Merge branch 'next' into guile.

Passes distcheck.

* next: (32 commits)
  volk: Fix for popcnt's 64/32-bit issues.
  Using a copy of config.guess and config.sub instead of sym links.
  Including time header to qa files.
  Changed python env variable to more globally usable version.
  gr_uhd: Quick fix for make distcheck failures if UHD is not installed.
  PFB resampler: fix it this way to avoid the signed/unsigned warning.
  PFB resampler: fixes bug where filter could be looking past the number of inputs.
  resampler PFB: Sets relative rate when rate is changed.
  Create method to set rate on pfb_arb_resamp after it has been created. Allow it to be called from GRC.
  volk: fix for running 32-bit OS on 64-bit processor. System is correctly identified as 32-bit and compiles with the correct flags.
  volk: changing the path variables again. This works on my various systems tested. Using abs_ path names failed on Ubuntu 8.04 32-bit.
  volk: May be a hack, but it was required for my 32-bit Fedora 13 to work.
  volk: Removing unnecessary shell script; last commit takes care if its functions.
  volk: Fixing build system to handle making volk_mktables, volk_tables.h, and volk_config.h instead of a standalone shell script.
  volk: readding 16sc_magnitude_32f_sse with fix for SSE hadd_ps error.
  volk: Adding a few more generic-only test cases.
  volk: adding generic QA test for 16sc_magnitude_32f.
  volk: modified the configure scripts to output which architectures it will be building based on the configure tests.
  uhd: update notes in grc blocks for addressing scheme
  volk: Removing erroneous SSE function that actually usese an SSE3 intrin (mm_hadd_ps).
  ...

1 files changed, 89 insertions, 0 deletions

diff --git a/volk/lib/qa_16s_add_quad_aligned16.cc b/volk/lib/qa_16s_add_quad_aligned16.cc
new file mode 100644
index 000000000..154aa0f17
--- /dev/null
+++ b/volk/lib/qa_16s_add_quad_aligned16.cc
@@ -0,0 +1,89 @@
+#include <volk/volk.h>
+#include <qa_16s_add_quad_aligned16.h>
+#include <volk/volk_16s_add_quad_aligned16.h>
+#include <cstdlib>
+#include <ctime>
+//test for sse2
+
+#ifndef LV_HAVE_SSE2
+
+void qa_16s_add_quad_aligned16::t1() {
+  printf("sse2 not available... no test performed\n");
+}
+
+#else
+
+
+
+void qa_16s_add_quad_aligned16::t1() {
+  
+  volk_environment_init();
+  clock_t start, end;
+  double total;
+  const int vlen = 3200;
+  const int ITERS = 100000;
+  short input0[vlen] __attribute__ ((aligned (16)));
+  short input1[vlen] __attribute__ ((aligned (16)));
+  short input2[vlen] __attribute__ ((aligned (16)));
+  short input3[vlen] __attribute__ ((aligned (16)));
+  short input4[vlen] __attribute__ ((aligned (16)));
+  
+  short output0[vlen] __attribute__ ((aligned (16)));
+  short output1[vlen] __attribute__ ((aligned (16)));
+  short output2[vlen] __attribute__ ((aligned (16)));
+  short output3[vlen] __attribute__ ((aligned (16)));
+  short output01[vlen] __attribute__ ((aligned (16)));
+  short output11[vlen] __attribute__ ((aligned (16)));
+  short output21[vlen] __attribute__ ((aligned (16)));
+  short output31[vlen] __attribute__ ((aligned (16)));
+
+  for(int i = 0; i < vlen; ++i) {
+    short plus0 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short minus0 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short plus1 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short minus1 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short plus2 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short minus2 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short plus3 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short minus3 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short plus4 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    short minus4 = ((short) (rand() - (RAND_MAX/2))) >> 2;
+    
+    input0[i] = plus0 - minus0;
+    input1[i] = plus1 - minus1;
+    input2[i] = plus2 - minus2;
+    input3[i] = plus3 - minus3;
+    input4[i] = plus4 - minus4;
+    
+  }
+  printf("16s_add_quad_aligned\n");
+
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
+    volk_16s_add_quad_aligned16_manual(output0, output1, output2, output3, input0, input1, input2, input3, input4, vlen << 1 , "generic");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("generic_time: %f\n", total);
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
+    volk_16s_add_quad_aligned16_manual(output01, output11, output21, output31, input0, input1, input2, input3, input4, vlen << 1 , "sse2");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("sse2_time: %f\n", total);
+  for(int i = 0; i < 1; ++i) {
+    //printf("inputs: %d, %d\n", input0[i*2], input0[i*2 + 1]);
+    //printf("generic... %d, ssse3... %d\n", output0[i], output1[i]);
+  }
+  
+  for(int i = 0; i < vlen; ++i) {
+    //printf("%d...%d\n", output0[i], output01[i]);
+    CPPUNIT_ASSERT_EQUAL(output0[i], output01[i]);
+    CPPUNIT_ASSERT_EQUAL(output1[i], output11[i]);
+    CPPUNIT_ASSERT_EQUAL(output2[i], output21[i]);
+    CPPUNIT_ASSERT_EQUAL(output3[i], output31[i]);
+  }
+}
+
+#endif