Re: Segfault when using \bookpart

2008-12-19 Thread Mats Bengtsson



Patrick McCarty wrote:

Can anyone reproduce this with the latest git?
  
Yes, I get a SIGSEGV with todays GIT. I attach the last lines output by 
valgrind, in case it

can provide some hint on the problem:


Drawing systems...
Layout output to `s.ps'...
Calculating line breaks...
Drawing systems... ==5786==
==5786== Invalid read of size 4
==5786==at 0x82EA775: get_column_description(std::vectorGrob*, 
std::allocatorGrob*  const, unsigned, bool) (simple-spacer.cc:345)
==5786==by 0x82EBD75: get_line_forces(std::vectorGrob*, 
std::allocatorGrob*  const, double, double, bool) (simple-spacer.cc:408)
==5786==by 0x81694E9: Constrained_breaking::initialize() 
(constrained-breaking.cc:351)
==5786==by 0x816A4BF: 
Constrained_breaking::Constrained_breaking(Paper_score*, 
std::vectorunsigned, std::allocatorunsigned  const) 
(constrained-breaking.cc:310)
==5786==by 0x8264F92: Page_breaking::find_chunks_and_breaks(bool 
(*)(Grob*)) (page-breaking.cc:374)
==5786==by 0x826694D: Page_breaking::Page_breaking(Paper_book*, bool 
(*)(Grob*)) (page-breaking.cc:106)
==5786==by 0x8259790: 
Optimal_page_breaking::Optimal_page_breaking(Paper_book*) 
(optimal-page-breaking.cc:28)
==5786==by 0x8262064: ly_optimal_breaking(scm_unused_struct*) 
(page-breaking-scheme.cc:31)

==5786==by 0x406AEAE: deval (eval.c:4220)
==5786==by 0x406C5DF: deval (eval.c:3636)
==5786==by 0x4068A66: scm_dapply (eval.c:4998)
==5786==by 0x4068C52: scm_apply (eval.c:4797)
==5786==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==5786==
==5786== Process terminating with default action of signal 11 (SIGSEGV)
==5786==  Access not within mapped region at address 0x0
==5786==at 0x82EA775: get_column_description(std::vectorGrob*, 
std::allocatorGrob*  const, unsigned, bool) (simple-spacer.cc:345)
==5786==by 0x82EBD75: get_line_forces(std::vectorGrob*, 
std::allocatorGrob*  const, double, double, bool) (simple-spacer.cc:408)
==5786==by 0x81694E9: Constrained_breaking::initialize() 
(constrained-breaking.cc:351)
==5786==by 0x816A4BF: 
Constrained_breaking::Constrained_breaking(Paper_score*, 
std::vectorunsigned, std::allocatorunsigned  const) 
(constrained-breaking.cc:310)
==5786==by 0x8264F92: Page_breaking::find_chunks_and_breaks(bool 
(*)(Grob*)) (page-breaking.cc:374)
==5786==by 0x826694D: Page_breaking::Page_breaking(Paper_book*, bool 
(*)(Grob*)) (page-breaking.cc:106)
==5786==by 0x8259790: 
Optimal_page_breaking::Optimal_page_breaking(Paper_book*) 
(optimal-page-breaking.cc:28)
==5786==by 0x8262064: ly_optimal_breaking(scm_unused_struct*) 
(page-breaking-scheme.cc:31)

==5786==by 0x406AEAE: deval (eval.c:4220)
==5786==by 0x406C5DF: deval (eval.c:3636)
==5786==by 0x4068A66: scm_dapply (eval.c:4998)
==5786==by 0x4068C52: scm_apply (eval.c:4797)
==5786==
==5786== ERROR SUMMARY: 88182 errors from 88 contexts (suppressed: 41 
from 2)

==5786== malloc/free: in use at exit: 90,572,019 bytes in 177,183 blocks.
==5786== malloc/free: 2,600,832 allocs, 2,423,649 frees, 493,393,725 
bytes allocated.

==5786== For counts of detected errors, rerun with: -v
==5786== searching for pointers to 177,183 not-freed blocks.
==5786== checked 90,609,632 bytes.
==5786==
==5786== LEAK SUMMARY:
==5786==definitely lost: 103,234 bytes in 3,616 blocks.
==5786==  possibly lost: 139,613 bytes in 237 blocks.
==5786==still reachable: 90,329,172 bytes in 173,330 blocks.
==5786== suppressed: 0 bytes in 0 blocks.
==5786== Use --leak-check=full to see details of leaked memory.


  /Mats


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


Re: Segfault when using \bookpart

2008-12-19 Thread Mats Bengtsson
I just noticed some more lines in the output of valgrind, that might be 
related (or perhaps not,

since it's within some STL code):

...
Finding the ideal number of pages...
Fitting music on 1 or 2 pages...==5786==
==5786== Source and destination overlap in memcpy(0x7D9A5C8, 0x7D9A5C8, 93)
==5786==at 0x4006236: memcpy (mac_replace_strmem.c:394)
==5786==by 0x8168F03: Constrained_breaking::line_details(unsigned, 
unsigned, unsigned) (stl_algobase.h:110)
==5786==by 0x8263ECC: Page_breaking::cache_line_details(unsigned) 
(page-breaking.cc:570)
==5786==by 0x826443B: Page_breaking::min_page_count(unsigned, 
unsigned) (page-breaking.cc:641)
==5786==by 0x8259D84: Optimal_page_breaking::solve() 
(optimal-page-breaking.cc:93)
==5786==by 0x826206C: ly_optimal_breaking(scm_unused_struct*) 
(page-breaking-scheme.cc:32)

==5786==by 0x406AEAE: deval (eval.c:4220)
==5786==by 0x406C5DF: deval (eval.c:3636)
==5786==by 0x4068A66: scm_dapply (eval.c:4998)
==5786==by 0x4068C52: scm_apply (eval.c:4797)
==5786==by 0x406994B: scm_apply_0 (eval.c:4686)
==5786==by 0x827DB78: Paper_book::pages() (paper-book.cc:591)

Drawing systems...
...

   /Mats

Mats Bengtsson wrote:



Patrick McCarty wrote:

Can anyone reproduce this with the latest git?
  
Yes, I get a SIGSEGV with todays GIT. I attach the last lines output 
by valgrind, in case it

can provide some hint on the problem:


Drawing systems...
Layout output to `s.ps'...
Calculating line breaks...
Drawing systems... ==5786==
==5786== Invalid read of size 4
==5786==at 0x82EA775: get_column_description(std::vectorGrob*, 
std::allocatorGrob*  const, unsigned, bool) (simple-spacer.cc:345)
==5786==by 0x82EBD75: get_line_forces(std::vectorGrob*, 
std::allocatorGrob*  const, double, double, bool) 
(simple-spacer.cc:408)
==5786==by 0x81694E9: Constrained_breaking::initialize() 
(constrained-breaking.cc:351)
==5786==by 0x816A4BF: 
Constrained_breaking::Constrained_breaking(Paper_score*, 
std::vectorunsigned, std::allocatorunsigned  const) 
(constrained-breaking.cc:310)
==5786==by 0x8264F92: Page_breaking::find_chunks_and_breaks(bool 
(*)(Grob*)) (page-breaking.cc:374)
==5786==by 0x826694D: Page_breaking::Page_breaking(Paper_book*, 
bool (*)(Grob*)) (page-breaking.cc:106)
==5786==by 0x8259790: 
Optimal_page_breaking::Optimal_page_breaking(Paper_book*) 
(optimal-page-breaking.cc:28)
==5786==by 0x8262064: ly_optimal_breaking(scm_unused_struct*) 
(page-breaking-scheme.cc:31)

==5786==by 0x406AEAE: deval (eval.c:4220)
==5786==by 0x406C5DF: deval (eval.c:3636)
==5786==by 0x4068A66: scm_dapply (eval.c:4998)
==5786==by 0x4068C52: scm_apply (eval.c:4797)
==5786==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==5786==
==5786== Process terminating with default action of signal 11 (SIGSEGV)
==5786==  Access not within mapped region at address 0x0
==5786==at 0x82EA775: get_column_description(std::vectorGrob*, 
std::allocatorGrob*  const, unsigned, bool) (simple-spacer.cc:345)
==5786==by 0x82EBD75: get_line_forces(std::vectorGrob*, 
std::allocatorGrob*  const, double, double, bool) 
(simple-spacer.cc:408)
==5786==by 0x81694E9: Constrained_breaking::initialize() 
(constrained-breaking.cc:351)
==5786==by 0x816A4BF: 
Constrained_breaking::Constrained_breaking(Paper_score*, 
std::vectorunsigned, std::allocatorunsigned  const) 
(constrained-breaking.cc:310)
==5786==by 0x8264F92: Page_breaking::find_chunks_and_breaks(bool 
(*)(Grob*)) (page-breaking.cc:374)
==5786==by 0x826694D: Page_breaking::Page_breaking(Paper_book*, 
bool (*)(Grob*)) (page-breaking.cc:106)
==5786==by 0x8259790: 
Optimal_page_breaking::Optimal_page_breaking(Paper_book*) 
(optimal-page-breaking.cc:28)
==5786==by 0x8262064: ly_optimal_breaking(scm_unused_struct*) 
(page-breaking-scheme.cc:31)

==5786==by 0x406AEAE: deval (eval.c:4220)
==5786==by 0x406C5DF: deval (eval.c:3636)
==5786==by 0x4068A66: scm_dapply (eval.c:4998)
==5786==by 0x4068C52: scm_apply (eval.c:4797)
==5786==
==5786== ERROR SUMMARY: 88182 errors from 88 contexts (suppressed: 41 
from 2)

==5786== malloc/free: in use at exit: 90,572,019 bytes in 177,183 blocks.
==5786== malloc/free: 2,600,832 allocs, 2,423,649 frees, 493,393,725 
bytes allocated.

==5786== For counts of detected errors, rerun with: -v
==5786== searching for pointers to 177,183 not-freed blocks.
==5786== checked 90,609,632 bytes.
==5786==
==5786== LEAK SUMMARY:
==5786==definitely lost: 103,234 bytes in 3,616 blocks.
==5786==  possibly lost: 139,613 bytes in 237 blocks.
==5786==still reachable: 90,329,172 bytes in 173,330 blocks.
==5786== suppressed: 0 bytes in 0 blocks.
==5786== Use --leak-check=full to see details of leaked memory.


  /Mats


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


--

Re: Segfault when using \bookpart

2008-12-19 Thread Nicolas Sceaux

Le 18 déc. 08 à 21:02, Patrick McCarty a écrit :


Hello,

See the comments in the LY file below.  The backtrace ends at the same
place as this one:

http://lists.gnu.org/archive/html/bug-lilypond/2008-12/msg00043.html

i.e. in get_column_description at simple-spacer.cc:345


Ahhh, thanks, I get the bus error too!
I'm investigating.

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