Re: [Lilypond-auto] Issue 4360 in lilypond: Patch: Reorganize smob initialization to make it more reliable

[lilypond]

Updates:
        Status: Invalid

Comment #9 on issue 4360 by address@hidden: Patch: Reorganize smob  
initialization to make it more reliable
https://code.google.com/p/lilypond/issues/detail?id=4360

After additional debugging it turned out that my diagnosis had been  
incorrect.  I was thrown off by getting a memory protection error in  
guile-init.cc in
  for (vsize i = scm_init_funcs_->size (); i--;)
    {
      (scm_init_funcs_->at (i)) ();
    }
so I suspected that some initializations were registered after code  
rearrangements into scm_init_funcs_ only after the loop had started.   
Reversing the direction of the loop to forward such that  
scm_init_funcs_->size () was checked after each iteration fixed the crash.

However, the fixing of the crash actually seems to have been caused by  
getting a different _order_ of initializations rather than having any  
initializations omitted, the real problem being an unprotected symbol.  So  
it seems like this did not actually solve a real problem.  Even though it's  
unclear why it should trigger more of the undead objects than previously.

--
You received this message because this project is configured to send all  
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings