gain_analysis.c, fft.c: error: can't convert value to a vector, can't convert between vector values of different size
Opened this issue · 2 comments
barracuda156 commented
This is on 10.6.8 in Rosetta with Apple gcc 4.2:
Making all in libmp3lame
make[2]: Entering directory `/opt/local/var/macports/build/_opt_PPCSnowLeopardPorts_audio_lame/lame/work/lame-3.100/libmp3lame'
Making all in i386
make[3]: Entering directory `/opt/local/var/macports/build/_opt_PPCSnowLeopardPorts_audio_lame/lame/work/lame-3.100/libmp3lame/i386'
make[3]: Nothing to be done for `all'.
make[3]: Leaving directory `/opt/local/var/macports/build/_opt_PPCSnowLeopardPorts_audio_lame/lame/work/lame-3.100/libmp3lame/i386'
Making all in vector
make[3]: Entering directory `/opt/local/var/macports/build/_opt_PPCSnowLeopardPorts_audio_lame/lame/work/lame-3.100/libmp3lame/vector'
make[3]: Nothing to be done for `all'.
make[3]: Leaving directory `/opt/local/var/macports/build/_opt_PPCSnowLeopardPorts_audio_lame/lame/work/lame-3.100/libmp3lame/vector'
make[3]: Entering directory `/opt/local/var/macports/build/_opt_PPCSnowLeopardPorts_audio_lame/lame/work/lame-3.100/libmp3lame'
/bin/sh ../libtool --tag=CC --mode=compile /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c -o VbrTag.lo VbrTag.c
/bin/sh ../libtool --tag=CC --mode=compile /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c -o bitstream.lo bitstream.c
/bin/sh ../libtool --tag=CC --mode=compile /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c -o encoder.lo encoder.c
/bin/sh ../libtool --tag=CC --mode=compile /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c -o fft.lo fft.c
/bin/sh ../libtool --tag=CC --mode=compile /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c -o gain_analysis.lo gain_analysis.c
/bin/sh ../libtool --tag=CC --mode=compile /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c -o id3tag.lo id3tag.c
libtool: compile: /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c gain_analysis.c -fno-common -DPIC -o .libs/gain_analysis.o
libtool: compile: /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c bitstream.c -fno-common -DPIC -o .libs/bitstream.o
libtool: compile: /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c VbrTag.c -fno-common -DPIC -o .libs/VbrTag.o
libtool: compile: /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c id3tag.c -fno-common -DPIC -o .libs/id3tag.o
libtool: compile: /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c fft.c -fno-common -DPIC -o .libs/fft.o
libtool: compile: /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c encoder.c -fno-common -DPIC -o .libs/encoder.o
gain_analysis.c: In function ‘filterIntegrated’:
gain_analysis.c:222: error: can't convert value to a vector
gain_analysis.c:223: error: can't convert between vector values of different size
gain_analysis.c:224: error: can't convert between vector values of different size
make[3]: *** [gain_analysis.lo] Error 1
make[3]: *** Waiting for unfinished jobs....
fft.c: In function ‘fht’:
fft.c:80: error: can't convert between vector values of different size
fft.c:81: error: can't convert between vector values of different size
make[3]: *** [fft.lo] Error 1
I used -arch ppc, since multi-arch did not work, at least in Rosetta.
P. S. Not sure why at all i386 folder is included in sources built, seems to be hardcoded. I have set build to powerpc-apple-darwin10, and according to configure log it has been correctly recognized.
barracuda156 commented
gcc 12.2.0 also fails on the same spot:
libtool: compile: /opt/local/bin/gcc-mp-12 -DHAVE_CONFIG_H -I. -I.. -I../include -I. -I../mpglib -I.. -I/opt/local/include -arch ppc -O3 -fomit-frame-pointer -ffast-math -funroll-loops -isysroot /Developer/SDKs/MacOSX10.6.sdk -pipe -Os -arch ppc -fno-common -c fft.c -fno-common -DPIC -o .libs/fft.o
gain_analysis.c: In function 'filterIntegrated':
gain_analysis.c:222:9: error: cannot convert value to a vector
222 | vbase = (vector float)VINIT4ALL(1e-10f);
| ^~~~~
gain_analysis.c:223:9: error: cannot convert a value of type 'int' to vector type '__vector(16) unsigned char' which has different size
223 | vperm1 = (vector unsigned char)VINIT16(24,25,26,27,16,17,18,19,8,9,10,11,0,1,2,3);
| ^~~~~~
gain_analysis.c:224:9: error: cannot convert a value of type 'int' to vector type '__vector(16) unsigned char' which has different size
224 | vperm2 = (vector unsigned char)VINIT16(28,29,30,31,20,21,22,23,12,13,14,15,4,5,6,7);
| ^~~~~~
fft.c: In function 'fht':
fft.c:80:5: error: cannot convert a value of type 'int' to vector type '__vector(16) unsigned char' which has different size
80 | vperm1 = (vector unsigned char)VINIT16(16,17,18,19,12,13,14,15,8,9,10,11,4,5,6,7);
| ^~~~~~
fft.c:81:5: error: cannot convert a value of type 'int' to vector type '__vector(16) unsigned char' which has different size
81 | vperm2 = (vector unsigned char)VINIT16(16,17,18,19,4,5,6,7,8,9,10,11,12,13,14,15);
| ^~~~~~
make[3]: *** [fft.lo] Error 1
barracuda156 commented
Oh, passing -faltivec fixed it.