similar to: More BeOS woe

Hi LLVM hackers, when switching one of our projects to LLVM we discovered some strange behavior. We tracked the issue down to the following test case: int main() { double v = 0.84711; int r; __asm__("fistl %0": "=m"(r) : "t"(v)); return r; } The (relevant) clang generated asm looks like this: fldt 0.84711 ## InlineAsm Start fistl -4(%ebp) ## InlineAsm End

I'm not sure how this would affect other OS's (I have no real idea how autoconf, automake etc work) but this is a go at fixing the BeOS undefined link problem. Hope the diffs are okay P.S. is there a simple test program for the FP bug in the math inline header I'd like to see if BeOS suffers from the problem or not. Dave --- configure.in Sun Nov 19 14:42:16 2000 +++

I merged branch_beta3 (modules ogg and vorbis) onto the mainline. Monty --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-dev-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be

My ogg library is located in /home/warren/local. So when I configure and compile the vorbis module on my linux box using configure --with-ogg-prefix=/home/warren/local make I get the following error during the make: gcc -DPACKAGE=\"libvorbis\" -DVERSION=\"1.0.0\" -DHAVE_ALLOCA_H=1 -DHAVE_ALLOCA=1 -I. -I. -I../include -O20 -ffast-math -mno-ieee-fp -D_REENTRANT

This is also pretty simple. Just a small patch to configure.in to set the right compiler flags and changes to os.h to (a) not have deeply nested ifdefs, making the file needlessly hard to read, (b) break up the sections by functionality and (c) don't assume that every non-Win32 has the floating versions of math library functions. -tim Index: configure.in

Hi all, I was just trying to build myself a RC3 set of Ogg/Vorbis libs and tools, but the build of libvorbis bombs out on me while making all in examples. I've ./configure'd with no options, which gave no warnings or errors. After that a normal make gives the below attached output. Can anyone tell me what I've done wrong ? I've also tried removing the examples/ directory. After

This fixes a build problem with libvorbisfile. Aaron Index: lib/Makefile.am =================================================================== --- lib/Makefile.am (revision 9064) +++ lib/Makefile.am (working copy) @@ -19,7 +19,7 @@ libvorbis_la_LIBADD = @OGG_LIBS@ @VORBIS_LIBS@ libvorbisfile_la_SOURCES = vorbisfile.c -libvorbisfile_la_LDFLAGS = -no-undefined -version-info

After fixing a 'bug' in libtool, I've finally got ogg compiling and producing a shared library. Now I'm trying to build vorbis but I'm falling over at the final step. While trying to produce libvorbisfile. The make file is trying to compile with gcc -nostart vorbisfile.o -L/boot/home/config/lib -logg -Wl,-soname - Wl,libvorbisfile.so -o .libs/libvorbisfile.so with

Just checked out the vorbis CVS branch a couple minutes ago and got an internal compiler error with gcc 2.95.4 on envelope.c gcc -DPACKAGE=\"libvorbis\" -DVERSION=\"1.0rc2\" -DHAVE_DLFCN_H=1 -DHAVE_ALLOCA_H=1 -DHAVE_ALLOCA=1 -DHAVE_SQRTF=1 -DHAVE_LOGF=1 -DHAVE_EXPF=1 -DHAVE_ACOSF=1 -DHAVE_ATANF=1 -DHAVE_FREXPF=1 -DHAVE_RINTF=1 -I. -I. -I../include -O20 -ffast-math

current Vorbis cvs gives: gcc -DPACKAGE=\"libvorbis\" -DVERSION=\"1.0\" -DHAVE_DLFCN_H=1 -DHAVE_ALLOCA_H=1 -DHAVE_ALLOCA=1 -I. -I. -I../include -O20 --fast-math -mno-ieee-fp -D_REENTRANT -fsigned-char -O2 -march=i686 -mcpu=i686 -DUSE_MEMORY_H -Wp,-MD,.deps/envelope.cpp -c envelope.c -fPIC -DPIC -o .libs/envelope.lo gcc: Internal compiler error: program cc1 got fatal

Wow, I've never seen an FSF gcc crap out like this on a non-overheating system. When lib/envelope.c is compiled with -march=i686, gcc gets a signal 11, but not with -march=i586. I'm not sure whether to blame gcc or libvorbis. There are apparently three other files in libvorbis where this same behavior is seen. jeffrey@diddl:/usr/src/build/libvorbis-1.0/lib$ gcc

HI, how do I get mods applied to Vorbis, I'm not a cvs guru and I don't know much about configure scripts. BeOS has this little quirk of being 99% POSIX compliant so these mods are required for vorbis to compile. To define int64_t in codec.h #ifdef __BEOS__ #include <inttypes.h> #endif run.c in vq needs 'changing' around line 141 becuase S_IFSOCK isn't

I'm trying to build libvorbis-1.1.0 on Cygwin and I discovered a missing dependency that may not affect other platforms. I found that I had to add -logg to the libvorbisfile.la rule in the Makefile. Working its way back upstream through configure and autogen, it looks like the following line in trunk/vorbis/lib/Makefile.am libvorbisfile_la_LIBADD = libvorbis.la just needs to be changed

The latest CVS update of libvorbis fails to compile on SuSE Linux 6.4. The error is below. Also, what happened to the xmms plugin directory in cvs? It complains that it is missing when I do a 'cvs update'. gcc -DPACKAGE=\"libvorbis\" -DVERSION=\"1.0beta3\" -DHAVE_ALLOCA_H=1 -DHAVE_ ALLOCA=1 -I. -I. -I../include -O20 -ffast-math -mno-ieee-fp -D_REENTRANT -fs

This is a linux from scratch based on the 2.4.13 kernel. I'm using a GCC 2.95.3, autoconf 2.52, and automake 1.5. Libvorbis compilation fails with a Signal 11 error as shown below. I had the same results with 2.4.12, 2.4.10, and 2.4.9. I've compiled hundreds of other packages without getting this error, and I've even tried changing the memory. This is my configure line, I used

This is a linux from scratch based on the 2.4.13 kernel. I'm using a GCC 2.95.3, autoconf 2.52, and automake 1.5. Libvorbis compilation fails with a Signal 11 error as shown below. I had the same results with 2.4.12, 2.4.10, and 2.4.9. I've compiled hundreds of other packages without getting this error, and I've even tried changing the memory. This is my configure line, I used

On Fri, Jul 20, 2001 at 03:13:31PM -0700, Josh Coalson wrote: > --- Matt Zimmerman <mdz@debian.org> wrote: > > Basically, you are allowed to specify object code, whether in shared or > > static libraries, normal object files, libtool libraries, and libtool > > object files. libtool figures out any magic that needs to be done to > > actually link those objects

Hi, This is a repeat question. Earlier I had asked the same question and eventually found the ans. (http://lists.xiph.org/pipermail/vorbis-dev/2005-August/018179.html) However now when I try the same thing somehow I am not able to use debugging. To use debugging, i compiled the source using the folloing commands,. >make clean >./autogen.sh --enable-DEBUG >make The configure.in has

I'm haivng trouble compiling the libvoribis source. I'm getting errors I don't understand, could somebody please look at the below and tell me if they have any ideas?? <p>make[2]: Entering directory `/root/libvorbis-1.0/lib' source='mdct.c' object='mdct.lo' libtool=yes \ depfile='.deps/mdct.Plo' tmpdepfile='.deps/mdct.TPlo' \ depmode=gcc3

Hello, I have compiled R-1.7.1 on Beos R5 (x86) and got it running. The modules x11 and internet dont work (so the only working display is postscript()). The libraries all compile, but there is a problem with methods. I get the following error when i make methods: dumping R code in package 'methods' Error in .Call("R_initialize_methods_metadata", table, PACKAGE =