KNOWNBUG

Known bugs and suggested enhancements in libpng-1.0.7

1. July 2, 2000 -- BUG -- pnggccrd.c

   Some programs compiled with PNG_USE_PNGGCCRD crash in pnggccrd.c.
   Also, pnggccrd.c will not compile under gcc-2.95-2.

   STATUS: Under investigation.  Experts on MMX assembler code are
   invited to help with the debugging.  Note that pngvcrd.c works fine.

2. July 3, 2000 -- BUG -- MEMORY LEAK

   There is a memory leak in pngpread.c; it doesn't free "key".

   STATUS: Fixed in libpng-1.0.8beta1, by adding.

      png_free(png_ptr, text_ptr);

   after lines 1221 and 1038 in pngpread.c

3. July 3, 2000 -- BUG -- PNG_EXPORT_VAR

   The definition of PNG_EXPORT_VAR, in pngconf.h, should be
   #  define PNG_EXPORT_VAR(type) extern PNG_IMPEXP type

   STATUS: Fixed in libpng-1.0.8beta1.

4. July 3, 2000 -- BUG -- PNG_NO_STDIO

   Several places in pngrutil.c and pngwutil.c, there are printf statements
   inside PNG_NO_STDIO blocks (should be PNG_NO_CONSOLE_IO)

   STATUS: Fixed in libpng-1.0.8beta1.

5. July 3, 2000 -- ENHANCEMENT -- WindowsCE support

   Libpng-1.0.7 does not support WindowsCE.

   STATUS:  Libpng-1.0.8 and libpng-2.0.0 will support WindowsCE.  A
   new typedef, png_file_p, will be added which is normally FILE * but
   will be HANDLE if libpng is being built for WindowsCE.  Other patches
   required for WindowsCE support have been provided and will be applied.

6. March 15, 1998 -- OPTIMIZATION -- Kevin Bracey

   Loops need to be optimized everywhere

   Make them count down instead of up -- Kevin Bracey

   Optimizing compilers don't need this, and making
   the change would be error prone -- Tom Lane, Glenn R-P

   Question whether i-- or --i is better.

   STATUS: Under investigation, postponed until after
   libpng-1.1.0.  About 160 loops will be turned around
   in libpng-2.0.x betaxx, for testing.

7. July 4, 1998 -- ENHANCEMENT -- Glenn R-P

   libpng-1.0.5 and earlier transform colors to gamma=1.0 space for
   merging with background, and then back to the image's gamma.  The
   bit_depth of the intermediate (gamma=1.0) representation is probably
   not sufficient.  In the typical gamma=1/2.2 situation, the linear
   pixels need about 4 more bits than the gamma-encoded ones, to avoid
   loss of precision.  A similar situation exists with the rgb_to_gray
   operation.

   STATUS: under development.

8. September 1999 -- ENHANCEMENT --

   It should be possible to use libpng without floating-point aritmetic.

   STATUS: Under investigation, implementation postponed until after
   libpng-1.0.7.

   Much of this was completed in libpng-1.0.6, but gamma compensation
   is not yet done in fixed-point arithmetic.