On 29 April 2015 at 04:51, Benoît Ganne bga...@kalray.eu wrote:
We are also interested by IPC and I also think that pktio is the
way to go.
pktio because it is an already existing concept in ODP we can
potentially piggy back on. But IPC has some differences compared to the
vanilla packet
Hi,
Sorry I clicked send way too early in the previous mail :)
We can optimize the odp_packet_parse_flags_t struct to support
both layered parsing and individual parsing by the following way.
I believe this might be useful for both application and implementation.
typedef struct
-Original Message-
From: ext Bala Manoharan [mailto:bala.manoha...@linaro.org]
Sent: Wednesday, April 29, 2015 11:12 AM
To: Savolainen, Petri (Nokia - FI/Espoo)
Cc: ext Zoltan Kiss; lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [PATCH API-NEXT 4/5] api: packet_io: added parse
mode
From: Balasubramanian Manoharan bala.manoha...@linaro.org
This patch removes support for odp_pmr_create_range() function
in the classifier example application.
Signed-off-by: Balasubramanian Manoharan bala.manoha...@linaro.org
---
example/classifier/odp_classifier.c | 106
Hi,
We can optimize the odp_packet_parse_flags_t in the following way to
handle the layered approach for parsing
+typedef struct odp_packet_parse_flags_t {
+ uint32_t eth:1; /** See odp_packet_has_eth() */
+ uint32_t jumbo:1; /** See odp_packet_has_jumbo() */
+
On 29 April 2015 at 13:24, Savolainen, Petri (Nokia - FI/Espoo)
petri.savolai...@nokia.com wrote:
Hi,
Possibly. Although, grouping adds struct complexity, so there should be some
benefit on doing that. I was going to add union all there, but decided to
leave it out since unnamed
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of ext
Zoltan Kiss
Sent: Tuesday, April 28, 2015 8:59 PM
To: lng-odp
Subject: [lng-odp] how to receive and send on multiple threads with
multiple NIC queues?
Hi,
During fixing a bug in
Mike, Zoltan,
Thanks for sharing this.
--
Best regards,
Nikita Kalyazin,
n.kalya...@samsung.com
CE OS Group
Samsung RD Institute Russia
Tel: +7 (495) 797-25-00 #3816
Tel: +7 (495) 797-25-03
Office #1501, 12-1, Dvintsev str.,
Moscow, 127018, Russia
On Tue, Apr 28, 2015 at 08:51:01AM -0400,
Hi,
Possibly. Although, grouping adds struct complexity, so there should be some
benefit on doing that. I was going to add union all there, but decided to
leave it out since unnamed structs/unions are not defined in C99, so the both
the union and the bit field struct inside would have to be
Merged both patches.
Maxim.
On 04/27/2015 19:48, Stuart Haslam wrote:
Coverity complains that the return value of odp_pktio_close() isn't
being checked.
This fixes https://bugs.linaro.org/show_bug.cgi?id=1513
Signed-off-by: Stuart Haslam stuart.has...@linaro.org
---
Ola, Mike please add your review.
Maxim.
On 04/27/2015 19:19, Jerin Jacob wrote:
use the odp_shared_memory allocater instead of global variables to enable
the timer example to run on baremetal/linux process execution environments
Signed-off-by: Jerin Jacob jerin.ja...@caviumnetworks.com
---
It's (v2) on the list (since last Thu):
[lng-odp] [API-NEXT PATCH v2 4/5] api: packet_io: added parse mode
-Petri
-Original Message-
From: ext Zoltan Kiss [mailto:zoltan.k...@linaro.org]
Sent: Tuesday, April 28, 2015 9:17 PM
To: Savolainen, Petri (Nokia - FI/Espoo);
api-next should = master + proposed new API work.
I think this change is needed in api-next so that the merge to master just
migrates a complete working solution.
That is a requirement for any proposed API change making it into master,
it must have working tests, and no regressions in api-next
On 04/29/2015 11:36, Bala Manoharan wrote:
This patch needs to be applied on the master branch after the
following patches are merged from api-next to master branch.
api: classification: remove odp_pmr_create_range() function definition
linux-generic: classification: remove
On 29/04/15 16:41, Ola Liljedahl wrote:
On 29 April 2015 at 16:10, Bill Fischofer bill.fischo...@linaro.org
mailto:bill.fischo...@linaro.org wrote:
We need to be careful about trying to solve problems that haven't
arisen yet. There are several interrelated issues here:
1.
https://bugs.linaro.org/show_bug.cgi?id=1518
Bug ID: 1518
Summary: CID 88336: Wrong sizeof argument odp_l2fwd.c
Product: OpenDataPlane
Version: 1.0.3
Hardware: Other
OS: Linux
Status: UNCONFIRMED
On 29 April 2015 at 16:10, Bill Fischofer bill.fischo...@linaro.org wrote:
We need to be careful about trying to solve problems that haven't arisen
yet. There are several interrelated issues here:
1. Parsing (and classification) in SW takes cycles. This should only be
done once, not
I'll post a (crude) lazy eval patch to enable initial evaluation. Right
now linux-generic and linux-dpdk have drifted apart in this area but I'm
also working to re-converge things a bit to help keep non-DPDK-related
maintenance in sync.
On Wed, Apr 29, 2015 at 10:41 AM, Ola Liljedahl
BTW, what complicates doing lazy evaluation is the existence of the various
odp_packet_has_xxx_set() APIs we've defined. If you set a value you should
be able to retrieve it without triggering parsing since you've taking
control of that variable yourself. So each parse bit actually needs to
be a
It doesn't apply to api-next. Anyway, I've sent an implementation, see
[API-NEXT PATCH] packet: implement optional parsing. But if we go with
Ola's idea, it won't be needed.
On 29/04/15 07:45, Savolainen, Petri (Nokia - FI/Espoo) wrote:
It's (v2) on the list (since last Thu):
[lng-odp]
On 04/29/2015 19:35, Christophe Milard wrote:
It looks that Stuart and I now agree on what we hope would be an
improvment of the test environment.
I guess you (Maxim, Mike, Anders) have been following the
conversation, but here comes some summary.
I lost on some point. Sorry. But after
All the examples are setting the param structure to 0, and ODP_PKTIN_PARSE_ALL
is 0, so no need to change their default behavious.
Signed-off-by: Zoltan Kiss zoltan.k...@linaro.org
---
platform/linux-generic/odp_packet.c | 5 +
1 file changed, 5 insertions(+)
diff --git
Hey Petri, I like the API changes. A very minor nit, looks like a few
places where extra whitespace got injected if you happen to be making other
changes. The review is for the patch set.
Reviewed-by: Robbie King robk...@cisco.com
-Original Message-
From: lng-odp
Hi Bala, thank you! I added that to api-next and rebased it on current
master. Now api-next and master are synced again. And all api-next
patches are on top.
Thanks,
Maxim.
On 04/29/2015 11:36, Bala Manoharan wrote:
This patch needs to be applied on the master branch after the
following
Merged.
Discussed with Mike on his comment that there is no API change it's
define value change, i.e. implementation.
While api: patch prefix is good here to indicate that api file was touched.
Thank you,
Maxim.
On 04/28/2015 14:44, Savolainen, Petri (Nokia - FI/Espoo) wrote:
OK. It avoids
It looks that Stuart and I now agree on what we hope would be an improvment
of the test environment.
I guess you (Maxim, Mike, Anders) have been following the conversation, but
here comes some summary.
Does this looks reasonable for you as well?
Christophe.
1) validation would contain 1
On 29 April 2015 at 17:50, Zoltan Kiss zoltan.k...@linaro.org wrote:
On 29/04/15 16:41, Ola Liljedahl wrote:
On 29 April 2015 at 16:10, Bill Fischofer bill.fischo...@linaro.org
mailto:bill.fischo...@linaro.org wrote:
We need to be careful about trying to solve problems that haven't
On 29/04/15 18:19, Ola Liljedahl wrote:
On 29 April 2015 at 18:14, Zoltan Kiss zoltan.k...@linaro.org
mailto:zoltan.k...@linaro.org wrote:
It doesn't apply to api-next. Anyway, I've sent an implementation,
see [API-NEXT PATCH] packet: implement optional parsing. But if we
go with
On 04/29/2015 23:06, Anders Roxell wrote:
On 29 April 2015 at 21:56, Maxim Uvarov maxim.uva...@linaro.org wrote:
On 04/29/2015 21:55, Anders Roxell wrote:
On 2015-04-29 14:43, Mike Holmes wrote:
On 29 April 2015 at 12:25, Maxim Uvarov maxim.uva...@linaro.org wrote:
Merged.
Discussed with
On 04/29/2015 21:55, Anders Roxell wrote:
On 2015-04-29 14:43, Mike Holmes wrote:
On 29 April 2015 at 12:25, Maxim Uvarov maxim.uva...@linaro.org wrote:
Merged.
Discussed with Mike on his comment that there is no API change it's define
value change, i.e. implementation.
To clarify, we
On 29 April 2015 at 21:56, Maxim Uvarov maxim.uva...@linaro.org wrote:
On 04/29/2015 21:55, Anders Roxell wrote:
On 2015-04-29 14:43, Mike Holmes wrote:
On 29 April 2015 at 12:25, Maxim Uvarov maxim.uva...@linaro.org wrote:
Merged.
Discussed with Mike on his comment that there is no API
On 27 April 2015 at 12:19, Jerin Jacob jerin.ja...@caviumnetworks.com
wrote:
use the odp_shared_memory allocater instead of global variables to enable
the timer example to run on baremetal/linux process execution environments
Signed-off-by: Jerin Jacob jerin.ja...@caviumnetworks.com
On 29 April 2015 at 12:25, Maxim Uvarov maxim.uva...@linaro.org wrote:
Merged.
Discussed with Mike on his comment that there is no API change it's define
value change, i.e. implementation.
To clarify, we discussed if this is an API change and should come via
api-next.
My initial point was
On 29 April 2015 at 18:14, Zoltan Kiss zoltan.k...@linaro.org wrote:
It doesn't apply to api-next. Anyway, I've sent an implementation, see
[API-NEXT PATCH] packet: implement optional parsing. But if we go with
Ola's idea, it won't be needed.
Actually Bill's idea.
If parsing is made lazy,
Signed-off-by: Maxim Uvarov maxim.uva...@linaro.org
---
debian/changelog | 27 +++
include/odp/api/version.h | 2 +-
2 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/debian/changelog b/debian/changelog
index 2087289..c6c0ec8 100644
---
I'm currently playing around with this and as usual there's no free lunch.
The only way to know whether you need to parse is to check a bit, which at
least doubles the cost of the odp_packet_has_xxx() calls since normally
they are just returning a bit. So now the logic for each of these is:
if
On 2015-04-29 14:43, Mike Holmes wrote:
On 29 April 2015 at 12:25, Maxim Uvarov maxim.uva...@linaro.org wrote:
Merged.
Discussed with Mike on his comment that there is no API change it's define
value change, i.e. implementation.
To clarify, we discussed if this is an API change and
On 04/23/2015 13:29, Petri Savolainen wrote:
Application can indicate which packet parsing results it is
interested in (all or none).
Signed-off-by: Petri Savolainen petri.savolai...@nokia.com
---
include/odp/api/packet_flags.h | 1 +
include/odp/api/packet_io.h| 13 +
2
do we need odp_pktio_status() also?
Maxim.
On 04/23/2015 13:29, Petri Savolainen wrote:
Packet IO start and stop enable a controlled setup and
tear down phases.
Interface control sequence:
* odp_pktio_open() creates a pktio interface object
* potential interface configurations
Robbie, please also take a look at proposed pktio api change.
Thanks,
Maxim.
On 04/29/2015 09:56, Savolainen, Petri (Nokia - FI/Espoo) wrote:
Ping. Review for the series needed.
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of ext
Petri
Lazy parsing defers parsing until the results are actually needed.
This allows applications that do their own parsing and never reference
ODP parse results to avoid the overhead of SW parsing.
Signed-off-by: Bill Fischofer bill.fischo...@linaro.org
---
I disprove of this patch. There are ways to accomplish what we need without
changing the API.
On 29 April 2015 at 15:06, Maxim Uvarov maxim.uva...@linaro.org wrote:
On 04/23/2015 13:29, Petri Savolainen wrote:
Application can indicate which packet parsing results it is
interested in (all or
A lazy parser only parses when the results of parsing are asked for by the
application. So if the application never asks (as is the case for OVS
today) then nothing gets called, so no overhead.
There is a certain amount of overhead imposed by lazy parsing for programs
that do ask for parse
I've posted patch http://patches.opendataplane.org/patch/1511/ to add lazy
parsing support to linux-generic. Needs to be ported to odp-dpdk but that
should be straightforward.
This version implements all/none lazy parsing. The first reference to any
ODP parse results API causes a full parse to
Need to evaluate this against patch
http://patches.opendataplane.org/patch/1511/ which implements lazy parsing
as an alternative to the API change.
On Wed, Apr 29, 2015 at 12:05 PM, Zoltan Kiss zoltan.k...@linaro.org
wrote:
All the examples are setting the param structure to 0, and
We need to be careful about trying to solve problems that haven't arisen
yet. There are several interrelated issues here:
1. Parsing (and classification) in SW takes cycles. This should only be
done once, not duplicated between the app and the ODP implementation, and
ideally should only be done
46 matches
Mail list logo
