/linux-ge

[lng-odp] [API-NEXT PATCH 3/3] validation: pktio: add cunit tests for pktio user contexts

Signed-off-by: Bill Fischofer --- test/validation/pktio/pktio.c | 20 test/validation/pktio/pktio.h | 1 + 2 files changed, 21 insertions(+) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index cb403a6..1a90c81 100644 --- a/test/validation/pktio/

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

Add the APIs odp_pktio_context() and odp_pktio_context_set() to permit applications to associate contexts with PktIO interfaces. Suggested-by: Bogdan Pricope Signed-off-by: Bill Fischofer --- include/odp/api/spec/packet_io.h | 21 + 1 file changed, 21 insertions(+) diff --g

[lng-odp] [PATCH 2/2] linux-generic: classification: remove unused code

This function is not referred. Signed-off-by: Zoltan Kiss --- platform/linux-generic/include/odp_classification_internal.h | 9 ++--- platform/linux-generic/odp_classification.c | 10 -- 2 files changed, 2 insertions(+), 17 deletions(-) diff --git a/platform/linux-

[lng-odp] [PATCH 1/2] linux-generic: classification: fix error checking _odp_packet_classifier()

The callers are only interested whether there is a CoS or not. It is not an error if there isn't, so it has a positive return value. Any other case needs to release the packet, which is not handled after calling queue_enq(). Signed-off-by: Zoltan Kiss --- platform/linux-generic/odp_classificatio

[lng-odp] [PATCH 0/2] Minor classification fixes

The first one prevents memory leak/double-free situations though. Zoltan Kiss (2): linux-generic: classification: fix error checking _odp_packet_classifier() linux-generic: classification: remove unused code .../linux-generic/include/odp_classification_internal.h | 9 ++--- platform

Re: [lng-odp] [PATCHv3] validation: configure: move cflags guards to test/m4

Merged, Maxim. On 04/07/16 17:49, Anders Roxell wrote: On 2016-04-07 17:12, Maxim Uvarov wrote: Move cflags guards to test/m4 to fix cunit library search. https://bugs.linaro.org/show_bug.cgi?id=2161 Guess you should add that it failed when building --with-cunit-path=... Reported-by: Brian B

Re: [lng-odp] [PATCH] linux-generic/odp_impl: removing __DATE__ and __TIME__

On 7 April 2016 at 16:59, Ricardo Salveti wrote: > This is so we can get reproducible builds. Besides the library version, > the string already contains the git hash, which is enough to traceback > the build used. > > Bug-Link: https://bugs.linaro.org/show_bug.cgi?id=2154 > > Signed-off-by: Ricard

[lng-odp] [PATCH] linux-generic/odp_impl: removing __DATE__ and __TIME__

This is so we can get reproducible builds. Besides the library version, the string already contains the git hash, which is enough to traceback the build used. Bug-Link: https://bugs.linaro.org/show_bug.cgi?id=2154 Signed-off-by: Ricardo Salveti --- platform/linux-generic/odp_impl.c | 1 - 1 fil

Re: [lng-odp] [PATCHv3] validation: configure: move cflags guards to test/m4

On 2016-04-07 17:12, Maxim Uvarov wrote: > Move cflags guards to test/m4 to fix cunit > library search. > https://bugs.linaro.org/show_bug.cgi?id=2161 Guess you should add that it failed when building --with-cunit-path=... > > Reported-by: Brian Brooks > Suggested-by: Anders Roxell > Signed-of

Re: [lng-odp] [PATCH] linux-generic: packet_io: fix state handling

Merged, Maxim. On 04/07/16 16:57, Zoltan Kiss wrote: Ping, I think it only waits for merge On 03/04/16 20:41, Bill Fischofer wrote: On Fri, Apr 1, 2016 at 11:44 AM, Zoltan Kiss mailto:zoltan.k...@linaro.org>> wrote: The current two state doesn't tell odp_pktio_close() whether the d

[lng-odp] [API-NEXT PATCH 7/7] api: pool: add pool capability

Introduce pool capability structure as preparation for removing pool and segmentation definitions from odp_config.h. Signed-off-by: Petri Savolainen --- include/odp/api/spec/pool.h | 82 + 1 file changed, 82 insertions(+) diff --git a/include/odp/api/

[lng-odp] [API-NEXT PATCH 4/7] api: packet: remove seg_buf_addr and seg_buf_len

Removed definitions to access segment memory outside of segment data. A tighter segment specification makes segment manipulation definitions cleaner. Only packet level head-/tailrooms and segment data are moved with the packet, segment level head-/tailrooms are not. Signed-off-by: Petri Savolainen

[lng-odp] [API-NEXT PATCH 5/7] 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 | 114 +- 1 file cha

[lng-odp] [API-NEXT PATCH 6/7] api: packet: add packet data prefetch

For improved cache hit rate, prefetch may be used to prefetch packet data before accessing it the first time. Largest benefit may be expected for the data that is deep in the packet. Typically, packet head and metadata are automatically prefetched anyway. Signed-off-by: Petri Savolainen --- incl

[lng-odp] [API-NEXT PATCH 3/7] api: packet: add packet concat and split

Added odp_packet_concat() and _split(). Two packets may be concatenated at segment border 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 (independent of segment

[lng-odp] [API-NEXT PATCH 2/7] api: packet: re-organize file content

Grouped packet allocation, segmentation and packet head/tail manipulation functions together. Rest of the file are metadata read/write functions which do not modify packet data storage. Signed-off-by: Petri Savolainen --- include/odp/api/spec/packet.h | 478 +-

[lng-odp] [API-NEXT PATCH 1/7] api: packet: add extend and trunc

Added functions to extend / truncate packet head / tail more than current head/tailroom or segment lengths. Signed-off-by: Petri Savolainen --- include/odp/api/spec/packet.h | 130 ++ 1 file changed, 130 insertions(+) diff --git a/include/odp/api/spec/pac

[lng-odp] [PATCHv3] validation: configure: move cflags guards to test/m4

Move cflags guards to test/m4 to fix cunit library search. https://bugs.linaro.org/show_bug.cgi?id=2161 Reported-by: Brian Brooks Suggested-by: Anders Roxell Signed-off-by: Maxim Uvarov --- v3: drop patch 1. (btw there were no reject). configure.ac | 14 -- test/m4/valida

Re: [lng-odp] [PATCH] linux-generic: packet_io: fix state handling

Ping, I think it only waits for merge On 03/04/16 20:41, Bill Fischofer wrote: On Fri, Apr 1, 2016 at 11:44 AM, Zoltan Kiss mailto:zoltan.k...@linaro.org>> wrote: The current two state doesn't tell odp_pktio_close() whether the device has ever been started, which is important to d

[lng-odp] Change in lng/odp[master]: pkg/debian: build depend on libcunit1-dev

lava-bot has posted comments on this change. Change subject: pkg/debian: build depend on libcunit1-dev .. Patch Set 1: Code-Review-1 "Build Failed https://ci.linaro.org/jenkins/job/odp-api-check-gerrit/5/ : [OUTPUT]:" -- To

Re: [lng-odp] [PATCH] linux-generic: init: comment out last local stage to avoid scan issues

On 7 April 2016 at 08:44, Bill Fischofer wrote: > Resolve bug https://bugs.linaro.org/show_bug.cgi?id=2168 by commenting out > last stage setting in odp_init_local(). This will need to be uncommented > if further stages are added in subsequent updates. > > Signed-off-by: Bill Fischofer > --- >

Re: [lng-odp] [PATCH 1/2] configure: support out-of-tree CUnit

On 7 April 2016 at 15:17, Maxim Uvarov wrote: > From: Brian Brooks > > If --with-cunit-path=DIR is used, skip the AC_CHECK_LIB and modify the linker > flags accordingly. > > Signed-off-by: Brian Brooks > Signed-off-by: Maxim Uvarov > --- > test/m4/validation.m4 | 8 ++-- > 1 file changed,

Re: [lng-odp] [PATCH 2/2] configure: remove dead code

> On 07 Apr 2016, at 16:27, EXT Maxim Uvarov wrote: > > On 04/07/16 16:22, Elo, Matias (Nokia - FI/Espoo) wrote: >>> -Original Message- >>> From: EXT Maxim Uvarov [mailto:maxim.uva...@linaro.org] >>> Sent: Thursday, April 07, 2016 3:46 PM >>> To: lng-odp@lists.linaro.org >>> Cc: Elo, Ma

Re: [lng-odp] [PATCH 2/2] configure: remove dead code

On 04/07/16 16:22, Elo, Matias (Nokia - FI/Espoo) wrote: -Original Message- From: EXT Maxim Uvarov [mailto:maxim.uva...@linaro.org] Sent: Thursday, April 07, 2016 3:46 PM To: lng-odp@lists.linaro.org Cc: Elo, Matias (Nokia - FI/Espoo) Subject: Re: [lng-odp] [PATCH 2/2] configure: remove

Re: [lng-odp] [PATCH 2/2] configure: remove dead code

> -Original Message- > From: EXT Maxim Uvarov [mailto:maxim.uva...@linaro.org] > Sent: Thursday, April 07, 2016 3:46 PM > To: lng-odp@lists.linaro.org > Cc: Elo, Matias (Nokia - FI/Espoo) > Subject: Re: [lng-odp] [PATCH 2/2] configure: remove dead code > > Matias, if you tested that can y

[lng-odp] [PATCH 2/2] validation: configure: move cflags guards to test/m4

Move cflags guards to test/m4 to fix cunit library search. https://bugs.linaro.org/show_bug.cgi?id=2161 Reported-by: Brian Brooks Suggested-by: Anders Roxell Signed-off-by: Maxim Uvarov --- configure.ac | 14 -- test/m4/validation.m4 | 14 ++ 2 files changed, 1

[lng-odp] [PATCH 1/2] configure: support out-of-tree CUnit

From: Brian Brooks If --with-cunit-path=DIR is used, skip the AC_CHECK_LIB and modify the linker flags accordingly. Signed-off-by: Brian Brooks Signed-off-by: Maxim Uvarov --- test/m4/validation.m4 | 8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/test/m4/validation.

Re: [lng-odp] Build broken in master branch

I will send updated patches. Maxim. On 7 April 2016 at 15:35, Savolainen, Petri (Nokia - FI/Espoo) < petri.savolai...@nokia.com> wrote: > > > > -Original Message- > > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT > > Savolainen, Petri (Nokia - FI/Espoo) > > Sen

[lng-odp] [Bug 2136] CID 158516: Out-of-bounds access odp_crypto.c

https://bugs.linaro.org/show_bug.cgi?id=2136 --- Comment #1 from Krishna Garapati --- Sent a patch : http://patches.opendataplane.org/patch/5493/ -- You are receiving this mail because: You are on the CC list for the bug.___ lng-odp mailing list lng-o

[lng-odp] [Bug 2136] CID 158516: Out-of-bounds access odp_crypto.c

https://bugs.linaro.org/show_bug.cgi?id=2136 Mike Holmes changed: What|Removed |Added Status|UNCONFIRMED |IN_PROGRESS Ever confirmed|0

[lng-odp] [Bug 2131] Traffic Manager introduced valgrind failures

https://bugs.linaro.org/show_bug.cgi?id=2131 Bill Fischofer changed: What|Removed |Added Ever confirmed|0 |1 Assignee|lng-odp@lists.linaro.org

[lng-odp] [Bug 2151] Packet pool runs out of blocks for irregular traffic patterns

https://bugs.linaro.org/show_bug.cgi?id=2151 Mike Holmes changed: What|Removed |Added Status|CONFIRMED |RESOLVED Resolution|---

[lng-odp] [Bug 2154] ODP fails to build with GCC 5.3 (__DATE__ and __TIME__ macros, preventing reproducible builds)

https://bugs.linaro.org/show_bug.cgi?id=2154 Mike Holmes changed: What|Removed |Added CC||mike.hol...@linaro.org Assignee|anders

[lng-odp] [Bug 2168] platform/linux-generic/odp_init.c: stage is never used

https://bugs.linaro.org/show_bug.cgi?id=2168 Bill Fischofer changed: What|Removed |Added Status|UNCONFIRMED |IN_PROGRESS Ever confirmed|0

[lng-odp] [Bug 2160] Building natively on AARch64 fails

https://bugs.linaro.org/show_bug.cgi?id=2160 Mike Holmes changed: What|Removed |Added Status|UNCONFIRMED |CONFIRMED Ever confirmed|0

Re: [lng-odp] [PATCH 2/2] configure: remove dead code

Matias, if you tested that can you please send Teted-by: ? Maxim. On 03/25/16 22:25, Brian Brooks wrote: Remove dead code which potentially acted as a workaround for making AC_CHECK_LIB to work with --with-cunit-path. Signed-off-by: Brian Brooks --- configure.ac | 14 -- 1 file

[lng-odp] [PATCH] linux-generic: init: comment out last local stage to avoid scan issues

Resolve bug https://bugs.linaro.org/show_bug.cgi?id=2168 by commenting out last stage setting in odp_init_local(). This will need to be uncommented if further stages are added in subsequent updates. Signed-off-by: Bill Fischofer --- platform/linux-generic/odp_init.c | 2 +- 1 file changed, 1 ins

[lng-odp] Build broken in master branch

> -Original Message- > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT > Savolainen, Petri (Nokia - FI/Espoo) > Sent: Thursday, April 07, 2016 11:26 AM > To: EXT Maxim Uvarov ; Elo, Matias (Nokia - > FI/Espoo) ; EXT Anders Roxell > > Cc: lng-odp@lists.linaro.org

[lng-odp] Change in lng/odp[master]: pkg/debian: build depend on libcunit1-dev

Anders Roxell has posted comments on this change. Change subject: pkg/debian: build depend on libcunit1-dev .. Patch Set 1: Code-Review+1 -- To view, visit https://review.linaro.org/11249 To unsubscribe, visit https://review.l

[lng-odp] Change in lng/odp[master]: pkg/debian: build depend on libcunit1-dev

lava-bot has posted comments on this change. Change subject: pkg/debian: build depend on libcunit1-dev .. Patch Set 1: "Build Started https://ci.linaro.org/jenkins/job/odp-api-check-gerrit/5/ " -- To view, visit https://revie

[lng-odp] Change in lng/odp[master]: pkg/debian: build depend on libcunit1-dev

Riku Voipio has uploaded a new change for review. https://review.linaro.org/11249 Change subject: pkg/debian: build depend on libcunit1-dev .. pkg/debian: build depend on libcunit1-dev To run testsuite on package builds, libc

[lng-odp] [Bug 2151] Packet pool runs out of blocks for irregular traffic patterns

https://bugs.linaro.org/show_bug.cgi?id=2151 --- Comment #3 from Bill Fischofer --- v2 patch submitted by Oriol at http://patches.opendataplane.org/patch/5479/ and reviewed. -- You are receiving this mail because: You are on the CC list for the bug.__

[lng-odp] [Bug 2138] CID 158518: CID 158519: Unchecked return value: odp_packet_io.c

https://bugs.linaro.org/show_bug.cgi?id=2138 --- Comment #4 from Bill Fischofer --- v5 patch submitted: http://patches.opendataplane.org/patch/5479/ -- You are receiving this mail because: You are on the CC list for the bug.___ lng-odp mailing list ln

[lng-odp] [Bug 2089] What happens if the out_queue is not empty?

https://bugs.linaro.org/show_bug.cgi?id=2089 --- Comment #3 from Bill Fischofer --- Patch v2 posted at http://patches.opendataplane.org/patch/5467/ -- You are receiving this mail because: You are on the CC list for the bug.___ lng-odp mailing list lng

Re: [lng-odp] [PATCH 1/2] configure: support out-of-tree CUnit

CI is still failing and I think this is the needed fix, any progress ? On 5 April 2016 at 13:12, Brian Brooks wrote: > On 04/04 16:08:26, Anders Roxell wrote: > > On 26 March 2016 at 02:06, Brian Brooks wrote: > > > On 04/01 22:15:34, Anders Roxell wrote: > > >> On 25 March 2016 at 20:25, Brian

Re: [lng-odp] [openfastpath] User defined context pointer associated with pktio

That's what I was thinking as well. Expect a patch for review later today. Thanks. On Thu, Apr 7, 2016 at 3:24 AM, Bogdan Pricope wrote: > Hi, > > > > I would prefer to avoid odp_pktio_info() as ‘getting’ part will be called > for every input packet and it contains a memset() and a memcpy(). >

Re: [lng-odp] [PATCHv6] linux-generic: test: shmem: atomic check+open fifo

Merged, Maxim. On 04/06/16 21:42, Bill Fischofer wrote: On Wed, Apr 6, 2016 at 7:31 AM, Christophe Milard mailto:christophe.mil...@linaro.org>> wrote: Fixes: https://bugs.linaro.org/show_bug.cgi?id=2146 (CID 159395) The open system call is directely used to check the presence of

Re: [lng-odp] [PATCHv3] test: performance: Avoid possible out-of-bounds memory access

ping On 6 April 2016 at 08:40, Balakrishna Garapati < balakrishna.garap...@linaro.org> wrote: > Resolving https://bugs.linaro.org/show_bug.cgi?id=2136 by initializing > definite number of worker threads. > > Signed-off-by: Balakrishna Garapati > --- > v1: Fix for the https://bugs.linaro.org/sho

Re: [lng-odp] [PATCHv2] linux-generic: packet pool runs out of blocks

Merged, Maxim. On 04/06/16 16:47, Bill Fischofer wrote: On Wed, Apr 6, 2016 at 6:08 AM, Oriol Arcas mailto:or...@starflownetworks.com>> wrote: The packet pool may run out of blocks before reaching the low buffer watermark. A block low watermark is added to fix it. Signed-off-by

Re: [lng-odp] [PATCHv2] configure: remove separate so_version file

Merged, Maxim. On 04/05/16 01:03, Ricardo Salveti wrote: On Fri, Apr 1, 2016 at 5:31 PM, Anders Roxell wrote: Set versioning in configure.ac directly, folks familiar with autotools will expect it to be set there. Suggested-by: Fathi Boudra Signed-off-by: Anders Roxell --- .so_version

Re: [lng-odp] [PATCH] configure: reorder m4_includes

> -Original Message- > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT > Maxim Uvarov > Sent: Thursday, April 07, 2016 10:55 AM > To: Elo, Matias (Nokia - FI/Espoo) ; EXT Anders > Roxell > Cc: lng-odp@lists.linaro.org > Subject: Re: [lng-odp] [PATCH] configure: r

Re: [lng-odp] [openfastpath] User defined context pointer associated with pktio

Hi, I would prefer to avoid odp_pktio_info() as ‘getting’ part will be called for every input packet and it contains a memset() and a memcpy(). I was thinking to an API similar with queue API: int odp_pktio_context_set(odp_pktio_t pktio, void *context) { pktio_entry_t *entry = g

[lng-odp] [PATCH v2] linux-generic: dpdk: implement pktio statistics

Pktio statistics implementation. Signed-off-by: Matias Elo --- v2: - Use memset in stats_convert() - Simplified dpdk_stats() return value platform/linux-generic/pktio/dpdk.c | 36 1 file changed, 36 insertions(+) diff --git a/platform/linux-generic/pktio/dp

Re: [lng-odp] [PATCH] configure: reorder m4_includes

On 04/07/16 10:47, Elo, Matias (Nokia - FI/Espoo) wrote: -Original Message- From: EXT Anders Roxell [mailto:anders.rox...@linaro.org] Sent: Thursday, April 07, 2016 9:06 AM To: Elo, Matias (Nokia - FI/Espoo) Cc: lng-odp@lists.linaro.org; EXT Brian Brooks Subject: Re: [lng-odp] [PATCH]

Re: [lng-odp] [PATCH] configure: reorder m4_includes

> -Original Message- > From: EXT Anders Roxell [mailto:anders.rox...@linaro.org] > Sent: Thursday, April 07, 2016 9:06 AM > To: Elo, Matias (Nokia - FI/Espoo) > Cc: lng-odp@lists.linaro.org; EXT Brian Brooks > Subject: Re: [lng-odp] [PATCH] configure: reorder m4_includes > > On 7 April

Re: [lng-odp] [PATCHv2] configure: remove separate so_version file

Merged, Maxim. On 04/05/16 01:03, Ricardo Salveti wrote: On Fri, Apr 1, 2016 at 5:31 PM, Anders Roxell wrote: Set versioning in configure.ac directly, folks familiar with autotools will expect it to be set there. Suggested-by: Fathi Boudra Signed-off-by: Anders Roxell --- .so_version