Re: Segmentation fault with a bad systemStartDelimiter

[Simon Albrecht]

On 13.07.2016 22:54, Nathan Ho wrote:

Hi list,

A bad value for StaffGroup.systemStartDelimiter appears to cause a 
segmentation fault:


I can confirm that.



\version "2.19.45"
\new StaffGroup \with { systemStartDelimiter = #'aaa } { c'4 }


$ lilypond test.ly
GNU LilyPond 2.19.45
Processing `traiettorie_ohshit.ly'
Parsing...
Interpreting music...Segmentation fault (core dumped)


Is this the right place to report this sort of thing?


No, you may report on the bug list. Still, most of the bug squad are 
likely to hang around here as well, so – 
 :-)


Best, Simon

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Patrick Karl]

On 12/29/15 6:58 PM, Simon Albrecht wrote:

On 30.12.2015 01:06, Patrick Karl wrote:

On 12/29/15 5:53 PM, Simon Albrecht wrote:

Hello Pat,

this is definitely a bug, I’d say. I’ll write to the bug list.
Two or three policy issues with this:
1. Don’t hijack existing threads, if your message has nothing at all 
to do with their subject. Just compose a new message to the list – 
sorry, but what’s so difficult about that?


Just out of curiosity, exactly what existing thread do you think I 
hijacked?


I can only guess how you proceeded, but if you reply to a previous 
message and edit all of the visible information, hidden data will 
still link this mail to the previous thread.
Bingo! (Slang for "you hit the nail on the head", which itself is slang 
for "you got it exactly right".)  I have to confess that even after 30 
years of working with various unices, mail generally remains a black 
hole for me.


I get the lilypond-user digest.  If I want to reply to an individual 
message in one of the digests, I know of no other way than replying to 
the digest while editing out all the other messages in the digest and 
changing the reply's subject to the one in the individual message I'm 
responding to.  So what I did here was similar.



2. (Re)read and follow .

Again, out of curiosity, how is my snippet not a tiny-example?
When trying to create an example, try commenting out (|%| or |%{ … 
%}|) sections of your file. If you can comment something while still 
demonstrating the main idea, then remove the commented-material. 




Have a look at issue 4718 (link in my previous post) – there’s a tiny 
example.


When I look at issue 4718, I see: \new Staff \new Voice \music, which 
contains "\new Voice", which I didn't use in my original submission.  If 
I look at the "Tiny examples" link you sent me in your previous post, I see:


When trying to create an example, try commenting out (% or %{ … %}) 
sections of your file. If you can comment something while still 
demonstrating the main idea, then remove the commented-material.
But that applies to your "\new Voice" addition.  So, using your 
criteria, I don't see how issue 4718 is actually a tiny example. It 
looks like you added the "\new Voice" to avoid the issue discussed in 
the original thread I hijacked.  But it really doesn't add anything to 
the discussion of this issue.


I guess my approach might be called "minimal example" rather than "tiny 
example".  Especially since the problem was that lilypond terminated 
abnormally, I felt I could save the investigators some time if I 
included, in a minimal way, everything that I knew about the problem.  
In this case, my submission showed that there was nothing intrinsically 
wrong with either the quoted or the quoting music.  I then included a 
commented out section that would trigger the problem if compiled in 
uncommented form.





3. Code formatting: Generally, always surround {} and = and Scheme 
expressions (those with `#') with spaces, even if they’re not 
technically necessary. They make it easier to read.


I didn't realize that coding style was a policy issue.


Yes, it is. If you post code to the list, others have to read and work 
with that code. Privately, you may do whatever you like.
Where is this policy enunciated?  Shouldn't there be a link to it at the 
point a person subscribes to the list, i.e., at 
https://lists.gnu.org/mailman/listinfo/lilypond-user?


Offhand I don't see any {} that are not surrounded by white space, 
i.e., blank, tab, or newline.  That leaves Scheme expressions.  I 
would say I can easily find many examples of exactly the style I 
choose in the NR.


Then it would have been messed up on the way. Attachments tend to be 
safer there. Find attached the version which arrived here and one with 
proper use of whitespace.

Sorry for being so strict with this.
Many of the differences in the two versions you attached appear to be 
due to you thinking I have used too much whitespace.  There is one case 
(\score { \new Staff {\cueNotes } }) where I have inadvertently omitted 
a space after a "{".  I think we're back to the Scheme expressions.  I 
use Frescobaldi to write LilyPond code. When I enter "\score {  \new 
Staff { \rem", Frescobaldi presents a list of the possible completions, 
and I click on the one I want. Frescobaldi then supplies "removeWithTag" 
without a trailing space. I then add #' etc.  I wonder if it wouldn't be 
nice if Frescobaldi supplied "removeWithTag #' " instead.  I don't think 
there are any occasions where anything except "#' " can follow 
\removeWithTag.


I also wonder why LilyPond even accepts \removeWithTag#' without a space 
before #'.


I call nitpicking.  Mea minima culpa.




Yours, Simon


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Patrick Karl]

On 12/29/15 6:38 PM, Thomas Morley wrote:

2015-12-30 0:59 GMT+01:00 Simon Albrecht :

On 30.12.2015 00:53, Simon Albrecht wrote:

this is definitely a bug, I’d say. I’ll write to the bug list.


FWIW, I created a tracker issue directly:

How did you get that log output, i.e. how did you invoke LilyPond?

Yours, Simon




I once created a function to show/print cues/quotes known to the parser.
Applied to the current use case:


%%%

\version "2.19.32"

cueNotes = \relative c'' {
s4. 8 d' s4.
}

\addQuote "cueNotes" \cueNotes

#(define* (music-quotes-info #:optional (name "")(print #f))
  (let* ((music-quotes-list
   (hash-map->list cons (ly:parser-lookup 'musicQuotes)))
 (quote-vector (assoc-get name music-quotes-list #()))
 (quote-list (vector->list quote-vector)))
   (cond ((string-null? name)
  (display-scheme-music music-quotes-list))
 ((and (not (null? quote-list)) print)
  (make-sequential-music
(map
  (lambda (m) (ly:prob-property (caadr m) 'music-cause))
  quote-list)))
 (else (display-scheme-music
 (make-sequential-music
   (map
 (lambda (m) (ly:prob-property (caadr m) 'music-cause))
 quote-list)))

%% raw data of all quotes
%#(music-quotes-info)
%% `display-scheme-music'
#(music-quotes-info "cueNotes")
%% set `print' #t and use $ to get it printed
$(music-quotes-info "cueNotes" #t)

%%%

You'll see that only the chord's first note is quoted!
That's at least related to
NR
"
Known issues and warnings
Only the contents of the first Voice occurring in an \addQuote command
will be considered for quotation [...]
"

Though,
(1) It should not return a segfault
(2) The NR-warning talks about new Voices etc. Before testing I wasn't
aware that quoting chords is not supported as well. Should be
mentioned, imho
(3) I do understand why quoting more than a single voice is a problem,
but not how difficult it would be to support chords.

That confuses me somewhat.  Consider:

\version "2.19.32"

cueNotes = \relative c'' {
   s4. 8 d' s4.
}

\addQuote "cueNotes" \cueNotes

{ \cueDuring #"cueNotes" #DOWN R1 }

The output looks exactly as I would expect if chords were supported.  I think 
they are supported.  And tied notes and tied chords in the quoted music work as 
expected.  The only issue I see is lack of correct implementation of tied 
individual notes within quoted chords, resulting in a segmentation fault.

(Since quoted chords do work, why doesn't music-quotes-info show them?)





So I'm not sure whether it's a documentation issue or defect or an
enhancement or both.

Simon, I answered here on user-list to let Patrick Karl know, but
please add it on the tracker if you feel it might be helpful


Cheers,
   Harm



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Thomas Morley]

2015-12-30 13:37 GMT+01:00 Patrick Karl :
> On 12/29/15 6:38 PM, Thomas Morley wrote:

>
> That confuses me somewhat.  Consider:
>
> \version "2.19.32"
>
> cueNotes = \relative c'' {
>s4. 8 d' s4.
> }
>
> \addQuote "cueNotes" \cueNotes
>
> { \cueDuring #"cueNotes" #DOWN R1 }
>
> The output looks exactly as I would expect if chords were supported.  I
> think they are supported.  And tied notes and tied chords in the quoted
> music work as expected.  The only issue I see is lack of correct
> implementation of tied individual notes within quoted chords, resulting in a
> segmentation fault.
>
> (Since quoted chords do work, why doesn't music-quotes-info show them?)

Didn't you get
http://lists.gnu.org/archive/html/lilypond-user/2015-12/msg00721.html
?

`music-quotes-info' was buggy, updated version here:
http://lists.gnu.org/archive/html/lilypond-user/2015-12/msg00726.html

Ofcourse it only prints info, the problem persists...

Cheers,
  Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Simon Albrecht]

On 30.12.2015 01:06, Patrick Karl wrote:

On 12/29/15 5:53 PM, Simon Albrecht wrote:

Hello Pat,

this is definitely a bug, I’d say. I’ll write to the bug list.
Two or three policy issues with this:
1. Don’t hijack existing threads, if your message has nothing at all 
to do with their subject. Just compose a new message to the list – 
sorry, but what’s so difficult about that?


Just out of curiosity, exactly what existing thread do you think I 
hijacked?


I can only guess how you proceeded, but if you reply to a previous 
message and edit all of the visible information, hidden data will still 
link this mail to the previous thread.



2. (Re)read and follow .

Again, out of curiosity, how is my snippet not a tiny-example?


Have a look at issue 4718 (link in my previous post) – there’s a tiny 
example.


3. Code formatting: Generally, always surround {} and = and Scheme 
expressions (those with `#') with spaces, even if they’re not 
technically necessary. They make it easier to read.


I didn't realize that coding style was a policy issue.


Yes, it is. If you post code to the list, others have to read and work 
with that code. Privately, you may do whatever you like.


Offhand I don't see any {} that are not surrounded by white space, 
i.e., blank, tab, or newline.  That leaves Scheme expressions.  I 
would say I can easily find many examples of exactly the style I 
choose in the NR.


Then it would have been messed up on the way. Attachments tend to be 
safer there. Find attached the version which arrived here and one with 
proper use of whitespace.

Sorry for being so strict with this.

Yours, Simon
%%%   Begin snippet %%%
\version "2.19.32"

music =   {
\tag #'noCues R1
\tag #'Cues {\cueDuring #"cueNotes" #DOWN  R1 }% 1
d''8( e'' f'') r r2% 2
}

cueNotes = \relative c'' {
   s4. 8 d' s4.% 1
}

\addQuote "cueNotes" \cueNotes

\score { \new Staff {\cueNotes }  }

\score {  \new Staff { \removeWithTag#'Cues \music } }

%\score { \new Staff { \keepWithTag#'Cues \music }  }

%%%   End snippet %%% 

%%% Begin snippet %%%
\version "2.19.32"

music = {
  \tag #'noCues R1
  \tag #'Cues { \cueDuring #"cueNotes" #DOWN R1 }% 1
  d''8( e'' f'') r r2% 2
}

cueNotes = \relative c'' {
  s4. 8 d' s4.% 1
}

\addQuote "cueNotes" \cueNotes

\score { \new Staff { \cueNotes } }

\score {  \new Staff { \removeWithTag #'Cues \music } }

%\score { \new Staff { \keepWithTag #'Cues \music } }

%%% End snippet %%%___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Patrick Karl]

On 12/29/15 5:53 PM, Simon Albrecht wrote:

Hello Pat,

this is definitely a bug, I’d say. I’ll write to the bug list.
Two or three policy issues with this:
1. Don’t hijack existing threads, if your message has nothing at all 
to do with their subject. Just compose a new message to the list – 
sorry, but what’s so difficult about that?


Just out of curiosity, exactly what existing thread do you think I hijacked?

2. (Re)read and follow .

Again, out of curiosity, how is my snippet not a tiny-example?
3. Code formatting: Generally, always surround {} and = and Scheme 
expressions (those with `#') with spaces, even if they’re not 
technically necessary. They make it easier to read.


I didn't realize that coding style was a policy issue.  Offhand I don't 
see any {} that are not surrounded by white space, i.e., blank, tab, or 
newline.  That leaves Scheme expressions.  I would say I can easily find 
many examples of exactly the style I choose in the NR.


Thank you,
Simon

On 29.12.2015 20:38, Patrick Karl wrote:
The following snippet contains three \score's, the third one 
commented out:


%%%   Begin snippet %%%
\version "2.19.32"

music =   {
\tag #'noCues R1
\tag #'Cues {\cueDuring #"cueNotes" #DOWN  R1 }% 1
d''8( e'' f'') r r2% 2
}

cueNotes = \relative c'' {
   s4. 8 d' s4.% 1
}

\addQuote "cueNotes" \cueNotes

\score { \new Staff {\cueNotes }  }

\score {  \new Staff { \removeWithTag#'Cues \music } }

%\score { \new Staff { \keepWithTag#'Cues \music }  }

%%%   End snippet %%%

This snippet compiles ok, but if the third \score is uncommented, 
then lilypond terminates with:


.
.
.
Preprocessing graphical objects...
Grob count 156
[lilypond_serif_3.068359375]
Interpreting music...Segmentation fault: 11

The problem appears to be the construct "8 d' " in the 
cueNotes.  Does anyone know why?



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user





___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[David Kastrup]

Patrick Karl  writes:

> On 12/29/15 5:53 PM, Simon Albrecht wrote:
>> Hello Pat,
>>
>> this is definitely a bug, I’d say. I’ll write to the bug list.
>> Two or three policy issues with this:
>> 1. Don’t hijack existing threads, if your message has nothing at all
>> to do with their subject. Just compose a new message to the list –
>> sorry, but what’s so difficult about that?
>
> Just out of curiosity, exactly what existing thread do you think I
> hijacked?

"cue note fontsize inherited by uncued notes?" by yourself.  The
reference header of your first "Segmentation Fault in music with
cueDuring" mail points to the first mail of that other thread.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Simon Albrecht]

Hello Pat,

this is definitely a bug, I’d say. I’ll write to the bug list.
Two or three policy issues with this:
1. Don’t hijack existing threads, if your message has nothing at all to 
do with their subject. Just compose a new message to the list – sorry, 
but what’s so difficult about that?

2. (Re)read and follow .
3. Code formatting: Generally, always surround {} and = and Scheme 
expressions (those with `#') with spaces, even if they’re not 
technically necessary. They make it easier to read.


Thank you,
Simon

On 29.12.2015 20:38, Patrick Karl wrote:
The following snippet contains three \score's, the third one commented 
out:


%%%   Begin snippet %%%
\version "2.19.32"

music =   {
\tag #'noCues R1
\tag #'Cues {\cueDuring #"cueNotes" #DOWN  R1 }% 1
d''8( e'' f'') r r2% 2
}

cueNotes = \relative c'' {
   s4. 8 d' s4.% 1
}

\addQuote "cueNotes" \cueNotes

\score { \new Staff {\cueNotes }  }

\score {  \new Staff { \removeWithTag#'Cues \music } }

%\score { \new Staff { \keepWithTag#'Cues \music }  }

%%%   End snippet %%%

This snippet compiles ok, but if the third \score is uncommented, then 
lilypond terminates with:


.
.
.
Preprocessing graphical objects...
Grob count 156
[lilypond_serif_3.068359375]
Interpreting music...Segmentation fault: 11

The problem appears to be the construct "8 d' " in the 
cueNotes.  Does anyone know why?



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Simon Albrecht]

On 30.12.2015 01:38, Thomas Morley wrote:

You'll see that only the chord's first note is quoted!


Still, the in-chord tie makes the difference, without that tie, the 
chord is properly quoted. So I’m not sure whether your testing function 
is useful here.


Yours, Simon

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Simon Albrecht]

On 30.12.2015 00:53, Simon Albrecht wrote:
this is definitely a bug, I’d say. I’ll write to the bug list. 


FWIW, I created a tracker issue directly: 


How did you get that log output, i.e. how did you invoke LilyPond?

Yours, Simon

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Patrick Karl]

On 12/29/15 5:59 PM, Simon Albrecht wrote:

On 30.12.2015 00:53, Simon Albrecht wrote:
this is definitely a bug, I’d say. I’ll write to the bug list. 


FWIW, I created a tracker issue directly: 


How did you get that log output, i.e. how did you invoke LilyPond?


lilypond -V myfile.ly


Yours, Simon



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Thomas Morley]

2015-12-30 0:59 GMT+01:00 Simon Albrecht :
> On 30.12.2015 00:53, Simon Albrecht wrote:
>>
>> this is definitely a bug, I’d say. I’ll write to the bug list.
>
>
> FWIW, I created a tracker issue directly:
> 
> How did you get that log output, i.e. how did you invoke LilyPond?
>
> Yours, Simon
>



I once created a function to show/print cues/quotes known to the parser.
Applied to the current use case:


%%%

\version "2.19.32"

cueNotes = \relative c'' {
   s4. 8 d' s4.
}

\addQuote "cueNotes" \cueNotes

#(define* (music-quotes-info #:optional (name "")(print #f))
 (let* ((music-quotes-list
  (hash-map->list cons (ly:parser-lookup 'musicQuotes)))
(quote-vector (assoc-get name music-quotes-list #()))
(quote-list (vector->list quote-vector)))
  (cond ((string-null? name)
 (display-scheme-music music-quotes-list))
((and (not (null? quote-list)) print)
 (make-sequential-music
   (map
 (lambda (m) (ly:prob-property (caadr m) 'music-cause))
 quote-list)))
(else (display-scheme-music
(make-sequential-music
  (map
(lambda (m) (ly:prob-property (caadr m) 'music-cause))
quote-list)))

%% raw data of all quotes
%#(music-quotes-info)
%% `display-scheme-music'
#(music-quotes-info "cueNotes")
%% set `print' #t and use $ to get it printed
$(music-quotes-info "cueNotes" #t)

%%%

You'll see that only the chord's first note is quoted!
That's at least related to
NR
"
Known issues and warnings
Only the contents of the first Voice occurring in an \addQuote command
will be considered for quotation [...]
"

Though,
(1) It should not return a segfault
(2) The NR-warning talks about new Voices etc. Before testing I wasn't
aware that quoting chords is not supported as well. Should be
mentioned, imho
(3) I do understand why quoting more than a single voice is a problem,
but not how difficult it would be to support chords.


So I'm not sure whether it's a documentation issue or defect or an
enhancement or both.

Simon, I answered here on user-list to let Patrick Karl know, but
please add it on the tracker if you feel it might be helpful


Cheers,
  Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Thomas Morley]

2015-12-30 1:45 GMT+01:00 Simon Albrecht :
> On 30.12.2015 01:38, Thomas Morley wrote:
>>
>> You'll see that only the chord's first note is quoted!
>
>
> Still, the in-chord tie makes the difference, without that tie, the chord is
> properly quoted. So I’m not sure whether your testing function is useful
> here.
>
> Yours, Simon

I stand corrected.
It's my function which disregards the chord not LilyPond! Looks like
I'll have to improve it

Thanks,
  Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in music with cueDuring

[Thomas Morley]

2015-12-30 1:56 GMT+01:00 Thomas Morley :
> 2015-12-30 1:45 GMT+01:00 Simon Albrecht :
>> On 30.12.2015 01:38, Thomas Morley wrote:
>>>
>>> You'll see that only the chord's first note is quoted!
>>
>>
>> Still, the in-chord tie makes the difference, without that tie, the chord is
>> properly quoted. So I’m not sure whether your testing function is useful
>> here.
>>
>> Yours, Simon
>
> I stand corrected.
> It's my function which disregards the chord not LilyPond! Looks like
> I'll have to improve it
>
> Thanks,
>   Harm

For the record, here an updated version of it.

Ofcourse only the option displaying the raw data is really accurate.
The other options rebuild stuff which may be handled different in
LilyPond internals. Though, studying the raw data looks like storing
the data happened correctly.

\version "2.19.32"

cueNotes = \relative c'' {
   8 d'
}

\addQuote "cueNotes" \cueNotes

#(define* (music-quotes-info #:optional (name "")(print #f))
 (let* ((music-quotes-list
  (hash-map->list cons (ly:parser-lookup 'musicQuotes)))
(quote-vector (assoc-get name music-quotes-list #()))
(quote-list (vector->list quote-vector)))
  (cond ((string-null? name)
 (display-scheme-music music-quotes-list))
((and (not (null? quote-list)) print)
 (make-sequential-music
   (map
 (lambda (m)
   (let ((elts
   (map
 (lambda (e) (ly:prob-property (car e) 'music-cause))
 (cdr m
  (if (= (length elts) 1)
  (car elts)
  (make-event-chord elts
 quote-list)))
(else
  (display-scheme-music
(make-sequential-music
  (map
(lambda (m)
  (let ((elts
  (map
(lambda (e) (ly:prob-property (car e) 'music-cause))
(cdr m
 (if (= (length elts) 1)
 (car elts)
 (make-event-chord elts
quote-list)))

%% raw data of all quotes
#(music-quotes-info)
%% `display-scheme-music'
%#(music-quotes-info "cueNotes")
%% set `print' #t and use $ to get it printed
%$(music-quotes-info "cueNotes" #t)

Cheers,
  Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: segmentation fault from 2.19.29

[Paul Scott]

On Sun, Oct 25, 2015 at 08:15:49PM +0100, David Kastrup wrote:
> Paul Scott  writes:
> 
> > I get a segmentation fault from 2.19.29 that I don't get with
> > 2.19.28.  I'm running 64-bit on Debian sid.  It looks to be size
> > related so a minimal example doesn't seem to be relevant.
> 
> Garbage collection while slurs are being processed during the iteration
> stage.  Yes, size-related (due to garbage collection needing to be
> triggered), yes, very very bad.
> 
> Phil is already preparing the release of 2.19.30 right now, one week
> early, in order to fix this awful blunder of mine that unfortunately
> escaped notice until after the release.
> 
> > I'd be happy to run any diagnostic that might help.
> 
> You can hopefully do so using 2.19.30 tomorrow at the latest.

Thank you for your fast response!  2.19.28 is fine for now.  I just like
to run the latest development version.  I will get 2.19.30 when I see it.

Thank you for all you do for Lily.  As soon as I see how we can contribute 
money I will do so.

Paul



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: segmentation fault from 2.19.29

[David Kastrup]

Paul Scott  writes:

> I get a segmentation fault from 2.19.29 that I don't get with
> 2.19.28.  I'm running 64-bit on Debian sid.  It looks to be size
> related so a minimal example doesn't seem to be relevant.

Garbage collection while slurs are being processed during the iteration
stage.  Yes, size-related (due to garbage collection needing to be
triggered), yes, very very bad.

Phil is already preparing the release of 2.19.30 right now, one week
early, in order to fix this awful blunder of mine that unfortunately
escaped notice until after the release.

> I'd be happy to run any diagnostic that might help.

You can hopefully do so using 2.19.30 tomorrow at the latest.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in 2.18 (was Re: Learn from Finale 2014 (seriously)?)

[Richard Shann]

I just came across a workaround for this LilyPond crash - adding

\once \override MultiMeasureRestNumber #'transparent = ##t

before the multi-measure rests in the second voice results in the
typesetting finishing correctly.

I came across this because I was getting the number printed twice in
some circumstances (even though the rest was only printed once), and
this fixed that bug.

The only remaining tweak is to adjust the vertical positioning of the
multi-measure rest, then Denemo will have a complete package of commands
for printing a score and parts where some parts have multi-measure rests
that are shared.

Richard


On Sat, 2014-02-01 at 15:04 +, Richard Shann wrote:
 On Thu, 2013-11-14 at 21:14 +0100, Jan-Peter Voigt wrote:
  On 14.11.2013 16:19, Joseph Rushton Wakeling wrote:
   there is an update of this snippet in the mail archives and I will post
   my version later.
  
   Fantastic, thank you! :-) 
  so, here is the version I regularly use ... in fact, this is taken from
  some e-mail on this list some time ago. It also merges MultiMeasure rests.
 
 I was just putting this code into Denemo and I noticed it merged a
 multi-measure rest in the top voice with a sequence of whole measure
 rests in the second voice, but it caused a segmentation fault in
 LilyPond when it tries to merge multi measure rests from both voices.
 
 The test file that shows this is attached. The include file is also
 attached, it is the mergeRests.ly from the original message with the
 example removed and convert-ly run on it to update it from 2.16.1 - the
 resultant version reads 2.17.97 which I recall is a known bug.
 
 I guess this part of the thread needs to switch to lilypond-devel or
 lilypond-bug, now - I originally started this email just to enquire
 about merging multi-measure rests ...
 
 Richard
 
 
 
 
 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 https://lists.gnu.org/mailman/listinfo/lilypond-user



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault in 2.18 (was Re: Learn from Finale 2014 (seriously)?)

[Paul Scott]

On Thu, Feb 27, 2014 at 10:27:53AM +, Richard Shann wrote:
 I just came across a workaround for this LilyPond crash - adding

I have been getting very occasional segfaults with Lily since at 
least 2.17.  I have never been able to create a minimal example 
because as soon as I change anything the code works fine.

Paul Scott

 
 \once \override MultiMeasureRestNumber #'transparent = ##t
 
 before the multi-measure rests in the second voice results in the
 typesetting finishing correctly.
 
 I came across this because I was getting the number printed twice in
 some circumstances (even though the rest was only printed once), and
 this fixed that bug.
 
 The only remaining tweak is to adjust the vertical positioning of the
 multi-measure rest, then Denemo will have a complete package of commands
 for printing a score and parts where some parts have multi-measure rests
 that are shared.
 
 Richard
 
 
 On Sat, 2014-02-01 at 15:04 +, Richard Shann wrote:
  On Thu, 2013-11-14 at 21:14 +0100, Jan-Peter Voigt wrote:
   On 14.11.2013 16:19, Joseph Rushton Wakeling wrote:
there is an update of this snippet in the mail archives and I will post
my version later.
   
Fantastic, thank you! :-) 
   so, here is the version I regularly use ... in fact, this is taken from
   some e-mail on this list some time ago. It also merges MultiMeasure rests.
  
  I was just putting this code into Denemo and I noticed it merged a
  multi-measure rest in the top voice with a sequence of whole measure
  rests in the second voice, but it caused a segmentation fault in
  LilyPond when it tries to merge multi measure rests from both voices.
  
  The test file that shows this is attached. The include file is also
  attached, it is the mergeRests.ly from the original message with the
  example removed and convert-ly run on it to update it from 2.16.1 - the
  resultant version reads 2.17.97 which I recall is a known bug.
  
  I guess this part of the thread needs to switch to lilypond-devel or
  lilypond-bug, now - I originally started this email just to enquire
  about merging multi-measure rests ...
  
  Richard
  
  
  
  
  ___
  lilypond-user mailing list
  lilypond-user@gnu.org
  https://lists.gnu.org/mailman/listinfo/lilypond-user
 
 
 
 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 https://lists.gnu.org/mailman/listinfo/lilypond-user
 


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[Trevor Daniels]

Paul Scott wrote Wednesday, March 20, 2013 7:56 AM


I get segmentation faults with 2.17.14 on a medium to large set of files 
 which compile fine with 2.17.13.  I have gotten these errors with some other 
 recent 2.17.xx versions.  I don't believe I will get errors with a minimal 
 example.
 
 What evidence can I provide that might help find the problem?

Well, start removing things from your code until the error goes away.
Start with large sections.  Put back what you have just removed and 
continue removing other things or smaller pieces.  When there is nothing 
left that can be removed without also removing the error you have a 
minimal example.  

The chances are in carrying out this process that you will discover
what is causing the problem.  Tell us what it is.

Trevor
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[David Kastrup]

Trevor Daniels t.dani...@treda.co.uk writes:

 Paul Scott wrote Wednesday, March 20, 2013 7:56 AM


I get segmentation faults with 2.17.14 on a medium to large set of files 
 which compile fine with 2.17.13.  I have gotten these errors with
 some other
 recent 2.17.xx versions.  I don't believe I will get errors with a
 minimal
 example.
 
 What evidence can I provide that might help find the problem?

 Well, start removing things from your code until the error goes away.
 Start with large sections.  Put back what you have just removed and 
 continue removing other things or smaller pieces.  When there is nothing 
 left that can be removed without also removing the error you have a 
 minimal example.  

 The chances are in carrying out this process that you will discover
 what is causing the problem.  Tell us what it is.

That's more or less cutpaste from the instructions for a minimal
example.  There is no guarantee that the code removed along with the
segfault is responsible for the segfault.  But there is a good chance
that you have arrived at a version where the segfault is easier to
diagnose, because we'll be searching for the needle in a quite smaller
haystack.

-- 
David Kastrup


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[David Kastrup]

Paul Scott waterho...@ultrasw.com writes:

 I get segmentation faults with 2.17.14 on a medium to large set of files 
 which compile fine with 2.17.13.  I have gotten these errors with some
 other
 recent 2.17.xx versions.  I don't believe I will get errors with a
 minimal example.

 What evidence can I provide that might help find the problem?

I tend to do something like

git log release/2.17.13-1..release/2.17.14-1

for problem occured between x and y to look for a likely candidate
based on the commit message, and I just did that.  Tell you what: easily
a third or more of all commits could be responsible.

There are two things for circling the problem: one is git bisect to
figure out the commit introducing the segfault.  If nothing else, we can
then revert the commit until figuring out the problem.  The other is
post-mortem-debugging the segfault to figure out the code where things
go wrong.

Either will require a developer being able to reproduce the problem.
Try reducing a file of yours to a point where the segfault still occurs
but most of the content could get thrown out.

-- 
David Kastrup


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[Paul Scott]

On Wed, Mar 20, 2013 at 08:37:38AM -, Trevor Daniels wrote:
 
 Paul Scott wrote Wednesday, March 20, 2013 7:56 AM
 
 
 I get segmentation faults with 2.17.14 on a medium to large set of files 
  which compile fine with 2.17.13.  I have gotten these errors with some 
  other 
  recent 2.17.xx versions.  I don't believe I will get errors with a minimal 
  example.
  
  What evidence can I provide that might help find the problem?
 
 Well, start removing things from your code until the error goes away.
 Start with large sections.  Put back what you have just removed and 
 continue removing other things or smaller pieces.  When there is nothing 
 left that can be removed without also removing the error you have a 
 minimal example.  

I know how to produce a minimal example.  I had been removing code and hadn't 
seen results yet and going back to 2.17.13 immediately solved the problem.  I 
have done exactly that with other music with some other recent versions.

I was wondering if some kind of trace might provide someone with a clue.

I have several parts to finish tonight which I am doing quite successfully 
with 2.17.13.

I will give it another try when I can. 

 The chances are in carrying out this process that you will discover
 what is causing the problem.  Tell us what it is.

I appreciate that possibility.  I have said the same thing to a number 
of programming classes I have taught including the assembly language 
class I was teaching tonight.

Thanks and thanks for all the work you all do on this magnificent project.  
I'll be back.

Paul

 
 Trevor


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[David Kastrup]

Paul Scott waterho...@ultrasw.com writes:

 I appreciate that possibility.  I have said the same thing to a number
 of programming classes I have taught including the assembly language
 class I was teaching tonight.

Sounds like you should be able to run lilypond in a debugger then and
get a useful backtrace.

How did you obtain your binary?  Is it compiled with the same procedure
and compiler version as the last one?

-- 
David Kastrup


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[Paul Scott]

Ok.  False alarm but hopefully useful info:

The error was triggered by two instances of \times which should have been 
\time .  Now my code compiles with either 2.17.13 or 2.17.14.

If it helps I am using lilypond-2.17.14-1.linux-64

Maybe this difference between 2.17.14 and 2.17.13 could be related to 
work on \tuplet .

Thanks,

Paul


On Wed, Mar 20, 2013 at 10:03:23AM +0100, David Kastrup wrote:
 Paul Scott waterho...@ultrasw.com writes:
 
  I get segmentation faults with 2.17.14 on a medium to large set of files 
  which compile fine with 2.17.13.  I have gotten these errors with some
  other
  recent 2.17.xx versions.  I don't believe I will get errors with a
  minimal example.
 
  What evidence can I provide that might help find the problem?
 
 I tend to do something like
 
 git log release/2.17.13-1..release/2.17.14-1
 
 for problem occured between x and y to look for a likely candidate
 based on the commit message, and I just did that.  Tell you what: easily
 a third or more of all commits could be responsible.
 
 There are two things for circling the problem: one is git bisect to
 figure out the commit introducing the segfault.  If nothing else, we can
 then revert the commit until figuring out the problem.  The other is
 post-mortem-debugging the segfault to figure out the code where things
 go wrong.
 
 Either will require a developer being able to reproduce the problem.
 Try reducing a file of yours to a point where the segfault still occurs
 but most of the content could get thrown out.
 
 -- 
 David Kastrup
 
 
 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 https://lists.gnu.org/mailman/listinfo/lilypond-user
 


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[Paul Scott]

On Wed, Mar 20, 2013 at 10:27:27AM +0100, David Kastrup wrote:
 Paul Scott waterho...@ultrasw.com writes:
 
  I appreciate that possibility.  I have said the same thing to a number
  of programming classes I have taught including the assembly language
  class I was teaching tonight.
 
 Sounds like you should be able to run lilypond in a debugger then and
 get a useful backtrace.
 
 How did you obtain your binary?  Is it compiled with the same procedure
 and compiler version as the last one?

I always run linux-x86 versions on my desktop and linux-64 versions 
on my laptop from http://lilypond.org/development.html on Debian Sid.

Thanks,

Paul



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with 2.17.14

[David Kastrup]

Paul Scott waterho...@ultrasw.com writes:

 Ok.  False alarm but hopefully useful info:

 The error was triggered by two instances of \times which should have
 been \time .  Now my code compiles with either 2.17.13 or 2.17.14.

That's no excuse for segfaulting, however.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault (core dumped)

[Thomas Morley]

2012/5/23 Jeff Barnes jbarnes...@yahoo.com:
 Ok, I checked the archives and saw another crash report on \shape, so I'll
 post my details here.

 Environments:
 Ubuntu 12.04
 Mac OSX Lion

 Versions:
 2.14.2 on ubu
 2.14.2-1 on mac


 Command on ubu:
 jbarnes@jbarnes-OptiPlex-780:~/mac/Documents/apc/music$ lilypond
 WhenILookIntoYourHoliness.ly
 GNU LilyPond 2.14.2
 Processing `WhenILookIntoYourHoliness.ly'
 Parsing...
 Interpreting music...
 Interpreting music... [8][16][24][32]Segmentation fault (core dumped)

 No core file in that directory. Ubuntu sent a crash report (attached).

 Description on mac:
 I used the ui to compose the music saved and used Command+R to compile. An
 older version of the pdf came up.

 The attached ly file causes the crash. (Sorry its so long)

 If you comment out lines 50 and 96, then uncomment line 49 and 95, the crash
 doesn't happen.

 Regards,
 Jeff

Hi Jeff,

got a segfault with 2.14.2
But it compiles fine with 2.15.39
Only a warning:
warning: ignoring too many clashing note columns


 Do I send crash reports to this list?

I've got 5 mails from you concerning the Segmentation fault.

-Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault (core dumped)

[Janek Warchoł]

Hi Jeff,

On Wed, May 23, 2012 at 9:40 PM, Jeff Barnes jbarnes...@yahoo.com wrote:
 Do I send crash reports to this list?

Not quite.  They /usually/ should go to bug-lilyp...@gnu.org .


On Wed, May 23, 2012 at 10:03 PM, Jeff Barnes jbarnes...@yahoo.com wrote:
 Ok, I checked the archives and saw another crash report on \shape,

That's because \shape is an external, user-defined function.  In other
words, the crash wasn't Lily's fault, it was \shape's fault.
But don't feel bad about this.

cheers,
Janek

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Colin Hall]

On Thu, Dec 08, 2011 at 12:23:23AM +, Carl Sorensen wrote:
 On the other hand, let say I send you the entire choral symphony in a
 folder.  Then you are left going through the syntax looking for the
 problem.  
 
 Do you have any suggestions for how this should be fixed?
 
 Post a link to your source.  I'll run it and see if it segfaults on my
 machine.  I might try a simple test or two.
 
 I won't spend hours debugging it, but at least we'll have more information.
 

Are you any further forward with this, Craig?

Send me what info you can and I'll do my best to help.

Cheers,
Colin.

-- 

Colin Hall

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[m...@apollinemike.com]

Le Dec 7, 2011 à 9:12 PM, Craig a écrit :

 Hi,
 
 I am working on a large score (choral symphony).  When I include my tuba
 part I am getting a segmentation fault.  What does this mean?
 
 Wouldn't segmentation have to do with a memory issue?
 
 Craig Bakalian
 

Hey Craig,

A segfault usually means that there's a bug in the program.  If you e-mail 
bug-lilyp...@gnu.org with a minimal example attached, they should be able to 
help you out.  By minimal example, I mean shave your score down until you find 
where the error is coming from.  For example, if you just use the tuba part, 
does it segfault?  Does the segfault occur when other parts are present?  If 
you make the tuba part smaller, does it still occur?  The smaller the example 
is, the more likely the problem will get solved soon.

Cheers,
MS
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[Tim Roberts]

Craig wrote:
 I am working on a large score (choral symphony).  When I include my tuba
 part I am getting a segmentation fault.  What does this mean?

 Wouldn't segmentation have to do with a memory issue?

Well, the term is a leftover from the earliest days of Unix, when it ran
on machines that did have segmented memory.  Today, that error code
means the application used an invalid address, or memory that had
already been released, or a null pointer, something like that.

-- 
Tim Roberts, t...@probo.com
Providenza  Boekelheide, Inc.


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[Craig]

Hi Mike,

I have found the spot in the tuba part that makes the segmentation
fault.  And, yes, if I delete or comment out to make the tuba part
smaller, lily will write a pdf.

Now that I have found the measure, what do I do?  I cannot make this
project smaller.  I would have to send the entire score to the bug
department.  

Any time I have an issue with a large score, that cannot be reduced, it
becomes problematic in the lilypond community.  It is like approaching
an American Bank in the year 2011. 

Craig Bakalian


On Wed, 2011-12-07 at 21:16 +0100, m...@apollinemike.com wrote:
 Le Dec 7, 2011 à 9:12 PM, Craig a écrit :
 
  Hi,
  
  I am working on a large score (choral symphony).  When I include my tuba
  part I am getting a segmentation fault.  What does this mean?
  
  Wouldn't segmentation have to do with a memory issue?
  
  Craig Bakalian
  
 
 Hey Craig,
 
 A segfault usually means that there's a bug in the program.  If you e-mail 
 bug-lilyp...@gnu.org with a minimal example attached, they should be able to 
 help you out.  By minimal example, I mean shave your score down until you 
 find where the error is coming from.  For example, if you just use the tuba 
 part, does it segfault?  Does the segfault occur when other parts are 
 present?  If you make the tuba part smaller, does it still occur?  The 
 smaller the example is, the more likely the problem will get solved soon.
 
 Cheers,
 MS




___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Craig]

Hi Mike,

I have found out what it is.  It has to do with the midi output.  If I
remove the \midi {} from my project there are not segmentation faults.
So this has to be a memory allocation issue inside the lily machine.

Craig Bakalian


On Wed, 2011-12-07 at 21:16 +0100, m...@apollinemike.com wrote:
 Le Dec 7, 2011 à 9:12 PM, Craig a écrit :
 
  Hi,
  
  I am working on a large score (choral symphony).  When I include my tuba
  part I am getting a segmentation fault.  What does this mean?
  
  Wouldn't segmentation have to do with a memory issue?
  
  Craig Bakalian
  
 
 Hey Craig,
 
 A segfault usually means that there's a bug in the program.  If you e-mail 
 bug-lilyp...@gnu.org with a minimal example attached, they should be able to 
 help you out.  By minimal example, I mean shave your score down until you 
 find where the error is coming from.  For example, if you just use the tuba 
 part, does it segfault?  Does the segfault occur when other parts are 
 present?  If you make the tuba part smaller, does it still occur?  The 
 smaller the example is, the more likely the problem will get solved soon.
 
 Cheers,
 MS




___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[m...@apollinemike.com]

Le Dec 7, 2011 à 9:48 PM, Craig a écrit :

 Hi Mike,
 
 I have found out what it is.  It has to do with the midi output.  If I
 remove the \midi {} from my project there are not segmentation faults.
 So this has to be a memory allocation issue inside the lily machine.
 
 Craig Bakalian
 
 
 On Wed, 2011-12-07 at 21:16 +0100, m...@apollinemike.com wrote:
 Le Dec 7, 2011 à 9:12 PM, Craig a écrit :
 
 Hi,
 
 I am working on a large score (choral symphony).  When I include my tuba
 part I am getting a segmentation fault.  What does this mean?
 
 Wouldn't segmentation have to do with a memory issue?
 
 Craig Bakalian
 
 
 Hey Craig,
 
 A segfault usually means that there's a bug in the program.  If you e-mail 
 bug-lilyp...@gnu.org with a minimal example attached, they should be able to 
 help you out.  By minimal example, I mean shave your score down until you 
 find where the error is coming from.  For example, if you just use the tuba 
 part, does it segfault?  Does the segfault occur when other parts are 
 present?  If you make the tuba part smaller, does it still occur?  The 
 smaller the example is, the more likely the problem will get solved soon.
 
 Cheers,
 MS
 
 

Hey Craig,

You can send as detailed a report as possible to the bug list.  They'll likely 
ask you for a smaller example than your score, but explain to them the 
situation.  Hopefully, if someone has time, they'll be able to work with you to 
reduce it down a bit.

Cheers,
MS


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[David Kastrup]

Craig craigbakal...@verizon.net writes:

 Hi Mike,

 I have found the spot in the tuba part that makes the segmentation
 fault.  And, yes, if I delete or comment out to make the tuba part
 smaller, lily will write a pdf.

 Now that I have found the measure, what do I do?  I cannot make this
 project smaller.  I would have to send the entire score to the bug
 department.  

 Any time I have an issue with a large score, that cannot be reduced, it
 becomes problematic in the lilypond community.

It becomes problematic to diagnose the problem, period, and it makes
little sense to blame the Lilypond community for that.  I mean, get
real.  You don't give us any information, not your operating system, not
your Lilypond version, not a bit of the code that causes the problem.

You only tell us that it occurs in a tuba part when it gets large.

So what explanation do you expect?  That Lilypond has a particular
musical taste that lets it throw up when the tuba gets too dominant?

That's about the only possible diagnosis to be given from the
information you deliver, and how likely does that seem?  And if there
was indeed a remote possibility, what would you do with that
information?  Use less tuba?

Really, up to now your report has been little more but venting.  If you
are interested in getting your problem fixed rather than just
complaining, you'll need to do better.  And that is not the fault of the
Lilypond community.  It is just the world being mean.

-- 
David Kastrup


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[Francisco Vila]

2011/12/7 Craig craigbakal...@verizon.net:
 Hi Mike,

 I have found the spot in the tuba part that makes the segmentation
 fault.  And, yes, if I delete or comment out to make the tuba part
 smaller, lily will write a pdf.

 Now that I have found the measure, what do I do?  I cannot make this
 project smaller.

Let me suggest,

  - less notes
  - less staves
  - remove all that is unnecessary

-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Carl Sorensen]

On 12/7/11 1:48 PM, Craig craigbakal...@verizon.net wrote:

Hi Mike,

I have found out what it is.  It has to do with the midi output.  If I
remove the \midi {} from my project there are not segmentation faults.
So this has to be a memory allocation issue inside the lily machine.

So now how could anybody at lilypond troubleshoot this?

There is no way for us to reproduce it, so we can't investigate it.

If we solved the problem, how would we know?

Will my system behave the same as your system?  If not, is it because of a
different operating system, a different memory size, or some other reason?

In order for us to even begin to look at the problem you're having, we
need access to a copy of the problematic file.  Even if the file is
*huge*, at least we can see if the problem occurs on our machines as well
as on yours.

If it works on mine, and fails on yours, then we'll need to compare
software and OS versions.  You haven't given that information, so we can't
do that.

Is this segmentation fault due to the number of staves, or the number of
notes, or something else?  Can you eliminate some of the non-tuba staves,
and see if you still get the problem?  Can you eliminate some of the
previous measures in all of the voices/staves and still get the problem?

Can you do something like

myMusic = \repeat unfold 100 {a4 a a a}

\score {
  \new StaffGroup 
 \new Staff { \myMusic }
 \new Staff { \myMusic }
 \new Staff { \myMusic }
  
  \layout {}
  \midi {}
}

and change the number of staves and the number of measures (via \repeat
unfold) and see if you get the error when the number of staves or the
number of measures gets too big?

Since including the \midi {} block causes the problem, can you explore
whether the problem is too many midi voices?

Can you create two separate \score blocks in your file, one with a \layout
{} and the other with a \midi {}, to see if you still get the segfault?

If we don¹t have access to your code, it's like my telling you My car
won't start.  Any idea what's wrong?

If we *do* have access to your code, and it's very complicated, we are not
very likely to do lots of detailed troubleshooting, but at least we can
test whether the problem is reproducible.  And we *might* be willing to
try one or two things to see if we can make it work.

If you come to the list with very little information and no code
available, and ask us what's wrong, we can't possibly tell you.  We don't
have any idea.

If you come to the list with that same little bit of information, and ask
us for ideas about how you can troubleshoot it, we might have some
suggestions.  If by following those suggestions, you are able to come up
with a simple (but not necessarily *short*) lilypond file that shows the
behavior, then we have a great chance to troubleshoot it.

For an example of how this worked on another bug that showed up only in a
complex score (Valentin's opera), see Issue 1475
http://code.google.com/p/lilypond/issues/detail?id=1475can=1q=unfoldcol
spec=ID%20Type%20Status%20Stars%20Owner%20Patch%20Needs%20Summary

This started out as a crash only on valentin's opera.  Pal then found a
way to use \repeat unfold to make a simple (but long) score that got the
same error.  Then we could start experimenting simply with the code to
find when the bug was introduced.  And it was a straightforward fix for
the person who knew the right code.

So we can't work miracles to fix your problem.  But we can work with you
to help you get to a state where there's a good enough bug report to fix
the problem.

I hope this helps,

Carl


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Craig]

On Wed, 2011-12-07 at 23:52 +, Carl Sorensen wrote:
 If you come to the list with very little information and no code
 available, and ask us what's wrong, we can't possibly tell you.  We
 don't
 have any idea. 

This is the main issue with reporting such bugs with lilypond.  Let's
say I start reducing this problematic file, I reduce the project down to
a single file, and then the bug goes away.  I spend hours trying to
reduce the project to its point of seg fault.  

On the other hand, let say I send you the entire choral symphony in a
folder.  Then you are left going through the syntax looking for the
problem.  

Hence, the issue becomes frozen.  Hence, my correlation to the American
Banking System.  

I can't come to the list with a little information, and I can't come to
the list with too much information.

Craig Bakalian

Let me know if you want the project.  


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Carl Sorensen]

On 12/7/11 5:05 PM, Craig craigbakal...@verizon.net wrote:

On Wed, 2011-12-07 at 23:52 +, Carl Sorensen wrote:
 If you come to the list with very little information and no code
 available, and ask us what's wrong, we can't possibly tell you.  We
 don't
 have any idea. 

This is the main issue with reporting such bugs with lilypond.  Let's
say I start reducing this problematic file, I reduce the project down to
a single file, and then the bug goes away.  I spend hours trying to
reduce the project to its point of seg fault.

On the other hand, let say I send you the entire choral symphony in a
folder.  Then you are left going through the syntax looking for the
problem.  

Do you have any suggestions for how this should be fixed?

When a problem only shows up on complex files, *somebody* has to go
through the complex file to figure out why.  The person with the most
benefit to gain is generally the one who would be expected to spend the
most effort.  But not without guidance, which is what I hope the list can
provide.

Post a link to your source.  I'll run it and see if it segfaults on my
machine.  I might try a simple test or two.

I won't spend hours debugging it, but at least we'll have more information.

Thanks,

Carl


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Graham Percival]

On Wed, Dec 07, 2011 at 07:05:35PM -0500, Craig wrote:
 This is the main issue with reporting such bugs with lilypond.  Let's
 say I start reducing this problematic file, I reduce the project down to
 a single file, and then the bug goes away.

Then you did a bad job of reducing it.  When constructing a tiny
example, you make the smallest possible file that still
demonstrates the problem.

 I spend hours trying to reduce the project to its point of seg
 fault.  

Sure.  That sounds fair to me -- what, you want *me* to spend
hours working on *your* file?

Somebody needs to reduce that segfault to less than 10 lines of
lilypond input.  You're more interested in this issue than I am,
and way more familiar with your symphony than me, so *I'm* not
going to spend time creating a tiny example for you.

 Hence, the issue becomes frozen.  Hence, my correlation to the American
 Banking System.  

WTF does the banking system have to do with this?

 I can't come to the list with a little information, and I can't come to
 the list with too much information.

Right.  You should come to the list with the *right amount* of
information.  And with a tiny bit of practice, it takes about 15
minutes to reach the point of right amount of information.

- do you see the problem if you compile the tuba part by itself?
  great, you've just eliminated 90% of your symphony.
- do you see the problem if you comment out the first half of the
  tuba part?  what about the second half?
  great, you've just cut your remanining problem area in half.
- repeat the process of commenting stuff out until you have 1 or 2
  bars.

- now look at the notes.  You probably have some accientals in
  those 2 bars.  What happens if you remove those accidentals?
  what about slurs, articulations?  try removing all of those,
  one by one, and see if the segfault still remains.

 Let me know if you want the project.  

No, I don't want the project.

- Graham

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[David Kastrup]

Craig craigbakal...@verizon.net writes:

 On Wed, 2011-12-07 at 23:52 +, Carl Sorensen wrote:
 If you come to the list with very little information and no code
 available, and ask us what's wrong, we can't possibly tell you.  We
 don't
 have any idea. 

 This is the main issue with reporting such bugs with lilypond.  Let's
 say I start reducing this problematic file, I reduce the project down to
 a single file, and then the bug goes away.  I spend hours trying to
 reduce the project to its point of seg fault.

That's not the issue with reporting such bugs with Lilypond.  That is
the issue with bugs.  So please stop whining and start working.  Only
the latter will make the bug go away eventually, and it is nobody's
fault here but your own.  Of course, the bug may be somebody's fault.
But we even recently had segfaults for which gcc-4.6 was responsible.
By now we have code in our code base circumventing those.  But since you
don't even bother telling us what version of Lilypond you use on what
platform, it would not be even possible to tell if that might be one
reason.

Stop whining.  Get working.

-- 
David Kastrup


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault (Midi News Flash)

[Martin Tarenskeen]

On Wed, 7 Dec 2011, Craig wrote:


This is the main issue with reporting such bugs with lilypond.  Let's
say I start reducing this problematic file, I reduce the project down to
a single file, and then the bug goes away.  I spend hours trying to
reduce the project to its point of seg fault.


Following this thread I can only come to the following conclusion:
You did not even seriously TRY to produce a smaller example to make it 
easier for the developers to fix your problem. And you don't really know 
if that will take hours of work. I think it will not take that much time.


Try before you cry.

--

MT

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: segmentation fault with 2.13.32

[Dmytro O. Redchuk]

On Sat 04 Sep 2010, 16:22 Paul Scott wrote:
 I have reduced the problem to the presence of a \tempo command:
 
 \version 2.13.32
 
 \score{ 
   
 \new StaffGroup 
   \new Staff  
   {
   %The following line causes a segmentation fault
 \tempo \markup\bold{ Andante } 
 \time 6/8 r4. dis4. 
   }
   
   \new Staff  { R2. } 
 
   
 }
 
 
 I'm copying this to the bug list.
Thank you, added as 1251:
http://code.google.com/p/lilypond/issues/detail?id=1251

-- 
  Dmytro O. Redchuk
  Bug Squad

___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: segmentation fault with 2.13.32

[Paul Scott]

On Fri, Sep 03, 2010 at 11:30:58PM -0700, Paul Scott wrote:
 I'm getting a seg. fault with 2.13.32.  I haven't got the exact
 cause yet but one variable is that I only see the fault when I add a
 second \new Staff in a StaffGroup.  I have two parts based on the
 same score which both succeed when I delete the second staff.  I
 have also attached a verbose output.

I have reduced the problem to the presence of a \tempo command:

\version 2.13.32

\score{ 
  
\new StaffGroup 
  \new Staff  
{
  %The following line causes a segmentation fault
  \tempo \markup\bold{ Andante } 
  \time 6/8 r4. dis4. 
}
  
  \new Staff  { R2. } 

  
}


I'm copying this to the bug list.

Paul




___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: segmentation fault with 2.13.32

[Neil Puttock]

On 5 September 2010 00:22, Paul Scott waterho...@ultrasw.com wrote:

 I have reduced the problem to the presence of a \tempo command:

Thanks for the report, Paul.

The segfault is triggered by the presence of the full-bar rest in the
lower stave: it prevents the setting of a horizontal parent for the
MetronomeMark, which means it never gets attached to a paper column.
Fortunately it's fixed by the patch I've already prepared here:
http://codereview.appspot.com/2042043/

I'll apply the patch once I've sorted out a few additions for
regression testing.

While you're waiting for this to be fixed, you can use the following workaround:

\once \override Score.MetronomeMark #'non-break-align-symbols = #'()

Cheers,
Neil

___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with \revert Stem #'(details beamed-lengths) in \voiceOne

[Neil Puttock]

On 22 April 2010 08:20, Rodolfo Zitellini xhero...@gmail.com wrote:

 Mmh interesting to debug! :)

Heh, I guess we're lucky it triggered a segfault as a side effect,
otherwise it could've gone unnoticed for even longer. :)

Here's the tracker entry:
http://code.google.com/p/lilypond/issues/detail?id=1063

As you can see from the snippet, it actually affects all nested
properties, not just lists.

 If you can give me some pointers on how to start diving in the sources
 I can try to track this down myself.

Well, assuming you're compiling with --disable-optimising and have the
.gdbinit file which allows you to peek at grobs and SCMs
(http://lilypond.org/doc/v2.13/Documentation/contributor/debugging-c_002b_002b-code#Debugging-C_002b_002b-code),
you'll probably want to follow the override/revert path from
property-iterator.cc - nested-property.cc.

If you grep for Override you'll see how the stream event for an
override is broadcast, then caught by the listener in
engraver-group.cc.

Cheers,
Neil


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with \revert Stem #'(details beamed-lengths) in \voiceOne

[Rodolfo Zitellini]

On Wed, Apr 21, 2010 at 11:03 PM, Neil Puttock n.putt...@gmail.com wrote:

 Nasty. :)

 Thanks for reporting this; I'll add it to the bug tracker once I've
 done a few more tests.

 It seems any override for Stem which precedes the nested override
 triggers the crash.  Looks like there's a bug in the nested property
 code which needs the following conditions to show up:

 1. The nested override must be preceded by another override for the same grob;
 2. The nested property being overridden is a list.

 With these two conditions satisfied, the default (i.e., immutable)
 property setting is evicted from the alist, which means once the
 revert happens, the property is missing.  This causes a problem in
 stem.cc since the code which checks 'beamed-lengths returns '() if it
 can't find it in the 'details list (this is what causes the segfault:
 the empty list is equivalent to a null pointer).

Mmh interesting to debug! :)
If you can give me some pointers on how to start diving in the sources
I can try to track this down myself.

Thanks,
Rodolfo


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with \revert Stem #'(details beamed-lengths) in \voiceOne

[Neil Puttock]

On 21 April 2010 15:20, Rodolfo Zitellini xhero...@gmail.com wrote:

 I get a  segmentation fault (bus error). I am using MacOS X 10.5.8
 x86. Tried it with lilypond 2.12.1, 2.13.11, 2.13.18. It seems that
 the problem is generated by the \voiceOne (or \voiveTwo for the
 matter) - removing that it compiles fine. It compiles file even if I
 remove the \revert.
 During some tests I noted that if I force the stems up or down in the
 same snippet I get a SIGSEGV. I can generate a complete backtrace if
 someone is interested.

Nasty. :)

Thanks for reporting this; I'll add it to the bug tracker once I've
done a few more tests.

It seems any override for Stem which precedes the nested override
triggers the crash.  Looks like there's a bug in the nested property
code which needs the following conditions to show up:

1. The nested override must be preceded by another override for the same grob;
2. The nested property being overridden is a list.

With these two conditions satisfied, the default (i.e., immutable)
property setting is evicted from the alist, which means once the
revert happens, the property is missing.  This causes a problem in
stem.cc since the code which checks 'beamed-lengths returns '() if it
can't find it in the 'details list (this is what causes the segfault:
the empty list is equivalent to a null pointer).

 BTW I am not fully sure I'm doing the right thing with beamed-lengths
 (I want to lengthen the beams): the manual just says List of stem
 lengths given beam multiplicity, and I interpreted it as a list of
 the maximum stem lengths beamed together. Am a wrong?

Sounds about right.

Regards,
Neil


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault with \ and \!

[James E. Bailey]

1) I cannot reproduce this error.
2) typically the numbered output you see Interpreting music... [8] 
[16][24][32][33] is lilypond's way of telling you where in the music  
it is. In this case, the error is likely in measure 33. Try compiling  
everything before measure 33 and add notes until you find the problem.


On 16.11.2009, at 18:46, Michael DePriest wrote:

This line (measure 11 of the piece I'm working on) works just fine  
if I compile

it without dynamics:

d fis16 d fis d fis8 cis e16 cis e cis e8 | b d16 e  
a, c e' b

d8 r |

But if I add a diminuendo,

d fis16 d fis d fis8 cis e16\ cis e cis e8\! | b d16  
e a, c e'

b d8 r |

I get this in the terminal:

GNU LilyPond 2.10.33
Processing `file.ly'
Parsing...
Interpreting music... [8][16][24][32][33]
Preprocessing graphical objects...Segmentation fault

I have other crescendos and diminuendos in the piece.



___
bug-lilypond mailing list
bug-lilyp...@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-lilypond


James E. Bailey



___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Arjan Bos]

On 15 dec 2008, at 20:11, Tim Reeves wrote:



Also ran successfully here on Win XP SP2, Lilypond 2.11.65, with  
many warnings. (different from above warnings!)


Here on Mac OSX 10.5.5 with lilypond 2.11.65 I also get a whole host  
of warnings and errors. Most of them are due to font issues:


lilypondoutput.log
Description: Binary data




HTH,
Arjan

___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Simon J Mackenzie]

\version 2.11.65

Segmentfaults with either command

lilypond lorca.ly
lilypond -V lorca.ly

Simon

On Mon, 15 Dec 2008 07:25:37 -0600, Jonathan Kulp
jonlancek...@gmail.com said:
 Nicolas Sceaux wrote:
  
  I'll try to investigate it soon. Currently my problem is that I can't
  reproduce these bus errors or seg faults on my machine.
  
  Jon, I have problems untaring your archive, only cancioncilla.ly gets
  untared, can you check if something is wrong with lorca.tar?
  
  nicolas
  
  
 
 Yes, I had trouble, too, when I tried downloading it from my site.  I 
 think my admin doesn't allow serving of .tar files.  Try this one:
 
 http://dl4.louisiana.edu/lorca.tar
 
 Run lilypond on the lorca.ly file and see what happens. Thanks,
 Nicolas,
 
 Jon
 -- 
 Jonathan Kulp
 http://www.jonathankulp.com
 
 
 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 http://lists.gnu.org/mailman/listinfo/lilypond-user


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Jonathan Kulp]

Simon J Mackenzie wrote:

\version 2.11.65

Segmentfaults with either command

lilypond lorca.ly
lilypond -V lorca.ly

Simon



Thanks for testing, Simon.  I tried the -V flag as soon as I read your 
post but it still segfaulted for me, too.  What you said about the 
\addlyrics command interests me.  I don't have any empty addlyrics 
contexts, but I do have a mixture of \addlyrics and lyricsto styles of 
affixing lyrics to the notes.  I think I'll try making them all 
consistent and see if that changes anything.  It would surprise me if 
this fixed it, though, because each song compiles fine on its own.  I'll 
experiment later.  Unfortunately I have to do actual work now :(


Jon

--
Jonathan Kulp
http://www.jonathankulp.com


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Jonathan Kulp]

Nicolas Sceaux wrote:


I'll try to investigate it soon. Currently my problem is that I can't
reproduce these bus errors or seg faults on my machine.

Jon, I have problems untaring your archive, only cancioncilla.ly gets
untared, can you check if something is wrong with lorca.tar?

nicolas




Yes, I had trouble, too, when I tried downloading it from my site.  I 
think my admin doesn't allow serving of .tar files.  Try this one:


http://dl4.louisiana.edu/lorca.tar

Run lilypond on the lorca.ly file and see what happens. Thanks, Nicolas,

Jon
--
Jonathan Kulp
http://www.jonathankulp.com


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Patrick McCarty]

On Mon, Dec 15, 2008 at 5:25 AM, Jonathan Kulp jonlancek...@gmail.com wrote:

 Yes, I had trouble, too, when I tried downloading it from my site.  I think
 my admin doesn't allow serving of .tar files.  Try this one:

 http://dl4.louisiana.edu/lorca.tar

 Run lilypond on the lorca.ly file and see what happens.

I am able to compile this file successfully.  Processing seemed
normal, and there were only three types of warnings:

warning: type check for `stencil' failed; value `#t' must be of type `unknown'
warning: ignoring too many clashing note columns
warning: too many colliding rests

Patrick


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Jonathan Kulp]

Patrick McCarty wrote:

On Mon, Dec 15, 2008 at 5:25 AM, Jonathan Kulp jonlancek...@gmail.com wrote:

Yes, I had trouble, too, when I tried downloading it from my site.  I think
my admin doesn't allow serving of .tar files.  Try this one:

http://dl4.louisiana.edu/lorca.tar

Run lilypond on the lorca.ly file and see what happens.


I am able to compile this file successfully.  Processing seemed
normal, and there were only three types of warnings:

warning: type check for `stencil' failed; value `#t' must be of type `unknown'
warning: ignoring too many clashing note columns
warning: too many colliding rests

Patrick



Thanks for testing, Patrick.  It's good to hear that someone can run it 
successfully :)


What OS are you running?  So far I've tried it unsuccessfully on Linux 
with kernel 2.6.24-21 and on Mac OSX 10.4.11 on a ppc G4.  I'll try on 
Windows in a few minutes but I have to reboot and install 2.11.65 before 
I can do it.  Best,


Jon

--
Jonathan Kulp
http://www.jonathankulp.com


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Patrick McCarty]

On Mon, Dec 15, 2008 at 7:06 AM, Jonathan Kulp jonlancek...@gmail.com wrote:
 Patrick McCarty wrote:

 On Mon, Dec 15, 2008 at 5:25 AM, Jonathan Kulp jonlancek...@gmail.com
 wrote:

 Yes, I had trouble, too, when I tried downloading it from my site.  I
 think
 my admin doesn't allow serving of .tar files.  Try this one:

 http://dl4.louisiana.edu/lorca.tar

 Run lilypond on the lorca.ly file and see what happens.

 I am able to compile this file successfully.  Processing seemed
 normal, and there were only three types of warnings:

 warning: type check for `stencil' failed; value `#t' must be of type
 `unknown'
 warning: ignoring too many clashing note columns
 warning: too many colliding rests

 Patrick


 Thanks for testing, Patrick.  It's good to hear that someone can run it
 successfully :)

 What OS are you running?  So far I've tried it unsuccessfully on Linux with
 kernel 2.6.24-21 and on Mac OSX 10.4.11 on a ppc G4.  I'll try on Windows in
 a few minutes but I have to reboot and install 2.11.65 before I can do it.

I'm running Arch Linux (x86) with kernel 2.6.27.8.  I can test a GUB
build of 2.11.65 later today too.

Patrick


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Simon J Mackenzie]

I'm using
Kubuntu Linux 8.04 Hardy Heron
KDE 3.5.10
Linux Kernel 2.6.24-22 generic (i386)

Simon


On Mon, 15 Dec 2008 09:06:35 -0600, Jonathan Kulp
jonlancek...@gmail.com said:
 Patrick McCarty wrote:
  On Mon, Dec 15, 2008 at 5:25 AM, Jonathan Kulp jonlancek...@gmail.com 
  wrote:
  Yes, I had trouble, too, when I tried downloading it from my site.  I think
  my admin doesn't allow serving of .tar files.  Try this one:
 
  http://dl4.louisiana.edu/lorca.tar
 
  Run lilypond on the lorca.ly file and see what happens.
  
  I am able to compile this file successfully.  Processing seemed
  normal, and there were only three types of warnings:
  
  warning: type check for `stencil' failed; value `#t' must be of type 
  `unknown'
  warning: ignoring too many clashing note columns
  warning: too many colliding rests
  
  Patrick
  
 
 Thanks for testing, Patrick.  It's good to hear that someone can run it 
 successfully :)
 
 What OS are you running?  So far I've tried it unsuccessfully on Linux 
 with kernel 2.6.24-21 and on Mac OSX 10.4.11 on a ppc G4.  I'll try on 
 Windows in a few minutes but I have to reboot and install 2.11.65 before 
 I can do it.  Best,
 
 Jon
 
 -- 
 Jonathan Kulp
 http://www.jonathankulp.com
 
 
 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 http://lists.gnu.org/mailman/listinfo/lilypond-user


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Jonathan Kulp]

Simon J Mackenzie wrote:

I'm using
Kubuntu Linux 8.04 Hardy Heron
KDE 3.5.10
Linux Kernel 2.6.24-22 generic (i386)

Simon



Run lilypond on the lorca.ly file and see what happens.

I am able to compile this file successfully.  Processing seemed
normal, and there were only three types of warnings:

warning: type check for `stencil' failed; value `#t' must be of type `unknown'
warning: ignoring too many clashing note columns
warning: too many colliding rests

Patrick

Thanks for testing, Patrick.  It's good to hear that someone can run it 
successfully :)


What OS are you running?  So far I've tried it unsuccessfully on Linux 
with kernel 2.6.24-21 and on Mac OSX 10.4.11 on a ppc G4.  I'll try on 
Windows in a few minutes but I have to reboot and install 2.11.65 before 
I can do it.  Best,


Jon



Well, score one for Windows, I guess.  I ran the file successfully on 
Windows Vista just now.  Maybe I should try it with a newer Linux kernel 
since Patrick compiled it successfully with kernel 2.6.27.8.  I'll look 
into it.


Jon
--
Jonathan Kulp
http://www.jonathankulp.com


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[James E . Bailey]

I just realise that I never sent my file publicly, to see if anyone  
else gets the bus error. Here is mine, along with terminal output.

I'm using OSX 10.4 intel

http://pastebin.com/f1a3a6fc0 -- score
http://pastebin.com/fe002d69 -- verbose output


Am 14.12.2008 um 20:29 schrieb Jonathan Kulp:


Running 2.11.65, Ubuntu Linux.

I'm having problems running a file made up of several bookparts.   
It looks like it's compiling fine and then it gets to the part  
where it draws systems and then there's a segmentation fault that  
kills it.  Each of the files runs fine on its own, but gathered  
into a series of bookparts it fails.  From what I've been able to  
find out reading old postings and issues, this happens in large  
files sometimes.  Is there a known way to avoid this?


If anyone would care to try the files out, I've posted a tarball on  
my website with all of the files in it.  Just run lilypond on the  
one called LorcaMaster.ly.


http://www.jonathankulp.com/lorca.tar

Thanks,

Jon

Terminal output (the various warnings are normal on these files and  
don't cause them to fail):


GNU LilyPond 2.11.65
Processing `LorcaMaster.ly'
Parsing...
Interpreting music... [8][16][24][32][40][48][56][64][72][80][88] 
[96][104][112][120][128]

Preprocessing graphical objects...
Interpreting music...
Interpreting music... [8][16][24][32][40][48][56][64]
Preprocessing graphical objects...
Interpreting music...
Interpreting music... [8][16]
warning: type check for `stencil' failed; value `#t' must be of  
type `unknown'

[24][32][40]
Preprocessing graphical objects...
Interpreting music...
Interpreting music... [8][16][24][32][40][48][56][64][64]
Preprocessing graphical objects...
Interpreting music...
Interpreting music... [8][16][24][32]
Preprocessing graphical objects...
cantada-notes.ly:130:8: warning: ignoring too many clashing note  
columns


ais4. gis   | % 26
cantada-notes.ly:128:8: warning: ignoring too many clashing note  
columns


c4. a   | % 24
cantada-notes.ly:126:8: warning: ignoring too many clashing note  
columns


bes4. c | % 22
cantada-notes.ly:123:8: warning: ignoring too many clashing note  
columns


fis4. g,| % 19
cantada-notes.ly:121:8: warning: ignoring too many clashing note  
columns


dis'4. c| % 17
Interpreting music...
Interpreting music... [8][16][24]
Preprocessing graphical objects...
Interpreting music...
MIDI output to `LorcaMaster.midi'...
Finding the ideal number of pages...
Fitting music on 3 or 4 pages...
Drawing systems...
Layout output to `LorcaMaster.ps'...
Calculating line breaks...
Drawing systems...
Calculating line breaks...
Drawing systems...
Calculating line breaks...
Drawing systems...
Calculating line breaks...
Drawing systems...
Calculating line breaks...
Drawing systems...
warning: too many colliding rests
warning: too many colliding rests
warning: too many colliding rests
warning: too many colliding rests
warning: too many colliding rests
/home/jon/bin/lily: line 144: 14291 Segmentation fault   
lilypond --include=$HOME $srcfile




___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user




___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Patrick McCarty]

On Mon, Dec 15, 2008 at 8:31 AM, Jonathan Kulp jonlancek...@gmail.com wrote:

 Well, score one for Windows, I guess.  I ran the file successfully on
 Windows Vista just now.  Maybe I should try it with a newer Linux kernel
 since Patrick compiled it successfully with kernel 2.6.27.8.  I'll look into
 it.

Ah, this is very interesting.  I just download the x86 GUB build for
2.11.65-1, and I get a segfault.  I have never tried using GDB with a
GUB build before, but since the debugging symbols are stripped, I
can't send a (useful) backtrace.

Could this be another issue with one of the dependencies, like we saw
earlier with FontConfig on Windows?  Maybe GCC versions earlier than
4.3 are to blame?

-Patrick


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Tim Reeves]

 
 Message: 1
 Date: Mon, 15 Dec 2008 07:01:36 -0800
 From: Patrick McCarty pnor...@gmail.com
 Subject: Re: Segmentation fault on file with bookparts
 To: Jonathan Kulp jonlancek...@gmail.com
 Cc: lilypond-user lilypond-user@gnu.org
 Message-ID:
d1c3ee7b0812150701o69707465yecb03f6fb9e6a...@mail.gmail.com
 Content-Type: text/plain; charset=UTF-8
 
 On Mon, Dec 15, 2008 at 5:25 AM, Jonathan Kulp jonlancek...@gmail.com 
wrote:
 
  Yes, I had trouble, too, when I tried downloading it from my site.  I 
think
  my admin doesn't allow serving of .tar files.  Try this one:
 
  http://dl4.louisiana.edu/lorca.tar
 
  Run lilypond on the lorca.ly file and see what happens.
 
 I am able to compile this file successfully.  Processing seemed
 normal, and there were only three types of warnings:
 
 warning: type check for `stencil' failed; value `#t' must be of 
type`unknown'
 warning: ignoring too many clashing note columns
 warning: too many colliding rests
 
 Patrick


Also ran successfully here on Win XP SP2, Lilypond 2.11.65, with many 
warnings. (different from above warnings!)

Tim Reeves___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Jonathan Kulp]

Graham Percival wrote:

See
http://www.mail-archive.com/bug-lilyp...@gnu.org/msg13523.html

If you could build with debug symbols and provide a backtrace,
and/or run lilypond with valgrind, it might help debug it.

Cheers,
- Graham




Graham,

I've never used tools like this before and can't find anything in the AU 
manual about using a debugging tool while compiling.  Can you give an 
example of how to do this?  I installed valgrind and did valgrind 
lilypond and it said there were no errors.  I have gdb installed but 
when I ran gdb lilypond filename.ly it said lilypond was not in 
executable format.  I'm probably doing something obviously wrong...


Thanks,

Jon

--
Jonathan Kulp
http://www.jonathankulp.com


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Graham Percival]

See
http://www.mail-archive.com/bug-lilyp...@gnu.org/msg13523.html

If you could build with debug symbols and provide a backtrace,
and/or run lilypond with valgrind, it might help debug it.

Cheers,
- Graham


On Sun, Dec 14, 2008 at 01:29:03PM -0600, Jonathan Kulp wrote:
 Running 2.11.65, Ubuntu Linux.

 I'm having problems running a file made up of several bookparts.  It  
 looks like it's compiling fine and then it gets to the part where it  
 draws systems and then there's a segmentation fault that kills it.  Each  
 of the files runs fine on its own, but gathered into a series of  
 bookparts it fails.  From what I've been able to find out reading old  
 postings and issues, this happens in large files sometimes.  Is there a  
 known way to avoid this?

 If anyone would care to try the files out, I've posted a tarball on my  
 website with all of the files in it.  Just run lilypond on the one  
 called LorcaMaster.ly.

 http://www.jonathankulp.com/lorca.tar

 Thanks,

 Jon

 Terminal output (the various warnings are normal on these files and  
 don't cause them to fail):

 GNU LilyPond 2.11.65
 Processing `LorcaMaster.ly'
 Parsing...
 Interpreting music...  
 [8][16][24][32][40][48][56][64][72][80][88][96][104][112][120][128]
 Preprocessing graphical objects...
 Interpreting music...
 Interpreting music... [8][16][24][32][40][48][56][64]
 Preprocessing graphical objects...
 Interpreting music...
 Interpreting music... [8][16]
 warning: type check for `stencil' failed; value `#t' must be of type  
 `unknown'
 [24][32][40]
 Preprocessing graphical objects...
 Interpreting music...
 Interpreting music... [8][16][24][32][40][48][56][64][64]
 Preprocessing graphical objects...
 Interpreting music...
 Interpreting music... [8][16][24][32]
 Preprocessing graphical objects...
 cantada-notes.ly:130:8: warning: ignoring too many clashing note columns
   
 ais4. gis | % 26
 cantada-notes.ly:128:8: warning: ignoring too many clashing note columns
   
 c4. a | % 24
 cantada-notes.ly:126:8: warning: ignoring too many clashing note columns
   
 bes4. c   | % 22
 cantada-notes.ly:123:8: warning: ignoring too many clashing note columns
   
 fis4. g,  | % 19
 cantada-notes.ly:121:8: warning: ignoring too many clashing note columns
   
 dis'4. c  | % 17
 Interpreting music...
 Interpreting music... [8][16][24]
 Preprocessing graphical objects...
 Interpreting music...
 MIDI output to `LorcaMaster.midi'...
 Finding the ideal number of pages...
 Fitting music on 3 or 4 pages...
 Drawing systems...
 Layout output to `LorcaMaster.ps'...
 Calculating line breaks...
 Drawing systems...
 Calculating line breaks...
 Drawing systems...
 Calculating line breaks...
 Drawing systems...
 Calculating line breaks...
 Drawing systems...
 Calculating line breaks...
 Drawing systems...
 warning: too many colliding rests
 warning: too many colliding rests
 warning: too many colliding rests
 warning: too many colliding rests
 warning: too many colliding rests
 /home/jon/bin/lily: line 144: 14291 Segmentation fault  lilypond  
 --include=$HOME $srcfile



 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 http://lists.gnu.org/mailman/listinfo/lilypond-user


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Nicolas Sceaux]

Le 14 déc. 08 à 20:29, Jonathan Kulp a écrit :


Running 2.11.65, Ubuntu Linux.

I'm having problems running a file made up of several bookparts.  It  
looks like it's compiling fine and then it gets to the part where it  
draws systems and then there's a segmentation fault that kills it.   
Each of the files runs fine on its own, but gathered into a series  
of bookparts it fails.  From what I've been able to find out reading  
old postings and issues, this happens in large files sometimes.  Is  
there a known way to avoid this?


If anyone would care to try the files out, I've posted a tarball on  
my website with all of the files in it.  Just run lilypond on the  
one called LorcaMaster.ly.


http://www.jonathankulp.com/lorca.tar


I'll try to investigate it soon. Currently my problem is that I can't
reproduce these bus errors or seg faults on my machine.

Jon, I have problems untaring your archive, only cancioncilla.ly gets
untared, can you check if something is wrong with lorca.tar?

nicolas



___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault on file with bookparts

[Simon J Mackenzie]

Hi

Just adding to this from my early experience...

\version 2.11.65

Yeah I had similar problems. When I removed the incomplete temporary
\addlyrics {} entries the problem cleared up (Noting in them but %
comment lines.)
The other interesting thing was if I ran lilypond with the -V (verbose)
option set the problem also seemed to clear up.

Simon


On Sun, 14 Dec 2008 22:36:10 +0100, Nicolas Sceaux
nicolas.sce...@free.fr said:
 Le 14 déc. 08 à 20:29, Jonathan Kulp a écrit :
 
  Running 2.11.65, Ubuntu Linux.
 
  I'm having problems running a file made up of several bookparts.  It  
  looks like it's compiling fine and then it gets to the part where it  
  draws systems and then there's a segmentation fault that kills it.   
  Each of the files runs fine on its own, but gathered into a series  
  of bookparts it fails.  From what I've been able to find out reading  
  old postings and issues, this happens in large files sometimes.  Is  
  there a known way to avoid this?
 
  If anyone would care to try the files out, I've posted a tarball on  
  my website with all of the files in it.  Just run lilypond on the  
  one called LorcaMaster.ly.
 
  http://www.jonathankulp.com/lorca.tar
 
 I'll try to investigate it soon. Currently my problem is that I can't
 reproduce these bus errors or seg faults on my machine.
 
 Jon, I have problems untaring your archive, only cancioncilla.ly gets
 untared, can you check if something is wrong with lorca.tar?
 
 nicolas
 
 
 
 ___
 lilypond-user mailing list
 lilypond-user@gnu.org
 http://lists.gnu.org/mailman/listinfo/lilypond-user


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault in Lily 2.7.32-3

[Cameron Horsburgh]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Thomas Scharkowski wrote:
 I get the same error here with 2.7.32.3 windows.
 
 Thomas
 
 
Hi folks,

I've just downloaded and installed the GNU/Linux GUB for 2.7.32-3.
I've convert-ly'ed a slightly older score (that compiled cleanly under
2.7.29) and compiled. It fails with this error:

[EMAIL PROTECTED]:~/Work_Folder/music/projects/IdaAndDot$ lilypond
IdaAndDot.ly GNU LilyPond 2.7.32 Processing `IdaAndDot.ly' Parsing...
error: unknown translator: `Parenthesis_engraver' Segmentation fault


SHould I forward this to lily-bug?

Cam



___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user
 
 
 
 
Good news--it all works with the new 2.7.33. Thanks team!

Cam
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFD68NevoajcVq9gkURAjSHAJ9cvxJPPzYdc10Im54Jqc6WEteIOgCfWd/7
iAVodr5kiR8Jt6t12TBZrOQ=
=XKSV
-END PGP SIGNATURE-



___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault

[Graham Percival]

On 7-Jan-06, at 10:08 AM, Arthur Dyck wrote:


I have tried installing both lilypond-2.6.4.3.x86.package and
lilypond-2.7.19.x86.package on Mandrake 10.1.  Both installs went 
smoothly
once I figured out that I had to change the permissions to make the 
file
executable.  However, when I try to process a file I get a 
segmentation fault
error.  I've looked around but can't seem to find a specific fix.  I 
did

reinstall guile.


Did this happen with a particular file, or just *any* file?  Like,
{ c'4 }
?

If it's anything, try running lilypond -V foo.ly  (where foo.ly is 
that small snippet above) and send the result to bug-lilypond.


Cheers,
- Graham



___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation fault

[Arthur Dyck]

On Saturday 07 January 2006 02:52 pm, Graham Percival wrote:
 On 7-Jan-06, at 10:08 AM, Arthur Dyck wrote:
  I have tried installing both lilypond-2.6.4.3.x86.package and
  lilypond-2.7.19.x86.package on Mandrake 10.1.  Both installs went
  smoothly
  once I figured out that I had to change the permissions to make the
  file
  executable.  However, when I try to process a file I get a
  segmentation fault
  error.  I've looked around but can't seem to find a specific fix.  I
  did
  reinstall guile.

 Did this happen with a particular file, or just *any* file?  Like,
 { c'4 }
 ?

 If it's anything, try running lilypond -V foo.ly  (where foo.ly is
 that small snippet above) and send the result to bug-lilypond.

 Cheers,
 - Graham

No, it failed on the snippet also with the following:
[EMAIL PROTECTED] Desktop]$ lilypond -V foo.ly
GNU LilyPond 2.7.19

LILYPOND_DATADIR=/opt/autopackage/share/lilypond/2.7.19
LILYPONDPREFIX=/opt/autopackage/share/lilypond/2.7.19
LOCALEDIR=/opt/autopackage/share/locale

Effective prefix: /opt/autopackage/share/lilypond/2.7.19
GS_LIB=/home/arthur/.fonts
PATH=/opt/autopackage/bin:/usr/bin:/bin:/usr/bin:/usr/local/bin:/usr/X11R6/bin/:/usr/games:/opt/autopackage/bin:/home/arthur/bin
Segmentation fault


Arthur


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[PGA]

At 00:20 15/03/2005, Han-Wen Nienhuys wrote:
[EMAIL PROTECTED] writes:
 I have just done a fresh install of Fedora 3 with new partitions, not 
over top
 of RH8. The main reason was to install Lilypond as RH8 was too big a 
hassle.

 Installation of the fonts and then Lilypond using rpm went without a 
hitch. I
 then followed the workaround to eliminate the feta20.afm problem, and it
 worked! I was delighted. The last thing I did before going to bed was 
to run
 Lilypond over my bass tablature .ly file.

 First thing in the morning, if I run Lilypond over ANY .ly file, I get:
 GNU LilyPond 2.4.2
 Segmentation fault


can you check if it helps to install
  guile-1.6.7
from
  http://lilypond.org/download/binaries/Fedora-3/
Thanks. In the meantime, I just re-installed Lilypond, expecting it to 
segfault again. It didn't.

On the other hand, I am not satisfied with Fedora 3. The font antialising 
is so bad by the end of the day I can't see clearly. I have actually been 
using Windoze for two days!!! Help!!! Am now trying Slackware and Debian. 
Any suggestions as to what is a good distro to be using?  


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Segmentation Fault

[PGA]

At 00:20 15/03/2005, Han-Wen Nienhuys wrote:
[EMAIL PROTECTED] writes:
 I have just done a fresh install of Fedora 3 with new partitions, not 
over top
 of RH8. The main reason was to install Lilypond as RH8 was too big a 
hassle.

 Installation of the fonts and then Lilypond using rpm went without a 
hitch. I
 then followed the workaround to eliminate the feta20.afm problem, and it
 worked! I was delighted. The last thing I did before going to bed was 
to run
 Lilypond over my bass tablature .ly file.

 First thing in the morning, if I run Lilypond over ANY .ly file, I get:
 GNU LilyPond 2.4.2
 Segmentation fault


can you check if it helps to install
  guile-1.6.7
from
  http://lilypond.org/download/binaries/Fedora-3/
Thanks. In the meantime, I just re-installed Lilypond, expecting it to 
segfault again. It didn't.

On the other hand, I am not satisfied with Fedora 3. The font antialising 
is so bad by the end of the day I can't see clearly. I have actually been 
using Windoze for two days!!! Help!!! Am now trying Slackware and Debian. 
Any suggestions as to what is a good distro to be using?  


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user