The previous fix (patch 6f7ec60c) had the undesireable effect of slowing
down encoding speed on 16 bit files where the arithmetic overflow was
less likely to happen.
This fix forces the use of a FLAC__uint64 accumulator for 24 bit files
and restores the use of a FLAC_uint32 accumulator for 16 (and less) bit
files.
Unfortunately, I have not been able to prove to myself that this overflow
*cannot* happen with 16 bit files.
For a specific 24 bit WAV file provided by Leigh Dyer
http://lists.xiph.org/pipermail/flac-dev/2013-July/004284.html
encoding with compression level 7 was generating a file a couple of
orders of magintude larger than the original.
Debugging showed that variable abs_residual_partition_sum (a FLAC__uint32)
in function precompute_partition_info_sums_() was suffering from an
arithmetic overflowing on some 24 bit input files although this value
overflowing did not always cause larger output files.
Since the value abs_residual_partition_sum is eventually stored in an
array of FLAC__uint64, it makes sense to make abs_residual_partition_sum
a FLAC__uint64 anyway.
Debugging this problem was made easier by use of the Clang compiler's
-fsanitize=integer option.
The files src/flac/encode.c and src/libFLAC/stream_encoder.c use
functions in libFLAC that are marked as 'unpublished debug routines'.
This patch moves these functions to new file include/share/private.h
and marks them as 'unpublished debug routines'.
The problem was that the function safe_malloc_mul_2op_() was originally
defined as static inline in inclide/share/alloc.h but had to be moved
because GCC was refusing to inline it. Once moved however, static linking
would fail when building the flac executable because the function ended
up beiong linked twice.
- INCLUDES is deprecated, and CPPFLAGS is an user-defined
variable, use the proper AM_CPPFLAGS instead
- Remove FLAC__INLINE definition, providing proper
replacement for MSVC compilers.
- Detect if we have C99 's lround and provide a replacement
for windows...
