[cairo] Access violation in _cairo_surface_acquire_dest_image

[Jacob (=Jouk) Jansen]

joukj at hrem.nano.tudelft.nl wrote on 1-NOV-2007 12:47:59.79
>chris at chris-wilson.co.uk wrote on 28-OCT-2007 14:09:12.15
>
>>Jacob (=Jouk) Jansen (joukj at hrem.nano.tudelft.nl) said: 
>>> When trying to run the cairo tests&libs, extracted from the git-repositry
>>> recently I get on my OpenVMS machine
>>> 
>>> %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=000000000000
>>> 001C, PC=000000000012A970, PS=0000001B
>>> %TRACE-F-TRACEBACK, symbolic stack dump follows
>>>   image    module    routine             line      rel PC           abs PC
>>>  libcairo  cairo-surface  _cairo_surface_acquire_dest_image
>>>                                         24365 0000000000001370 000000000012A970
>>
>>Hmm, this implies that surface->backend is NULL within
>>_cairo_surface_acquire_dest_image - 0x1c is the offset of
>>backend->acquire_dest_image. (I was thinking this was the patch
>>I made to remove the cairo_image_surface_backend from the error
>>surfaces, which reduces cairo's .data down to below 1 page, but it seems
>>I've yet to push that one...) 
>>
>>However, as this is impossible (!), a little more detective work is in
>>order. Could you try to git bisect the broken commit? Thanks.
>The first version not to work seems to be :
>
>aaee-jj ) git bisect good
>b5c36010edc92ca129a0498cf04592f8f692f800 is first bad commit
>commit b5c36010edc92ca129a0498cf04592f8f692f800
>Author: Adrian Johnson <ajohnson at redneon.com>
>Date:   Sat Oct 13 21:14:28 2007 +0930
>
>    Add cairo_ps_surface_restrict_to_level() API
>
>:040000 040000 f1e0b1009cc0160344b41985ea16d1405b66f6d7 3ab202b5599754134370bbfafacb9c41d6bfe265 M      doc
>:040000 040000 16e61d640dd9720f618bbf5bcd3a749e092e6c47 8387378105aaa4def767b99ef43836c9daeae309 M      src
>
>However, it gives an error on a different location. So I will try to redo
>the bisect to obatain the first version to introduce the error in the
>original E-mail. (it may take a while, since OpenVMS does not has git, so I
>have to do git on one machine, copy the data to the VMS system and then
>start compiling from scratch.)

The patch introducing exactly the same error as reported was also include 13
November by Adrian Johnson:

Author: Adrian Johnson <ajohnson at redneon.com>  2007-10-13 16:34:21
Committer: Adrian Johnson <ajohnson at redneon.com>  2007-10-13 16:41:57
Parent: e347a7a7c394fc2638faa2ff52e4b96545ac1ee3 (PS: Add support for images with bilevel alpha)
Branches: bisect, master, origin
Follows: 1.4.10
Precedes: 

    Make PostScript output text selectable
    
    The glyph names used in the Type42 and Type1 fallback fonts are now of
    the form "/uniXXXX" where XXXX is the unicode character for each
    glyph. When converted to pdf (eg using ps2pdf), pdf viewers are now
    able to correctly extract the text.



How to proceed from here?

                   Jouk
		   


Bush : All votes are equal but some votes are more equal than others.

>------------------------------------------------------------------------------<

  Jouk Jansen
		 
  joukj at hrem.nano.tudelft.nl

  Technische Universiteit Delft        tttttttttt  uu     uu  ddddddd
  Kavli Institute of Nanoscience       tttttttttt  uu     uu  dd    dd
  Nationaal centrum voor HREM              tt      uu     uu  dd     dd
  Lorentzweg 1                             tt      uu     uu  dd     dd
  2628 CJ Delft                            tt      uu     uu  dd     dd
  Nederland                                tt      uu     uu  dd    dd
  tel. 31-15-2782272                       tt       uuuuuuu   ddddddd

>------------------------------------------------------------------------------<