diff options
Diffstat (limited to 'src/com/mbien')
| -rw-r--r-- | src/com/mbien/opencl/CLProgram.java | 121 |
1 files changed, 118 insertions, 3 deletions
diff --git a/src/com/mbien/opencl/CLProgram.java b/src/com/mbien/opencl/CLProgram.java index d1645213..4b7fac7c 100644 --- a/src/com/mbien/opencl/CLProgram.java +++ b/src/com/mbien/opencl/CLProgram.java @@ -143,7 +143,7 @@ public class CLProgram implements CLResource { * @return this */ public CLProgram build() { - build(null, null); + build(null, (CLDevice[])null); return this; } @@ -152,7 +152,16 @@ public class CLProgram implements CLResource { * @return this */ public CLProgram build(String options) { - build(null, options); + build(options, (CLDevice[])null); + return this; + } + + /** + * Builds this program for all devices associated with the context using the specified build options. + * @return this + */ + public CLProgram build(Option... options) { + build(Option.optionsOf(options), (CLDevice[])null); return this; } @@ -162,7 +171,7 @@ public class CLProgram implements CLResource { * @return this * @param devices A list of devices this program should be build on or null for all devices of its context. */ - public CLProgram build(CLDevice[] devices, String options) { + public CLProgram build(String options, CLDevice... devices) { if(kernels != null) { //No changes to the program executable are allowed while there are @@ -421,4 +430,110 @@ public class CLProgram implements CLResource { } } + /** + * Common compiler optons. + */ + public enum Option { + + /** + * Treat double precision floating-point constant as single precision constant. + */ + SINGLE_PRECISION_CONSTANTS("-cl-single-precision-constant"), + + /** + * This option controls how single precision and double precision denormalized numbers are handled. + * If specified as a build option, the single precision denormalized numbers may be flushed to zero + * and if the optional extension for double precision is supported, double precision denormalized numbers + * may also be flushed to zero. This is intended to be a performance hint and the OpenCL compiler can choose + * not to flush denorms to zero if the device supports single precision (or double precision) denormalized numbers.<br> + * This option is ignored for single precision numbers if the device does not support single precision denormalized + * numbers i.e. CL_FP_DENORM bit is not set in CL_DEVICE_SINGLE_FP_CONFIG<br> + * This option is ignored for double precision numbers if the device does not support double precision or if it does support + * double precison but CL_FP_DENORM bit is not set in CL_DEVICE_DOUBLE_FP_CONFIG.<br> + * This flag only applies for scalar and vector single precision floating-point variables and computations on + * these floating-point variables inside a program. It does not apply to reading from or writing to image objects. + */ + DENORMS_ARE_ZERO("-cl-denorms-are-zero"), + + /** + * This option disables all optimizations. The default is optimizations are enabled. + */ + DISABLE_OPT("-cl-opt-disable"), + + /** + * This option allows the compiler to assume the strictest aliasing rules. + */ + STRICT_ALIASING("-cl-strict-aliasing"), + + /** + * Allow a * b + c to be replaced by a mad. The mad computes a * b + c with reduced accuracy. + * For example, some OpenCL devices implement mad as truncate the result of a * b before adding it to c. + */ + ENABLE_MAD("-cl-mad-enable"), + + /** + * Allow optimizations for floating-point arithmetic that ignore the signedness of zero. + * IEEE 754 arithmetic specifies the behavior of distinct +0.0 and -0.0 values, which then prohibits + * simplification of expressions such as x+0.0 or 0.0*x (even with -cl-finite-math-only ({@link #FINITE_MATH_ONLY})). + * This option implies that the sign of a zero result isn't significant. + */ + NO_SIGNED_ZEROS("-cl-no-signed-zeros"), + + /** + * Allow optimizations for floating-point arithmetic that<br> + * (a) assume that arguments and results are valid,<br> + * (b) may violate IEEE 754 standard and<br> + * (c) may violate the OpenCL numerical compliance requirements as defined in section + * 7.4 for single-precision floating-point, section 9.3.9 for double-precision floating-point, + * and edge case behavior in section 7.5.<br> + * This option includes the -cl-no-signed-zeros ({@link #NO_SIGNED_ZEROS}) + * and -cl-mad-enable ({@link #ENABLE_MAD}) options. + */ + UNSAFE_MATH("-cl-unsafe-math-optimizations"), + + /** + * Allow optimizations for floating-point arithmetic that assume that arguments and results are not NaNs or ±∞. + * This option may violate the OpenCL numerical compliance requirements defined in in section 7.4 for + * single-precision floating-point, section 9.3.9 for double-precision floating-point, and edge case behavior in section 7.5. + */ + FINITE_MATH_ONLY("-cl-finite-math-only"), + + /** + * Sets the optimization options -cl-finite-math-only ({@link #FINITE_MATH_ONLY}) and -cl-unsafe-math-optimizations ({@link #UNSAFE_MATH}). + * This allows optimizations for floating-point arithmetic that may violate the IEEE 754 + * standard and the OpenCL numerical compliance requirements defined in the specification + * in section 7.4 for single-precision floating-point, section 9.3.9 for double-precision + * floating-point, and edge case behavior in section 7.5. This option causes the preprocessor + * macro __FAST_RELAXED_MATH__ to be defined in the OpenCL program. + */ + FAST_RELAXED_MATH("-cl-fast-relaxed-math"), + + /** + * Inhibit all warning messages. + */ + DISABLE_WARNINGS("-w"), + + /** + * Make all warnings into errors. + */ + WARNINGS_ARE_ERRORS("-Werror"); + + private final String option; + + private Option(String option) { + this.option = option; + } + + public final static String optionsOf(Option... options) { + StringBuilder sb = new StringBuilder(options.length * 24); + for (int i = 0; i < options.length; i++) { + sb.append(options[i].option); + if(i!= options.length-1) + sb.append(" "); + } + return sb.toString(); + } + + } + } |