Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-04-16 Thread Kevin Rushforth 

OK, in that case once the OCA is in place we can evaluate the proposed fix.

Thanks.

-- Kevin


On 4/16/2018 1:10 PM, Michael Ennen wrote:

I note in the issue Kevin said:

" I note that if this really is Monocle-specific, then we will likely need
someone from the community to provide a fix. "

Just want to make sure it is clear, John has already provided the fix:

https://gitlab.com/openjfxepd/jfxpatch/commit/f7c341775e5258
e790a049f3fdce4a956ef665c7



On Mon, Apr 16, 2018 at 11:41 AM, Nir Lisker  wrote:


It was moved to https://bugs.openjdk.java.net/browse/JDK-8201567.

On Mon, Apr 16, 2018, 20:26 John Neffenger  wrote:


On 02/23/2018 03:12 PM, Michael Ennen wrote:

Good point about the OCA, I have sent an email just now to
John Neffenger. Hopefully he is still interested in this topic :).

I filed the bug report yesterday, along with two others. It was assigned
the internal review ID 9053385 and has the synopsis, "QuantumRenderer
modifies buffer in use by JavaFX Application Thread."

The problem happens on the Monocle VNC and Headless platforms, along
with any other NativeScreen implementation that happens to use a
non-direct buffer to back the Framebuffer.

I created two short videos showing the problem before and after a fix is
applied:

Before: Waving and Jumping Duke (one minute)
https://youtu.be/mfSzSMDIEIM

After: Waving Duke (10 seconds)
https://youtu.be/pUXPL_Gm8Qw

Thank you,
John

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-04-16 Thread Michael Ennen 
I note in the issue Kevin said:

" I note that if this really is Monocle-specific, then we will likely need
someone from the community to provide a fix. "

Just want to make sure it is clear, John has already provided the fix:

https://gitlab.com/openjfxepd/jfxpatch/commit/f7c341775e5258
e790a049f3fdce4a956ef665c7



On Mon, Apr 16, 2018 at 11:41 AM, Nir Lisker  wrote:

> It was moved to https://bugs.openjdk.java.net/browse/JDK-8201567.
>
> On Mon, Apr 16, 2018, 20:26 John Neffenger  wrote:
>
> > On 02/23/2018 03:12 PM, Michael Ennen wrote:
> > > Good point about the OCA, I have sent an email just now to
> > > John Neffenger. Hopefully he is still interested in this topic :).
> >
> > I filed the bug report yesterday, along with two others. It was assigned
> > the internal review ID 9053385 and has the synopsis, "QuantumRenderer
> > modifies buffer in use by JavaFX Application Thread."
> >
> > The problem happens on the Monocle VNC and Headless platforms, along
> > with any other NativeScreen implementation that happens to use a
> > non-direct buffer to back the Framebuffer.
> >
> > I created two short videos showing the problem before and after a fix is
> > applied:
> >
> > Before: Waving and Jumping Duke (one minute)
> > https://youtu.be/mfSzSMDIEIM
> >
> > After: Waving Duke (10 seconds)
> > https://youtu.be/pUXPL_Gm8Qw
> >
> > Thank you,
> > John
> >
>



-- 
Michael Ennen

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-04-16 Thread Nir Lisker 
It was moved to https://bugs.openjdk.java.net/browse/JDK-8201567.

On Mon, Apr 16, 2018, 20:26 John Neffenger  wrote:

> On 02/23/2018 03:12 PM, Michael Ennen wrote:
> > Good point about the OCA, I have sent an email just now to
> > John Neffenger. Hopefully he is still interested in this topic :).
>
> I filed the bug report yesterday, along with two others. It was assigned
> the internal review ID 9053385 and has the synopsis, "QuantumRenderer
> modifies buffer in use by JavaFX Application Thread."
>
> The problem happens on the Monocle VNC and Headless platforms, along
> with any other NativeScreen implementation that happens to use a
> non-direct buffer to back the Framebuffer.
>
> I created two short videos showing the problem before and after a fix is
> applied:
>
> Before: Waving and Jumping Duke (one minute)
> https://youtu.be/mfSzSMDIEIM
>
> After: Waving Duke (10 seconds)
> https://youtu.be/pUXPL_Gm8Qw
>
> Thank you,
> John
>

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-04-16 Thread John Neffenger 

On 02/23/2018 03:12 PM, Michael Ennen wrote:

Good point about the OCA, I have sent an email just now to
John Neffenger. Hopefully he is still interested in this topic :).


I filed the bug report yesterday, along with two others. It was assigned 
the internal review ID 9053385 and has the synopsis, "QuantumRenderer 
modifies buffer in use by JavaFX Application Thread."


The problem happens on the Monocle VNC and Headless platforms, along 
with any other NativeScreen implementation that happens to use a 
non-direct buffer to back the Framebuffer.


I created two short videos showing the problem before and after a fix is 
applied:


Before: Waving and Jumping Duke (one minute)
https://youtu.be/mfSzSMDIEIM

After: Waving Duke (10 seconds)
https://youtu.be/pUXPL_Gm8Qw

Thank you,
John

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-02-23 Thread Michael Ennen 
>From my limited testing, the issue is still present in 8u172 b03 (early
access)
and 9.0.1+11 (that is, assuming it is the same bug which I am crossing my
fingers
about!).

Thanks for taking this back up, John.

On Fri, Feb 23, 2018 at 4:45 PM, Kevin Rushforth  wrote:

>
> In general, what should we do when we find bugs in version 8 but the
>> latest development is happening on version 11? Open the bug, or test it
>> against the latest version first?
>>
>
> The latter is preferable if it is possible to test it on the latest
> version.
>
> Thanks.
>
> -- Kevin
>
>
>
> John Neffenger wrote:
>
>> On 02/23/2018 02:29 PM, Michael Ennen wrote:
>>
>>> Here is the issue:
>>>
>>> https://gitlab.com/openjfxepd/jfxpatch/issues/1
>>>
>>> Thanks to John Neffenger for taking the time to put this together! I
>>> wanted to make this work was known to Kevin and the team.
>>>
>>
>> I'm so glad you found it! I posted a link to the issue, along with two
>> others, on this mailing list a year ago, but I think people were too busy
>> with all the Java 9 changes happening at the time (rightfully so).
>>
>> I was reluctant to open the official bugs on the Oracle page before
>> testing whether they still occur in the latest version. Actually, I haven't
>> even had the time to check whether they still occur in the latest OpenJFX 8.
>>
>> In general, what should we do when we find bugs in version 8 but the
>> latest development is happening on version 11? Open the bug, or test it
>> against the latest version first?
>>
>> Thanks,
>> John
>>
>


-- 
Michael Ennen

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-02-23 Thread Kevin Rushforth 


In general, what should we do when we find bugs in version 8 but the 
latest development is happening on version 11? Open the bug, or test 
it against the latest version first?


The latter is preferable if it is possible to test it on the latest version.

Thanks.

-- Kevin


John Neffenger wrote:

On 02/23/2018 02:29 PM, Michael Ennen wrote:

Here is the issue:

https://gitlab.com/openjfxepd/jfxpatch/issues/1

Thanks to John Neffenger for taking the time to put this together! I
wanted to make this work was known to Kevin and the team.


I'm so glad you found it! I posted a link to the issue, along with two 
others, on this mailing list a year ago, but I think people were too 
busy with all the Java 9 changes happening at the time (rightfully so).


I was reluctant to open the official bugs on the Oracle page before 
testing whether they still occur in the latest version. Actually, I 
haven't even had the time to check whether they still occur in the 
latest OpenJFX 8.


In general, what should we do when we find bugs in version 8 but the 
latest development is happening on version 11? Open the bug, or test 
it against the latest version first?


Thanks,
John

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-02-23 Thread John Neffenger 

On 02/23/2018 02:29 PM, Michael Ennen wrote:

Here is the issue:

https://gitlab.com/openjfxepd/jfxpatch/issues/1

Thanks to John Neffenger for taking the time to put this together! I
wanted to make this work was known to Kevin and the team.


I'm so glad you found it! I posted a link to the issue, along with two 
others, on this mailing list a year ago, but I think people were too 
busy with all the Java 9 changes happening at the time (rightfully so).


I was reluctant to open the official bugs on the Oracle page before 
testing whether they still occur in the latest version. Actually, I 
haven't even had the time to check whether they still occur in the 
latest OpenJFX 8.


In general, what should we do when we find bugs in version 8 but the 
latest development is happening on version 11? Open the bug, or test it 
against the latest version first?


Thanks,
John

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-02-23 Thread Michael Ennen 
Good point about the OCA, I have sent an email just now to
John Neffenger. Hopefully he is still interested in this topic :).

I will file the bug report after hopefully getting into contact with
John.

Thanks Kevin,

And no rush on the Robot review, just didn't want you to totally
forget - I know how it is,

Michael

On Fri, Feb 23, 2018 at 3:36 PM, Kevin Rushforth  wrote:

> I haven't seen this bug, and I did a quick search and don't find it. Go
> ahead and file a bug, but unless it is a trivial fix, you will need to
> contact the author of the fix and ask him to sign the OCA before we could
> take it.
>
> > (P.S. When is ramp down phase 2 over? :) ).
>
> Oh, right. We're actually at GAC for JDK 10 now. I owe you a review of the
> Robot API...
>
> -- Kevin
>
>
>
>
> Michael Ennen wrote:
>
>> Hello,
>>
>> When doing some testing with TestFX running in headless mode
>> via Monocle I came across a BufferOverflowException.
>>
>> After some google searching it seems that someone has come
>> across this before me and created a GitLab repository with the
>> full issue details and a suggested fix! I was wondering if this
>> has ever been brought to the attention of the JavaFX devs and if
>> there was a corresponding bug in JBS.
>>
>> Here's the commit:
>>
>> https://gitlab.com/openjfxepd/jfxpatch/commit/f7c341775e5258
>> e790a049f3fdce4a
>> 956ef665c7
>>
>> Here is the issue:
>>
>> https://gitlab.com/openjfxepd/jfxpatch/issues/1
>>
>> Thanks to John Neffenger for taking the time to put this together! I
>> wanted to make this work was known to Kevin and the team.
>>
>> Thanks very much,
>>
>> (P.S. When is ramp down phase 2 over? :) ).
>>
>>
>>
>

Re: Monocle Headless BufferOverflowException (Issue + Fix)

2018-02-23 Thread Kevin Rushforth 
I haven't seen this bug, and I did a quick search and don't find it. Go 
ahead and file a bug, but unless it is a trivial fix, you will need to 
contact the author of the fix and ask him to sign the OCA before we 
could take it.


> (P.S. When is ramp down phase 2 over? :) ).

Oh, right. We're actually at GAC for JDK 10 now. I owe you a review of 
the Robot API...


-- Kevin



Michael Ennen wrote:

Hello,

When doing some testing with TestFX running in headless mode
via Monocle I came across a BufferOverflowException.

After some google searching it seems that someone has come
across this before me and created a GitLab repository with the
full issue details and a suggested fix! I was wondering if this
has ever been brought to the attention of the JavaFX devs and if
there was a corresponding bug in JBS.

Here's the commit:

https://gitlab.com/openjfxepd/jfxpatch/commit/f7c341775e5258e790a049f3fdce4a
956ef665c7

Here is the issue:

https://gitlab.com/openjfxepd/jfxpatch/issues/1

Thanks to John Neffenger for taking the time to put this together! I
wanted to make this work was known to Kevin and the team.

Thanks very much,

(P.S. When is ramp down phase 2 over? :) ).

Monocle Headless BufferOverflowException (Issue + Fix)

2018-02-23 Thread Michael Ennen 
Hello,

When doing some testing with TestFX running in headless mode
via Monocle I came across a BufferOverflowException.

After some google searching it seems that someone has come
across this before me and created a GitLab repository with the
full issue details and a suggested fix! I was wondering if this
has ever been brought to the attention of the JavaFX devs and if
there was a corresponding bug in JBS.

Here's the commit:

https://gitlab.com/openjfxepd/jfxpatch/commit/f7c341775e5258e790a049f3fdce4a
956ef665c7

Here is the issue:

https://gitlab.com/openjfxepd/jfxpatch/issues/1

Thanks to John Neffenger for taking the time to put this together! I
wanted to make this work was known to Kevin and the team.

Thanks very much,

(P.S. When is ramp down phase 2 over? :) ).

-- 
Michael Ennen