[libpng16] Reverted the fix of byte order in png_do_read_filler() with 16-bit

input that was made in version 1.6.17beta01, to preserve legacy
behavior even though it was incorrect.  Instead, added new API
png_set_filter_16() and png_set_add_alpha_16() that set a flag to
make png_do_read_filter() interpret the filler bytes properly.

1 files changed, 1 insertions, 1 deletions

diff --git a/pngread.c b/pngread.c
index 48aae8488..1b1bdbfe5 100644
--- a/pngread.c
+++ b/pngread.c
@@ -3860,7 +3860,7 @@ png_image_read_direct(png_voidp argument)
 #           endif
                where = PNG_FILLER_AFTER;
 
-            png_set_add_alpha(png_ptr, filler, where);
+            png_set_add_alpha_16(png_ptr, filler, where);
          }
 
          /* This stops the (irrelevant) call to swap_alpha below. */