GitLab

This will also work on systems where libm is not required.

This is similar to the change we did in libogg some time back,
it means the generated header is less likely to vary unnecessarily
between system architectures.

There's lots more that could and possibly should be done here, but this
is a simple first pass.  We could probably just replace SPEEX_VERSION
with PACKAGE_VERSION everywhere, but that would tangle too many other
things into this set.

Use 'foreign' so the INSTALL file doesn't get stomped every time autogen
is run.

This is a partial revert of 785e61a5.

The _BUILD_SPEEX special case isn't needed in the public headers, and
we don't need '../include' relative paths in the internal files.
Drop a few of the latter altogether where foo.c was including headers
already included by its foo.h.

The OUTSIDE_SPEEX guard isn't needed here anymore either, since that
was only for the resampler which is now in a separate source tree.

The previous version didn't actually update everything needed
(like config.{guess,sub} at the very least).

Signed-off-by: Tristan Matthews <le.businessman@gmail.com>

studio compiler options for SSE/SSE2 are specified or if compiling for x86-64.

There are multiple problems with the elf2flt usage in the configure script:
 - clobbers LDFLAGS
 - is used for all targets so breaks non-FLAT targets
 - is only used for Blackfin FLAT targets
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

Looks like I didn't test the `make dist` target and so broke it by not
listing the bfin.h in one of the headers vars.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

Most asm statements clobber ASTAT bits (shifts, maxes, etc...) but do
declare the register as clobbered.  Same thing with CC in a few places.
Some places make an attempt at clobbering some hardware loop registers,
but it's very incomplete compared with how many asm statements actually
use hardware loops.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

fft.

even though it wasn't an actual bug.

hi

here is just a small patch for speex doc, fixing some typos..
I was reading through the whole manual a couple of times.

could we also update the Internet-Draft in the Manual?

latest version is here:

  http://tools.ietf.org/html/draft-ietf-avt-rtp-speex-05

/alfred

_USE_SSE2 defined because it refuses to cast __m128 to __m128d and vice
versa. While there are intrinsics to do the casts, I thought it would be
simpler to just use an intrinsic that accomplishes the same thing
without all the casting.