blockStatus uninitialised

11 messages Options
Embed this post
Permalink
Martyn Shaw-2

blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
Just ran into this when zooming in to the end of a track that I had
just applied an effect to:
Crash: blockStatus uninitialized
Sequence::GetWaveDisplay line 1248
       bl[pixel - 1] = blockStatus;//mchinen


Traces back to
WaveClip::GetWaveDisplay 673
TrackArtist::DrawClipWaveform 1263
TrackArtist::DrawWaveform 1084
TrackArtist::DrawTrack  256
etc.

May be related to crashes with applying effects when zoomed in?

No time to look into it right now.

TTFN
Martyn

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Martyn Shaw-2

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
Actually, that's annoyingly repeatable here (XP VS 2008):
Generate chirp, 30s, 0->0.8 amplitude
Effect -> Change Speed... -> 10 -> OK
Mouse over the RH end then mouse-wheel forward
Just as it looks like it might get to display samples...
boink!

Martyn

Martyn Shaw wrote:

> Just ran into this when zooming in to the end of a track that I had just
> applied an effect to:
> Crash: blockStatus uninitialized
> Sequence::GetWaveDisplay line 1248
>       bl[pixel - 1] = blockStatus;//mchinen
>
>
> Traces back to
> WaveClip::GetWaveDisplay 673
> TrackArtist::DrawClipWaveform 1263
> TrackArtist::DrawWaveform 1084
> TrackArtist::DrawTrack  256
> etc.
>
> May be related to crashes with applying effects when zoomed in?
>
> No time to look into it right now.
>
> TTFN
> Martyn
>

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Vidyashankar Vellal

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
Ha! This is related to the problem of creation of extra clips.

On Wed, Oct 21, 2009 at 5:02 AM, Martyn Shaw <[hidden email]> wrote:
Actually, that's annoyingly repeatable here (XP VS 2008):
Generate chirp, 30s, 0->0.8 amplitude
Effect -> Change Speed... -> 10 -> OK
Mouse over the RH end then mouse-wheel forward
Just as it looks like it might get to display samples...
boink!

The reason this happens is a split is made just 1 sample before the track end (atleast in this example). So there is a clip (not visible to the eye) at the end of the track that is 1 sample long. So I am guessing that when there is only one sample, it can't display the waveform?

I have been looking at this problem of extra clips for some time now. Will start a separate thread soon.

Vidyashankar
 

Martyn

Martyn Shaw wrote:
> Just ran into this when zooming in to the end of a track that I had just
> applied an effect to:
> Crash: blockStatus uninitialized
> Sequence::GetWaveDisplay line 1248
>       bl[pixel - 1] = blockStatus;//mchinen
>
>
> Traces back to
> WaveClip::GetWaveDisplay 673
> TrackArtist::DrawClipWaveform 1263
> TrackArtist::DrawWaveform 1084
> TrackArtist::DrawTrack  256
> etc.
>
> May be related to crashes with applying effects when zoomed in?
>
> No time to look into it right now.
>
> TTFN
> Martyn
>

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel


------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Michael Chinen

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
In reply to this post by Martyn Shaw-2
On Wed, Oct 21, 2009 at 1:09 AM, Martyn Shaw
<[hidden email]> wrote:

> Just ran into this when zooming in to the end of a track that I had
> just applied an effect to:
> Crash: blockStatus uninitialized
> Sequence::GetWaveDisplay line 1248
>       bl[pixel - 1] = blockStatus;//mchinen
>
>
> Traces back to
> WaveClip::GetWaveDisplay 673
> TrackArtist::DrawClipWaveform 1263
> TrackArtist::DrawWaveform 1084
> TrackArtist::DrawTrack  256
> etc.
>
> May be related to crashes with applying effects when zoomed in?
>
> No time to look into it right now.
Ditto.  But as a placeholder I think blockStatus should be initialized
to 1.  Will look into it next week.

Michael

>
> TTFN
> Martyn
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> audacity-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/audacity-devel
>

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Martyn Shaw-2

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
In reply to this post by Vidyashankar Vellal
So is there a fix for this or a change that we should revert?

Martyn

Vidyashankar Vellal wrote:

> Ha! This is related to the problem of creation of extra clips.
>
> On Wed, Oct 21, 2009 at 5:02 AM, Martyn Shaw
> <[hidden email] <mailto:[hidden email]>> wrote:
>
>     Actually, that's annoyingly repeatable here (XP VS 2008):
>     Generate chirp, 30s, 0->0.8 amplitude
>     Effect -> Change Speed... -> 10 -> OK
>     Mouse over the RH end then mouse-wheel forward
>     Just as it looks like it might get to display samples...
>     boink!
>
>
> The reason this happens is a split is made just 1 sample before the
> track end (atleast in this example). So there is a clip (not visible to
> the eye) at the end of the track that is 1 sample long. So I am guessing
> that when there is only one sample, it can't display the waveform?
>
> I have been looking at this problem of extra clips for some time now.
> Will start a separate thread soon.
>
> Vidyashankar
>  
>
>
>     Martyn
>
>     Martyn Shaw wrote:
>      > Just ran into this when zooming in to the end of a track that I
>     had just
>      > applied an effect to:
>      > Crash: blockStatus uninitialized
>      > Sequence::GetWaveDisplay line 1248
>      >       bl[pixel - 1] = blockStatus;//mchinen
>      >
>      >
>      > Traces back to
>      > WaveClip::GetWaveDisplay 673
>      > TrackArtist::DrawClipWaveform 1263
>      > TrackArtist::DrawWaveform 1084
>      > TrackArtist::DrawTrack  256
>      > etc.
>      >
>      > May be related to crashes with applying effects when zoomed in?
>      >
>      > No time to look into it right now.
>      >
>      > TTFN
>      > Martyn
>      >
>
>     ------------------------------------------------------------------------------
>     Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>     is the only developer event you need to attend this year. Jumpstart your
>     developing skills, take BlackBerry mobile applications to market and
>     stay
>     ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>     http://p.sf.net/sfu/devconference
>     _______________________________________________
>     audacity-devel mailing list
>     [hidden email]
>     <mailto:[hidden email]>
>     https://lists.sourceforge.net/lists/listinfo/audacity-devel
>
>
>
> ------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> audacity-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/audacity-devel

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Gale (Audacity Team)

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink

| From Martyn Shaw <[hidden email]>
| Sat, 31 Oct 2009 01:11:44 +0000
| Subject: [Audacity-devel] blockStatus uninitialised

> So is there a fix for this or a change that we should revert?
>
> Martyn
>
> Vidyashankar Vellal wrote:
> > Ha! This is related to the problem of creation of extra clips.
> >
> > On Wed, Oct 21, 2009 at 5:02 AM, Martyn Shaw
> > <[hidden email] <mailto:[hidden email]>> wrote:
> >
> >     Actually, that's annoyingly repeatable here (XP VS 2008):
> >     Generate chirp, 30s, 0->0.8 amplitude
> >     Effect -> Change Speed... -> 10 -> OK
> >     Mouse over the RH end then mouse-wheel forward
> >     Just as it looks like it might get to display samples...
> >     boink!
> >
> >
> > The reason this happens is a split is made just 1 sample before the
> > track end (atleast in this example). So there is a clip (not visible to
> > the eye) at the end of the track that is 1 sample long. So I am guessing
> > that when there is only one sample, it can't display the waveform?
> >
> > I have been looking at this problem of extra clips for some time now.
> > Will start a separate thread soon.

Is it zooming causing this again (CTRL + Wheel rotate according to the
Mouse Preferences)? Wheel Rotate is scroll up or down.  

FWIW I only have a wheel-less mouse. I get the superfluous clip at the
end of the track after applying Change Speed, but I don't get any crash
clicking near the track end and zooming (in Win XP Unicode Release).  



Gale



> >     Martyn Shaw wrote:
> >      > Just ran into this when zooming in to the end of a track that I
> >     had just
> >      > applied an effect to:
> >      > Crash: blockStatus uninitialized
> >      > Sequence::GetWaveDisplay line 1248
> >      >       bl[pixel - 1] = blockStatus;//mchinen
> >      >
> >      >
> >      > Traces back to
> >      > WaveClip::GetWaveDisplay 673
> >      > TrackArtist::DrawClipWaveform 1263
> >      > TrackArtist::DrawWaveform 1084
> >      > TrackArtist::DrawTrack  256
> >      > etc.
> >      >
> >      > May be related to crashes with applying effects when zoomed in?
> >      >
> >      > No time to look into it right now.
> >      >
> >      > TTFN
> >      > Martyn
> >      >
> >
> >     ------------------------------------------------------------------------------
> >     Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> >     is the only developer event you need to attend this year. Jumpstart your
> >     developing skills, take BlackBerry mobile applications to market and
> >     stay
> >     ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> >     http://p.sf.net/sfu/devconference
> >     _______________________________________________
> >     audacity-devel mailing list
> >     [hidden email]
> >     <mailto:[hidden email]>
> >     https://lists.sourceforge.net/lists/listinfo/audacity-devel
> >
> >
> >
> > ------------------------------------------------------------------------
> >
> > ------------------------------------------------------------------------------
> > Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> > is the only developer event you need to attend this year. Jumpstart your
> > developing skills, take BlackBerry mobile applications to market and stay
> > ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> > http://p.sf.net/sfu/devconference
> >
> >
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > audacity-devel mailing list
> > [hidden email]
> > https://lists.sourceforge.net/lists/listinfo/audacity-devel
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> audacity-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/audacity-devel



------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Michael Chinen

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
In reply to this post by Martyn Shaw-2
On Sat, Oct 31, 2009 at 2:11 AM, Martyn Shaw
<[hidden email]> wrote:
> So is there a fix for this or a change that we should revert?
It doesn't crash for me on mac, but treatment of uninited variables is
compiler specific (if I recall).
My fix (which just initializes the variable to 1) probably will work
if you could test it.
Vidyashankar, if you've got more insight or some larger fix please let us know.

Michael

>
> Martyn
>
> Vidyashankar Vellal wrote:
>> Ha! This is related to the problem of creation of extra clips.
>>
>> On Wed, Oct 21, 2009 at 5:02 AM, Martyn Shaw
>> <[hidden email] <mailto:[hidden email]>> wrote:
>>
>>     Actually, that's annoyingly repeatable here (XP VS 2008):
>>     Generate chirp, 30s, 0->0.8 amplitude
>>     Effect -> Change Speed... -> 10 -> OK
>>     Mouse over the RH end then mouse-wheel forward
>>     Just as it looks like it might get to display samples...
>>     boink!
>>
>>
>> The reason this happens is a split is made just 1 sample before the
>> track end (atleast in this example). So there is a clip (not visible to
>> the eye) at the end of the track that is 1 sample long. So I am guessing
>> that when there is only one sample, it can't display the waveform?
>>
>> I have been looking at this problem of extra clips for some time now.
>> Will start a separate thread soon.
>>
>> Vidyashankar
>>
>>
>>
>>     Martyn
>>
>>     Martyn Shaw wrote:
>>      > Just ran into this when zooming in to the end of a track that I
>>     had just
>>      > applied an effect to:
>>      > Crash: blockStatus uninitialized
>>      > Sequence::GetWaveDisplay line 1248
>>      >       bl[pixel - 1] = blockStatus;//mchinen
>>      >
>>      >
>>      > Traces back to
>>      > WaveClip::GetWaveDisplay 673
>>      > TrackArtist::DrawClipWaveform 1263
>>      > TrackArtist::DrawWaveform 1084
>>      > TrackArtist::DrawTrack  256
>>      > etc.
>>      >
>>      > May be related to crashes with applying effects when zoomed in?
>>      >
>>      > No time to look into it right now.
>>      >
>>      > TTFN
>>      > Martyn
>>      >
>>
>>     ------------------------------------------------------------------------------
>>     Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>>     is the only developer event you need to attend this year. Jumpstart your
>>     developing skills, take BlackBerry mobile applications to market and
>>     stay
>>     ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>>     http://p.sf.net/sfu/devconference
>>     _______________________________________________
>>     audacity-devel mailing list
>>     [hidden email]
>>     <mailto:[hidden email]>
>>     https://lists.sourceforge.net/lists/listinfo/audacity-devel
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> ------------------------------------------------------------------------------
>> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>> is the only developer event you need to attend this year. Jumpstart your
>> developing skills, take BlackBerry mobile applications to market and stay
>> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>> http://p.sf.net/sfu/devconference
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> audacity-devel mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/audacity-devel
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> audacity-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/audacity-devel
>

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Vidyashankar Vellal

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
Hi

On Sat, Oct 31, 2009 at 7:49 PM, Michael Chinen <[hidden email]> wrote:
On Sat, Oct 31, 2009 at 2:11 AM, Martyn Shaw
<[hidden email]> wrote:
> So is there a fix for this or a change that we should revert?
It doesn't crash for me on mac, but treatment of uninited variables is
compiler specific (if I recall).
My fix (which just initializes the variable to 1) probably will work
if you could test it.
Vidyashankar, if you've got more insight or some larger fix please let us know.

Michael

>
>>
>>     Martyn Shaw wrote:
>>      > Just ran into this when zooming in to the end of a track that I
>>     had just
>>      > applied an effect to:
>>      > Crash: blockStatus uninitialized
>>      > Sequence::GetWaveDisplay line 1248
>>      >       bl[pixel - 1] = blockStatus;//mchinen
>>      >
>>      >
>>      > Traces back to
>>      > WaveClip::GetWaveDisplay 673
>>      > TrackArtist::DrawClipWaveform 1263
>>      > TrackArtist::DrawWaveform 1084
>>      > TrackArtist::DrawTrack  256
>>      > etc.


I just built a fresh unicode debug from HEAD. The bug is repeatable for me in winXP. But surprisingly, my stack trace looks different.
Crash: wxASSERT( t0 >= 0, msg) (The t0 value is approx. -0.029)
Envelope.cpp line 1105

Traces back to
Envelope::GetValue 1031
Envelope::GetValueAtX 1040
TrackArtist::DrawIndividualSamples 953
TrackArtist::DrawClipWaveform 1295
TrackArtist:DrawWaveform 1084

In Unicode Release mode, obviously there is no wxASSERT. So no crash is seen.

I still feel the problem is with the spurious clip at the end. When the spurious clip is present, the crash happens when I zoom in for the 10th time. (Generate Chirp 30 sec, Change speed 10% and start zooming at the end)

I recently started a separate thread on the spurious clips. Without the spurious clip, I can zoom in to the maximum extent without any problems.

However, looking at the different stack trace in Martyn's case, it may be a different bug in itself. Martyn, if you have time, can you apply the patch I submitted in the thread on spurious clips and see if the 'blockStatus uninitialised' bug is still there?

Thanks

Vidyashankar

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Vidyashankar Vellal

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
On Sun, Nov 1, 2009 at 2:37 PM, Vidyashankar Vellal
<[hidden email]> wrote:

> Hi
>
> On Sat, Oct 31, 2009 at 7:49 PM, Michael Chinen <[hidden email]> wrote:
>>
>> On Sat, Oct 31, 2009 at 2:11 AM, Martyn Shaw
>> <[hidden email]> wrote:
>> > So is there a fix for this or a change that we should revert?
>> It doesn't crash for me on mac, but treatment of uninited variables is
>> compiler specific (if I recall).
>> My fix (which just initializes the variable to 1) probably will work
>> if you could test it.
>> Vidyashankar, if you've got more insight or some larger fix please let us
>> know.
>>
>> Michael
>>
>> >
>> >>
>> >>     Martyn Shaw wrote:
>> >>      > Just ran into this when zooming in to the end of a track that I
>> >>     had just
>> >>      > applied an effect to:
>> >>      > Crash: blockStatus uninitialized
>> >>      > Sequence::GetWaveDisplay line 1248
>> >>      >       bl[pixel - 1] = blockStatus;//mchinen
>> >>      >
>> >>      >
>> >>      > Traces back to
>> >>      > WaveClip::GetWaveDisplay 673
>> >>      > TrackArtist::DrawClipWaveform 1263
>> >>      > TrackArtist::DrawWaveform 1084
>> >>      > TrackArtist::DrawTrack  256
>> >>      > etc.
>>
>
> I just built a fresh unicode debug from HEAD. The bug is repeatable for me
> in winXP. But surprisingly, my stack trace looks different.
> Crash: wxASSERT( t0 >= 0, msg) (The t0 value is approx. -0.029)
> Envelope.cpp line 1105
> Traces back to
> Envelope::GetValue 1031
> Envelope::GetValueAtX 1040
> TrackArtist::DrawIndividualSamples 953
> TrackArtist::DrawClipWaveform 1295
> TrackArtist:DrawWaveform 1084
> In Unicode Release mode, obviously there is no wxASSERT. So no crash is
> seen.
> I still feel the problem is with the spurious clip at the end.

[snip]

Sorry I spoke a bit too early. This wxAssert problem seems to be
present when continuously zooming in on any split line.
(For example, generate chirp 30 sec, split at 5 sec, keep zooming in
on the split line). So it is not related to the presence of spurious
clips. This wxAssert bug is different from the subject of this thread.

I am still not able to replicate the uninitialised block bug that
Martyn got. As he mentioned, it is probably related to the P2 'crashes
with applying repeated effects when zoomed in'. I will look into it
more and see if I find anything.

Vidyashankar

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Martyn Shaw-2

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
In reply to this post by Vidyashankar Vellal


Vidyashankar Vellal wrote:

> Hi
>
> On Sat, Oct 31, 2009 at 7:49 PM, Michael Chinen <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     On Sat, Oct 31, 2009 at 2:11 AM, Martyn Shaw
>     <[hidden email] <mailto:[hidden email]>>
>     wrote:
>      > So is there a fix for this or a change that we should revert?
>     It doesn't crash for me on mac, but treatment of uninited variables is
>     compiler specific (if I recall).
>     My fix (which just initializes the variable to 1) probably will work
>     if you could test it.
>     Vidyashankar, if you've got more insight or some larger fix please
>     let us know.
>
>     Michael
>
>      >
>      >>
>      >>     Martyn Shaw wrote:
>      >>      > Just ran into this when zooming in to the end of a track
>     that I
>      >>     had just
>      >>      > applied an effect to:
>      >>      > Crash: blockStatus uninitialized
>      >>      > Sequence::GetWaveDisplay line 1248
>      >>      >       bl[pixel - 1] = blockStatus;//mchinen
>      >>      >
>      >>      >
>      >>      > Traces back to
>      >>      > WaveClip::GetWaveDisplay 673
>      >>      > TrackArtist::DrawClipWaveform 1263
>      >>      > TrackArtist::DrawWaveform 1084
>      >>      > TrackArtist::DrawTrack  256
>      >>      > etc.
>
>
> I just built a fresh unicode debug from HEAD. The bug is repeatable for
> me in winXP. But surprisingly, my stack trace looks different.
> Crash: wxASSERT( t0 >= 0, msg) (The t0 value is approx. -0.029)
> Envelope.cpp line 1105

I have that assert commented out - it is not really a problem, as you
can see a bit further on in the code.

Martyn

> Traces back to
> Envelope::GetValue 1031
> Envelope::GetValueAtX 1040
> TrackArtist::DrawIndividualSamples 953
> TrackArtist::DrawClipWaveform 1295
> TrackArtist:DrawWaveform 1084
>
> In Unicode Release mode, obviously there is no wxASSERT. So no crash is
> seen.
>
> I still feel the problem is with the spurious clip at the end. When the
> spurious clip is present, the crash happens when I zoom in for the 10th
> time. (Generate Chirp 30 sec, Change speed 10% and start zooming at the end)
>
> I recently started a separate thread on the spurious clips. Without the
> spurious clip, I can zoom in to the maximum extent without any problems.
>
> However, looking at the different stack trace in Martyn's case, it may
> be a different bug in itself. Martyn, if you have time, can you apply
> the patch I submitted in the thread on spurious clips and see if the
> 'blockStatus uninitialised' bug is still there?
>
> Thanks
>
> Vidyashankar
>
>
> ------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> audacity-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/audacity-devel

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Martyn Shaw-2

Re: blockStatus uninitialised

Reply Threaded More More options
Print post
Permalink
In reply to this post by Michael Chinen
Initialising blockStatus to 1 works here.

Martyn

Michael Chinen wrote:

> On Sat, Oct 31, 2009 at 2:11 AM, Martyn Shaw
> <[hidden email]> wrote:
>> So is there a fix for this or a change that we should revert?
> It doesn't crash for me on mac, but treatment of uninited variables is
> compiler specific (if I recall).
> My fix (which just initializes the variable to 1) probably will work
> if you could test it.
> Vidyashankar, if you've got more insight or some larger fix please let us know.
>
> Michael
>
>> Martyn
>>
>> Vidyashankar Vellal wrote:
>>> Ha! This is related to the problem of creation of extra clips.
>>>
>>> On Wed, Oct 21, 2009 at 5:02 AM, Martyn Shaw
>>> <[hidden email] <mailto:[hidden email]>> wrote:
>>>
>>>     Actually, that's annoyingly repeatable here (XP VS 2008):
>>>     Generate chirp, 30s, 0->0.8 amplitude
>>>     Effect -> Change Speed... -> 10 -> OK
>>>     Mouse over the RH end then mouse-wheel forward
>>>     Just as it looks like it might get to display samples...
>>>     boink!
>>>
>>>
>>> The reason this happens is a split is made just 1 sample before the
>>> track end (atleast in this example). So there is a clip (not visible to
>>> the eye) at the end of the track that is 1 sample long. So I am guessing
>>> that when there is only one sample, it can't display the waveform?
>>>
>>> I have been looking at this problem of extra clips for some time now.
>>> Will start a separate thread soon.
>>>
>>> Vidyashankar
>>>
>>>
>>>
>>>     Martyn
>>>
>>>     Martyn Shaw wrote:
>>>      > Just ran into this when zooming in to the end of a track that I
>>>     had just
>>>      > applied an effect to:
>>>      > Crash: blockStatus uninitialized
>>>      > Sequence::GetWaveDisplay line 1248
>>>      >       bl[pixel - 1] = blockStatus;//mchinen
>>>      >
>>>      >
>>>      > Traces back to
>>>      > WaveClip::GetWaveDisplay 673
>>>      > TrackArtist::DrawClipWaveform 1263
>>>      > TrackArtist::DrawWaveform 1084
>>>      > TrackArtist::DrawTrack  256
>>>      > etc.
>>>      >
>>>      > May be related to crashes with applying effects when zoomed in?
>>>      >
>>>      > No time to look into it right now.
>>>      >
>>>      > TTFN
>>>      > Martyn
>>>      >
>>>
>>>     ------------------------------------------------------------------------------
>>>     Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>>>     is the only developer event you need to attend this year. Jumpstart your
>>>     developing skills, take BlackBerry mobile applications to market and
>>>     stay
>>>     ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>>>     http://p.sf.net/sfu/devconference
>>>     _______________________________________________
>>>     audacity-devel mailing list
>>>     [hidden email]
>>>     <mailto:[hidden email]>
>>>     https://lists.sourceforge.net/lists/listinfo/audacity-devel
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> ------------------------------------------------------------------------------
>>> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>>> is the only developer event you need to attend this year. Jumpstart your
>>> developing skills, take BlackBerry mobile applications to market and stay
>>> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>>> http://p.sf.net/sfu/devconference
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> audacity-devel mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/audacity-devel
>> ------------------------------------------------------------------------------
>> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>> is the only developer event you need to attend this year. Jumpstart your
>> developing skills, take BlackBerry mobile applications to market and stay
>> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>> http://p.sf.net/sfu/devconference
>> _______________________________________________
>> audacity-devel mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/audacity-devel
>>
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> audacity-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/audacity-devel

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
audacity-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/audacity-devel