Multistream encode was failing to add the length of the extra length for
self-delimited packets causing corrupted output. Multistream decode was
not properly handling lost frames (and potentially reading out of bounds
as a result).

GET_FINAL_RANGE has been implemented as the xor of the final range of all
the streams in the packet.

test_opus_encode now does the mono narrowband tests using dual-mono
multistream.

Newly introduced check_encoder_option failed unconditionally instead of only when you used the wrong options.

And vice versa

Koen Vos authored 13 years ago and Jean-Marc Valin committed 13 years ago

- Calling the decoder for FEC when there's no LBRR should no longer crash
- There should no longer be an issue with the PLC on frame size switching

__GNUC_PREREQ is generally defined by glibc; ecintrin.h
was checking to see if it was defined before use for
portability to non-glibc systems. However, we now have
a fallback definition in arch.h.

Files are already compiled with -I./include, the relative
path in the #include line isn't necessary.

Our arch.h makes sure it's always defined. Thanks to derf
for the suggestion.

We use the __GNUC_PREREQ macro from celt/arch.h to check
the required minimum gcc version. The previous check would
have failed for gcc 5.n with n from 0 to 5.

Thanks to pcgod for pointing out the problem and derf for
pointing out we had a macro for this.

This array is only used by the malloc failure test, and
defining it unconditionally creates an unused variable
warning on platforms where MALLOC_FAIL isn't set.

This test only works on glibc, but the important thing is support
for the __malloc_hook api, so that's what we should depend on,
especially since not all glibc versions have it.

These pragmas were adding in 4.6, and earlier versions warn
about them being undefined.

The glibc 2.14 NEWS file says __malloc_hook will be removed
in the next release, so future-proof our use by checking
for this symbol at configure time and only compiling the
malloc failure tests if it is present.

This just works around the deprecation in glibc 2.14. I
do worry what will happen if they follow through and
remove it in 2.15.

Fixes a bug in ec_enc_done when no range coded data was coded and adds test cases to test_unit_entropy.

Timothy B. Terriberry authored 13 years ago and Gregory Maxwell committed 13 years ago

The test to make sure that _nbits had been encoded was checking for
 8-_nbits instead of _nbits.
This only affected cases when less than 8 bits had ever been
 encoded by the time the initial bits were patched.

This avoids potential alignment issues with allocating a char
array on the stack.