The gcov/lcov are a common toolchain for visualizing code coverage with
the GNU/Toolchain. The documentation and implementation of this
integration was heavily inspired from the blog entry by Mike Melanson:
http://multimedia.cx/eggs/using-lcov-with-ffmpeg/
The "suncc" atomics implementation uses a suncc specific memory
barrier, but also relies on a few atomic functions from atomic.h,
that are not suncc specific but specific to solaris. This made
the current implementation fail on suncc on linux.
This makes standalone compilation of the eatqi decoder
succeed. The dependency comes from the shared mpeg12dec.o file.
Signed-off-by: Martin Storsjö <martin@martin.st>
Error resilience is enabled by the h264 decoder, unless explicitly
disabled. --disable-everything --enable-decoder=h264 will produce
a h264 decoder with error resilience enabled, while
--disable-everything --enable-decoder=h264 --disable-error-resilience
will produce a h264 decoder with error resilience disabled.
Signed-off-by: Martin Storsjö <martin@martin.st>
This allows dropping the mpegvideo dependency from a number of
components.
This also fixes standalone building of the h264 parser, which
was broken in 64e438697.
Signed-off-by: Martin Storsjö <martin@martin.st>
Not all gcc configurations have an implementation of all the atomic
operations, and some gcc configurations have some atomic builtins
implemented but not all.
Thus check for the most essential function, whose presence should
indicate that all others are present as well, since it can be used
to implement all the other ones.
Signed-off-by: Martin Storsjö <martin@martin.st>
These could be used for reference counting, or for keeping track of
decoding progress in references in multithreaded decoders.
Support is provided by gcc/msvc/suncc intrinsics, with a fallback using
pthread mutexes.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This makes the decoder independent of mpegvideo.
This copy of the draw_horiz_band code is simplified compared to
the "generic" mpegvideo one which still has a number of special
cases for different codecs.
Signed-off-by: Martin Storsjö <martin@martin.st>
On Cygwin systems MinGW headers can be present if the corresponding
packages have been installed. Since the MinGW libc is checked for
first, this results in newlib getting misdetected as MinGW libc.
Previously PIC was enabled as a magic workaround for binaries that
built fine, but failed to function at all. This problem no longer
exists, possibly since the introduction of symbol versioning.
These flags are as linker-specific as other LDFLAGS and thus
need to be translated to the correct linker syntax.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Also fixes linking in various configs with only individual parts enabled
because the RTP muxer chaining code depends on the general RTP code,
which is now accounted for.
Move some functions from dsputil. The idea is that videodsp contains
functions that are useful for a large and varied set of video decoders.
Currently, it contains emulated_edge_mc() and prefetch().
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
This fixes the automatic use of $foo_extralibs when feature foo
is enabled indirectly through a _select or _suggest.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This allows compiling optimised functions for features not enabled
in the core build and selecting these at runtime if the system has
the necessary support.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This is consistent with usual ARM nomenclature as well as with the
VFPV3 and NEON symbols which both lack the ARM prefix.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This tests instruction set support in both inline and external asm.
If both fail, the base config option is disabled.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The check_insn function tests an instruction in both inline asm and
standalone assembly, and sets _external/_inline config properties
accordingly.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The check_inline_asm function should check the actual C compiler,
not the one used for assembly files. Usually these are the same,
but they might be different, typically when using a compiler other
than gcc.
The check_as should, as its name suggests, test the type of input
the AS command is used with, i.e. a standalond assembly (.S) file.
Finally, check for gnu assembler using the modified check_as as
this reflects actual usage.
Signed-off-by: Mans Rullgard <mans@mansr.com>
These are properties of the targeted core and do not depend on
specific assembly support in the toolchain which if missing will
render the controlling options here disabled.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Probe for the toolchain default architecture version if no --cpu flag
is present or an unknown cpu is specified. Works with gcc, clang and
armcc.
This allows configuring based on the arch version even if it is not
explicitly specified to configure. It also causes an explicit -march
flag to be added to CFLAGS and ASFLAGS, which in turn lets us do
proper instruction set tests with the assembler.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This will allow arch-specific ways of determining the target
variant when none is specified on the command line.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The def files are used for generating import libraries for
other toolchains (in particular, for generating import libraries
for MSVC for DLLs built with mingw).
The def files produced by mingw/gcc contains ordinals for each
exported function. When MSVC tools generate import libraries
from such a def file, MSVC links to the DLL by the ordinals
instead of linking by name.
Since the def files aren't maintained by hand, the ordinal
numbers are assigned (more or less) randomly and any caller
linking to the libs by ordinals will break as soon as the libraries
export more/fewer functions.
Therefore, strip out the ordinals from the generated def files,
to make users link to the libraries by name.
Callers linking to the DLLs using the gcc provided import library
link by name as they should.
Signed-off-by: Martin Storsjö <martin@martin.st>
When targeting the metro API subset, this function still exists in
the link libraries, but is excluded from the headers. This makes
sure w32threads is automatically disabled when targeting this API
subset (since not all the necessary functions for it are available).
Signed-off-by: Martin Storsjö <martin@martin.st>
This simplifies the condition to avoid hardcoding the systems
where the function exists. This also simplifies support for
newer Windows API subsets where this function doesn't exist,
such as Windows Phone 8 and the "metro" API subset of Windows 8.
Signed-off-by: Martin Storsjö <martin@martin.st>
This makes sure minimal configurations such as
"--disable-everything --enable-avconv" will enable the filters
necessary for running avconv, instead of just keeping avconv
disabled (even if the user specified "--enable-avconv").
Signed-off-by: Martin Storsjö <martin@martin.st>
GCC does not appear to have a -march= string for Westmere, which is a
bit surprising as it has a few more instructions than a Nehalem, but
a few less than a Sandy Bridge.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
This is useful for debugging. Dependencies for these files are not
generated due to limitations in many compilers.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This is (erroneously) required to enable various things in the
newlib headers. As cygwin uses newlib, it is covered by this.
Signed-off-by: Mans Rullgard <mans@mansr.com>
A bug in tail call optimisation in gcc 4.3 and later on parisc causes
numerous tests to fail. Disabling this optimisation gives a working
build. See http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55023
Signed-off-by: Mans Rullgard <mans@mansr.com>
The -mpowerpc64 and -mpowerpc-gfxopt flags are implicitly set by
-mcpu as needed. Passing them explicitly is redundant and can
conflict with user-supplied flags.
Signed-off-by: Mans Rullgard <mans@mansr.com>
All modern assemblers have this capability. Older NASM versions
that lack the capability produce code that crashes at runtime,
so it's better to error out during the build process instead.
CC: libav-stable@libav.org
Some awk versions do not treat the result of unary + on a (numeric)
string as numeric, giving wrong results when used in a boolean context
Using unary - instead is logically equivalent works as expected.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This isn't too useful as a normal codec, but can be used in
voip style applications. The decoder updates the noise
generator parameters when a packet is given to it for decoding,
but if called with an empty packet, it generates more noise
according to the last parameters.
Signed-off-by: Martin Storsjö <martin@martin.st>
This adds support for building on Plan 9 x86-32. The compat/plan9
directory contains these items:
- replacements for the 'head' and 'printf' shell commands
- wrapper for main() to disable FPU exceptions
Larger required changes to the system are described in the
documentation.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This allows targets to include special objects when linking
executables without including them in (shared) libraries.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The absence of this function will only give a less informative
string back from our fallback implementation of getnameinfo().
Signed-off-by: Mans Rullgard <mans@mansr.com>
This requires the makedef perl script by Derek, from the
c89-to-c99 repo. That scripts produces a .def file, listing
the symbols to be exported, based on the gcc version scripts
and the built object files.
To properly load non-function symbols from DLL files, the
data symbol declarations need to have the attribute
__declspec(dllimport) when building the calling code. (On mingw,
the linker can fix this up automatically, which is why it has not
been an issue so far. If this attribute is omitted, linking
actually succeeds, but reads from the table will not produce the
desired results at runtime.)
MSVC seems to manage to link DLLs (and run properly) even if
this attribute is present while building the library itself
(which normally isn't recommended) - other object files in the
same library manage to link to the symbol (with a small warning
at link time, like "warning LNK4049: locally defined symbol
_avpriv_mpa_bitrate_tab imported" - it doesn't seem to be possible
to squelch this warning), and the definition of the tables
themselves produce a warning that can be squelched ("warning C4273:
'avpriv_mpa_bitrate_tab' : inconsistent dll linkage, see previous
definition of 'avpriv_mpa_bitrate_tab').
In this setup, mingw isn't able to link object files that refer to
data symbols with __declspec(dllimport) without those symbols
actually being linked via a DLL (linking avcodec.dll ends up with
errors like "undefined reference to `__imp__avpriv_mpa_freq_tab'").
The dllimport declspec isn't needed at all in mingw, so we simply
choose not to declare it for other compilers than MSVC that requires
it. (If ICL support later requires it, the condition can be extended
later to include both of them.)
This also implies that code that is built to link to a certain
library as a DLL can't link to the same library as a static library.
Therefore, we only allow building either static or shared but not
both at the same time. (That is, static libraries as such can be,
and actually are, built - this is used for linking the test tools to
internal symbols in the libraries - but e.g. libavformat built to
link to libavcodec as a DLL cannot link statically to libavcodec.)
Also, linking to DLLs is slightly different from linking to shared
libraries on other platforms. DLLs use a thing called import
libraries, which is basically a stub library allowing the linker
to know which symbols exist in the DLL and what name the DLL will
have at runtime.
In mingw/gcc, the import library is usually named libfoo.dll.a,
which goes next to a static library named libfoo.a. This allows
gcc to pick the dynamic one, if available, from the normal -lfoo
switches, just as it does for libfoo.a vs libfoo.so on Unix. On
MSVC however, you need to literally specify the name of the import
library instead of the static library.
Signed-off-by: Martin Storsjö <martin@martin.st>
The name mingw32 as target OS is both misleading, and very little
of the target OS specific settings actually match.
Since the target OS default is set based on uname, the default
(which on MSYS is set to mingw) is overridden by --toolchain=msvc.
Signed-off-by: Martin Storsjö <martin@martin.st>
Add a configure function to pull in a compat object and set up
redirects in one operation. This avoids duplicating conditions
across configure and makefiles.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The glibc definitions of INFINITY and NAN do not work with the
tms470 compiler, nor do our usual fallbacks.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Apply flags to work around glibc quirks only if glibc is detected,
and add a few more such flags.
Do not mess with as/ld settings in probe_cc. This is not the
proper place.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Solaris defaults to non-standard utilities (grep, sed, ...) with
proper ones being in /usr/xpg4/bin. Prefixing PATH with this
directory when it exists ensures we get correct variants.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Enable dead store elimination. The last few releases work no worse
with this flag than without. Older versions failed to build some
source files when using this flag.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Now that there is CPU detection in YASM, there will always be one of
inline or external assembly enabled, which obviates the need to fall
back on CPU detection through compiler intrinsics.
v5.0 of the TI ARM compiler changes the version string.
This updates the detection to check for both the old and
the new strings.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This simplifies ensuring proper flags are used when the default
is overridden by the system or on the command line.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Some compilers are extra strict about register usage in main(),
disallowing ebp in inline asm there while allowing it elsewhere.
This change makes the test better reflect actual usage.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The 64 bit cl.exe version 16.00.30319.01 crashes with an internal
compiler error on the current check (and thus deduces it isn't
supported, even if the actual usage in libavuil/x86/cpu.c works
fine), but by assigning the value from the intrinsic to a variable,
or returning it, it works fine.
This error is fixed in cl.exe version 16.00.40219.01.
Signed-off-by: Martin Storsjö <martin@martin.st>
Some msvcrt versions (the static 64 bit libc in MSVC 10) have
a log2 function, but there is no declaration for it in the headers.
Therefore, the normal configure check might find it, but it can fail
during build or at runtime, depending on whether implicit function
declarations are an error or not.
Therefore simply ignore this function on this platform.
Signed-off-by: Martin Storsjö <martin@martin.st>
This disables the following warnings:
C4100: unreferenced formal parameter
(1035 occurrances)
C4214: nonstandard extension used : bit field types other than int
(609 occurances)
C4996: 'avpriv_snprintf': This function or variable may be unsafe.
Consider using _snprintf_s instead. To disable deprecation,
use _CRT_SECURE_NO_WARNINGS. See online help for details.
(351 occurrances)
Signed-off-by: Martin Storsjö <martin@martin.st>
Using ranlib is not required but prevents using the libraries with
msvc.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
This is necessary to avoid spuriously enabling _external or _inline
variants of arch extensions when they should be disabled.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This enables replacing the -l and -L flags used to specify the
just-built libraries when linking the tools and shared libs with
non-standard syntaxes. System library flags are already handled
by the filtering mechanism in configure.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Some tools use different command line syntax for specifying output
when compiling and linking. To accomodate these, separate variables
must be used. No currently supported compilers/linkers are affected
by the change.
Signed-off-by: Mans Rullgard <mans@mansr.com>
There is no point in having the user disable any fastdiv macros.
Besides the condition implementation was broken and only disabled
the C implementation, but no platform specific assembly versions.
There used to be one test for Altivec intrinsics support and a
separate test to determine which of two possible syntaxes to use
for vector literals. Since 2008, we only support the more common
of these so the split test no longer makes sense.
This combines the tests into one and also changes the hard error on
failure to a warning. The test can reasonably fail if no --cpu flag
is provided (or is provided with an unknown CPU) and the compiler
default target does not support Altivec. Aborting in this case is
probably over-reacting.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This fixes build failures on debian/kfreebsd, which has the
sctp.h header, but it is currently broken (a cpp test succeeds,
but a compile test fails), see http://bugs.debian.org/684330 for
details.
Also remove the checked item from HAVE_LIST, since the corresponding
HAVE_* define isn't used by the source code.
Signed-off-by: Martin Storsjö <martin@martin.st>
This adds a hidden config variable for the mpegvideo.o dependency
and selects from the codecs which require it.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Some compilers do not support the Q/R modifiers used to access
the low/high parts of a 64-bit register pair. Check for this
and disable all uses of it when not supported.
Fixes bug #337.
Signed-off-by: Mans Rullgard <mans@mansr.com>
nasm does not support 'CPU foonop' directives. This adds a configure
test for the directive and uses it only if supported.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This installs libraries using the proper names and locations,
generates an import lib for the DLL, and drops no longer needed
linker flags.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This simplifies adding extra flags for individual programs
and also allows more than one object file per program.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Refactoring mmx2/mmxext YASM code with cpuflags will force renames.
So switching to a consistent naming scheme beforehand is sensible.
The name "mmxext" is more official and widespread and also the name
of the CPU flag, as reported e.g. by the Linux kernel.
This allows non-standard replacements for the -c compiler flag.
Some compilers use other flags or no flag at all in place of
the usual one.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This allows using non-standard flags for running the C preprocessor.
The -o flag must be included in this setting due to strange syntax
required by some compilers.
Set the correct flags for tms470.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This adds a full identification probe of CC, AS, LD and HOSTCC,
and sets up correct flags and dependency tracking for each.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The rtmpts protocol uses https implicitly, via the ffrtmphttp
protocol, but the ffrtmphttp protocol is also useable for plain
rtmpt without https, so the dependency needs to be added here instead.
Signed-off-by: Martin Storsjö <martin@martin.st>
This adds two protocols, but one of them is an internal implementation
detail just used as an abstraction layer/generalization in the code. The
RTMPE protocol implementation uses ffrtmpcrypt:// as an alternative to the
tcp:// protocol. This allows moving most of the lower level logic out
from the higher level generic rtmp code.
Signed-off-by: Martin Storsjö <martin@martin.st>
Use compiler-specific replacements for the -o flag in check_cc.
This makes tests work properly with compilers using non-standard
flags.
The tms470 flags are updated to work with this scheme.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The incompatibility with librtmp is already expressed in the ffrtmphttp
protocol dependency declaration, which both RTMPT and RTMPTS depend on.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
This allows filtering of linker flags the same was as already
supported for CFLAGS. The filter must be initialised to 'echo'
early since it is invoked by --extra-ldflags.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This is not used and the current design would not work properly
if mixing tools needing different filters.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This is only used for checking for a certain library, but the code
doesn't need to know whether the function was found.
Signed-off-by: Martin Storsjö <martin@martin.st>
The actual result of the check isn't used anywhere (since we
use this function unconditionally within #ifdef _WIN32), but it
makes sure we explicitly link to shell32 (which is linked in
implictly on mingw).
Signed-off-by: Martin Storsjö <martin@martin.st>
This function is only available in the headers if unconditionally
targeting a version >= XP. It is not strictly necessary, since we
try loading these functions dynamically at runtime on windows in
the fallback, but this makes things a bit faster and more
straightforward.
On mingw32, this function isn't visible by default, while it is
on mingw64 (on both 32 and 64 bit).
Signed-off-by: Martin Storsjö <martin@martin.st>
This allows compiling and running these tests on systems lacking a built-
in version of getopt(), such as MSVC.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This simplifies testing arbitrary code fragments within a function
body.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Currently, --enable-small turns av_always_inline into plain inline,
which is more or less ignored by the compiler. While the intent of
this is probably to reduce code size by avoiding some inlining, it
has more far-reaching effects.
We use av_always_inline in two situations:
1. The body of a function is smaller than the call overhead.
Instances of these are abundant in libavutil, the bswap.h
functions being good examples.
2. The function is a template relying on constant propagation
through inlined calls for sane code generation. These are
often found in motion compensation code.
Both of these types of functions should be inlined even if targeting
small code size.
Although GCC has heuristics for detecting the first of these types,
it is not always reliable, especially when the function uses inline
assembler, which is often the reason for having those functions in
the first place, so making it explicit is generally a good idea.
The size increase from inlining template-type functions is usually
much smaller than it seems due to different branches being mutually
exclusive between the different invocations. The dead branches can,
however, only be removed after inlining and constant propagation have
been performed, which means the initial cost estimate for inlining
these is much higher than is actually the case, resulting in GCC
often making bad choices if left to its own devices.
Furthermore, the GCC inliner limits how much it allows a function to
grow due to automatic inlining of calls, and this appears to not take
call overhead into account. When nested inlining is used, the limit
may be hit before the innermost level is reached. In some cases, this
has prevented inlining of type 1 functions as defined above, resulting
in significant performance loss.
Signed-off-by: Mans Rullgard <mans@mansr.com>
We need to include winsock2.h here, to make sure we have the
real pollfd struct definition, if one exists, before defining the
fallback poll function.
Signed-off-by: Martin Storsjö <martin@martin.st>
It is included for the open/read/write/close functions. On
MSVC, where this header does not exist, the same functions
are provided by io.h, which is already included.
On windows, these functions are provided by io.h. Make sure
io.h is included if it exists, regardless of the setmode
function.
Signed-off-by: Martin Storsjö <martin@martin.st>
Apparently, some build environments require dxva.h even for dxva2,
while others lack this header entirely. Including it conditionally
allows building in both cases.
Signed-off-by: Martin Storsjö <martin@martin.st>
This adds a fallback for cbrtf() using powf(x, 1/3). Since
powf() with a non-integer exponent requires a non-negative
base, special handling of negative inputs is needed.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Apparently this include is needed on some systems for building the
poll fallback (for the timeval struct for select?), but it isn't
available on all systems. Thus only include it if it exists.
Signed-off-by: Martin Storsjö <martin@martin.st>
This function implements a delay using the first available
of the following functions:
- nanosleep()
- usleep()
- Sleep() (Windows)
The conditional #includes in time.c are simplified by including
unistd.h and windows.h whenever they are available rather than
having these lines triggered by specific functions.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The check uses check_func_header, since this function is
conditionally available depending on the targeted MSVCRT
version.
Signed-off-by: Martin Storsjö <martin@martin.st>
This adds two protocols, but one of them is an internal implementation
detail just used as an abstraction layer/generalization in the code. The
RTMPT protocol implementation uses rtmphttp:// as an alternative to the
tcp:// protocol. This allows moving most of the lower level logic out
from the higher level generic rtmp code.
Signed-off-by: Martin Storsjö <martin@martin.st>
This adds the avoption mpegts_flags and converts the existing
resend_headers option into a flag, keeping the old option as
fallback for now.
Signed-off-by: Jindrich Makovicka <makovick@gmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
This removes the dependency on adts.c internals, and simplifies
adding other packetization formats.
Signed-off-by: Jindrich Makovicka <makovick@gmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
FATE_SAMPLES is now used directly by the Makefiles, which induces the test
system to run a test with the value of the environment variable as name.
Renaming the environment variable to LIBAV_SAMPLES avoids this problem.
This disables the warning "external declaration in primary source file"
which is issued when a prototype for an extern function is found in a
.c file rather than a header file. We have such prototypes for asm
functions where a separate header file would be pointless.
Signed-off-by: Mans Rullgard <mans@mansr.com>
The fields in config.fate are colon-separated so any colons
within the fields should be escaped to prevent confusion.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Since the mandatory memcpy in vsrc_buffer has been eliminated, there
shouldn't be any significant reason to build without lavfi anymore.
This will make upcoming support for complex filtergraphs easier to do.
This one is available both in SDL 1.2 and in 1.3 (which is the current
version available e.g. in macports), while 1.3 doesn't contain
SDL_Linked_Version().
The current check for SDL_Linked_Version() (available since SDL 1.2.13)
was added in 8f1b06c8, because including the headers for SDL_Init()
redirects the main() function, requiring the main function signature
to match the one of SDL_main (including argc/argv).
Signed-off-by: Martin Storsjö <martin@martin.st>
An obscure Japanese lossless video codec, originally intended
for use with a remote desktop application.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
The were broken since August of 2010 without anyone noticing until
three weeks ago. Nobody cares about it anymore and hopefully Marvell
will support NEON like in the PXA978 from now on.
This will cause linkers to link against the major lib names, instead of the
base names, allowing multiple major versions of the libraries to co-exist.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
This library does not fit into Libav as a whole and its code is just a
maintenance burden. Furthermore it is now available as an external project,
which completely obviates any reason to keep it around.
URL: http://git.videolan.org/?p=libpostproc.git
This structure is well defined by Microsoft at:
http://msdn.microsoft.com/en-us/library/windows/hardware/ff564012(v=vs.85).aspx
Thus, the wDecodedPictureIndex member is guaranteed to exist.
Also, both the MPEG-2 and VC-1 hwaccel decoders depend on this struct member,
but only the VC-1 decoder was disabled if the check failed.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
This allows easily differentiating between both implementations within the build
system and combining the native implementation for plain RTMP with librtmp for
the RTMPE, RTMPS, RTMPT, RTMPTE protocol variants.
This will be useful to test more aggressively for failures to mark XMM
registers as clobbered in Win64 builds, and prevent regressions thereof.
Based on a patch by Ramiro Polla <ramiro.polla@gmail.com>
Now that 0.8 is out we can reapply this commit. It breaks shared
avserver builds due to avserver using internal libavformat symbols,
which are now hidden, so this commit also disables avserver with
--enable-shared.
Use sched_getaffinity to determine the number of logical CPUs.
Limits the number of threads to 16 since slice threading of H.264
seems to be buggy with more than 16 threads.
When turned on, H264/CAVLC gets ~15% (CVPCMNL1_SVA_C.264) slower for
ultra-high-bitrate files, or ~2.5% (CVFI1_SVA_C.264) for lower-bitrate
files. Other codecs are affected to a lesser extent because they are
less optimized; e.g., VC-1 slows down by less than 1% (all on x86).
The patch generated 3 extra instructions (cmp, cmovae and mov) per
call to get_bits().
The performance penalty on ARM is within the error margin for most
files, up to 4% in extreme cases such as CVPCMNL1_SVA_C.264.
Based on work (for GCI) by Aneesh Dogra <lionaneesh@gmail.com>, and
inspired by patch in Chromium by Chris Evans <cevans@chromium.org>.
