Re: [E-devel] Re: E CVS: libs/imlib2 raster

[Adam K. <ki...@if...>]

Hi Tom,

> * enl...@so... (enl...@so...) wrote:
> > Enlightenment CVS committal
> >
> > Author  : raster
> > Project : e17
> > Module  : libs/imlib2
> >
> > Dir     : e17/libs/imlib2/src
> >
> >
> > Modified Files:
> > 	Imlib2.h api.c font.c font.h
> >
> >
> > Log Message:
> >
> >
> > adam's patches... :)
>
> Has anyone actually tested these? They cause every imlib2 program I know
> that uses fonts to dump core when they were fine before (they don't do
> anything wrong - I checked), eg:
>
> #0  0x4015fc20 in TT_Get_Font_Data () from /usr/lib/libttf.so.2
> (gdb) bt
> #0  0x4015fc20 in TT_Get_Font_Data () from /usr/lib/libttf.so.2
> #1  0x4015f077 in TT_Done_Glyph () from /usr/lib/libttf.so.2
> #2  0x40123f3d in __imlib_free_ttf_font_hash (h=0x80d19d0) at font.c:2187
> #3  0x40123d8b in __imlib_nuke_font (fn=0x80d01f0) at font.c:2082
> #4  0x40123bcc in __imlib_flush_font_cache () at font.c:1965
> #5  0x40123cd6 in __imlib_free_font (font=0x80d01f0) at font.c:2024
> #6  0x4011b829 in imlib_free_font () at api.c:1791
>
> That's a font that just got successfully loaded being freed after use.
  Fixed (I believe). Patch applied. Tetsed with imlib2_test and efm with
no problems. Let me know if your applications work too.
  And thanks for pointing this out.

	Adam

--------------------------------------------------------------------------
Adam Kisiel                          | e-mail: ki...@if...
Faculty of Physics                   | homepage: www.if.pw.edu.pl/~kisiel
Warsaw University Of Technology      |
--------------------------------------------------------------------------