https://bugs.linaro.org/show_bug.cgi?id=2494
--- Comment #8 from Maciej Czekaj ---
(In reply to Bala Manoharan from comment #7)
> Regards,
> Bala
>
>
> On 3 November 2016 at 19:36, wrote:
> >
On Fri, 11 Nov 2016 11:13:27 +0100 Francois Ozog wrote:
On 11 November 2016 at 10:10, Brian Brooks wrote:
On 11/10 18:52:49, Christophe Milard wrote:
Hi,
My hope was that packet segments would all be smaller than one page
(either normal pages or huge pages)
When is
https://bugs.linaro.org/show_bug.cgi?id=2571
Mike Holmes changed:
What|Removed |Added
Resolution|--- |FIXED
--
Mike Holmes
Program Manager - Linaro Networking Group
Linaro.org │ Open source software for ARM SoCs
"Work should be fun and collaborative, the rest follows"
That post showed up for me.
On Thu, Nov 17, 2016 at 8:30 AM, Mike Holmes wrote:
> --
> Mike Holmes
> Program Manager - Linaro Networking Group
> Linaro.org │ Open source software for ARM SoCs
> "Work should be fun and collaborative, the rest follows"
>
On Thu, Nov 17, 2016 at 9:42 AM, Joe Savage wrote:
> Hey,
>
> I have recently been looking into the suitability of the ODP APIs for the
> fragmentation and reassembly of IP packets. Particularly, I've been focused
> on the split and concat APIs — using the former to break
For this series:
Reviewed-and-tested-by: Bill Fischofer
On Thu, Nov 10, 2016 at 5:07 AM, Petri Savolainen <
petri.savolai...@nokia.com> wrote:
> Pool performance is optimized by using a ring as the global buffer storage.
> IPC build is disabled, since it needs large
Done. My v2 patch applies on top of your v3. Still need your review on the
APIs for packet references.
On Tue, Nov 15, 2016 at 8:50 AM, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia-bell-labs.com> wrote:
>
> <2nd reply to HTML mail>
>
> Actually, the definition of '#define
ODP is about abstracting things so that applications can concentrate on the
logic.
Exposing drivers on this northbound interface contradicts the essence of
ODP.
ODP implementers will decide how they will accomodate driver "additions".
One may even decide to retrieve drivers through TFTP ;-)
If
Trying to reproduce this I'm seeing sporadic failures in the scheduler
validation test that don't seem to appear in the base api-next branch.
Issue seems to be failures in the ordered queue tests:
Test: scheduler_test_multi_mq_mt_prio_n
...linux.c:273:odpthread_run_start_routine():helper: ODP
A new flag called _ODP_ISHM_EXPORT is added to _ishm. When this flag is
specified at reserve() time, an extra file
("/tmp/odp--shm-", where is the process ID of the
main ODP instatiation process and is the block name given at
reserve time) is created, describing to the underlying block
File platform/linux-generic/include/odp_shm_internal.h exposes shm
internals used by IPC. The bits used by the internal flags are moved
to make room for more "official" values.
The platform/linux-generic/include/odp_shm_internal.h file should really
be removed when _ishm is used, but as long as we
Implemented by calling the related functions from _ishm.
Signed-off-by: Christophe Milard
---
platform/linux-generic/odp_shared_memory.c | 31 +++---
1 file changed, 28 insertions(+), 3 deletions(-)
diff --git
The shmem "sunnydays" tests for the north interface API are replaced with
proper tests, testing memory allocation at different time (before and
after ODP thread creation, i.e. the tests make sure shmem behaves
the same regardless of fork time). The tests also include stress testing
trying to
Signed-off-by: Christophe Milard
---
doc/users-guide/users-guide.adoc | 68 ++--
1 file changed, 66 insertions(+), 2 deletions(-)
diff --git a/doc/users-guide/users-guide.adoc b/doc/users-guide/users-guide.adoc
index
Generated binaries should be marked as test programs to be
installed by autotools in test directory.
Signed-off-by: Maxim Uvarov
---
test/linux-generic/mmap_vlan_ins/Makefile.am | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
On 14 November 2016 at 06:02, Yi He wrote:
> One comment inline:
>
> On 11 November 2016 at 22:50, Christophe Milard
> wrote:
>>
>> _odp_ishmphy_memmap_print() prints the virtual to physical address
>> mapping of some memory region (defined by
On Tue, Nov 15, 2016 at 6:20 AM, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia-bell-labs.com> wrote:
>
>
> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
> Sent: Monday, November 14, 2016 3:37 AM
> To: Savolainen, Petri (Nokia - FI/Espoo)
<2nd reply to HTML mail>
Actually, the definition of '#define ODP_UNUSED(x) (void)(x)' implementation
was never discussed. We just wrapped "__attribute__((__unused__))" of GCC.
That's why I'm questioning it here now. What if your compiler X checks for
unused params but does not support the
Why not just scanning a directory or give a command line parameter?
On 16 November 2016 at 12:05, Christophe Milard <
christophe.mil...@linaro.org> wrote:
> So what?
> If we don't allow to load a driver from the north interface, where should
> they be loaded from??
> Or are we saying that we go
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> Christophe Milard
> Sent: Monday, October 31, 2016 10:52 AM
> To: mike.hol...@linaro.org; bill.fischo...@linaro.org; lng-
> o...@lists.linaro.org
> Subject: [lng-odp] [API-NEXT PATCHv6 05/13]
ok, looks like you changed something in the pool and long packets can
not be allocated:
passed
Test: pktio_test_pktio_config ...passed
Test: pktio_test_info ...pktio 0
name pktiop0p1
driver socket
pktio 1
name pktiop1p0
driver socket
passed
Test:
functionality to export and map memory between ODP instance is added:
This includes:
- a bit of simplification in _odp_ishm_reserve() for externaly provided
file descriptors.
- a new function, _odp_ishm_find_exported() to map memory from
other ODP instances (On same OS)
Signed-off-by:
Remaining (forgotten, not freed) blocks are gathered and related files
cleaned when odp_term_global() is called. An error message is also issued
so the application writters get to know about these blocks
Signed-off-by: Christophe Milard
---
On 16 November 2016 at 22:06, Bill Fischofer wrote:
> Signed-off-by: Bill Fischofer
> ---
> CHANGELOG | 211
> ++
> 1 file changed, 211 insertions(+)
>
> diff --git a/CHANGELOG
Ping. Review needed.
Regards,
Bala
On 10 November 2016 at 19:58, Balasubramanian Manoharan
wrote:
> Fixes https://bugs.linaro.org/show_bug.cgi?id=2496
>
> Signed-off-by: Balasubramanian Manoharan
> ---
> v2: Incorporate review comments
>
Reposting this since it doesn't seem to have made it to the mailing list.
For this series:
Reviewed-and-tested-by: Bill Fischofer
On Tue, Nov 15, 2016 at 8:41 AM, Bill Fischofer
wrote:
> For this series:
>
> Reviewed-and-tested-by: Bill
Ping.
> On 1 Nov 2016, at 10:53, Matias Elo wrote:
>
> Add new test application for ordered queue functionality and performance
> validation. The application sets sequence numbers to incoming packets using
> ordered pktin queues and ordered context locks. After being
The odp shared_memory API is changed to use the ODP internal memory
allocator: _ishm.
_ishm supports memory sharing between processes, regardless of fork time.
The test testing the ODP_SHM_PROC flag is also changed to cope with the
new OS sharing interface used by _ishm (link in /tmp).
_ishm.c assumed that both _ishm_init_global() and _ishm_init_local()
had been run to work properly. This assumption turns out the be a problem
if _ishm is to be used as main memory allocator, as many modules
init_global functions assume the availability of the odp_reserve()
function before any
The ODP_SHM_SINGLE_VA flag is created: when set (at odp_shm_reserve()),
this flag guarantees that all ODP threads sharing this memory
block will see the block at the same address (regadless of ODP
thread type -pthread vs process- or fork time)
The flag ODP_SHM_EXPORT is added: when passed at
This flag guarentess the unicity the the block address on all ODP threads.
The patch just exposes the _ODP_ISHM_SINGLE_VA flag of the internal memory
allocator, ishm.
Signed-off-by: Christophe Milard
---
platform/linux-generic/odp_shared_memory.c | 1 +
1 file
From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
Sent: Tuesday, November 15, 2016 3:03 PM
To: Savolainen, Petri (Nokia - FI/Espoo)
Cc: lng-odp-forward
Subject: Re: [lng-odp] [API-NEXT PATCH v3 15/19] linux-gen: packet:
On Tue, Nov 15, 2016 at 8:13 AM, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia-bell-labs.com> wrote:
>
>
> From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
> Sent: Tuesday, November 15, 2016 3:03 PM
> To: Savolainen, Petri (Nokia - FI/Espoo)
Implement the APIs:
- odp_packet_ref_static()
- odp_packet_ref()
- odp_packet_ref_pkt()
- odp_packet_is_ref()
- odp_packet_has_ref()
- odp_packet_unshared_len()
This also involves functional upgrades to the existing packet manipulation
APIs to work with packet references as input.
Signed-off-by:
Ping. This patch set has been reviewed and tested.
Reviewed-and-tested-by: Bill Fischofer
>
Reviewed-by: Petri Savolainen
>
On 24 Oct 2016, at 13:44, Elo, Matias (Nokia -
Signed-off-by: Bill Fischofer
---
doc/users-guide/users-guide-packet.adoc | 261 +++-
1 file changed, 260 insertions(+), 1 deletion(-)
diff --git a/doc/users-guide/users-guide-packet.adoc
b/doc/users-guide/users-guide-packet.adoc
index
Something like this would not need application to load any drivers, just open
pktio with correct name
1) Place all driver libs into a well know directory e.g. odp/drv/libs (user can
which drivers to install into this dir)
2) Only drivers in the directory are considered available
3) Names and
Merged,
Maxim.
On 11/16/16 13:43, Elo, Matias (Nokia - FI/Espoo) wrote:
Ping. This patch set has been reviewed and tested.
Reviewed-and-tested-by: Bill Fischofer >
Reviewed-by: Petri Savolainen
> ODP specifies functional behavior of it's APIs. Implementations are free to
> choose whatever techniques or other internals they wish to realize this
> behavior. So things like lazy/deferred/batched operations are entirely
> possible as long as the external semantics are matched.
Sure, and
For this series of patch:
Reviewed-by: Yi He
On 11 November 2016 at 23:05, Christophe Milard <
christophe.mil...@linaro.org> wrote:
> NOTE: Must be applied on top of "getting physical addresses from
> _ishmphy", V2
>
> since V1:
> -name changes (Maxim, Francois)
>
Sorry, strange to reply into a blank mail thread, my two comments are
solved and for this series of patch:
Reviewed-by: Yi He
On 11 November 2016 at 22:50, Christophe Milard <
christophe.mil...@linaro.org> wrote:
> since V1:
> -changes to match names suggested by Maxim.
From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
Sent: Monday, November 14, 2016 3:37 AM
To: Savolainen, Petri (Nokia - FI/Espoo)
Cc: lng-odp-forward
Subject: Re: [lng-odp] [API-NEXT PATCH v3 15/19] linux-gen: packet:
Hello,
If the north API is the one visible by ODP applications then I don't think
it is a good idea to expose that.
DPDK exposed it at the beginning and is now internal.
That said there must be a standard way to manage drivers fir the benefit of
the device framework.
I don't think the idea of
Command line parameters will go to the application. As far as I know, we
have no ODP parameters at this time, and they would require some command
line processing cooperation with the application (main(argv, argc)) is in
the app, we are just a lib). That put apps using getopts in a bad position.
Since we did make some minor changes to the Traffic Manager API in this
release, need to discuss whether this should be v1.10.2.0 or v1.12.0.0.
I'll send a v2 if we want to pick a different release number.
It would be good to bump this to v1.12 since Monarch-LTS is v1.11 and this
is a higher
Ping.
> On 1 Nov 2016, at 10:53, Matias Elo wrote:
>
> Add new test application for ordered queue functionality and performance
> validation. The application sets sequence numbers to incoming packets using
> ordered pktin queues and ordered context locks. After being
Signed-off-by: Christophe Milard
---
platform/linux-generic/_ishm.c | 22 --
1 file changed, 16 insertions(+), 6 deletions(-)
diff --git a/platform/linux-generic/_ishm.c b/platform/linux-generic/_ishm.c
index 0586a96..782c32f 100644
---
Regards,
Bala
On 11 November 2016 at 13:26, Brian Brooks wrote:
> On 11/10 15:17:15, Bala Manoharan wrote:
>> On 10 November 2016 at 13:26, Brian Brooks wrote:
>> > On 11/07 16:46:12, Bala Manoharan wrote:
>> >> Hi,
>> >
>> > Hiya
>> >
>> >>
Introduce three new APIs that support efficient sharing of portions of
packets.
odp_packet_ref_static() creates an alias for a base packet
odp_packet_ref() creates a reference to a base packet
odp_packet_ref_pkt() creates a reference to a base packet from a supplied
header packet
In addition,
Trying again as the repost doesn't seem to show up on the list either.
For this series:
Reviewed-and-tested-by: Bill Fischofer
On Tue, Nov 15, 2016 at 5:55 PM, Bill Fischofer
wrote:
> Reposting this since it doesn't seem to have made it
git bisect shows this problem starts surfacing with this commit:
eea929d61e25106adc2598448c865f40e4a6f13b is the first bad commit
commit eea929d61e25106adc2598448c865f40e4a6f13b
Author: Petri Savolainen
Date: Thu Nov 10 13:07:39 2016 +0200
linux-gen: pool:
One comment inline:
On 11 November 2016 at 22:50, Christophe Milard <
christophe.mil...@linaro.org> wrote:
> The function _odp_ishmphy_virt_to_phys() is added to query for physical
> addresses (given a virtual address)
> This function is meant to be used to populate the physical address
> of
> /**
> + * Driver loading
> + *
> + * This function is used by the application to load NIC drivers into ODP.
> + * Calls to this function are optional, but should be performed (if any)
> + * after odp_init_global
> + *
> + *
> + * @param filenameDriver shared lib (dynamic library)
> +
Ping. This set has been on the list two months now.
We use l2fwd to measure raw packet IO performance with and without scheduler.
As long as the scheduler mode of the application is un-optimized, scheduler
mode results are worse than those should be. A perfect HW scheduler does not
help, if
So what?
If we don't allow to load a driver from the north interface, where should
they be loaded from??
Or are we saying that we go for a predefined list of driver only?
So drivers have to be .so files (because of autotools), but we don't give
the possibility to load a free driver??
Christophe
Didn't make it to the list. Sending again.
> -Original Message-
> From: Savolainen, Petri (Nokia - FI/Espoo)
> Sent: Wednesday, November 16, 2016 11:21 AM
> To: lng-odp@lists.linaro.org
> Subject: RE: [lng-odp] [PATCH 1/3] test: l2fwd: lookup table for sched
> mode
>
> Ping. This set has
Regards,
Bala
On 15 November 2016 at 22:43, Brian Brooks wrote:
> On Mon, Nov 14, 2016 at 2:12 AM, Bala Manoharan
> wrote:
>> Regards,
>> Bala
>>
>>
>> On 11 November 2016 at 13:26, Brian Brooks wrote:
>>> On 11/10
I have reported a bug https://bugs.linaro.org/show_bug.cgi?id=2595 about the
new VLAN test hang. Could this be the same issue? It hangs for me sometimes,
most times not. It happens both at the tip of the master and api-next.
-Petri
> -Original Message-
> From: Maxim Uvarov
since v6:
- All points according to Petri's request i.e.:
Odp_shm_find_external() changed again: now odp_shm_import().
Function description updated.
since v5:
-fixed doxygen issue (Bill)
since v4:
- All points of v3 according to Petri's request i.e.:
-All API
On Thu, Nov 10, 2016 at 5:07 AM, Petri Savolainen <
petri.savolai...@nokia.com> wrote:
> Added support for multi-segmented packets. The first segments
> is the packet descriptor, which contains all metadata and
> pointers to other segments.
>
> Signed-off-by: Petri Savolainen
Add support for packet reference APIs. Note that this patch series applies
on top of Petri's v3 patch series for pool enhancements:
http://patches.opendataplane.org/patch/7299/
Changes for v2:
- Rebased on top of Petri's pool restructure patch series
Bill Fischofer (5):
api: packet: add
Add validation tests for the new packet reference APIs:
- odp_packet_ref_static()
- odp_packet_ref()
- odp_packet_ref_pkt()
- odp_packet_is_ref()
- odp_packet_has_ref()
- odp_packet_unshared_len()
Signed-off-by: Bill Fischofer
---
On 16 November 2016 at 10:07, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>
>> /**
>> + * Driver loading
>> + *
>> + * This function is used by the application to load NIC drivers into ODP.
>> + * Calls to this function are optional, but should be performed
On 11/17/16 11:36, Savolainen, Petri (Nokia - FI/Espoo) wrote:
I have reported a bug https://bugs.linaro.org/show_bug.cgi?id=2595 about the
new VLAN test hang. Could this be the same issue? It hangs for me sometimes,
most times not. It happens both at the tip of the master and api-next.
The platform tests odp/test/linux-generic/validation/api/shmem
are updated to both test ODP<->linux process memory sharing, but also test
ODP to ODP (different instances) memory sharing.
shmem_linux is the main test process, and shmem_linux.c contains (at
file top) a chart flow of the test
Signed-off-by: Bill Fischofer
---
CHANGELOG | 211 ++
1 file changed, 211 insertions(+)
diff --git a/CHANGELOG b/CHANGELOG
index d8230cd..3ab7354 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,214 @@
+==
One comment inline:
On 11 November 2016 at 22:50, Christophe Milard <
christophe.mil...@linaro.org> wrote:
> _odp_ishmphy_memmap_print() prints the virtual to physical address
> mapping of some memory region (defined by virtuall address + length).
>
> Signed-off-by: Christophe Milard
Th page must be locked, indeed: if the page is swapped out its
physical address does not exist...
_ishm (the internal ODP memory allocator has a flag to lock memory:
_ODP_ISHM_LOCK
On the north API (odp_*), any memory reserved is locked anyway. This
was a requirement from Petri: that is any
On Mon, Nov 14, 2016 at 2:12 AM, Bala Manoharan
wrote:
> Regards,
> Bala
>
>
> On 11 November 2016 at 13:26, Brian Brooks wrote:
>> On 11/10 15:17:15, Bala Manoharan wrote:
>>> On 10 November 2016 at 13:26, Brian Brooks
Signed-off-by: Bill Fischofer
---
CHANGELOG | 211 ++
1 file changed, 211 insertions(+)
diff --git a/CHANGELOG b/CHANGELOG
index d8230cd..3ab7354 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,214 @@
+==
Signed-off-by: Bill Fischofer
---
CHANGELOG | 211 ++
1 file changed, 211 insertions(+)
diff --git a/CHANGELOG b/CHANGELOG
index d8230cd..3ab7354 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,214 @@
+==
On Thu, Nov 17, 2016 at 3:05 AM, Bala Manoharan
wrote:
> Regards,
> Bala
>
>
> On 15 November 2016 at 22:43, Brian Brooks
> wrote:
> > On Mon, Nov 14, 2016 at 2:12 AM, Bala Manoharan
> > wrote:
> >> Regards,
> >>
On 7 November 2016 at 10:49, Maxim Uvarov wrote:
> Turn on linux-generic dpdk and pcap pktios with their validation
> tests.
>
> Signed-off-by: Maxim Uvarov
Reviewed-by: Mike Holmes
We need to make this complete and add
These are good questions. Answers inline.
On Thu, Nov 17, 2016 at 10:52 AM, Joe Savage wrote:
> > ODP specifies functional behavior of it's APIs. Implementations are free
> to
> > choose whatever techniques or other internals they wish to realize this
> > behavior. So things
Hey,
I have recently been looking into the suitability of the ODP APIs for the
fragmentation and reassembly of IP packets. Particularly, I've been focused
on the split and concat APIs — using the former to break larger packets into
smaller fragments, and the latter to glue such fragments back
Hi Bala,
I am trying to catch up on this conversation. I have few questions here,
not sure if they are discussed already.
1) Why is there a need for such large number of queues (Million) in a
system? If the system meets the line rate, the packets (and the flows)
waiting to be processed is
OK, sent v2 with level changed to v1.11.1.0
On Thu, Nov 17, 2016 at 3:17 PM, Mike Holmes wrote:
> On 16 November 2016 at 22:06, Bill Fischofer
> wrote:
> > Signed-off-by: Bill Fischofer
> > ---
> > CHANGELOG | 211
78 matches
Mail list logo