Hi All,

I just tried some Blue projects of mine and got some crashes.  It's
happening in useropcdset.  I've attached a CSD that demonstrates the
crash.

I'll look at this tonight. I think we may need to issue a 6.03.1 soon.

steven

it was not crashing here with a build from april 30. With today’s develop code I get

new alloc for instr 1:
PERF ERROR in instr 1 (opcode yi_env_orch): Division by zero
   note aborted
Csound tidy up: Segmentation fault

========================
Dr Victor Lazzarini
Senior Lecturer
NUI Maynooth, Ireland
victor dot lazzarini at nuim dot ie




On 5 May 2014, at 18:09, Steven Yi  wrote:

> 


------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

This line seems to be culprit:

529:   iEndAmp = iNewAmp / iOldAmp


========================
Dr Victor Lazzarini
Senior Lecturer
NUI Maynooth, Ireland
victor dot lazzarini at nuim dot ie




On 5 May 2014, at 20:04, Victor Lazzarini  wrote:

> it was not crashing here with a build from april 30. With today’s develop code I get
> 
> new alloc for instr 1:
> PERF ERROR in instr 1 (opcode yi_env_orch): Division by zero
>   note aborted
> Csound tidy up: Segmentation fault
> 
> ========================
> Dr Victor Lazzarini
> Senior Lecturer
> NUI Maynooth, Ireland
> victor dot lazzarini at nuim dot ie
> 
> 
> 
> 
> On 5 May 2014, at 18:09, Steven Yi  wrote:
> 
>> 
> 
> 
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

But I think what is not right is the fact that the perferror is causing a crash. I looked and it is in the usropcdset() code, and
I am wondering if this is to do with the new udo code at all.
========================
Dr Victor Lazzarini
Senior Lecturer
NUI Maynooth, Ireland
victor dot lazzarini at nuim dot ie




On 5 May 2014, at 20:11, jpff@cs.bath.ac.uk wrote:

> This must be my fault -- I was adding checks for division by zero in  
> an attempt to chase the Josep/clarinet problem
> 
> ==Jojm ff
> 
> Quoting Victor Lazzarini :
> 
>> it was not crashing here with a build from april 30. With today?s  
>> develop code I get
>> 
>> new alloc for instr 1:
>> PERF ERROR in instr 1 (opcode yi_env_orch): Division by zero
>>   note aborted
>> Csound tidy up: Segmentation fault
>> 
>> ========================
>> Dr Victor Lazzarini
>> Senior Lecturer
>> NUI Maynooth, Ireland
>> victor dot lazzarini at nuim dot ie
>> 
>> 
>> 
>> 
>> On 5 May 2014, at 18:09, Steven Yi  wrote:
>> 
>>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>> • 3 signs your SCM is hindering your productivity
>> • Requirements for releasing software faster
>> • Expert tips and advice for migrating your SCM now
>> http://p.sf.net/sfu/perforce
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

The trouble is that PerfError() is being called at init time and that causes the crash. Is it possible to adjust the
code to call InitError() on div.ii and PerfError elsewhere?

PerfError() cannot be called at i-time otherwise crashes might ensue (as it gets rid of the instance).

========================
Dr Victor Lazzarini
Senior Lecturer
NUI Maynooth, Ireland
victor dot lazzarini at nuim dot ie




On 5 May 2014, at 20:36, Victor Lazzarini  wrote:

> But I think what is not right is the fact that the perferror is causing a crash. I looked and it is in the usropcdset() code, and
> I am wondering if this is to do with the new udo code at all.
> ========================
> Dr Victor Lazzarini
> Senior Lecturer
> NUI Maynooth, Ireland
> victor dot lazzarini at nuim dot ie
> 
> 
> 
> 
> On 5 May 2014, at 20:11, jpff@cs.bath.ac.uk wrote:
> 
>> This must be my fault -- I was adding checks for division by zero in  
>> an attempt to chase the Josep/clarinet problem
>> 
>> ==Jojm ff
>> 
>> Quoting Victor Lazzarini :
>> 
>>> it was not crashing here with a build from april 30. With today?s  
>>> develop code I get
>>> 
>>> new alloc for instr 1:
>>> PERF ERROR in instr 1 (opcode yi_env_orch): Division by zero
>>>  note aborted
>>> Csound tidy up: Segmentation fault
>>> 
>>> ========================
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> NUI Maynooth, Ireland
>>> victor dot lazzarini at nuim dot ie
>>> 
>>> 
>>> 
>>> 
>>> On 5 May 2014, at 18:09, Steven Yi  wrote:
>>> 
>>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>>> • 3 signs your SCM is hindering your productivity
>>> • Requirements for releasing software faster
>>> • Expert tips and advice for migrating your SCM now
>>> http://p.sf.net/sfu/perforce
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>> • 3 signs your SCM is hindering your productivity
>> • Requirements for releasing software faster
>> • Expert tips and advice for migrating your SCM now
>> http://p.sf.net/sfu/perforce
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> 
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

This is all a bit odd as the code did not previous crash or have any
divisions by zero.  This is for both the crashing example but also for
a blue demo project I had done a year ago.  So I suspect that maybe it
might be something where UDO's are called from UDO's.

I just got in and it's already a bit late, so I think I'll tackle this
in the morning.

On Mon, May 5, 2014 at 10:31 PM,   wrote:
> Eak...!!!!!  div.ii does not exist -- uses div.kk code
> Can fix but not tonight
>
> ==John
> Quoting Victor Lazzarini :
>
>> The trouble is that PerfError() is being called at init time and
>> that causes the crash. Is it possible to adjust the
>> code to call InitError() on div.ii and PerfError elsewhere?
>>
>> PerfError() cannot be called at i-time otherwise crashes might ensue
>> (as it gets rid of the instance).
>>
>> ========================
>> Dr Victor Lazzarini
>> Senior Lecturer
>> NUI Maynooth, Ireland
>> victor dot lazzarini at nuim dot ie
>>
>>
>>
>>
>> On 5 May 2014, at 20:36, Victor Lazzarini  wrote:
>>
>>> But I think what is not right is the fact that the perferror is
>>> causing a crash. I looked and it is in the usropcdset() code, and
>>> I am wondering if this is to do with the new udo code at all.
>>> ========================
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> NUI Maynooth, Ireland
>>> victor dot lazzarini at nuim dot ie
>>>
>>>
>>>
>>>
>>> On 5 May 2014, at 20:11, jpff@cs.bath.ac.uk wrote:
>>>
>>>> This must be my fault -- I was adding checks for division by zero in
>>>> an attempt to chase the Josep/clarinet problem
>>>>
>>>> ==Jojm ff
>>>>
>>>> Quoting Victor Lazzarini :
>>>>
>>>>> it was not crashing here with a build from april 30. With today?s
>>>>> develop code I get
>>>>>
>>>>> new alloc for instr 1:
>>>>> PERF ERROR in instr 1 (opcode yi_env_orch): Division by zero
>>>>>  note aborted
>>>>> Csound tidy up: Segmentation fault
>>>>>
>>>>> ========================
>>>>> Dr Victor Lazzarini
>>>>> Senior Lecturer
>>>>> NUI Maynooth, Ireland
>>>>> victor dot lazzarini at nuim dot ie
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 5 May 2014, at 18:09, Steven Yi  wrote:
>>>>>
>>>>>> 
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> Is your legacy SCM system holding you back? Join Perforce May 7
>>>>> to find out:
>>>>> • 3 signs your SCM is hindering your productivity
>>>>> • Requirements for releasing software faster
>>>>> • Expert tips and advice for migrating your SCM now
>>>>> http://p.sf.net/sfu/perforce
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Is your legacy SCM system holding you back? Join Perforce May 7 to
>>>> find out:
>>>> • 3 signs your SCM is hindering your productivity
>>>> • Requirements for releasing software faster
>>>> • Expert tips and advice for migrating your SCM now
>>>> http://p.sf.net/sfu/perforce
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>>> • 3 signs your SCM is hindering your productivity
>>> • Requirements for releasing software faster
>>> • Expert tips and advice for migrating your SCM now
>>> http://p.sf.net/sfu/perforce
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>>
>> ------------------------------------------------------------------------------
>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>> • 3 signs your SCM is hindering your productivity
>> • Requirements for releasing software faster
>> • Expert tips and advice for migrating your SCM now
>> http://p.sf.net/sfu/perforce
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
>
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

It did run with no crashes with code from April 30.
========================
Dr Victor Lazzarini
Senior Lecturer
NUI Maynooth, Ireland
victor dot lazzarini at nuim dot ie




On 5 May 2014, at 23:48, Steven Yi  wrote:

> This is all a bit odd as the code did not previous crash or have any
> divisions by zero.  This is for both the crashing example but also for
> a blue demo project I had done a year ago.  So I suspect that maybe it
> might be something where UDO's are called from UDO's.
> 
> I just got in and it's already a bit late, so I think I'll tackle this
> in the morning.
> 
> On Mon, May 5, 2014 at 10:31 PM,   wrote:
>> Eak...!!!!!  div.ii does not exist -- uses div.kk code
>> Can fix but not tonight
>> 
>> ==John
>> Quoting Victor Lazzarini :
>> 
>>> The trouble is that PerfError() is being called at init time and
>>> that causes the crash. Is it possible to adjust the
>>> code to call InitError() on div.ii and PerfError elsewhere?
>>> 
>>> PerfError() cannot be called at i-time otherwise crashes might ensue
>>> (as it gets rid of the instance).
>>> 
>>> ========================
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> NUI Maynooth, Ireland
>>> victor dot lazzarini at nuim dot ie
>>> 
>>> 
>>> 
>>> 
>>> On 5 May 2014, at 20:36, Victor Lazzarini  wrote:
>>> 
>>>> But I think what is not right is the fact that the perferror is
>>>> causing a crash. I looked and it is in the usropcdset() code, and
>>>> I am wondering if this is to do with the new udo code at all.
>>>> ========================
>>>> Dr Victor Lazzarini
>>>> Senior Lecturer
>>>> NUI Maynooth, Ireland
>>>> victor dot lazzarini at nuim dot ie
>>>> 
>>>> 
>>>> 
>>>> 
>>>> On 5 May 2014, at 20:11, jpff@cs.bath.ac.uk wrote:
>>>> 
>>>>> This must be my fault -- I was adding checks for division by zero in
>>>>> an attempt to chase the Josep/clarinet problem
>>>>> 
>>>>> ==Jojm ff
>>>>> 
>>>>> Quoting Victor Lazzarini :
>>>>> 
>>>>>> it was not crashing here with a build from april 30. With today?s
>>>>>> develop code I get
>>>>>> 
>>>>>> new alloc for instr 1:
>>>>>> PERF ERROR in instr 1 (opcode yi_env_orch): Division by zero
>>>>>> note aborted
>>>>>> Csound tidy up: Segmentation fault
>>>>>> 
>>>>>> ========================
>>>>>> Dr Victor Lazzarini
>>>>>> Senior Lecturer
>>>>>> NUI Maynooth, Ireland
>>>>>> victor dot lazzarini at nuim dot ie
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> On 5 May 2014, at 18:09, Steven Yi  wrote:
>>>>>> 
>>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> ------------------------------------------------------------------------------
>>>>>> Is your legacy SCM system holding you back? Join Perforce May 7
>>>>>> to find out:
>>>>>> • 3 signs your SCM is hindering your productivity
>>>>>> • Requirements for releasing software faster
>>>>>> • Expert tips and advice for migrating your SCM now
>>>>>> http://p.sf.net/sfu/perforce
>>>>>> _______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> ------------------------------------------------------------------------------
>>>>> Is your legacy SCM system holding you back? Join Perforce May 7 to
>>>>> find out:
>>>>> • 3 signs your SCM is hindering your productivity
>>>>> • Requirements for releasing software faster
>>>>> • Expert tips and advice for migrating your SCM now
>>>>> http://p.sf.net/sfu/perforce
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>> 
>>>> 
>>>> ------------------------------------------------------------------------------
>>>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>>>> • 3 signs your SCM is hindering your productivity
>>>> • Requirements for releasing software faster
>>>> • Expert tips and advice for migrating your SCM now
>>>> http://p.sf.net/sfu/perforce
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>>> • 3 signs your SCM is hindering your productivity
>>> • Requirements for releasing software faster
>>> • Expert tips and advice for migrating your SCM now
>>> http://p.sf.net/sfu/perforce
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>> • 3 signs your SCM is hindering your productivity
>> • Requirements for releasing software faster
>> • Expert tips and advice for migrating your SCM now
>> http://p.sf.net/sfu/perforce
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks John and Victor for looking at this!

The iEndAmp line looks like something I used to use in older pieces
but I had changed how it was used later.  It seems then that this is a
case that I had bad Csound code for a while now without realizing it.

In that case,  Csound crashing on error is the bug, but with valid
code that won't be a problem.  I think since we already released
6.03.0, we should wait to finish off the release, then commit fixed
code to a hotfix branch (6.03.1).  That way we won't have any
confusion over the releases and what's "really" in the release.  Sound
good?

On Tue, May 6, 2014 at 8:32 AM,   wrote:
>
> Quoting Victor Lazzarini :
>
>> This line seems to be culprit:
>>
>> 529:   iEndAmp = iNewAmp / iOldAmp
>>
>>
>
> even odder is that the value of iEndAmp in not used!
>
>
>
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Just FYI: I was looking through my pieces and it looks like I've had
this bit of errant code for many years now (earliest piece I found
that has it crashing is from 2006).

I guess this is a bigger issue to deal with as the behavior of Csound
has changed.  I imagine other pieces by others might be affected.  I'm
wondering what is the best course of action.  Obviously division by
zero is allowed in C, as we were using that before, though the result
is undefined.  In other languages, a division by zero throws an
exception.

On the one hand, I like that a division by zero would give an error to
the user.  It'd help to debug Csound code.  On the other hand, I
dislike that my pieces won't compile and thus backwards compatibility
is broken.

I'd like to then make the case that we return to the older behavior
and simply return the division by zero result that C gives, with the
idea that backwards compatibility is more important here. We could
issue warnings if division by zero is found. For older pieces like
mine, it would cause lots of errors especially if done a k-time, but
at least the piece would run.



On Tue, May 6, 2014 at 10:39 AM, Steven Yi  wrote:
> Thanks John and Victor for looking at this!
>
> The iEndAmp line looks like something I used to use in older pieces
> but I had changed how it was used later.  It seems then that this is a
> case that I had bad Csound code for a while now without realizing it.
>
> In that case,  Csound crashing on error is the bug, but with valid
> code that won't be a problem.  I think since we already released
> 6.03.0, we should wait to finish off the release, then commit fixed
> code to a hotfix branch (6.03.1).  That way we won't have any
> confusion over the releases and what's "really" in the release.  Sound
> good?
>
> On Tue, May 6, 2014 at 8:32 AM,   wrote:
>>
>> Quoting Victor Lazzarini :
>>
>>> This line seems to be culprit:
>>>
>>> 529:   iEndAmp = iNewAmp / iOldAmp
>>>
>>>
>>
>> even odder is that the value of iEndAmp in not used!
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>> • 3 signs your SCM is hindering your productivity
>> • Requirements for releasing software faster
>> • Expert tips and advice for migrating your SCM now
>> http://p.sf.net/sfu/perforce
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks John,

I found the commits.  Regarding my other email, can we change this to
a warning instead of an error so that we can preserve backwards
compatibility?

steven

On Tue, May 6, 2014 at 12:08 PM,   wrote:
> yes; revied code in git now
> Quoting Steven Yi :
>
>> Thanks John and Victor for looking at this!
>>
>> The iEndAmp line looks like something I used to use in older pieces
>> but I had changed how it was used later.  It seems then that this is a
>> case that I had bad Csound code for a while now without realizing it.
>>
>> In that case,  Csound crashing on error is the bug, but with valid
>> code that won't be a problem.  I think since we already released
>> 6.03.0, we should wait to finish off the release, then commit fixed
>> code to a hotfix branch (6.03.1).  That way we won't have any
>> confusion over the releases and what's "really" in the release.  Sound
>> good?
>>
>> On Tue, May 6, 2014 at 8:32 AM,   wrote:
>>>
>>> Quoting Victor Lazzarini :
>>>
>>>> This line seems to be culprit:
>>>>
>>>> 529:   iEndAmp = iNewAmp / iOldAmp
>>>>
>>>>
>>>
>>> even odder is that the value of iEndAmp in not used!
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>>> • 3 signs your SCM is hindering your productivity
>>> • Requirements for releasing software faster
>>> • Expert tips and advice for migrating your SCM now
>>> http://p.sf.net/sfu/perforce
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
>> • 3 signs your SCM is hindering your productivity
>> • Requirements for releasing software faster
>> • Expert tips and advice for migrating your SCM now
>> http://p.sf.net/sfu/perforce
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
>
> ------------------------------------------------------------------------------
> Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
> • 3 signs your SCM is hindering your productivity
> • Requirements for releasing software faster
> • Expert tips and advice for migrating your SCM now
> http://p.sf.net/sfu/perforce
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net