Linux App changed to new branch for PullRequest

README.md

@@ -130,6 +130,8 @@ the _Fancier_ API. The `copy_libs.sh` script copies the already-compiled `.jar`
 their build directories into the paths where these test applications expect them to be placed in
 order to be able to link and run their code.
 
+For using the Linux test application it is necessary to have “at” installed and enabled. Moreover, the program must be executed as sudo.
+
 ## License
 
 See [COPYING](COPYING.txt) and [COPYING.LESSER](COPYING.LESSER.txt).

gen/generate.sh

@@ -85,7 +85,11 @@ if word_in_list "$ACTION" all java math; then
 fi
 
 if word_in_list "$ACTION" all opencl math; then
-  python "$FILL" "$OCL_TEMPLATES/math_lib.mako" "$OCL_DIR/math_lib.cl"
+  python "$FILL" "$OCL_TEMPLATES/fc_math.mako" "$OCL_DIR/fc_math.cl"
+fi
+
+if word_in_list "$ACTION" all opencl image; then
+  python "$FILL" "$OCL_TEMPLATES/fc_image.mako" "$OCL_DIR/fc_image.cl"
 fi
 
 for cls in Byte Double Float Int Long Short; do

gen/template/include/array.mako

@@ -87,8 +87,8 @@ FANCIER_API fcError fc${type|c}Array_setArray(fc${type|c}Array* self, fcInt len,
 FANCIER_API fcError fc${type|c}Array_setCopy(fc${type|c}Array* self, const fc${type|c}Array* array);
 FANCIER_API fcError fc${type|c}Array_setBuffer(fc${type|c}Array* self, fcLong len, const void* v);
 
-FANCIER_API fcError fc${type|c}Array_syncToNative(fc${type|c}Array* self);
-FANCIER_API fcError fc${type|c}Array_syncToOCL(fc${type|c}Array* self);
+FANCIER_API fcError fc${type|c}Array_syncToHost(fc${type|c}Array* self);
+FANCIER_API fcError fc${type|c}Array_syncToDevice(fc${type|c}Array* self);
 FANCIER_API fcBool fc${type|c}Array_valid(const fc${type|c}Array* self);
 
 % endfor

gen/template/include/vector.mako

@@ -93,7 +93,6 @@ FANCIER_API void fcVector_releaseJNI(JNIEnv* env);
 
 % for type in types:
 % for vlen in vlens:
-FANCIER_API fc${type|c}${vlen}* fc${type|c}${vlen}_getJava(JNIEnv* env, jobject obj);
 FANCIER_API jobject fc${type|c}${vlen}_wrap(JNIEnv* env, fc${type|c}${vlen} vec);
 FANCIER_API fc${type|c}${vlen} fc${type|c}${vlen}_unwrap(JNIEnv* env, jobject vec, fcError* err);
 

gen/template/include/vector_array.mako

@@ -86,8 +86,8 @@ FANCIER_API fcError fc${type|c}${vlen}Array_setArray(fc${type|c}${vlen}Array* se
 FANCIER_API fcError fc${type|c}${vlen}Array_setCopy(fc${type|c}${vlen}Array* self, const fc${type|c}${vlen}Array* array);
 FANCIER_API fcError fc${type|c}${vlen}Array_setBuffer(fc${type|c}${vlen}Array* self, fcLong len, const void* v);
 
-FANCIER_API fcError fc${type|c}${vlen}Array_syncToNative(fc${type|c}${vlen}Array* self);
-FANCIER_API fcError fc${type|c}${vlen}Array_syncToOCL(fc${type|c}${vlen}Array* self);
+FANCIER_API fcError fc${type|c}${vlen}Array_syncToHost(fc${type|c}${vlen}Array* self);
+FANCIER_API fcError fc${type|c}${vlen}Array_syncToDevice(fc${type|c}${vlen}Array* self);
 FANCIER_API fcBool fc${type|c}${vlen}Array_valid(const fc${type|c}${vlen}Array* self);
 
 % endfor

gen/template/java/Array.mako

@@ -20,6 +20,8 @@ package es.ull.pcg.hpc.fancier.array;
 
 import java.nio.ByteBuffer;
 import java.nio.ByteOrder;
+import es.ull.pcg.hpc.fancier.Translatable;
+
 
 public class ${type|c}Array implements AutoCloseable {
   private long nativeInstancePtr = 0L;
@@ -87,8 +89,11 @@ public class ${type|c}Array implements AutoCloseable {
   private native void releaseNative();
   private native void releaseNativeRef();
 
+  @Translatable
   public native ${type|l} get(int i);
+  @Translatable
   public native void set(int i, ${type|l} x);
+  @Translatable
   public native long length();
 
   public native ${type|l}[] getArray();
@@ -97,6 +102,6 @@ public class ${type|c}Array implements AutoCloseable {
   private native ByteBuffer getBufferImpl();
   public native void setBuffer(ByteBuffer buffer);
 
-  public native void syncToNative();
-  public native void syncToOCL();
+  public native void syncToHost();
+  public native void syncToDevice();
 }

gen/template/java/Math.mako

@@ -18,106 +18,177 @@
 <%include file="/auto-gen.txt"/>
 package es.ull.pcg.hpc.fancier;
 
+
 public class Math {
+  @Translatable
   public static final int CHAR_BIT = Byte.SIZE;
+  @Translatable
   public static final byte SCHAR_MAX = Byte.MAX_VALUE;
+  @Translatable
   public static final byte SCHAR_MIN = Byte.MIN_VALUE;
+  @Translatable
   public static final byte CHAR_MAX = SCHAR_MAX;
+  @Translatable
   public static final byte CHAR_MIN = SCHAR_MIN;
+  @Translatable
   public static final short SHRT_MAX = Short.MAX_VALUE;
+  @Translatable
   public static final short SHRT_MIN = Short.MIN_VALUE;
+  @Translatable
   public static final int INT_MAX = Integer.MAX_VALUE;
+  @Translatable
   public static final int INT_MIN = Integer.MIN_VALUE;
+  @Translatable
   public static final long LONG_MAX = Long.MAX_VALUE;
+  @Translatable
   public static final long LONG_MIN = Long.MIN_VALUE;
 
+  @Translatable
   public static final int FLT_DIG =        6;
+  @Translatable
   public static final int FLT_MANT_DIG =   24;
+  @Translatable
   public static final int FLT_MAX_10_EXP = +38;
+  @Translatable
   public static final int FLT_MAX_EXP =    +128;
+  @Translatable
   public static final int FLT_MIN_10_EXP = -37;
+  @Translatable
   public static final int FLT_MIN_EXP =    -125;
+  @Translatable
   public static final int FLT_RADIX =      2;
+  @Translatable
   public static final float FLT_MAX =      Float.MAX_VALUE;
+  @Translatable
   public static final float FLT_MIN =      Float.MIN_NORMAL;
+  @Translatable
   public static final float FLT_EPSILON =  0x1.0p-23f;
 
+  @Translatable
   public static final int DBL_DIG =        15;
+  @Translatable
   public static final int DBL_MANT_DIG =   53;
+  @Translatable
   public static final int DBL_MAX_10_EXP = +308;
+  @Translatable
   public static final int DBL_MAX_EXP =    +1024;
+  @Translatable
   public static final int DBL_MIN_10_EXP = -307;
+  @Translatable
   public static final int DBL_MIN_EXP =    -1021;
+  @Translatable
   public static final int DBL_RADIX =      2;
+  @Translatable
   public static final double DBL_MAX =     Double.MAX_VALUE;
+  @Translatable
   public static final double DBL_MIN =     Double.MIN_NORMAL;
+  @Translatable
   public static final double DBL_EPSILON = 0x1.0p-52;
 
+  @Translatable
   public static final float M_E_F =        2.71828174591064f;
+  @Translatable
   public static final float M_LOG2E_F =    1.44269502162933f;
+  @Translatable
   public static final float M_LOG10E_F =   0.43429449200630f;
+  @Translatable
   public static final float M_LN2_F =      0.69314718246460f;
+  @Translatable
   public static final float M_LN10_F =     2.30258512496948f;
+  @Translatable
   public static final float M_PI_F =       3.14159274101257f;
+  @Translatable
   public static final float M_PI_2_F =     1.57079637050629f;
+  @Translatable
   public static final float M_PI_4_F =     0.78539818525314f;
+  @Translatable
   public static final float M_1_PI_F =     0.31830987334251f;
+  @Translatable
   public static final float M_2_PI_F =     0.63661974668503f;
+  @Translatable
   public static final float M_2_SQRTPI_F = 1.12837922573090f;
+  @Translatable
   public static final float M_SQRT2_F =    1.41421353816986f;
+  @Translatable
   public static final float M_SQRT1_2_F =  0.70710676908493f;
 
+  @Translatable
   public static final double M_E =        2.718281828459045090796;
+  @Translatable
   public static final double M_LOG2E =    1.442695040888963387005;
+  @Translatable
   public static final double M_LOG10E =   0.434294481903251816668;
+  @Translatable
   public static final double M_LN2 =      0.693147180559945286227;
+  @Translatable
   public static final double M_LN10 =     2.302585092994045901094;
+  @Translatable
   public static final double M_PI =       3.141592653589793115998;
+  @Translatable
   public static final double M_PI_2 =     1.570796326794896557999;
+  @Translatable
   public static final double M_PI_4 =     0.785398163397448278999;
+  @Translatable
   public static final double M_1_PI =     0.318309886183790691216;
+  @Translatable
   public static final double M_2_PI =     0.636619772367581382433;
+  @Translatable
   public static final double M_2_SQRTPI = 1.128379167095512558561;
+  @Translatable
   public static final double M_SQRT2 =    1.414213562373095145475;
+  @Translatable
   public static final double M_SQRT1_2 =  0.707106781186547572737;
 
+  @Translatable
   public static final float MAXFLOAT =  Float.MAX_VALUE;
+  @Translatable
   public static final float HUGE_VALF = Float.POSITIVE_INFINITY;
+  @Translatable
   public static final double HUGE_VAL = Double.POSITIVE_INFINITY;
+  @Translatable
   public static final float INFINITY =  Float.POSITIVE_INFINITY;
+  @Translatable
   public static final float NAN =       Float.NaN;
 
   // All types
 
 % for type in types:
+  @Translatable
   public static ${type|l} abs(${type|l} a) {
     return a < ${defaults[type.lower()]}? (${type|l}) -a : a;
   }
 
+  @Translatable
   public static ${type|l} max(${type|l} a, ${type|l} b) {
     return a > b? a : b;
   }
 
+  @Translatable
   public static ${type|l} min(${type|l} a, ${type|l} b) {
     return a < b? a : b;
   }
 
+  @Translatable
   public static ${type|l} clamp(${type|l} a, ${type|l} b, ${type|l} c) {
     return a < b? b : (a > c? c : a);
   }
 
+  @Translatable
   public static ${type|l} mix(${type|l} a, ${type|l} b, ${type|l} c) {
     return (${type|l}) (a + (b - a) * c);
   }
 
+  @Translatable
   public static ${type|l} maxMag(${type|l} a, ${type|l} b) {
     return abs(a) > abs(b)? a : b;
   }
 
+  @Translatable
   public static ${type|l} minMag(${type|l} a, ${type|l} b) {
     return abs(a) < abs(b)? a : b;
   }
 
+  @Translatable
   public static ${type|l} select(${type|l} a, ${type|l} b, int c) {
     return c == 0? a : b;
   }
@@ -127,31 +198,38 @@ public class Math {
 
 % for fname, num_params in math_int_functions.items():
 % for type in inttypes:
+  @Translatable
   public static native ${type|l} ${fname}(${', '.join([f'{type.lower()} {chr(i + ord("a"))}' for i in range(num_params)])});
 % endfor
 % endfor
+  @Translatable
   public static native int mad24 (int a, int b, int c);
+  @Translatable
   public static native int mul24 (int a, int b);
 
   // Float types
 
 % for fname, num_params in math_float_functions.items():
 % for type in floattypes:
+  @Translatable
   public static native ${type|l} ${fname}(${', '.join([f'{type.lower()} {chr(i + ord("a"))}' for i in range(num_params)])});
 % endfor
 % endfor
 % for fname in ('scalb', 'ldexp', 'pown', 'rootn'):
 % for type in floattypes:
+  @Translatable
   public static native ${type|l} ${fname}(${type|l} a, int n);
 % endfor
 % endfor
 % for fname in ('isFinite', 'isInf', 'isNaN', 'isNormal'):
 % for type in floattypes:
+  @Translatable
   public static native int ${fname}(${type|l} x);
 % endfor
 % endfor
 % for fname in ('isOrdered', 'isUnordered'):
 % for type in floattypes:
+  @Translatable
   public static native int ${fname}(${type|l} x, ${type|l} y);
 % endfor
 % endfor