search for: _use_sse

On Sat, Nov 1, 2008 at 2:40 PM, Aymeric Moizard <jack at atosc.org> wrote: > The vcproj file provided for 2005 in 1.2rc1 does not include > _USE_SSE when compiling the project for "Release_SSE". This > define is only used when compiling under "Release_Dynamic_SSE". > > Is this just because the vcproj file is wrong or is it because > this #define must not be used in that mode? Just want to make > sure I got the...

Does anyone have a tip for compiling for SSE extensions under Visual Studio and Windows? If I add _USE_SSE to the preprocessor directives, I get a successful compile, but speex_encode-int blows up when it's working (if it tries to compress anything other than silence). If I remove the _USE_SSE from the preprocessor directives everything works perfectly. I'm using VC6 under windows. Than...

Hi, The attached patch changes the configure.ac logic as follows: It tests if the current compiler/cflags combination supports SSE, and if so enables _USE_SSE. This can be overridden by the '--enable-sse' and '--disable-sse' flags. Using '--enable-sse' will additionally add '-O3 -msse' to CFLAGS, just as it used to. I've tested this using the following combinations: x86_64: ./configure: Detected and used. x86_64: ./...

Dear all, I have just implemented a gstreamer plugin for the Speex AEC, but now I would like to extend it to a multichannel echo cancellation I have been searching documentation about that but I could not find it Where it is? If it doesn't exist? Could you add a short example showing how do you think it should work? or It's "just" necessary to: init the multichannel echo

...rd16_t *resp, float *resp2, spx_word32_t *E, int shape_cb_size, int subvect_size, char *stack) But then calls it with resp2 being defined as type spx_word16_t * in line 185: compute_weighted_codebook(shape_cb, r, resp, resp2, E, shape_cb_size, subvect_size, stack); Defined on line 103: #ifdef _USE_SSE __m128 *resp2; __m128 *E; #else spx_word16_t *resp2; spx_word32_t *E; #endif I have downloaded the speex version 1.1.6 and have made no changes; I used a command line option to define FIXED_POINT. *** Is there a fix for this discrepancy? *** Change to float (a site search of "r...

> Am I right with the assumption, that currently you have to enable > processor specific optimizations with compile/configure options? > > How difficult would it be to add support for runtime CPU detection? > Is this a feature you might consider adding? Pretty complicated because of some annoying decisions taken by the gcc team. The problem is that gcc won't let you use

...ich are superfluous (i.e. test/demo programs with their own > main()). A minimal Makefile would help to document that, and enable > people create projects, with minimum effort, for all sorts of > environments nobody ever heard of. You sure missed with this approach that you have to define _USE_SSE to have SSE optimized build, that you could use USE_ALLOCA if you do not use _USE_SSE, etc, etc. You have to dig around for some amount of time to find all such things. Just to indicate my position: 1) MinGW build support is good. It should be there. 2) I _will_ maintain VS6, VS2003 and VS2005 pro...

...2 bytes. If you use larger frames, this will scale accordingly. Unless anyone can spot any glaring mistakes I've made, the plan is to fix the double versions, correct the int->float (and vice versa) conversions and make sure the magic bytes work. Then it's time for some unrolling and _USE_SSE improvements ;) -------------- next part -------------- A non-text attachment was scrubbed... Name: resample-both-test.patch Type: text/x-diff Size: 9068 bytes Desc: not available Url : http://lists.xiph.org/pipermail/speex-dev/attachments/20080404/eedc0f5a/attachment.patch

Skipped content of type multipart/alternative-------------- next part -------------- A non-text attachment was scrubbed... Name: preproc_patch_dth_10_26_05.patch Type: application/octet-stream Size: 8774 bytes Desc: not available Url : http://lists.xiph.org/pipermail/speex-dev/attachments/20051026/c27a3ed6/preproc_patch_dth_10_26_05.obj

...ool so I built libspeex.dll and libspeexdsp.dll with the > DUMP_ECHO_CANCEL_DATA flag, but I'm not getting any output files at runtime. > Any ideas? > Build information: > Speex version: 1.2rc1 > Target Win32 x86 > MS VC++2008 EE > Configuration: Release, Full optimization, _USE_SSE > DUMP_ECHO_CANCEL_DATA > > -- > Greger Burman > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.xiph.org/pipermail/speex-dev/attachments/20090811/9a5dade1/attachment.htm

> Is it a problem if all the files are compiled with -march=pentium3 > ? The patch that we sent in already detects in the configure.in script > which system you are on and sets a define correctly, i.e. _USE_SSE. Well, if what you want is auto-detection, turning on -march=pentium3 means that the code will crash on anything lower than a pentium3. Not really useful. Of course, you can make it auto-detect at compile-time, but that means you can't use the same binary on SSE and non-SSE machines....

What's wrong with a plain old .bat file, or even an NMAKE .mak file? Ship two files, debug.bat and release.bat, and call it good. It is best to leave project-file creation up to individual users, in my opinion. MS changes their IDEs and project-file formats more often than most people change their underwear. The odds that any given open-source project will actually compile without any

...ROG_LIBTOOL AC_C_BIGENDIAN @@ -52,8 +52,18 @@ AC_CHECK_LIB(gnugetopt, getopt_long) AC_DEFINE_UNQUOTED(VERSION, "${VERSION}") +AC_DEFINE(GCC_COMPILER) +echo -n "Checking for Platform ASM Speedups..." +case "$target" in + i?86*) + CFLAGS="$CFLAGS -D_USE_SSE" + echo "Found SSE and SSE2" + ;; + *) + echo "None Found" + ;; +esac -AC_ARG_ENABLE(sse, [ --enable-sse enable SSE support], [if test "$enableval" = yes; then AC_DEFINE(_USE_SSE) fi]) dnl Output the makefiles and v...

...ERY_SMALLs where they seem to fit well. Although I got CPU usage as low as 10%, I seriously lack knowledge of how things work inside speex. So just changing some code is not the best idea for me. My second attempt was to follow Jeff's suggestion to modify the MXCSR register and recompile with _USE_SSE. This works very well (CPU < 3%). However I would still prefer the first method (VERY_SMALL) because not all CPUs, my app is going to run on, have the SSE instruction set available. Hopefully someone with more insight is able to fix this some day :-) Thanks Mark -----Urspr?ngliche Nachricht-...

The reasons I have posted these questions are: 1) To find out if Speex can take advantage of SIMD extensions. 2) To maybe learn from someone with previous experience in optimizing Speex for moderns x86 architectures before I set off trying all kinds of things on my own. See answers inline: 2009/6/15 Tom Grandgent <tom at grandgent.com> > Why haven't you tried using release build

...f * `ENABLE_PACKAGE_CONFIG` - on, [CMake specific](https://cmake.org/cmake/help/latest/manual/cmake-packages.7.html#package-configuration-file) * Configured conditionals (`config.h`): * `WORDS_BIGENDIAN` * `FLOATING_POINT` * `FIXED_POINT` * `FIXED_POINT_DEBUG` * `DISABLE_FLOAT_API` * `_USE_SSE` * `ARM4_ASM` * `ARM5E_ASM` * `BFIN_ASM` * `TI_C55X` * `DISABLE_VBR` * `USE_SPEEXDSP` * `VORBIS_PSYCHO` * `HAVE_GETOPT_H` * `HAVE_GETOPT_LONG` * `VAR_ARRAYS` * `USE_ALLOCA` * `HAVE_ALLOCA_H` * `USE_SMALLFT` * `USE_KISS_FFT` * `USE_GPL_FFTW3` * `EXPORT` * Installation...

> Because of that, I believe that it's possible to do a much better job by > using a bit of inline assembly. Note that two of these sections of code > have already been coded with SSE assembly. By whom? Which functions are these? Will this eventually be part of an official Speex release? <p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage:

Jean-Marc, Is it a problem if all the files are compiled with -march=pentium3 ? The patch that we sent in already detects in the configure.in script which system you are on and sets a define correctly, i.e. _USE_SSE. Aron Rosenberg SightSpeed At 11:05 AM 1/21/2004, you wrote: > > Am I right with the assumption, that currently you have to enable > > processor specific optimizations with compile/configure options? > > > > How difficult would it be to add support for runtime CPU detectio...

...need to add AC_PROG_CXX into configure.in). That way you add both #ifdef inside those files and you have runtime selection of the function to use. ex (configure.in): echo -n "Checking for Platform ASM Speedups..." case "$target" in i?86*) CFLAGS="$CFLAGS -D_USE_SSE" CXXFLAGS="$CXXFLAGS -march=pentium3 -D_USE_SSE" echo "Found x86 - Adding SSE." ;; powerpc-apple-darwin*) CFLAGS="-D_USE_ALTIVEC" CXXFLAGS="$CXXFLAGS -faltivec -O3 -fPIC -D_USE_ALTIVEC" echo &quo...

...I should try the echo_diagnostic tool so I built libspeex.dll and libspeexdsp.dll with the DUMP_ECHO_CANCEL_DATA flag, but I'm not getting any output files at runtime. Any ideas? Build information: Speex version: 1.2rc1 Target Win32 x86 MS VC++2008 EE Configuration: Release, Full optimization, _USE_SSE DUMP_ECHO_CANCEL_DATA -- Greger Burman -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.xiph.org/pipermail/speex-dev/attachments/20090811/f29c75ae/attachment.htm