tsuites.h
+++ b/test/validation/classification/odp_classification_testsuites.h
@@ -13,11 +13,24 @@
extern CU_TestInfo classification_suite[];
extern CU_TestInfo classification_suite_basic[];
+extern CU_TestInfo classification_suite_pmr[];
int classification_suite_init(void);
int classification_suite_term(void
http://www.linaro.org/>***│ *Open source software for ARM SoCs
__
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
--
Regards,
Ivan Khoronzhuk
___
The word "cycle" is left from old API time names. The "cycle" is
ambiguous word, especially when it can be used for other purposes.
So better to use "tick" or "time" word or just "t" symbol.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz.
On 15.10.15 16:39, Bala Manoharan wrote:
Hi Ivan,
Thanks for pointing out the issues. Since this patch is merged I will
create a bug and add the missing points.
Pls provide your inputs on the comments.
On 15 October 2015 at 16:53, Ivan Khoronzhuk <ivan.khoronz...@linaro.org> wrote:
Hi
It shouldn't depend on catalog the project is built.
On 15.10.15 18:10, Ivan Khoronzhuk wrote:
This reverts commit a9cc0fc700a4a8b9589404a18136b01974ca4aa3.
This revert helps me to revert messages like:
"fatal: No names found, cannot describe anything." in case if I build
ODP outside
inaro.org>
Reviewed-and-tested-by: Mike Holmes <mike.hol...@linaro.org>
Signed-off-by: Maxim Uvarov <maxim.uva...@linaro.org>
Call to git_hash now adds a path to top_srcdir and handles non git dir
On 10/15/2015 05:10 PM, Ivan Khoronzhuk wro
This reverts commit a9cc0fc700a4a8b9589404a18136b01974ca4aa3.
This revert helps me to revert messages like:
"fatal: No names found, cannot describe anything." in case if I build
ODP outside of source catalog by:
../../odp/configure --enable-test-perf --enable-test-vald
make
Seems it's connected
if git --git-dir=${ROOTDIR}/.git diff-index --name-only HEAD
&>/dev/null ; then
dirty=-dirty
fi
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
On 15.10.15 19:51, Ivan Khoronzhuk wrote:
On 15.10.15 19:31, Mike Holmes wrote:
On 15 October 2015 at 14:04, Ivan Khoronzhuk <ivan.khoronz...@linaro.org
<mailto:ivan.khoronz...@linaro.org>> wrote:
Please, no Camel casemaybe we can duplicate the struct with na
On 15.10.15 19:31, Mike Holmes wrote:
On 15 October 2015 at 14:04, Ivan Khoronzhuk <ivan.khoronz...@linaro.org
<mailto:ivan.khoronz...@linaro.org>> wrote:
Please, no Camel casemaybe we can duplicate the struct with names and
use convenient one.
So we adop
When counter reaches it's counter value, the next clock cycle will
cause it to roll over to zero. As result the counter spends 1
additional cycle to switch from UINT64_MAX to 0, it should be
taken into account with diff function.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.
Currently, time API reuses cpu API to get ticks, but uses it's own
function to count diff but it can differ from cpu diff function.
So, better to use the same function in order to eliminate possible
difference in counting.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
pl
This series fixes cycle lost while counter overflow.
First patch move diff function in one place.
Second corrects diff function.
Based on api-next as it corrects new cpu cycle api.
Ivan Khoronzhuk (2):
linux-generic: use cycles_diff for time API also
linux-generic: cpu: fix cycle lost while
/platform/linux-generic/include/odp/packet_io.h
b/platform/linux-generic/include/odp/packet_io.h
index 1d690f5..18f8e78 100644
--- a/platform/linux-generic/include/odp/packet_io.h
+++ b/platform/linux-generic/include/odp/packet_io.h
@@ -33,6 +33,7 @@ extern "C" {
The direct comparing of "cur_cycles" and "end_cycles" is not valid,
as "end_cycles" can be overflowed and comparison will give wrong
result. So use odp_time_diff_cycles() for that, as it takes in
account cycles overflow.
Signed-off-by: Ivan Khoronzhuk <iva
add any functional changes.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
example/ipsec/odp_ipsec.c | 57 +--
1 file changed, 20 insertions(+), 37 deletions(-)
diff --git a/example/ipsec/odp_ipsec.c b/example/ipsec/odp_ipsec.c
When times are equal the difference should be 0. Currently it's equal
to UINT64_MAX and doesn't allow to compare ranges beginning from
start time. The validation test to check it will be added later.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
platform/linux-g
The direct comparing of "cur_cycles" and "next_tx_cycles" is not
valid, as "next_tx_cycles" can be overflowed and comparison will
give wrong result. So use odp_time_diff_cycles() for that, as it
takes into account ticks overflow.
Signed-off-by: Ivan Khoronzh
On 16.09.15 12:16, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 16, 2015 11:19 AM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp
- included as new patch
Based on master.
CC:
stuart.has...@linaro.org
ola.liljed...@linaro.org
petri.savolai...@nokia.com
Ivan Khoronzhuk (5):
example: timer: print timer ticks/ns table instead of cycles/ns
example: ipsec: avoid mixing of scheduler wait time and time API time
linux-generic
The timer API can have nothing common with CPU cycles or time API.
Thus timer test shouldn't print conversion cycles/ns table.
More correct to print conversion table for timer ticks/ns.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
example/timer/odp_timer_test.
On 14.09.15 16:35, Maxim Uvarov wrote:
Use full default word in api to make function
name more clear.
https://bugs.linaro.org/show_bug.cgi?id=1745
Signed-off-by: Maxim Uvarov <maxim.uva...@linaro.org>
Reviewed-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
v2: - merge 1
Maxim,
On 16.09.15 11:05, Maxim Uvarov wrote:
On 09/15/15 18:03, Ivan Khoronzhuk wrote:
ping
are you planning v3 with removed
CU_ASSERT(wait_time > 0) ?
Maxim.
Not here.
It will be removed along with ODP-190 implementation.
On 10.09.15 19:18, Ivan Khoronzhuk wrote:
These patc
Petri,
On 16.09.15 10:02, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 15, 2015 5:07 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp
On 16.09.15 10:02, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 15, 2015 5:07 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [PATCH
On 16.09.15 10:52, Ivan Khoronzhuk wrote:
On 16.09.15 10:02, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 15, 2015 5:07 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp
Petri,
What about this fix? It's similar to to CPU API.
On 11.09.15 13:04, Ivan Khoronzhuk wrote:
It's better to describe by example:
cur = 15;
start = 15;
diff = 2;
while (odp_time_cycles_diff(start, cur) < diff) {
cur = odp_time_cycles();
}
This example has to work. It's possi
On 16.09.15 14:39, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 16, 2015 1:47 PM
To: lng-odp@lists.linaro.org; Savolainen, Petri (Nokia - FI/Espoo)
Subject: Re: [lng-odp
On 16.09.15 13:23, Ivan Khoronzhuk wrote:
Petri,
What about this fix? It's similar to to CPU API.
On 11.09.15 13:04, Ivan Khoronzhuk wrote:
It's better to describe by example:
cur = 15;
start = 15;
diff = 2;
while (odp_time_cycles_diff(start, cur) < diff) {
cur = odp_time_cyc
Petri,
Could you please review this patch?
This patch fixes bad ipsec example we were talked about some time ago.
On 11.09.15 13:04, Ivan Khoronzhuk wrote:
It's not correct to mix wall time and scheduler wait time,
used timers can have different rates. As in this example scheduler
is used
On 15.09.15 14:42, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
EXT Ivan Khoronzhuk
Sent: Friday, September 11, 2015 1:05 PM
To: lng-odp@lists.linaro.org
Subject: [lng-odp] [PATCH v2 5/5
ping
On 10.09.15 19:18, Ivan Khoronzhuk wrote:
These patches corrects ODP_SCHED_NO_WAIT corner case when scheduler time
has resolution more then 1ns.
Since v2:
- linux-generic: odp_schedule: fix odp_schdule_wait_time
Correct commit message
- validation: schedule: don't check schedule
Petri,
On 15.09.15 16:01, Ivan Khoronzhuk wrote:
On 15.09.15 15:54, Ivan Khoronzhuk wrote:
On 15.09.15 15:45, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 15, 2015 3:08 PM
fo.page_size;
}
-const char *odp_sys_cpu_model_str(void)
+const char *odp_cpu_model_str(void)
{
return odp_global_data.system_info.model_str[0];
}
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://list
On 15.09.15 15:45, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 15, 2015 3:08 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [PATCH
On 15.09.15 15:54, Ivan Khoronzhuk wrote:
On 15.09.15 15:45, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 15, 2015 3:08 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp
Stuart,
Could you please review this patch.
It uses start_cycles as it was proposed.
On 11.09.15 13:04, Ivan Khoronzhuk wrote:
The direct comparing of "cur_cycles" and "end_cycles" is not valid,
as "end_cycles" can be overflowed and comparison
Petri,
On 11.09.15 12:47, Savolainen, Petri (Nokia - FI/Espoo) wrote:
Ping.
It be good schedule some task to update/add validation test for cpu API.
-Original Message-
From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 08, 2015 1:04 PM
The direct comparing of "cur_cycles" and "next_tx_cycles" is not
valid, as "next_tx_cycles" can be overflowed and comparison will
give wrong result. So use odp_time_diff_cycles() for that, as it
takes in account ticks overflow.
Signed-off-by: Ivan Khoronzhuk <ivan
to review it.
On 10.09.15 19:18, Ivan Khoronzhuk wrote:
The ODP_SCHED_NO_WAIT now corresponds to 0, not 1.
So no need to check it anymore.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
test/validation/scheduler/scheduler.c | 2 --
1 file changed, 2 deletions(-)
diff
The timer API can have nothing common with CPU cycles or time API.
Thus timer test shouldn't print here conversion cycles/ns table.
More correct to print here conversion table for timer ticks/ns.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
example/timer/odp_timer_test.
It's better to describe by example:
cur = 15;
start = 15;
diff = 2;
while (odp_time_cycles_diff(start, cur) < diff) {
cur = odp_time_cycles();
}
This example has to work. It's possible only when t2 - t1 = 0 if t2 = t1.
The validation test on it will be added later.
Signed-off-by: I
functional changes.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
example/ipsec/odp_ipsec.c | 57 +--
1 file changed, 20 insertions(+), 37 deletions(-)
diff --git a/example/ipsec/odp_ipsec.c b/example/ipsec/odp_ipsec.c
index 9
The direct comparing of "cur_cycles" and "end_cycles" is not valid,
as "end_cycles" can be overflowed and comparison will give wrong
result. So use odp_time_diff_cycles() for that, as it takes in
account cycles overflow.
Signed-off-by: Ivan Khoronzhuk <ivan
s://lists.linaro.org/pipermail/lng-odp/2015-August/014850.html
Since v1:
- included new patch
"example: timer: print timer ticks/ns table instead of cycles/ns"
Based on master.
CC:
stuart.haslam at linaro.org
ola.liljed...@linaro.org
petri.savolai...@nokia.com
Ivan Khoronzhuk (5):
Ola,
The patch with your corrections is included in the series that I sent recently:
"[lng-odp] [PATCH v2 0/5] preparation series before updating odp time API"
On 07.09.15 14:55, Ivan Khoronzhuk wrote:
Ola,
On 07.09.15 14:45, Ola Liljedahl wrote:
On 7 September 2015 at 11:45, Ivan
On 11.09.15 13:04, Ivan Khoronzhuk wrote:
It's not correct to mix wall time and scheduler wait time,
It's not wall time, rather time from time API.
Should be corrected.
used timers can have different rates. As in this example scheduler
is used only for poll purposes, using wait time
The ODP_SCHED_NO_WAIT now corresponds to 0, not 1.
So no need to check it anymore.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
test/validation/scheduler/scheduler.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/test/validation/scheduler/scheduler.c
b/test/vali
Nicolas,
On 09.09.15 16:38, Ivan Khoronzhuk wrote:
On 09.09.15 16:27, Ivan Khoronzhuk wrote:
On 09.09.15 15:44, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 09, 2015 3:35
15093.html
Ivan Khoronzhuk (2):
linux-generic: odp_schedule: fix odp_schdule_wait_time
validation: schedule: don't check schedule time on 0
platform/linux-generic/include/odp/plat/schedule_types.h | 4 ++--
platform/linux-generic/odp_schedule.c| 3 ---
test/validation
at all.
Also ODP_SCHED_NO_WAIT corresponds to 1 tick, it's rarely but can
wait a little, when shced time has slower rate. It should correspond
to schedule only once.
So, change ODP_SCHED_NO_WAIT to 0 and ODP_SCHED_NO_WAIT to UINT64_MAX.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.
On 10.09.15 15:21, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 12:49 PM
To: lng-odp@lists.linaro.org; Savolainen, Petri (Nokia - FI/Espoo);
nmo...@kalray.eu
Cc: Ivan
On 10.09.15 16:03, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 3:34 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org;
nmo...@kalray.eu
Subject: Re
Petri,
Are you OK with this change?
On 10.09.15 12:48, Ivan Khoronzhuk wrote:
The resolution of schedule time can be more than 1ns. As result
can happen that 1ns corresponds 0 ticks of timer counter, but if
passed 1ns it's obvious that user wanted to schedule at least once.
Currently it can
_
lng-odp mailing list
lng-odp@lists.linaro.org <mailto:lng-odp@lists.linaro.org>
https://lists.linaro.org/mailman/listinfo/lng-odp
_______
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/l
ing 1/2. I prefer to hold the same numbering within series.
But, next time I will bear it in mind.
The log entry could be clarified a bit, see under.
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 12:49 PM
T
Petri,
On 10.09.15 17:35, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 5:13 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org;
nmo...@kalray.eu
All,
On 10.09.15 16:15, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 4:00 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org;
nmo...@kalray.eu
Subject
= odp_schedule_wait_time(1);
Ivan Khoronzhuk (2):
linux-generic: odp_schedule: fix odp_schdule_wait_time
validation: schedule: don't check schedule time on 0
platform/linux-generic/include/odp/plat/schedule_types.h | 4 ++--
platform/linux-generic/odp_schedule.c| 3 ---
test
to schedule only once.
So, change ODP_SCHED_NO_WAIT to 0 and ODP_SCHED_NO_WAIT to UINT64_MAX.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
platform/linux-generic/include/odp/plat/schedule_types.h | 4 ++--
platform/linux-generic/odp_schedule.c| 3
The ODP_SCHED_NO_WAIT now corresponds to 0, not 1.
So no need to check it anymore.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
test/validation/scheduler/scheduler.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/test/validation/scheduler/scheduler.c
b/test/vali
On 10.09.15 21:08, Ivan Khoronzhuk wrote:
Petri,
On 10.09.15 18:16, Ivan Khoronzhuk wrote:
Petri,
On 10.09.15 17:35, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 5
Petri,
On 10.09.15 18:16, Ivan Khoronzhuk wrote:
Petri,
On 10.09.15 17:35, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 10, 2015 5:13 PM
To: Savolainen, Petri (Nokia - FI/Espoo
On 09.09.15 15:34, Ivan Khoronzhuk wrote:
On 09.09.15 15:00, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
ext Ivan Khoronzhuk
Sent: Monday, September 07, 2015 11:51 AM
To: lng-odp
On 09.09.15 15:00, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
ext Ivan Khoronzhuk
Sent: Monday, September 07, 2015 11:51 AM
To: lng-odp@lists.linaro.org
Subject: [lng-odp] [Patch] linux-generic
(which don’t
produce events), or driving back ground processing (started only when app is
idling).
-Petri
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
--
Regards,
Iva
On 09.09.15 15:44, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 09, 2015 3:35 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp
On 09.09.15 16:27, Ivan Khoronzhuk wrote:
On 09.09.15 15:44, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Wednesday, September 09, 2015 3:35 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp
nt64_t odp_time_cycles(void)
+{
+ return odp_cpu_cycles();
+}
+
uint64_t odp_time_diff_cycles(uint64_t t1, uint64_t t2)
{
if (odp_likely(t2 > t1))
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.l
On 08.09.15 12:37, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 08, 2015 12:20 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [API
On 08.09.15 11:35, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 08, 2015 10:26 AM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [API
On 08.09.15 11:31, Petri Savolainen wrote:
Added implementation for CPU cycle diff, max and resolution.
Signed-off-by: Petri Savolainen <petri.savolai...@nokia.com>
Reviewed-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
platform/linux-generic/Makefile.am
On 08.09.15 14:03, Maxim Uvarov wrote:
On 09/08/15 13:55, Ivan Khoronzhuk wrote:
Maxim,
Could you please clarify.
I expect that after patch adding some implementation should build.
If you adding new API it doesn't have impact.
But if API is modified, probably that project has build problem
UNUSED)
{
odp_cpumask_zero(mask);
/* By default all control threads on CPU 0 */
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
On 08.09.15 11:31, Petri Savolainen wrote:
Renamed time_cycles files to cpu_cycles, since those files
implement cpu cycle counter read needed by cpu API but not
needed by new time API.
Signed-off-by: Petri Savolainen <petri.savolai...@nokia.com>
Reviewed-by: Ivan Khoronzhuk <iva
at all.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
Prerequisit for this patch was taken from:
"[lng-odp] [Patch] validation: scheduler: increase time check"
https://lists.linaro.org/pipermail/lng-odp/2015-August/014738.html
platform/linux-generic/od
something
common with timer API, but that's not true.
Maybe it can be deleted at all, just want to leave it somewhere.
On 03.09.15 11:50, Ivan Khoronzhuk wrote:
The linux-generic implements timers supposing that timers are based
on CPU cycle counter. It's not always true and timer API can have
nothing
On 07.09.15 14:08, Hongbo Zhang wrote:
On 3 September 2015 at 22:47, Ivan Khoronzhuk
<ivan.khoronz...@linaro.org> wrote:
Hi, Hongbo
On 11.08.15 10:33, hongbo.zh...@freescale.com wrote:
From: Hongbo Zhang <hongbo.zh...@linaro.org>
Currently the API to get CPU frequency is vag
c1 = odp_cpu_cycles();
....
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
version: %s\n",odp_version_api_str());
printf("CPU model: %s\n",odp_sys_cpu_model_str());
- printf("CPU freq (hz): %"PRIu64"\n", odp_sys_cpu_hz());
+ printf("CPU freq (hz): %" PRIu64 "\n", odp_sys_cpu_hz
s
+ * resolution of odp_cpu_cycles() in CPU cycles.
+ *
+ * @return CPU cycle count resolution in CPU cycles
+ */
+uint64_t odp_cpu_cycles_resolution(void);
+
+/**
* @}
*/
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.
; !cycle)
cycle = 1;
ODP_SCHED_NO_WAIT (1) like check once, so if we passed ~ 1ns we wanted to
schedule at least once.
Currently it can lead to wait forever ODP_SCHED_WAIT (0).
On Thu, Sep 3, 2015 at 1:28 AM, Ivan Khoronzhuk <ivan.khoronz...@linaro.org
<mailto:ivan.khoronz...@lina
Petri,
On 04.09.15 12:14, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Thursday, September 03, 2015 6:26 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp
ectly? Check it's 32 bit or 64 in application? Check
time wrap period? It's not
very friendly on my opinion. Always use only for ranges is more correct variant.
-Petri
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lis
odp_queue_lookup(name);
if (q == ODP_QUEUE_INVALID)
return -1;
+ qctx = odp_queue_context(q);
+ if (qctx)
+ odp_buffer_free(qctx->ctx_handle);
return odp_queue_destroy(q);
}
@@ -952,6 +1083,9 @@ static int destroy_queues(void)
z_max();
CU_ASSERT(0 < hz);
}
@@ -90,7 +90,7 @@ CU_TestInfo system_suite[] = {
{"odp_cpu_model_str", system_test_odp_cpu_model_str},
{"odp_sys_page_size", system_test_odp_sys_page_size},
{"odp_sys_huge_page_size",
Petri,
On 03.09.15 16:13, Ivan Khoronzhuk wrote:
On 03.09.15 15:29, Savolainen, Petri (Nokia - FI/Espoo) wrote:
There are many example apps that actually want to measure execution time in CPU
cycles, not in nsec.
Time API and CPU (cycle) API updates are linked together. Real CPU cycle
u_id_hz_max(int id)
return -1;
}
+uint64_t odp_cpu_hz(void)
+{
+ int id = sched_getcpu();
+
+ return arch_cpu_hz_current(id);
+}
+
+uint64_t odp_cpu_id_hz(int id)
+{
+ return arch_cpu_hz_current(id);
+}
+
uint64_t odp_sys_huge_page_size(void)
{
return od
);
void system_test_odp_cpu_hz_max(void);
+void system_test_odp_cpu_id_hz_max(void);
+void system_test_odp_cpu_hz(void);
+void system_test_odp_cpu_id_hz(void);
/* test arrays: */
extern CU_TestInfo system_suite[];
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
e_page_size(void);
void system_test_odp_cpu_hz_max(void);
+void system_test_odp_cpu_id_hz_max(void);
+void system_test_odp_cpu_hz(void);
+void system_test_odp_cpu_id_hz(void);
/* test arrays: */
extern CU_TestInfo system_suite[];
--
Regards,
Ivan Khoronzhuk
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp
On 03.09.15 09:20, Nicolas Morey-Chaisemartin wrote:
On 09/02/2015 11:16 PM, Ivan Khoronzhuk wrote:
On 02.09.15 12:42, Nicolas Morey-Chaisemartin wrote:
On 08/26/2015 05:47 PM, Ivan Khoronzhuk wrote:
On 26.08.15 18:22, Stuart Haslam wrote:
On Wed, Aug 26, 2015 at 06:11:13PM +0300
The linux-generic implements timers supposing that timers are based
on CPU cycle counter. It's not always true and timer API can have
nothing common with CPU cycles. Thus timer test shouldn't print
here conversion cycles/ns table for time API.
Signed-off-by: Ivan Khoronzhuk <ivan.khor
One correction, that also makes me worry a little.
On 03.09.15 01:29, Ivan Khoronzhuk wrote:
Hi, Petri
We have to look at it proceeding from performance, platform portability and
simplicity
If you want to split on hi-res time and low-res they must have separate
functions and
be not under
ed instead of odp_time_from_ns(0) for
comparison and initialization.
This patch only changes used time API, it doesn't change used var
names for simplicity.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
example/timer/odp_timer_test.c| 22 +++
includ
ng-odp/2015-August/014850.html
v2:
api: time: change API to use ticks instead of cycles
https://lists.linaro.org/pipermail/lng-odp/2015-August/014198.html
Since v1:
- changed series name a little bit
- added opaque type odp_time_t
- added missed spaces in printf
- removed legacy "count"
On 02.09.15 12:42, Nicolas Morey-Chaisemartin wrote:
On 08/26/2015 05:47 PM, Ivan Khoronzhuk wrote:
On 26.08.15 18:22, Stuart Haslam wrote:
On Wed, Aug 26, 2015 at 06:11:13PM +0300, Ivan Khoronzhuk wrote:
It's needed because time resolution can be a little more than 1ns
and in this case
lign linux-generic implementation with
new time API.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
platform/linux-generic/Makefile.am | 1 +
.../linux-generic/arch/linux/odp_time_cycles.c | 4 +-
.../linux-generic/arch/mips64/odp_time_cycles.c| 4 +-
ich causes the test to take nearly 2 minutes.
Signed-off-by: Nicolas Morey-Chaisemartin <nmo...@kalray.eu>
Cc: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
---
Reviewed-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org>
test/validation/pktio/pktio.c | 2 +-
1 file changed, 1 inser
On 02.09.15 12:35, Nicolas Morey-Chaisemartin wrote:
The odp_scheduler() requires time value in its own ticks, so
pass scheduler wait time instead of ns.
Signed-off-by: Nicolas Morey-Chaisemartin <nmo...@kalray.eu>
---
Reviewed-by: Ivan Khoronzhuk <ivan.khoronz...@linaro.org&
The word "cycle" is left from old API time names. The "cycle" is
ambiguous word, especially when it can be used for other purposes.
So better to use "tick" or "time" word or just "t" symbol.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronz.
401 - 500 of 612 matches
Mail list logo