Displaying 7 results from an estimated 7 matches for "__vector".
Did you mean:
_vector
2019 Oct 24
2
Failed PPC64 compile when using Power7 loads and stores?
...-MT libppc_power7_la-ppc_power7.lo -MD -MP -MF
.deps/libppc_power7_la-ppc_power7.Tpo -c -o
libppc_power7_la-ppc_power7.lo `test -f 'ppc_power7.cpp' || echo
'./'`ppc_power7.cpp
ppc_power7.cpp:71:21: error: use of undeclared identifier 'vec_xl'
vec_xst(vec_xl(0, (__vector unsigned int*)(b1+3)), 0, (__vec...
The source file includes <altivec.h>. The source file is compiled with
-mcpu=power7 -mvsx -maltivec.
==========
Here is the source code of interest:
// POWER7 added unaligned loads and store operations
byte b1[19] = {255, 255, 255, 1,1,1,1,1,1,1,1,1,1,1...
2005 Dec 02
0
run time assembler patch for altivec, sse + bug fixes
...OF SUCH DAMAGE.
*/
#ifdef _USE_ALTIVEC
#include "ltp_altivec.h"
spx_word32_t inner_prod_altivec(const float *a, const float *b, int len)
{
int i;
float sum;
int a_aligned = (((unsigned long)a) & 15) ? 0 : 1;
int b_aligned = (((unsigned long)b) & 15) ? 0 : 1;
__vector float MSQa, LSQa, MSQb, LSQb;
__vector unsigned char maska, maskb;
__vector float vec_a, vec_b;
__vector float vec_result;
vec_result = (__vector float)vec_splat_u8(0);
if ((!a_aligned) && (!b_aligned))
{
// This (unfortunately) is the common case....
2011 Feb 15
3
[LLVMdev] Possible LLVM or DragonEgg bug
.../lib/libcxcore.so.2.0.0 ../../3rdparty/lib/libopencv_lapack.a ../../3rdparty/lib/libzlib.a ../../3rdparty/lib/libflann.a -ldl -lm -lpthread -lrt -lgomp -Wl,-rpath,/home/Gytis/Documents/Project/OpenCV/new2/OpenCV-2.0.0/bin/lib:
../../lib/libcv.so.2.0.0: undefined reference to `_mm_cvtsd_si32(double __vector)'
collect2: ld returned 1 exit status
As I understand the linker can not find intrinsic SSE2 function "_mm_cvtsd_si32" while trying to link "opencv_createsamples" executable to "libcv.so.2.0.0".
If compiled with the same setup but with OpenMP disabled(i.e. no &...
2004 Aug 06
2
[PATCH] Make SSE Run Time option. Add Win32 SSE code
Jean-Marc,
>I'm still not sure I get it. On an Athlon XP, I can do something like
>"mulps xmm0, xmm1", which means that the xmm registers are indeed
>supported. Besides, without the xmm registers, you can't use much of
>SSE.
In the Atholon XP 2400+ that we have in our QA lab (Win2000 ) if you run
that code it generates an Illegal Instruction Error. In addition,
2004 Aug 06
6
[PATCH] Make SSE Run Time option.
...od(float *a, float *b, int len)
{
if (!(global_use_mmx_sse & CPU_MODE_ALTIVEC ))
{
#ifdef _USE_ALTIVEC
int i;
float sum;
int a_aligned = (((unsigned long)a) & 15) ? 0 : 1;
int b_aligned = (((unsigned long)b) & 15) ? 0 : 1;
__vector float MSQa, LSQa, MSQb, LSQb;
__vector unsigned char maska, maskb;
__vector float vec_a, vec_b;
__vector float vec_result;
vec_result = (__vector float)vec_splat_u8(0);
if ((!a_aligned) && (!b_aligned)) {
// This (unfortunately) is...
2011 Feb 16
0
[LLVMdev] Possible LLVM or DragonEgg bug
.../libcxcore.so.2.0.0 ../../3rdparty/lib/libopencv_lapack.a ../../3rdparty/lib/libzlib.a ../../3rdparty/lib/libflann.a -ldl -lm -lpthread -lrt -lgomp -Wl,-rpath,/home/Gytis/Documents/Project/OpenCV/new2/OpenCV-2.0.0/bin/lib:
> ../../lib/libcv.so.2.0.0: undefined reference to `_mm_cvtsd_si32(double __vector)'
> collect2: ld returned 1 exit status
>
> As I understand the linker can not find intrinsic SSE2 function "_mm_cvtsd_si32" while trying to link "opencv_createsamples" executable to "libcv.so.2.0.0".
>
> If compiled with the same setup but with Op...
2011 Feb 17
1
[LLVMdev] Possible LLVM or DragonEgg bug
.../libcxcore.so.2.0.0 ../../3rdparty/lib/libopencv_lapack.a ../../3rdparty/lib/libzlib.a ../../3rdparty/lib/libflann.a -ldl -lm -lpthread -lrt -lgomp -Wl,-rpath,/home/Gytis/Documents/Project/OpenCV/new2/OpenCV-2.0.0/bin/lib:
> ../../lib/libcv.so.2.0.0: undefined reference to `_mm_cvtsd_si32(double __vector)'
> collect2: ld returned 1 exit status
>
> As I understand the linker can not find intrinsic SSE2 function "_mm_cvtsd_si32" while trying to link "opencv_createsamples" executable to "libcv.so.2.0.0".
>
> If compiled with the same setup but with Op...