On Thu, 17 Nov 2022 19:29:22 GMT, Alexander Zuev wrote:
>> Still have a question why we skip this duplicates and do not consider them
>> as a bug?
>
>> Still have a question why we skip this duplicates and do not consider them
>> as a bug?
>
> I do not have enough information to act on this is
> On m1 mac sometimes (in very rare instance) it takes a long time to get the
> sequencer so increasing timeout helps with that.
> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
> reason is unclear but watching for that fixes the problem. Also on Linux
> sometimes p
On Thu, 17 Nov 2022 18:05:09 GMT, Sergey Bylokhov wrote:
> Still have a question why we skip this duplicates and do not consider them as
> a bug?
I do not have enough information to act on this issue because while not being
problem listed it only occurred twice on CI system and in more than a
On Thu, 17 Nov 2022 03:19:31 GMT, Prasanta Sadhukhan
wrote:
>>> Do we not need to do seq.stop() before close() in l136
>>
>> On that line we never call start() on the sequencer so calling stop() is
>> redundant.
>
> ok, understood...
Still have a question why we skip this duplicates and do no
On Wed, 16 Nov 2022 19:46:03 GMT, Alexander Zuev wrote:
> we do not access counter on another thread until the flag is changed.
But we access it on a different thread w/o any synchronization, which we should
add.
-
PR: https://git.openjdk.org/jdk/pull/11157
On Wed, 16 Nov 2022 19:52:26 GMT, Alexander Zuev wrote:
>> If this code is needed can we check first where we create duplicate events?
>
>> Do we not need to do seq.stop() before close() in l136
>
> On that line we never call start() on the sequencer so calling stop() is
> redundant.
ok, under
On Wed, 16 Nov 2022 05:37:15 GMT, Alexander Zuev wrote:
>> On m1 mac sometimes (in very rare instance) it takes a long time to get the
>> sequencer so increasing timeout helps with that.
>> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
>> reason is unclear but wat
On Wed, 16 Nov 2022 08:50:57 GMT, Sergey Bylokhov wrote:
>> test/jdk/javax/sound/midi/Sequencer/MetaCallback.java line 119:
>>
>>> 117: finished = true;
>>> 118: } else if (msg.getData().length > 0 && msg.getType() == 1 &&
>>> !received.contains(msg)) {
>>> 119:
On Wed, 16 Nov 2022 08:51:01 GMT, Sergey Bylokhov wrote:
> One of the issues is that the flags are updated in the "meta()" on one thread
> and checked on another thread.
I can make counter and boolean flag volatile but looking at the logic of the
updates i do not think it is a problem since bo
On Tue, 15 Nov 2022 19:16:52 GMT, Alexander Zuev wrote:
>> test/jdk/javax/sound/midi/Sequencer/MetaCallback.java line 69:
>>
>>> 67: // On M1 Mac sometimes system notifies listener about the same
>>> message twice
>>> 68: static List received =
>>> Collections.synchronizedList(new Arra
On Wed, 16 Nov 2022 06:45:28 GMT, Prasanta Sadhukhan
wrote:
>> Alexander Zuev has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Fixed typo in copyright year
>
> test/jdk/javax/sound/midi/Sequencer/MetaCallback.java line 119:
>
>> 117:
On Wed, 16 Nov 2022 05:37:15 GMT, Alexander Zuev wrote:
>> On m1 mac sometimes (in very rare instance) it takes a long time to get the
>> sequencer so increasing timeout helps with that.
>> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
>> reason is unclear but wat
On Wed, 16 Nov 2022 04:11:21 GMT, Prasanta Sadhukhan
wrote:
> futuristic year? or you didnt have confidence that this test will be
> integrated this year :-)
Is this still 2022? I guess i just wish this year is over already. Will fix
that though.
-
PR: https://git.openjdk.org/jd
> On m1 mac sometimes (in very rare instance) it takes a long time to get the
> sequencer so increasing timeout helps with that.
> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
> reason is unclear but watching for that fixes the problem. Also on Linux
> sometimes p
On Tue, 15 Nov 2022 20:04:07 GMT, Alexander Zuev wrote:
>> On m1 mac sometimes (in very rare instance) it takes a long time to get the
>> sequencer so increasing timeout helps with that.
>> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
>> reason is unclear but wat
On Tue, 15 Nov 2022 20:04:07 GMT, Alexander Zuev wrote:
>> On m1 mac sometimes (in very rare instance) it takes a long time to get the
>> sequencer so increasing timeout helps with that.
>> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
>> reason is unclear but wat
> On m1 mac sometimes (in very rare instance) it takes a long time to get the
> sequencer so increasing timeout helps with that.
> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
> reason is unclear but watching for that fixes the problem. Also on Linux
> sometimes p
On Tue, 15 Nov 2022 17:57:13 GMT, Phil Race wrote:
> Can you check other jtreg sound tests for similar potential problems
Sure, i will check and if i find something i will create a separate bug listing
all the cases where it might be a problem.
-
PR: https://git.openjdk.org/jdk/pu
On Tue, 15 Nov 2022 17:55:40 GMT, Phil Race wrote:
> Is this our bug ?
Unlikely - it happens on one platform only, looks more like for some reason
system restarts the midi stream after first message. But i can not be sure
because i only saw it on the CI system once in more than a thousand runs
On Tue, 15 Nov 2022 17:55:17 GMT, Phil Race wrote:
> Hmm. 5 minutes is a really long time for something that should take less than
> 1 second (getting the sequencer). Is that because something else (perhaps
> another process) is holding a needed resource ?
I tried to find reasons or any kind o
On Tue, 15 Nov 2022 07:31:46 GMT, Alexander Zuev wrote:
> On m1 mac sometimes (in very rare instance) it takes a long time to get the
> sequencer so increasing timeout helps with that.
> Also on m1 mac sequencer got notified about the first MetaMessage twice. The
> reason is unclear but watchin
On m1 mac sometimes (in very rare instance) it takes a long time to get the
sequencer so increasing timeout helps with that.
Also on m1 mac sequencer got notified about the first MetaMessage twice. The
reason is unclear but watching for that fixes the problem. Also on Linux
sometimes process han
22 matches
Mail list logo
