instead of directly.
This makes clear that the code assumes the maximum sample rate to be
a multiple of 48000 and also removes the division from the MAX_BLOCKSIZE
macros, which causes an issue with the Solaris assembler where "/" is
a comment marker unless the --divide option is used.
Originally committed as revision 20026 to svn://svn.ffmpeg.org/ffmpeg/trunk
Now that max channels and primitive matrices are properly validated, there is
no need to be paranoid that random data will be overwritten.
As a bonus this makes matrix_coeff 16-byte aligned between matrices.
Originally committed as revision 18651 to svn://svn.ffmpeg.org/ffmpeg/trunk
This way the decoder does not have to depend on the parser being initialized
before.
Patch by Laurent Aimar <fenrir at via dot ecp dot fr>.
Originally committed as revision 15986 to svn://svn.ffmpeg.org/ffmpeg/trunk
Consistently apply this rule: the guard name is obtained from the
filename by stripping the leading "lib", converting '/' and '.' to
'_' and uppercasing the resulting name. Guard names in the root
directory have to be prefixed by "FFMPEG_".
Originally committed as revision 15120 to svn://svn.ffmpeg.org/ffmpeg/trunk
