This avoids the danger that get_bits.h might get indirectly #included before
BITSTREAM_READER_LE is defined.

Also sort headers into canonical order where appropriate.

Express bitfields more simply.

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

Convert doxygen to multiline and express bitfields more simply.

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

The valid returned values are always at most 11bit.
Remove the previous check that assumed larger values plausible and
use a signed integer to check get_vlc2 return values.

CC: libav-stable@libav.org

Makes the code a little simpler.

Those decoders have been switched to float output and so do not use
fmtconvert anymore.

The use of ff_inverse speeds up slightly arches + compilers that
do not provide a division faster than the whole machinery, such
as ppc32 + gcc4.7, but has operational limits.

Drop the always-enable assert and provide a fallback.

No need to use fixed-size integers.

This is consistent with the rest of libavcodec.

The specification does not prevent an encoder to write the amplitude 0
as 0 amplitude_bits.

Our get_bits() implementation might not support a zero sized read
properly, thus the additional branch.

Rate and order must not be 0 even if the specification does not say that
explicitly.

Use the commonly used "if (!var)" instead of "if (var == 0)".

The value is used to calculate output LSP curve and a division by zero
and out of array accesses would occur.

CVE-2013-0894

CC: libav-stable@libav.org

Reported-by: Dale Curtis <dalecurtis@chromium.org>
Found-by: inferno@chromium.org
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

This saves one instruction in the x86-64 assembly.

Conveniently (together with Justin's earlier patches), this makes
our vorbis decoder entirely independent of dsputil.

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

It will be useful in the upcoming transition to refcounted AVFrames.

In both usages of FASTDIV the denominator might be 1.

Using a branch could make the function slower than using a normal
division.

Both denominator and numerator can be multiplied by 2 safely and
using shifts is faster than using a branch.

Signed-off-by: Mans Rullgard <mans@mansr.com>

2x speedup of surround decoding on Cortex-A9.

Signed-off-by: Mans Rullgard <mans@mansr.com>

Move vector_fmul() from DSPContext to AVFloatDSPContext.

Duplicate values in this vector are explicitly banned by the Vorbis I spec
and cause divide-by-zero crashes later on.

Also break some long lines, remove codec function placeholder comments
and add spaces in sample/pixel format lists.

Signed-off-by: Martin Storsjö <martin@martin.st>

This fixes some invalid memory access caused later in the function
by res_chan[] not being set for all channels. This happens when a
channel doesn't appear a submap. This change simply returns a
decoder error when this situation is detected.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

This also allows for removing some of the Vorbis-related hacks.

When reading sequentially, we are using the actual flag from the previous
frame, but when seeking we do not know what the previous window flag was, so
we need to read it from the bitstream.

clear MDCT overlap buffer and reset previous window mode when seeking

Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Martin Storsjö <martin@martin.st>

Fixes Bug: #190
Chromium Bug: #100543
Related to CVE-2011-3893

Signed-off-by: Reinhard Tartler <siretart@tauware.de>

Fixes Bug: #191
Chromium Bug: #101458
CVE-2011-3895

Signed-off-by: Reinhard Tartler <siretart@tauware.de>